diff options
561 files changed, 14305 insertions, 5776 deletions
diff --git a/Documentation/gpu/amdgpu/amd-hardware-list-info.rst b/Documentation/gpu/amdgpu/amd-hardware-list-info.rst index 1786544fe7c1..e72f4ff770c4 100644 --- a/Documentation/gpu/amdgpu/amd-hardware-list-info.rst +++ b/Documentation/gpu/amdgpu/amd-hardware-list-info.rst @@ -10,7 +10,7 @@ Accelerated Processing Units (APU) Info .. csv-table:: :header-rows: 1 - :widths: 3, 2, 2, 1, 1, 1, 1 + :widths: 3, 2, 2, 1, 1, 1, 1, 1 :file: ./apu-asic-info-table.csv Discrete GPU Info @@ -18,6 +18,6 @@ Discrete GPU Info .. csv-table:: :header-rows: 1 - :widths: 3, 2, 2, 1, 1, 1 + :widths: 3, 2, 2, 1, 1, 1, 1, 1 :file: ./dgpu-asic-info-table.csv diff --git a/Documentation/gpu/amdgpu/apu-asic-info-table.csv b/Documentation/gpu/amdgpu/apu-asic-info-table.csv index 1d50b539677f..b479c5629146 100644 --- a/Documentation/gpu/amdgpu/apu-asic-info-table.csv +++ b/Documentation/gpu/amdgpu/apu-asic-info-table.csv @@ -1,17 +1,17 @@ -Product Name, Code Reference, DCN/DCE version, GC version, VCE/UVD/VCN version, SDMA version, MP0 version -Radeon R* Graphics, CARRIZO/STONEY, DCE 11, 8, VCE 3 / UVD 6, 3, n/a -Ryzen 3000 series / AMD Ryzen Embedded V1*/R1* with Radeon Vega Gfx, RAVEN/PICASSO, DCN 1.0, 9.1.0, VCN 1.0, 4.1.0, 10.0.0 -Ryzen 4000 series, RENOIR, DCN 2.1, 9.3, VCN 2.2, 4.1.2, 11.0.3 -Ryzen 3000 series / AMD Ryzen Embedded V1*/R1* with Radeon Vega Gfx, RAVEN2, DCN 1.0, 9.2.2, VCN 1.0.1, 4.1.1, 10.0.1 -SteamDeck, VANGOGH, DCN 3.0.1, 10.3.1, VCN 3.1.0, 5.2.1, 11.5.0 -Ryzen 5000 series / Ryzen 7x30 series, GREEN SARDINE / Cezanne / Barcelo / Barcelo-R, DCN 2.1, 9.3, VCN 2.2, 4.1.1, 12.0.1 -Ryzen 6000 series / Ryzen 7x35 series / Ryzen 7x36 series, YELLOW CARP / Rembrandt / Rembrandt-R, 3.1.2, 10.3.3, VCN 3.1.1, 5.2.3, 13.0.3 -Ryzen 7000 series (AM5), Raphael, 3.1.5, 10.3.6, 3.1.2, 5.2.6, 13.0.5 -Ryzen 9000 series (AM5), Granite Ridge, 3.1.5, 10.3.6, 3.1.2, 5.2.6, 13.0.5 -Ryzen 7x45 series (FL1), Dragon Range, 3.1.5, 10.3.6, 3.1.2, 5.2.6, 13.0.5 -Ryzen 7x20 series, Mendocino, 3.1.6, 10.3.7, 3.1.1, 5.2.7, 13.0.8 -Ryzen 7x40 series, Phoenix, 3.1.4, 11.0.1 / 11.0.4, 4.0.2, 6.0.1, 13.0.4 / 13.0.11 -Ryzen 8x40 series, Hawk Point, 3.1.4, 11.0.1 / 11.0.4, 4.0.2, 6.0.1, 13.0.4 / 13.0.11 -Ryzen AI 300 series, Strix Point, 3.5.0, 11.5.0, 4.0.5, 6.1.0, 14.0.0 -Ryzen AI 350 series, Krackan Point, 3.5.0, 11.5.2, 4.0.5, 6.1.2, 14.0.4 -Ryzen AI Max 300 series, Strix Halo, 3.5.1, 11.5.1, 4.0.6, 6.1.1, 14.0.1 +Product Name, Code Reference, DCN/DCE version, GC version, VCE/UVD/VCN version, SDMA version, MP0 version, MP1 version +Radeon R* Graphics, CARRIZO/STONEY, DCE 11, 8, VCE 3 / UVD 6, 3, n/a, 8 +Ryzen 3000 series / AMD Ryzen Embedded V1*/R1* with Radeon Vega Gfx, RAVEN/PICASSO, DCN 1.0, 9.1.0, VCN 1.0, 4.1.0, 10.0.0, 10.0.0 +Ryzen 4000 series, RENOIR, DCN 2.1, 9.3, VCN 2.2, 4.1.2, 11.0.3, 12.0.1 +Ryzen 3000 series / AMD Ryzen Embedded V1*/R1* with Radeon Vega Gfx, RAVEN2, DCN 1.0, 9.2.2, VCN 1.0.1, 4.1.1, 10.0.1, 10.0.1 +SteamDeck, VANGOGH, DCN 3.0.1, 10.3.1, VCN 3.1.0, 5.2.1, 11.5.0, 11.5.0 +Ryzen 5000 series / Ryzen 7x30 series, GREEN SARDINE / Cezanne / Barcelo / Barcelo-R, DCN 2.1, 9.3, VCN 2.2, 4.1.1, 12.0.1, 12.0.1 +Ryzen 6000 series / Ryzen 7x35 series / Ryzen 7x36 series, YELLOW CARP / Rembrandt / Rembrandt-R, 3.1.2, 10.3.3, VCN 3.1.1, 5.2.3, 13.0.3, 13.0.3 +Ryzen 7000 series (AM5), Raphael, 3.1.5, 10.3.6, 3.1.2, 5.2.6, 13.0.5, 13.0.5 +Ryzen 9000 series (AM5), Granite Ridge, 3.1.5, 10.3.6, 3.1.2, 5.2.6, 13.0.5, 13.0.5 +Ryzen 7x45 series (FL1), Dragon Range, 3.1.5, 10.3.6, 3.1.2, 5.2.6, 13.0.5, 13.0.5 +Ryzen 7x20 series, Mendocino, 3.1.6, 10.3.7, 3.1.1, 5.2.7, 13.0.8, 13.0.8 +Ryzen 7x40 series, Phoenix, 3.1.4, 11.0.1 / 11.0.4, 4.0.2, 6.0.1, 13.0.4 / 13.0.11, 13.0.4 / 13.0.11 +Ryzen 8x40 series, Hawk Point, 3.1.4, 11.0.1 / 11.0.4, 4.0.2, 6.0.1, 13.0.4 / 13.0.11, 13.0.4 / 13.0.11 +Ryzen AI 300 series, Strix Point, 3.5.0, 11.5.0, 4.0.5, 6.1.0, 14.0.0, 14.0.0 +Ryzen AI 350 series, Krackan Point, 3.5.0, 11.5.2, 4.0.5, 6.1.2, 14.0.4, 14.0.4 +Ryzen AI Max 300 series, Strix Halo, 3.5.1, 11.5.1, 4.0.6, 6.1.1, 14.0.1, 14.0.1 diff --git a/Documentation/gpu/amdgpu/debugfs.rst b/Documentation/gpu/amdgpu/debugfs.rst index 5150d0a95658..151d8bfc79e2 100644 --- a/Documentation/gpu/amdgpu/debugfs.rst +++ b/Documentation/gpu/amdgpu/debugfs.rst @@ -94,7 +94,7 @@ amdgpu_error_<name> ------------------- Provides an interface to set an error code on the dma fences associated with -ring <name>. The error code specified is propogated to all fences associated +ring <name>. The error code specified is propagated to all fences associated with the ring. Use this to inject a fence error into a ring. amdgpu_pm_info @@ -165,7 +165,7 @@ GTT memory. amdgpu_regs_* ------------- -Provides direct access to various register aperatures on the GPU. Used +Provides direct access to various register apertures on the GPU. Used by tools like UMR to access GPU registers. amdgpu_regs2 diff --git a/Documentation/gpu/amdgpu/dgpu-asic-info-table.csv b/Documentation/gpu/amdgpu/dgpu-asic-info-table.csv index d2f10ee69dfc..bfd44c6e052a 100644 --- a/Documentation/gpu/amdgpu/dgpu-asic-info-table.csv +++ b/Documentation/gpu/amdgpu/dgpu-asic-info-table.csv @@ -1,28 +1,30 @@ -Product Name, Code Reference, DCN/DCE version, GC version, VCN version, SDMA version -AMD Radeon (TM) HD 8500M/ 8600M /M200 /M320 /M330 /M335 Series, HAINAN, --, 6, --, -- -AMD Radeon HD 7800 /7900 /FireGL Series, TAHITI, DCE 6, 6, VCE 1 / UVD 3, -- -AMD Radeon R7 (TM|HD) M265 /M370 /8500M /8600 /8700 /8700M, OLAND, DCE 6, 6, VCE 1 / UVD 3, -- -AMD Radeon (TM) (HD|R7) 7800 /7970 /8800 /8970 /370/ Series, PITCAIRN, DCE 6, 6, VCE 1 / UVD 3, -- -AMD Radeon (TM|R7|R9|HD) E8860 /M360 /7700 /7800 /8800 /9000(M) /W4100 Series, VERDE, DCE 6, 6, VCE 1 / UVD 3, -- -AMD Radeon HD M280X /M380 /7700 /8950 /W5100, BONAIRE, DCE 8, 7, VCE 2 / UVD 4.2, 1 -AMD Radeon (R9|TM) 200 /390 /W8100 /W9100 Series, HAWAII, DCE 8, 7, VCE 2 / UVD 4.2, 1 -AMD Radeon (TM) R(5|7) M315 /M340 /M360, TOPAZ, *, 8, --, 2 -AMD Radeon (TM) R9 200 /380 /W7100 /S7150 /M390 /M395 Series, TONGA, DCE 10, 8, VCE 3 / UVD 5, 3 -AMD Radeon (FirePro) (TM) R9 Fury Series, FIJI, DCE 10, 8, VCE 3 / UVD 6, 3 -Radeon RX 470 /480 /570 /580 /590 Series - AMD Radeon (TM) (Pro WX) 5100 /E9390 /E9560 /E9565 /V7350 /7100 /P30PH, POLARIS10, DCE 11.2, 8, VCE 3.4 / UVD 6.3, 3 -Radeon (TM) (RX|Pro WX) E9260 /460 /V5300X /550 /560(X) Series, POLARIS11, DCE 11.2, 8, VCE 3.4 / UVD 6.3, 3 -Radeon (RX/Pro) 500 /540(X) /550 /640 /WX2100 /WX3100 /WX200 Series, POLARIS12, DCE 11.2, 8, VCE 3.4 / UVD 6.3, 3 -Radeon (RX|TM) (PRO|WX) Vega /MI25 /V320 /V340L /8200 /9100 /SSG MxGPU, VEGA10, DCE 12, 9.0.1, VCE 4.0.0 / UVD 7.0.0, 4.0.0 -AMD Radeon (Pro) VII /MI50 /MI60, VEGA20, DCE 12, 9.4.0, VCE 4.1.0 / UVD 7.2.0, 4.2.0 -MI100, ARCTURUS, *, 9.4.1, VCN 2.5.0, 4.2.2 -MI200 Series, ALDEBARAN, *, 9.4.2, VCN 2.6.0, 4.4.0 -MI300 Series, AQUA_VANJARAM, *, 9.4.3, VCN 4.0.3, 4.4.2 -AMD Radeon (RX|Pro) 5600(M|XT) /5700 (M|XT|XTB) /W5700, NAVI10, DCN 2.0.0, 10.1.10, VCN 2.0.0, 5.0.0 -AMD Radeon (Pro) 5300 /5500XTB/5500(XT|M) /W5500M /W5500, NAVI14, DCN 2.0.0, 10.1.1, VCN 2.0.2, 5.0.2 -AMD Radeon RX 6800(XT) /6900(XT) /W6800, SIENNA_CICHLID, DCN 3.0.0, 10.3.0, VCN 3.0.0, 5.2.0 -AMD Radeon RX 6700 XT / 6800M / 6700M, NAVY_FLOUNDER, DCN 3.0.0, 10.3.2, VCN 3.0.0, 5.2.2 -AMD Radeon RX 6600(XT) /6600M /W6600 /W6600M, DIMGREY_CAVEFISH, DCN 3.0.2, 10.3.4, VCN 3.0.16, 5.2.4 -AMD Radeon RX 6500M /6300M /W6500M /W6300M, BEIGE_GOBY, DCN 3.0.3, 10.3.5, VCN 3.0.33, 5.2.5 -AMD Radeon RX 7900 XT /XTX, , DCN 3.2.0, 11.0.0, VCN 4.0.0, 6.0.0 -AMD Radeon RX 7800 XT, , DCN 3.2.0, 11.0.3, VCN 4.0.0, 6.0.3 -AMD Radeon RX 7600M (XT) /7700S /7600S, , DCN 3.2.1, 11.0.2, VCN 4.0.4, 6.0.2 +Product Name, Code Reference, DCN/DCE version, GC version, VCN version, SDMA version, MP0 version, MP1 version +AMD Radeon (TM) HD 8500M/ 8600M /M200 /M320 /M330 /M335 Series, HAINAN, --, 6, --, --, --, 6 +AMD Radeon HD 7800 /7900 /FireGL Series, TAHITI, DCE 6, 6, VCE 1 / UVD 3, --, --, 6 +AMD Radeon R7 (TM|HD) M265 /M370 /8500M /8600 /8700 /8700M, OLAND, DCE 6, 6, -- / UVD 3, --, --, 6 +AMD Radeon (TM) (HD|R7) 7800 /7970 /8800 /8970 /370/ Series, PITCAIRN, DCE 6, 6, VCE 1 / UVD 3, --, --, 6 +AMD Radeon (TM|R7|R9|HD) E8860 /M360 /7700 /7800 /8800 /9000(M) /W4100 Series, VERDE, DCE 6, 6, VCE 1 / UVD 3, --, --, 6 +AMD Radeon HD M280X /M380 /7700 /8950 /W5100, BONAIRE, DCE 8, 7, VCE 2 / UVD 4.2, 1, --, 7 +AMD Radeon (R9|TM) 200 /390 /W8100 /W9100 Series, HAWAII, DCE 8, 7, VCE 2 / UVD 4.2, 1, --, 7 +AMD Radeon (TM) R(5|7) M315 /M340 /M360, TOPAZ, *, 8, --, 2, n/a, 7 +AMD Radeon (TM) R9 200 /380 /W7100 /S7150 /M390 /M395 Series, TONGA, DCE 10, 8, VCE 3 / UVD 5, 3, n/a, 7 +AMD Radeon (FirePro) (TM) R9 Fury Series, FIJI, DCE 10, 8, VCE 3 / UVD 6, 3, n/a, 7 +Radeon RX 470 /480 /570 /580 /590 Series - AMD Radeon (TM) (Pro WX) 5100 /E9390 /E9560 /E9565 /V7350 /7100 /P30PH, POLARIS10, DCE 11.2, 8, VCE 3.4 / UVD 6.3, 3, n/a, 7 +Radeon (TM) (RX|Pro WX) E9260 /460 /V5300X /550 /560(X) Series, POLARIS11, DCE 11.2, 8, VCE 3.4 / UVD 6.3, 3, n/a, 7 +Radeon (RX/Pro) 500 /540(X) /550 /640 /WX2100 /WX3100 /WX200 Series, POLARIS12, DCE 11.2, 8, VCE 3.4 / UVD 6.3, 3, n/a, 7 +Radeon (RX|TM) (PRO|WX) Vega /MI25 /V320 /V340L /8200 /9100 /SSG MxGPU, VEGA10, DCE 12, 9.0.1, VCE 4.0.0 / UVD 7.0.0, 4.0.0, 9.0.0, 9.0.0 +AMD Radeon (Pro) VII /MI50 /MI60, VEGA20, DCE 12, 9.4.0, VCE 4.1.0 / UVD 7.2.0, 4.2.0, 11.0.2, 11.0.2 +MI100, ARCTURUS, *, 9.4.1, VCN 2.5.0, 4.2.2, 11.0.4, 11.0.2 +MI200 Series, ALDEBARAN, *, 9.4.2, VCN 2.6.0, 4.4.0, 13.0.2, 13.0.2 +MI300 Series, AQUA_VANJARAM, *, 9.4.3, VCN 4.0.3, 4.4.2, 13.0.6, 13.0.6 +AMD Radeon (RX|Pro) 5600(M|XT) /5700 (M|XT|XTB) /W5700, NAVI10, DCN 2.0.0, 10.1.10, VCN 2.0.0, 5.0.0, 11.0.0, 11.0.0 +AMD Radeon (Pro) 5300 /5500XTB/5500(XT|M) /W5500M /W5500, NAVI14, DCN 2.0.0, 10.1.1, VCN 2.0.2, 5.0.2, 11.0.5, 11.0.5 +AMD Radeon RX 6800(XT) /6900(XT) /W6800, SIENNA_CICHLID, DCN 3.0.0, 10.3.0, VCN 3.0.0, 5.2.0, 11.0.7, 11.0.7 +AMD Radeon RX 6700 XT / 6800M / 6700M, NAVY_FLOUNDER, DCN 3.0.0, 10.3.2, VCN 3.0.0, 5.2.2, 11.0.11, 11.0.11 +AMD Radeon RX 6600(XT) /6600M /W6600 /W6600M, DIMGREY_CAVEFISH, DCN 3.0.2, 10.3.4, VCN 3.0.16, 5.2.4, 11.0.12, 11.0.12 +AMD Radeon RX 6500M /6300M /W6500M /W6300M, BEIGE_GOBY, DCN 3.0.3, 10.3.5, VCN 3.0.33, 5.2.5, 11.0.13, 11.0.13 +AMD Radeon RX 7900 XT /XTX, , DCN 3.2.0, 11.0.0, VCN 4.0.0, 6.0.0, 13.0.0, 13.0.0 +AMD Radeon RX 7800 XT, , DCN 3.2.0, 11.0.3, VCN 4.0.0, 6.0.3, 13.0.10, 13.0.10 +AMD Radeon RX 7600M (XT) /7700S /7600S, , DCN 3.2.1, 11.0.2, VCN 4.0.4, 6.0.2, 13.0.7, 13.0.7 +AMD Radeon RX 9070 (XT), , DCN 4.0.1, 12.0.1, VCN 5.0.0, 7.0.1, 14.0.3, 14.0.3 +AMD Radeon RX 9060 XT, , DCN 4.0.1, 12.0.0, VCN 5.0.0, 7.0.0, 14.0.2, 14.0.2 diff --git a/Documentation/gpu/amdgpu/display/dc-glossary.rst b/Documentation/gpu/amdgpu/display/dc-glossary.rst index 7dc034e9e586..cbe737d1fcea 100644 --- a/Documentation/gpu/amdgpu/display/dc-glossary.rst +++ b/Documentation/gpu/amdgpu/display/dc-glossary.rst @@ -5,7 +5,7 @@ DC Glossary On this page, we try to keep track of acronyms related to the display component. If you do not find what you are looking for, look at the 'Documentation/gpu/amdgpu/amdgpu-glossary.rst'; if you cannot find it anywhere, -consider asking in the amdgfx and update this page. +consider asking on the amd-gfx mailing list and update this page. .. glossary:: diff --git a/Documentation/gpu/amdgpu/display/display-contributing.rst b/Documentation/gpu/amdgpu/display/display-contributing.rst index 36f3077eee00..2f741c52dce5 100644 --- a/Documentation/gpu/amdgpu/display/display-contributing.rst +++ b/Documentation/gpu/amdgpu/display/display-contributing.rst @@ -9,8 +9,8 @@ contribution to the display code, and for that, we say thank you :) This page summarizes some of the issues you can help with; keep in mind that this is a static page, and it is always a good idea to try to reach developers -in the amdgfx or some of the maintainers. Finally, this page follows the DRM -way of creating a TODO list; for more information, check +on the amd-gfx mailing list or some of the maintainers. Finally, this page +follows the DRM way of creating a TODO list; for more information, check 'Documentation/gpu/todo.rst'. Gitlab issues diff --git a/Documentation/gpu/amdgpu/display/programming-model-dcn.rst b/Documentation/gpu/amdgpu/display/programming-model-dcn.rst index c1b48d49fb0b..bc7de97a746f 100644 --- a/Documentation/gpu/amdgpu/display/programming-model-dcn.rst +++ b/Documentation/gpu/amdgpu/display/programming-model-dcn.rst @@ -100,7 +100,7 @@ represents the connected display. For historical reasons, we used the name `dc_link`, which gives the wrong impression that this abstraction only deals with physical connections that the developer can easily manipulate. However, this also covers - conections like eDP or cases where the output is connected to other devices. + connections like eDP or cases where the output is connected to other devices. There are two structs that are not represented in the diagram since they were elaborated in the DCN overview page (check the DCN block diagram :ref:`Display diff --git a/Documentation/gpu/amdgpu/driver-core.rst b/Documentation/gpu/amdgpu/driver-core.rst index 81256318e93c..bd4be32f2725 100644 --- a/Documentation/gpu/amdgpu/driver-core.rst +++ b/Documentation/gpu/amdgpu/driver-core.rst @@ -65,7 +65,7 @@ SDMA (System DMA) GC (Graphics and Compute) This is the graphics and compute engine, i.e., the block that - encompasses the 3D pipeline and and shader blocks. This is by far the + encompasses the 3D pipeline and shader blocks. This is by far the largest block on the GPU. The 3D pipeline has tons of sub-blocks. In addition to that, it also contains the CP microcontrollers (ME, PFP, CE, MEC) and the RLC microcontroller. It's exposed to userspace for user mode diff --git a/Documentation/gpu/amdgpu/process-isolation.rst b/Documentation/gpu/amdgpu/process-isolation.rst index 6b6d70e357a7..25b06ffefc33 100644 --- a/Documentation/gpu/amdgpu/process-isolation.rst +++ b/Documentation/gpu/amdgpu/process-isolation.rst @@ -26,7 +26,7 @@ Example of enabling enforce isolation on a GPU with multiple partitions: $ cat /sys/class/drm/card0/device/enforce_isolation 1 0 1 0 -The output indicates that enforce isolation is enabled on zeroth and second parition and disabled on first and fourth parition. +The output indicates that enforce isolation is enabled on zeroth and second partition and disabled on first and third partition. For devices with a single partition or those that do not support partitions, there will be only one element: diff --git a/Documentation/gpu/drm-uapi.rst b/Documentation/gpu/drm-uapi.rst index 843facf01b2d..d98428a592f1 100644 --- a/Documentation/gpu/drm-uapi.rst +++ b/Documentation/gpu/drm-uapi.rst @@ -418,13 +418,12 @@ needed. Recovery -------- -Current implementation defines three recovery methods, out of which, drivers +Current implementation defines four recovery methods, out of which, drivers can use any one, multiple or none. Method(s) of choice will be sent in the uevent environment as ``WEDGED=<method1>[,..,<methodN>]`` in order of less to -more side-effects. If driver is unsure about recovery or method is unknown -(like soft/hard system reboot, firmware flashing, physical device replacement -or any other procedure which can't be attempted on the fly), ``WEDGED=unknown`` -will be sent instead. +more side-effects. See the section `Vendor Specific Recovery`_ +for ``WEDGED=vendor-specific``. If driver is unsure about recovery or +method is unknown, ``WEDGED=unknown`` will be sent instead. Userspace consumers can parse this event and attempt recovery as per the following expectations. @@ -435,6 +434,7 @@ following expectations. none optional telemetry collection rebind unbind + bind driver bus-reset unbind + bus reset/re-enumeration + bind + vendor-specific vendor specific recovery method unknown consumer policy =============== ======================================== @@ -446,6 +446,35 @@ telemetry information (devcoredump, syslog). This is useful because the first hang is usually the most critical one which can result in consequential hangs or complete wedging. + +Vendor Specific Recovery +------------------------ + +When ``WEDGED=vendor-specific`` is sent, it indicates that the device requires +a recovery procedure specific to the hardware vendor and is not one of the +standardized approaches. + +``WEDGED=vendor-specific`` may be used to indicate different cases within a +single vendor driver, each requiring a distinct recovery procedure. +In such scenarios, the vendor driver must provide comprehensive documentation +that describes each case, include additional hints to identify specific case and +outline the corresponding recovery procedure. The documentation includes: + +Case - A list of all cases that sends the ``WEDGED=vendor-specific`` recovery method. + +Hints - Additional Information to assist the userspace consumer in identifying and +differentiating between different cases. This can be exposed through sysfs, debugfs, +traces, dmesg etc. + +Recovery Procedure - Clear instructions and guidance for recovering each case. +This may include userspace scripts, tools needed for the recovery procedure. + +It is the responsibility of the admin/userspace consumer to identify the case and +verify additional identification hints before attempting a recovery procedure. + +Example: If the device uses the Xe driver, then userspace consumer should refer to +:ref:`Xe Device Wedging <xe-device-wedging>` for the detailed documentation. + Task information ---------------- @@ -472,8 +501,12 @@ erroring out, all device memory should be unmapped and file descriptors should be closed to prevent leaks or undefined behaviour. The idea here is to clear the device of all user context beforehand and set the stage for a clean recovery. -Example -------- +For ``WEDGED=vendor-specific`` recovery method, it is the responsibility of the +consumer to check the driver documentation and the usecase before attempting +a recovery. + +Example - rebind +---------------- Udev rule:: diff --git a/Documentation/gpu/xe/index.rst b/Documentation/gpu/xe/index.rst index 42ba6c263cd0..88b22fad880e 100644 --- a/Documentation/gpu/xe/index.rst +++ b/Documentation/gpu/xe/index.rst @@ -25,5 +25,6 @@ DG2, etc is provided to prototype the driver. xe_tile xe_debugging xe_devcoredump + xe_device xe-drm-usage-stats.rst xe_configfs diff --git a/Documentation/gpu/xe/xe_device.rst b/Documentation/gpu/xe/xe_device.rst new file mode 100644 index 000000000000..39a937b97cd3 --- /dev/null +++ b/Documentation/gpu/xe/xe_device.rst @@ -0,0 +1,10 @@ +.. SPDX-License-Identifier: (GPL-2.0+ OR MIT) + +.. _xe-device-wedging: + +================== +Xe Device Wedging +================== + +.. kernel-doc:: drivers/gpu/drm/xe/xe_device.c + :doc: Xe Device Wedging diff --git a/Documentation/gpu/xe/xe_pcode.rst b/Documentation/gpu/xe/xe_pcode.rst index 5937ef3599b0..2a43601123cb 100644 --- a/Documentation/gpu/xe/xe_pcode.rst +++ b/Documentation/gpu/xe/xe_pcode.rst @@ -13,9 +13,11 @@ Internal API .. kernel-doc:: drivers/gpu/drm/xe/xe_pcode.c :internal: +.. _xe-survivability-mode: + ================== -Boot Survivability +Survivability Mode ================== .. kernel-doc:: drivers/gpu/drm/xe/xe_survivability_mode.c - :doc: Xe Boot Survivability + :doc: Survivability Mode diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index ef3af170dda4..01f53700694b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -819,6 +819,20 @@ struct amdgpu_ip_map_info { uint32_t mask); }; +enum amdgpu_uid_type { + AMDGPU_UID_TYPE_XCD, + AMDGPU_UID_TYPE_AID, + AMDGPU_UID_TYPE_SOC, + AMDGPU_UID_TYPE_MAX +}; + +#define AMDGPU_UID_INST_MAX 8 /* max number of instances for each UID type */ + +struct amdgpu_uid { + uint64_t uid[AMDGPU_UID_TYPE_MAX][AMDGPU_UID_INST_MAX]; + struct amdgpu_device *adev; +}; + struct amd_powerplay { void *pp_handle; const struct amd_pm_funcs *pp_funcs; @@ -896,6 +910,9 @@ struct amdgpu_pcie_reset_ctx { bool in_link_reset; bool occurs_dpc; bool audio_suspended; + struct pci_dev *swus; + struct pci_saved_state *swus_pcistate; + struct pci_saved_state *swds_pcistate; }; /* @@ -929,12 +946,6 @@ enum amdgpu_enforce_isolation_mode { AMDGPU_ENFORCE_ISOLATION_NO_CLEANER_SHADER = 3, }; - -/* - * Non-zero (true) if the GPU has VRAM. Zero (false) otherwise. - */ -#define AMDGPU_HAS_VRAM(_adev) ((_adev)->gmc.real_vram_size) - struct amdgpu_device { struct device *dev; struct pci_dev *pdev; @@ -1302,6 +1313,7 @@ struct amdgpu_device { struct list_head userq_mgr_list; struct mutex userq_mutex; bool userq_halt_for_enforce_isolation; + struct amdgpu_uid *uid_info; }; static inline uint32_t amdgpu_ip_version(const struct amdgpu_device *adev, @@ -1785,4 +1797,9 @@ static inline int amdgpu_device_bus_status_check(struct amdgpu_device *adev) return 0; } +void amdgpu_device_set_uid(struct amdgpu_uid *uid_info, + enum amdgpu_uid_type type, uint8_t inst, + uint64_t uid); +uint64_t amdgpu_device_get_uid(struct amdgpu_uid *uid_info, + enum amdgpu_uid_type type, uint8_t inst); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.c index cbc40cad581b..9b3180449150 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.c @@ -76,6 +76,7 @@ static void aca_banks_release(struct aca_banks *banks) list_for_each_entry_safe(node, tmp, &banks->list, node) { list_del(&node->node); kvfree(node); + banks->nr_banks--; } } @@ -130,6 +131,27 @@ static void aca_smu_bank_dump(struct amdgpu_device *adev, int idx, int total, st RAS_EVENT_LOG(adev, event_id, HW_ERR "hardware error logged by the scrubber\n"); } +static bool aca_bank_hwip_is_matched(struct aca_bank *bank, enum aca_hwip_type type) +{ + + struct aca_hwip *hwip; + int hwid, mcatype; + u64 ipid; + + if (!bank || type == ACA_HWIP_TYPE_UNKNOW) + return false; + + hwip = &aca_hwid_mcatypes[type]; + if (!hwip->hwid) + return false; + + ipid = bank->regs[ACA_REG_IDX_IPID]; + hwid = ACA_REG__IPID__HARDWAREID(ipid); + mcatype = ACA_REG__IPID__MCATYPE(ipid); + + return hwip->hwid == hwid && hwip->mcatype == mcatype; +} + static int aca_smu_get_valid_aca_banks(struct amdgpu_device *adev, enum aca_smu_type type, int start, int count, struct aca_banks *banks, struct ras_query_context *qctx) @@ -168,6 +190,15 @@ static int aca_smu_get_valid_aca_banks(struct amdgpu_device *adev, enum aca_smu_ bank.smu_err_type = type; + /* + * Poison being consumed when injecting a UE while running background workloads, + * which are unexpected. + */ + if (type == ACA_SMU_TYPE_UE && + ACA_REG__STATUS__POISON(bank.regs[ACA_REG_IDX_STATUS]) && + !aca_bank_hwip_is_matched(&bank, ACA_HWIP_TYPE_UMC)) + continue; + aca_smu_bank_dump(adev, i, count, &bank, qctx); ret = aca_banks_add_bank(banks, &bank); @@ -178,27 +209,6 @@ static int aca_smu_get_valid_aca_banks(struct amdgpu_device *adev, enum aca_smu_ return 0; } -static bool aca_bank_hwip_is_matched(struct aca_bank *bank, enum aca_hwip_type type) -{ - - struct aca_hwip *hwip; - int hwid, mcatype; - u64 ipid; - - if (!bank || type == ACA_HWIP_TYPE_UNKNOW) - return false; - - hwip = &aca_hwid_mcatypes[type]; - if (!hwip->hwid) - return false; - - ipid = bank->regs[ACA_REG_IDX_IPID]; - hwid = ACA_REG__IPID__HARDWAREID(ipid); - mcatype = ACA_REG__IPID__MCATYPE(ipid); - - return hwip->hwid == hwid && hwip->mcatype == mcatype; -} - static bool aca_bank_is_valid(struct aca_handle *handle, struct aca_bank *bank, enum aca_smu_type type) { const struct aca_bank_ops *bank_ops = handle->bank_ops; @@ -229,6 +239,7 @@ static struct aca_bank_error *new_bank_error(struct aca_error *aerr, struct aca_ mutex_lock(&aerr->lock); list_add_tail(&bank_error->node, &aerr->list); + aerr->nr_errors++; mutex_unlock(&aerr->lock); return bank_error; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c index 260165bbe373..d478acb4568a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c @@ -494,7 +494,8 @@ static int vm_update_pds(struct amdgpu_vm *vm, struct amdgpu_sync *sync) return amdgpu_sync_fence(sync, vm->last_update, GFP_KERNEL); } -static uint64_t get_pte_flags(struct amdgpu_device *adev, struct kgd_mem *mem) +static uint64_t get_pte_flags(struct amdgpu_device *adev, struct amdgpu_vm *vm, + struct kgd_mem *mem) { uint32_t mapping_flags = AMDGPU_VM_PAGE_READABLE | AMDGPU_VM_MTYPE_DEFAULT; @@ -504,7 +505,7 @@ static uint64_t get_pte_flags(struct amdgpu_device *adev, struct kgd_mem *mem) if (mem->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_EXECUTABLE) mapping_flags |= AMDGPU_VM_PAGE_EXECUTABLE; - return amdgpu_gem_va_map_flags(adev, mapping_flags); + return mapping_flags; } /** @@ -961,7 +962,7 @@ static int kfd_mem_attach(struct amdgpu_device *adev, struct kgd_mem *mem, goto unwind; } attachment[i]->va = va; - attachment[i]->pte_flags = get_pte_flags(adev, mem); + attachment[i]->pte_flags = get_pte_flags(adev, vm, mem); attachment[i]->adev = adev; list_add(&attachment[i]->list, &mem->attachments); @@ -2969,9 +2970,22 @@ int amdgpu_amdkfd_gpuvm_restore_process_bos(void *info, struct dma_fence __rcu * struct amdgpu_device *adev = amdgpu_ttm_adev( peer_vm->root.bo->tbo.bdev); + struct amdgpu_fpriv *fpriv = + container_of(peer_vm, struct amdgpu_fpriv, vm); + + ret = amdgpu_vm_bo_update(adev, fpriv->prt_va, false); + if (ret) { + dev_dbg(adev->dev, + "Memory eviction: handle PRT moved failed, pid %8d. Try again.\n", + pid_nr(process_info->pid)); + goto validate_map_fail; + } + ret = amdgpu_vm_handle_moved(adev, peer_vm, &exec.ticket); if (ret) { - pr_debug("Memory eviction: handle moved failed. Try again\n"); + dev_dbg(adev->dev, + "Memory eviction: handle moved failed, pid %8d. Try again.\n", + pid_nr(process_info->pid)); goto validate_map_fail; } } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c index 5e375e9c4f5d..a381de8648e5 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c @@ -1195,29 +1195,60 @@ static void amdgpu_connector_dvi_force(struct drm_connector *connector) amdgpu_connector->use_digital = true; } +/** + * Returns the maximum supported HDMI (TMDS) pixel clock in KHz. + */ +static int amdgpu_max_hdmi_pixel_clock(const struct amdgpu_device *adev) +{ + if (adev->asic_type >= CHIP_POLARIS10) + return 600000; + else if (adev->asic_type >= CHIP_TONGA) + return 300000; + else + return 297000; +} + +/** + * Validates the given display mode on DVI and HDMI connectors, + * including analog signals on DVI-I. + */ static enum drm_mode_status amdgpu_connector_dvi_mode_valid(struct drm_connector *connector, const struct drm_display_mode *mode) { struct drm_device *dev = connector->dev; struct amdgpu_device *adev = drm_to_adev(dev); struct amdgpu_connector *amdgpu_connector = to_amdgpu_connector(connector); + const int max_hdmi_pixel_clock = amdgpu_max_hdmi_pixel_clock(adev); + const int max_dvi_single_link_pixel_clock = 165000; + int max_digital_pixel_clock_khz; /* XXX check mode bandwidth */ - if (amdgpu_connector->use_digital && (mode->clock > 165000)) { - if ((amdgpu_connector->connector_object_id == CONNECTOR_OBJECT_ID_DUAL_LINK_DVI_I) || - (amdgpu_connector->connector_object_id == CONNECTOR_OBJECT_ID_DUAL_LINK_DVI_D) || - (amdgpu_connector->connector_object_id == CONNECTOR_OBJECT_ID_HDMI_TYPE_B)) { - return MODE_OK; - } else if (connector->display_info.is_hdmi) { - /* HDMI 1.3+ supports max clock of 340 Mhz */ - if (mode->clock > 340000) - return MODE_CLOCK_HIGH; - else - return MODE_OK; - } else { - return MODE_CLOCK_HIGH; + if (amdgpu_connector->use_digital) { + switch (amdgpu_connector->connector_object_id) { + case CONNECTOR_OBJECT_ID_HDMI_TYPE_A: + max_digital_pixel_clock_khz = max_hdmi_pixel_clock; + break; + case CONNECTOR_OBJECT_ID_SINGLE_LINK_DVI_I: + case CONNECTOR_OBJECT_ID_SINGLE_LINK_DVI_D: + max_digital_pixel_clock_khz = max_dvi_single_link_pixel_clock; + break; + case CONNECTOR_OBJECT_ID_DUAL_LINK_DVI_I: + case CONNECTOR_OBJECT_ID_DUAL_LINK_DVI_D: + case CONNECTOR_OBJECT_ID_HDMI_TYPE_B: + max_digital_pixel_clock_khz = max_dvi_single_link_pixel_clock * 2; + break; } + + /* When the display EDID claims that it's an HDMI display, + * we use the HDMI encoder mode of the display HW, + * so we should verify against the max HDMI clock here. + */ + if (connector->display_info.is_hdmi) + max_digital_pixel_clock_khz = max_hdmi_pixel_clock; + + if (mode->clock > max_digital_pixel_clock_khz) + return MODE_CLOCK_HIGH; } /* check against the max pixel clock */ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cper.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cper.c index 25252231a68a..6c266f18c598 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cper.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cper.c @@ -206,6 +206,7 @@ int amdgpu_cper_entry_fill_bad_page_threshold_section(struct amdgpu_device *adev { struct cper_sec_desc *section_desc; struct cper_sec_nonstd_err *section; + uint32_t socket_id; section_desc = (struct cper_sec_desc *)((uint8_t *)hdr + SEC_DESC_OFFSET(idx)); section = (struct cper_sec_nonstd_err *)((uint8_t *)hdr + @@ -224,6 +225,9 @@ int amdgpu_cper_entry_fill_bad_page_threshold_section(struct amdgpu_device *adev section->ctx.reg_arr_size = sizeof(section->ctx.reg_dump); /* Hardcoded Reg dump for bad page threshold CPER */ + socket_id = (adev->smuio.funcs && adev->smuio.funcs->get_socket_id) ? + adev->smuio.funcs->get_socket_id(adev) : + 0; section->ctx.reg_dump[CPER_ACA_REG_CTL_LO] = 0x1; section->ctx.reg_dump[CPER_ACA_REG_CTL_HI] = 0x0; section->ctx.reg_dump[CPER_ACA_REG_STATUS_LO] = 0x137; @@ -234,8 +238,8 @@ int amdgpu_cper_entry_fill_bad_page_threshold_section(struct amdgpu_device *adev section->ctx.reg_dump[CPER_ACA_REG_MISC0_HI] = 0x0; section->ctx.reg_dump[CPER_ACA_REG_CONFIG_LO] = 0x2; section->ctx.reg_dump[CPER_ACA_REG_CONFIG_HI] = 0x1ff; - section->ctx.reg_dump[CPER_ACA_REG_IPID_LO] = 0x0; - section->ctx.reg_dump[CPER_ACA_REG_IPID_HI] = 0x96; + section->ctx.reg_dump[CPER_ACA_REG_IPID_LO] = (socket_id / 4) & 0x01; + section->ctx.reg_dump[CPER_ACA_REG_IPID_HI] = 0x096 | (((socket_id % 4) & 0x3) << 12); section->ctx.reg_dump[CPER_ACA_REG_SYND_LO] = 0x0; section->ctx.reg_dump[CPER_ACA_REG_SYND_HI] = 0x0; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c index d3f220be2ef9..2ac9729e4c86 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c @@ -396,7 +396,7 @@ static int amdgpu_cs_p2_ib(struct amdgpu_cs_parser *p, chunk_ib->ib_bytes : 0, AMDGPU_IB_POOL_DELAYED, ib); if (r) { - DRM_ERROR("Failed to get ib !\n"); + drm_err(adev_to_drm(p->adev), "Failed to get ib !\n"); return r; } @@ -468,7 +468,7 @@ static int amdgpu_syncobj_lookup_and_add(struct amdgpu_cs_parser *p, r = drm_syncobj_find_fence(p->filp, handle, point, flags, &fence); if (r) { - DRM_ERROR("syncobj %u failed to find fence @ %llu (%d)!\n", + drm_err(adev_to_drm(p->adev), "syncobj %u failed to find fence @ %llu (%d)!\n", handle, point, r); return r; } @@ -902,7 +902,7 @@ static int amdgpu_cs_parser_bos(struct amdgpu_cs_parser *p, sizeof(struct page *), GFP_KERNEL); if (!e->user_pages) { - DRM_ERROR("kvmalloc_array failure\n"); + drm_err(adev_to_drm(p->adev), "kvmalloc_array failure\n"); r = -ENOMEM; goto out_free_user_pages; } @@ -983,7 +983,7 @@ static int amdgpu_cs_parser_bos(struct amdgpu_cs_parser *p, r = amdgpu_vm_validate(p->adev, &fpriv->vm, NULL, amdgpu_cs_bo_validate, p); if (r) { - DRM_ERROR("amdgpu_vm_validate() failed.\n"); + drm_err(adev_to_drm(p->adev), "amdgpu_vm_validate() failed.\n"); goto out_free_user_pages; } @@ -1061,13 +1061,13 @@ static int amdgpu_cs_patch_ibs(struct amdgpu_cs_parser *p, va_start = ib->gpu_addr & AMDGPU_GMC_HOLE_MASK; r = amdgpu_cs_find_mapping(p, va_start, &aobj, &m); if (r) { - DRM_ERROR("IB va_start is invalid\n"); + drm_err(adev_to_drm(p->adev), "IB va_start is invalid\n"); return r; } if ((va_start + ib->length_dw * 4) > (m->last + 1) * AMDGPU_GPU_PAGE_SIZE) { - DRM_ERROR("IB va_start+ib_bytes is invalid\n"); + drm_err(adev_to_drm(p->adev), "IB va_start+ib_bytes is invalid\n"); return -EINVAL; } @@ -1238,7 +1238,7 @@ static int amdgpu_cs_sync_rings(struct amdgpu_cs_parser *p) r = amdgpu_ctx_wait_prev_fence(p->ctx, p->entities[p->gang_leader_idx]); if (r) { if (r != -ERESTARTSYS) - DRM_ERROR("amdgpu_ctx_wait_prev_fence failed.\n"); + drm_err(adev_to_drm(p->adev), "amdgpu_ctx_wait_prev_fence failed.\n"); return r; } @@ -1451,7 +1451,7 @@ int amdgpu_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) r = amdgpu_cs_parser_init(&parser, adev, filp, data); if (r) { - DRM_ERROR_RATELIMITED("Failed to initialize parser %d!\n", r); + drm_err_ratelimited(dev, "Failed to initialize parser %d!\n", r); return r; } @@ -1466,9 +1466,9 @@ int amdgpu_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) r = amdgpu_cs_parser_bos(&parser, data); if (r) { if (r == -ENOMEM) - DRM_ERROR("Not enough memory for command submission!\n"); + drm_err(dev, "Not enough memory for command submission!\n"); else if (r != -ERESTARTSYS && r != -EAGAIN) - DRM_DEBUG("Failed to process the buffer list %d!\n", r); + drm_dbg(dev, "Failed to process the buffer list %d!\n", r); goto error_fini; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index 01d234cf8156..b9e2f0293d61 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -178,6 +178,8 @@ struct amdgpu_init_level amdgpu_init_minimal_xgmi = { BIT(AMD_IP_BLOCK_TYPE_PSP) }; +static void amdgpu_device_load_switch_state(struct amdgpu_device *adev); + static inline bool amdgpu_ip_member_of_hwini(struct amdgpu_device *adev, enum amd_ip_block_type block) { @@ -2445,6 +2447,33 @@ int amdgpu_device_ip_block_version_cmp(struct amdgpu_device *adev, return 1; } +static const char *ip_block_names[] = { + [AMD_IP_BLOCK_TYPE_COMMON] = "common", + [AMD_IP_BLOCK_TYPE_GMC] = "gmc", + [AMD_IP_BLOCK_TYPE_IH] = "ih", + [AMD_IP_BLOCK_TYPE_SMC] = "smu", + [AMD_IP_BLOCK_TYPE_PSP] = "psp", + [AMD_IP_BLOCK_TYPE_DCE] = "dce", + [AMD_IP_BLOCK_TYPE_GFX] = "gfx", + [AMD_IP_BLOCK_TYPE_SDMA] = "sdma", + [AMD_IP_BLOCK_TYPE_UVD] = "uvd", + [AMD_IP_BLOCK_TYPE_VCE] = "vce", + [AMD_IP_BLOCK_TYPE_ACP] = "acp", + [AMD_IP_BLOCK_TYPE_VCN] = "vcn", + [AMD_IP_BLOCK_TYPE_MES] = "mes", + [AMD_IP_BLOCK_TYPE_JPEG] = "jpeg", + [AMD_IP_BLOCK_TYPE_VPE] = "vpe", + [AMD_IP_BLOCK_TYPE_UMSCH_MM] = "umsch_mm", + [AMD_IP_BLOCK_TYPE_ISP] = "isp", +}; + +static const char *ip_block_name(struct amdgpu_device *adev, enum amd_ip_block_type type) +{ + int idx = (int)type; + + return idx < ARRAY_SIZE(ip_block_names) ? ip_block_names[idx] : "unknown"; +} + /** * amdgpu_device_ip_block_add * @@ -2473,8 +2502,13 @@ int amdgpu_device_ip_block_add(struct amdgpu_device *adev, break; } - dev_info(adev->dev, "detected ip block number %d <%s>\n", - adev->num_ip_blocks, ip_block_version->funcs->name); + dev_info(adev->dev, "detected ip block number %d <%s_v%d_%d_%d> (%s)\n", + adev->num_ip_blocks, + ip_block_name(adev, ip_block_version->type), + ip_block_version->major, + ip_block_version->minor, + ip_block_version->rev, + ip_block_version->funcs->name); adev->ip_blocks[adev->num_ip_blocks].adev = adev; @@ -2674,6 +2708,24 @@ out: return err; } +static void amdgpu_uid_init(struct amdgpu_device *adev) +{ + /* Initialize the UID for the device */ + adev->uid_info = kzalloc(sizeof(struct amdgpu_uid), GFP_KERNEL); + if (!adev->uid_info) { + dev_warn(adev->dev, "Failed to allocate memory for UID\n"); + return; + } + adev->uid_info->adev = adev; +} + +static void amdgpu_uid_fini(struct amdgpu_device *adev) +{ + /* Free the UID memory */ + kfree(adev->uid_info); + adev->uid_info = NULL; +} + /** * amdgpu_device_ip_early_init - run early init for hardware IPs * @@ -2857,6 +2909,8 @@ static int amdgpu_device_ip_early_init(struct amdgpu_device *adev) if (adev->gmc.xgmi.supported) amdgpu_xgmi_early_init(adev); + if (amdgpu_is_multi_aid(adev)) + amdgpu_uid_init(adev); ip_block = amdgpu_device_ip_get_ip_block(adev, AMD_IP_BLOCK_TYPE_GFX); if (ip_block->status.valid != false) amdgpu_amdkfd_device_probe(adev); @@ -3389,7 +3443,7 @@ static int amdgpu_device_enable_mgpu_fan_boost(void) for (i = 0; i < mgpu_info.num_dgpu; i++) { gpu_ins = &(mgpu_info.gpu_ins[i]); adev = gpu_ins->adev; - if (!(adev->flags & AMD_IS_APU) && + if (!(adev->flags & AMD_IS_APU || amdgpu_sriov_multi_vf_mode(adev)) && !gpu_ins->mgpu_fan_enabled) { ret = amdgpu_dpm_enable_mgpu_fan_boost(adev); if (ret) @@ -3648,6 +3702,7 @@ static int amdgpu_device_ip_fini(struct amdgpu_device *adev) } amdgpu_ras_fini(adev); + amdgpu_uid_fini(adev); return 0; } @@ -4992,7 +5047,8 @@ void amdgpu_device_fini_sw(struct amdgpu_device *adev) adev->reset_domain = NULL; kfree(adev->pci_state); - + kfree(adev->pcie_reset_ctx.swds_pcistate); + kfree(adev->pcie_reset_ctx.swus_pcistate); } /** @@ -5701,7 +5757,7 @@ int amdgpu_device_link_reset(struct amdgpu_device *adev) dev_info(adev->dev, "GPU link reset\n"); - if (!adev->pcie_reset_ctx.occurs_dpc) + if (!amdgpu_reset_in_dpc(adev)) ret = amdgpu_dpm_link_reset(adev); if (ret) @@ -5830,6 +5886,7 @@ int amdgpu_device_reinit_after_reset(struct amdgpu_reset_context *reset_context) amdgpu_set_init_level(tmp_adev, init_level); if (full_reset) { /* post card */ + amdgpu_reset_set_dpc_status(tmp_adev, false); amdgpu_ras_clear_err_state(tmp_adev); r = amdgpu_device_asic_init(tmp_adev); if (r) { @@ -6136,12 +6193,11 @@ static int amdgpu_device_health_check(struct list_head *device_list_handle) return ret; } -static int amdgpu_device_recovery_prepare(struct amdgpu_device *adev, +static void amdgpu_device_recovery_prepare(struct amdgpu_device *adev, struct list_head *device_list, struct amdgpu_hive_info *hive) { struct amdgpu_device *tmp_adev = NULL; - int r; /* * Build list of devices to reset. @@ -6153,7 +6209,7 @@ static int amdgpu_device_recovery_prepare(struct amdgpu_device *adev, list_add_tail(&tmp_adev->reset_list, device_list); if (adev->shutdown) tmp_adev->shutdown = true; - if (adev->pcie_reset_ctx.occurs_dpc) + if (amdgpu_reset_in_dpc(adev)) tmp_adev->pcie_reset_ctx.in_link_reset = true; } if (!list_is_first(&adev->reset_list, device_list)) @@ -6161,14 +6217,6 @@ static int amdgpu_device_recovery_prepare(struct amdgpu_device *adev, } else { list_add_tail(&adev->reset_list, device_list); } - - if (!amdgpu_sriov_vf(adev) && (!adev->pcie_reset_ctx.occurs_dpc)) { - r = amdgpu_device_health_check(device_list); - if (r) - return r; - } - - return 0; } static void amdgpu_device_recovery_get_reset_lock(struct amdgpu_device *adev, @@ -6237,9 +6285,8 @@ static void amdgpu_device_halt_activities(struct amdgpu_device *adev, drm_client_dev_suspend(adev_to_drm(tmp_adev), false); /* disable ras on ALL IPs */ - if (!need_emergency_restart && - (!adev->pcie_reset_ctx.occurs_dpc) && - amdgpu_device_ip_need_full_reset(tmp_adev)) + if (!need_emergency_restart && !amdgpu_reset_in_dpc(adev) && + amdgpu_device_ip_need_full_reset(tmp_adev)) amdgpu_ras_suspend(tmp_adev); for (i = 0; i < AMDGPU_MAX_RINGS; ++i) { @@ -6267,11 +6314,7 @@ static int amdgpu_device_asic_reset(struct amdgpu_device *adev, retry: /* Rest of adevs pre asic reset from XGMI hive. */ list_for_each_entry(tmp_adev, device_list, reset_list) { - if (adev->pcie_reset_ctx.occurs_dpc) - tmp_adev->no_hw_access = true; r = amdgpu_device_pre_asic_reset(tmp_adev, reset_context); - if (adev->pcie_reset_ctx.occurs_dpc) - tmp_adev->no_hw_access = false; /*TODO Should we stop ?*/ if (r) { dev_err(tmp_adev->dev, "GPU pre asic reset failed with err, %d for drm dev, %s ", @@ -6449,8 +6492,9 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev, emergency_restart(); } - dev_info(adev->dev, "GPU %s begin!\n", - need_emergency_restart ? "jobs stop":"reset"); + dev_info(adev->dev, "GPU %s begin!. Source: %d\n", + need_emergency_restart ? "jobs stop" : "reset", + reset_context->src); if (!amdgpu_sriov_vf(adev)) hive = amdgpu_get_xgmi_hive(adev); @@ -6461,8 +6505,13 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev, reset_context->hive = hive; INIT_LIST_HEAD(&device_list); - if (amdgpu_device_recovery_prepare(adev, &device_list, hive)) - goto end_reset; + amdgpu_device_recovery_prepare(adev, &device_list, hive); + + if (!amdgpu_sriov_vf(adev)) { + r = amdgpu_device_health_check(&device_list); + if (r) + goto end_reset; + } /* We need to lock reset domain only once both for XGMI and single device */ amdgpu_device_recovery_get_reset_lock(adev, &device_list); @@ -6890,11 +6939,6 @@ pci_ers_result_t amdgpu_pci_error_detected(struct pci_dev *pdev, pci_channel_sta dev_info(adev->dev, "PCI error: detected callback!!\n"); - if (!amdgpu_dpm_is_link_reset_supported(adev)) { - dev_warn(adev->dev, "No support for XGMI hive yet...\n"); - return PCI_ERS_RESULT_DISCONNECT; - } - adev->pci_channel_state = state; switch (state) { @@ -6904,10 +6948,23 @@ pci_ers_result_t amdgpu_pci_error_detected(struct pci_dev *pdev, pci_channel_sta case pci_channel_io_frozen: /* Fatal error, prepare for slot reset */ dev_info(adev->dev, "pci_channel_io_frozen: state(%d)!!\n", state); + if (hive) { + /* Hive devices should be able to support FW based + * link reset on other devices, if not return. + */ + if (!amdgpu_dpm_is_link_reset_supported(adev)) { + dev_warn(adev->dev, + "No support for XGMI hive yet...\n"); + return PCI_ERS_RESULT_DISCONNECT; + } + /* Set dpc status only if device is part of hive + * Non-hive devices should be able to recover after + * link reset. + */ + amdgpu_reset_set_dpc_status(adev, true); - if (hive) mutex_lock(&hive->hive_lock); - adev->pcie_reset_ctx.occurs_dpc = true; + } memset(&reset_context, 0, sizeof(reset_context)); INIT_LIST_HEAD(&device_list); @@ -6966,22 +7023,34 @@ pci_ers_result_t amdgpu_pci_slot_reset(struct pci_dev *pdev) struct amdgpu_device *tmp_adev; struct amdgpu_hive_info *hive; struct list_head device_list; - int r = 0, i; + struct pci_dev *link_dev; + int r = 0, i, timeout; u32 memsize; - - /* PCI error slot reset should be skipped During RAS recovery */ - if ((amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(9, 4, 3) || - amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(9, 4, 4)) && - amdgpu_ras_in_recovery(adev)) - return PCI_ERS_RESULT_RECOVERED; + u16 status; dev_info(adev->dev, "PCI error: slot reset callback!!\n"); memset(&reset_context, 0, sizeof(reset_context)); - /* wait for asic to come out of reset */ - msleep(700); + if (adev->pcie_reset_ctx.swus) + link_dev = adev->pcie_reset_ctx.swus; + else + link_dev = adev->pdev; + /* wait for asic to come out of reset, timeout = 10s */ + timeout = 10000; + do { + usleep_range(10000, 10500); + r = pci_read_config_word(link_dev, PCI_VENDOR_ID, &status); + timeout -= 10; + } while (timeout > 0 && (status != PCI_VENDOR_ID_ATI) && + (status != PCI_VENDOR_ID_AMD)); + + if ((status != PCI_VENDOR_ID_ATI) && (status != PCI_VENDOR_ID_AMD)) { + r = -ETIME; + goto out; + } + amdgpu_device_load_switch_state(adev); /* Restore PCI confspace */ amdgpu_device_load_pci_state(pdev); @@ -7076,7 +7145,6 @@ void amdgpu_pci_resume(struct pci_dev *pdev) amdgpu_device_sched_resume(&device_list, NULL, NULL); amdgpu_device_gpu_resume(adev, &device_list, false); amdgpu_device_recovery_put_reset_lock(adev, &device_list); - adev->pcie_reset_ctx.occurs_dpc = false; if (hive) { mutex_unlock(&hive->hive_lock); @@ -7084,6 +7152,58 @@ void amdgpu_pci_resume(struct pci_dev *pdev) } } +static void amdgpu_device_cache_switch_state(struct amdgpu_device *adev) +{ + struct pci_dev *parent = pci_upstream_bridge(adev->pdev); + int r; + + if (parent->vendor != PCI_VENDOR_ID_ATI) + return; + + /* If already saved, return */ + if (adev->pcie_reset_ctx.swus) + return; + /* Upstream bridge is ATI, assume it's SWUS/DS architecture */ + r = pci_save_state(parent); + if (r) + return; + adev->pcie_reset_ctx.swds_pcistate = pci_store_saved_state(parent); + + parent = pci_upstream_bridge(parent); + r = pci_save_state(parent); + if (r) + return; + adev->pcie_reset_ctx.swus_pcistate = pci_store_saved_state(parent); + + adev->pcie_reset_ctx.swus = parent; +} + +static void amdgpu_device_load_switch_state(struct amdgpu_device *adev) +{ + struct pci_dev *pdev; + int r; + + if (!adev->pcie_reset_ctx.swds_pcistate || + !adev->pcie_reset_ctx.swus_pcistate) + return; + + pdev = adev->pcie_reset_ctx.swus; + r = pci_load_saved_state(pdev, adev->pcie_reset_ctx.swus_pcistate); + if (!r) { + pci_restore_state(pdev); + } else { + dev_warn(adev->dev, "Failed to load SWUS state, err:%d\n", r); + return; + } + + pdev = pci_upstream_bridge(adev->pdev); + r = pci_load_saved_state(pdev, adev->pcie_reset_ctx.swds_pcistate); + if (!r) + pci_restore_state(pdev); + else + dev_warn(adev->dev, "Failed to load SWDS state, err:%d\n", r); +} + bool amdgpu_device_cache_pci_state(struct pci_dev *pdev) { struct drm_device *dev = pci_get_drvdata(pdev); @@ -7108,6 +7228,8 @@ bool amdgpu_device_cache_pci_state(struct pci_dev *pdev) return false; } + amdgpu_device_cache_switch_state(adev); + return true; } @@ -7494,3 +7616,53 @@ ssize_t amdgpu_show_reset_mask(char *buf, uint32_t supported_reset) size += sysfs_emit_at(buf, size, "\n"); return size; } + +void amdgpu_device_set_uid(struct amdgpu_uid *uid_info, + enum amdgpu_uid_type type, uint8_t inst, + uint64_t uid) +{ + if (!uid_info) + return; + + if (type >= AMDGPU_UID_TYPE_MAX) { + dev_err_once(uid_info->adev->dev, "Invalid UID type %d\n", + type); + return; + } + + if (inst >= AMDGPU_UID_INST_MAX) { + dev_err_once(uid_info->adev->dev, "Invalid UID instance %d\n", + inst); + return; + } + + if (uid_info->uid[type][inst] != 0) { + dev_warn_once( + uid_info->adev->dev, + "Overwriting existing UID %llu for type %d instance %d\n", + uid_info->uid[type][inst], type, inst); + } + + uid_info->uid[type][inst] = uid; +} + +u64 amdgpu_device_get_uid(struct amdgpu_uid *uid_info, + enum amdgpu_uid_type type, uint8_t inst) +{ + if (!uid_info) + return 0; + + if (type >= AMDGPU_UID_TYPE_MAX) { + dev_err_once(uid_info->adev->dev, "Invalid UID type %d\n", + type); + return 0; + } + + if (inst >= AMDGPU_UID_INST_MAX) { + dev_err_once(uid_info->adev->dev, "Invalid UID instance %d\n", + inst); + return 0; + } + + return uid_info->uid[type][inst]; +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c index 395c6be901ce..17f754d1135d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c @@ -886,7 +886,7 @@ module_param_named(dcfeaturemask, amdgpu_dc_feature_mask, uint, 0444); /** * DOC: dcdebugmask (uint) - * Override display features enabled. See enum DC_DEBUG_MASK in drivers/gpu/drm/amd/include/amd_shared.h. + * Display debug options. See enum DC_DEBUG_MASK in drivers/gpu/drm/amd/include/amd_shared.h. */ MODULE_PARM_DESC(dcdebugmask, "all debug options disabled (default))"); module_param_named(dcdebugmask, amdgpu_dc_debug_mask, uint, 0444); @@ -2597,6 +2597,7 @@ static int amdgpu_pmops_suspend(struct device *dev) else if (amdgpu_acpi_is_s3_active(adev)) adev->in_s3 = true; if (!adev->in_s0ix && !adev->in_s3) { +#if IS_ENABLED(CONFIG_SUSPEND) /* don't allow going deep first time followed by s2idle the next time */ if (adev->last_suspend_state != PM_SUSPEND_ON && adev->last_suspend_state != pm_suspend_target_state) { @@ -2604,11 +2605,14 @@ static int amdgpu_pmops_suspend(struct device *dev) pm_suspend_target_state); return -EINVAL; } +#endif return 0; } +#if IS_ENABLED(CONFIG_SUSPEND) /* cache the state last used for suspend */ adev->last_suspend_state = pm_suspend_target_state; +#endif return amdgpu_device_suspend(drm_dev, true); } @@ -2933,11 +2937,14 @@ static int amdgpu_drm_release(struct inode *inode, struct file *filp) { struct drm_file *file_priv = filp->private_data; struct amdgpu_fpriv *fpriv = file_priv->driver_priv; + struct drm_device *dev = file_priv->minor->dev; + int idx; - if (fpriv) { + if (fpriv && drm_dev_enter(dev, &idx)) { fpriv->evf_mgr.fd_closing = true; amdgpu_eviction_fence_destroy(&fpriv->evf_mgr); amdgpu_userq_mgr_fini(&fpriv->userq_mgr); + drm_dev_exit(idx); } return drm_release(inode, filp); @@ -2964,15 +2971,15 @@ out: } static const struct dev_pm_ops amdgpu_pm_ops = { - .prepare = amdgpu_pmops_prepare, - .complete = amdgpu_pmops_complete, - .suspend = amdgpu_pmops_suspend, - .suspend_noirq = amdgpu_pmops_suspend_noirq, - .resume = amdgpu_pmops_resume, - .freeze = amdgpu_pmops_freeze, - .thaw = amdgpu_pmops_thaw, - .poweroff = amdgpu_pmops_poweroff, - .restore = amdgpu_pmops_restore, + .prepare = pm_sleep_ptr(amdgpu_pmops_prepare), + .complete = pm_sleep_ptr(amdgpu_pmops_complete), + .suspend = pm_sleep_ptr(amdgpu_pmops_suspend), + .suspend_noirq = pm_sleep_ptr(amdgpu_pmops_suspend_noirq), + .resume = pm_sleep_ptr(amdgpu_pmops_resume), + .freeze = pm_sleep_ptr(amdgpu_pmops_freeze), + .thaw = pm_sleep_ptr(amdgpu_pmops_thaw), + .poweroff = pm_sleep_ptr(amdgpu_pmops_poweroff), + .restore = pm_sleep_ptr(amdgpu_pmops_restore), .runtime_suspend = amdgpu_pmops_runtime_suspend, .runtime_resume = amdgpu_pmops_runtime_resume, .runtime_idle = amdgpu_pmops_runtime_idle, @@ -3117,7 +3124,7 @@ static struct pci_driver amdgpu_kms_pci_driver = { .probe = amdgpu_pci_probe, .remove = amdgpu_pci_remove, .shutdown = amdgpu_pci_shutdown, - .driver.pm = &amdgpu_pm_ops, + .driver.pm = pm_ptr(&amdgpu_pm_ops), .err_handler = &amdgpu_pci_err_handler, .dev_groups = amdgpu_sysfs_groups, }; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c index 9e7506965cab..fd8cca241da6 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c @@ -120,7 +120,6 @@ int amdgpu_fence_emit(struct amdgpu_ring *ring, struct dma_fence **f, am_fence = kzalloc(sizeof(*am_fence), GFP_KERNEL); if (!am_fence) return -ENOMEM; - am_fence->context = 0; } else { am_fence = af; } @@ -738,7 +737,7 @@ void amdgpu_fence_driver_force_completion(struct amdgpu_ring *ring) } -/** +/* * Kernel queue reset handling * * The driver can reset individual queues for most engines, but those queues diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c index 6626a6e64ff5..d5e685c5e28b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c @@ -790,36 +790,6 @@ error: return fence; } -/** - * amdgpu_gem_va_map_flags - map GEM UAPI flags into hardware flags - * - * @adev: amdgpu_device pointer - * @flags: GEM UAPI flags - * - * Returns the GEM UAPI flags mapped into hardware for the ASIC. - */ -uint64_t amdgpu_gem_va_map_flags(struct amdgpu_device *adev, uint32_t flags) -{ - uint64_t pte_flag = 0; - - if (flags & AMDGPU_VM_PAGE_EXECUTABLE) - pte_flag |= AMDGPU_PTE_EXECUTABLE; - if (flags & AMDGPU_VM_PAGE_READABLE) - pte_flag |= AMDGPU_PTE_READABLE; - if (flags & AMDGPU_VM_PAGE_WRITEABLE) - pte_flag |= AMDGPU_PTE_WRITEABLE; - if (flags & AMDGPU_VM_PAGE_PRT) - pte_flag |= AMDGPU_PTE_PRT_FLAG(adev); - if (flags & AMDGPU_VM_PAGE_NOALLOC) - pte_flag |= AMDGPU_PTE_NOALLOC; - - if (adev->gmc.gmc_funcs->map_mtype) - pte_flag |= amdgpu_gmc_map_mtype(adev, - flags & AMDGPU_VM_MTYPE_MASK); - - return pte_flag; -} - int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) { @@ -840,7 +810,6 @@ int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data, struct dma_fence_chain *timeline_chain = NULL; struct dma_fence *fence; struct drm_exec exec; - uint64_t va_flags; uint64_t vm_size; int r = 0; @@ -944,10 +913,9 @@ int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data, switch (args->operation) { case AMDGPU_VA_OP_MAP: - va_flags = amdgpu_gem_va_map_flags(adev, args->flags); r = amdgpu_vm_bo_map(adev, bo_va, args->va_address, args->offset_in_bo, args->map_size, - va_flags); + args->flags); break; case AMDGPU_VA_OP_UNMAP: r = amdgpu_vm_bo_unmap(adev, bo_va, args->va_address); @@ -959,10 +927,9 @@ int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data, args->map_size); break; case AMDGPU_VA_OP_REPLACE: - va_flags = amdgpu_gem_va_map_flags(adev, args->flags); r = amdgpu_vm_bo_replace_map(adev, bo_va, args->va_address, args->offset_in_bo, args->map_size, - va_flags); + args->flags); break; default: break; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.h index 3a8f57900a3a..b51e8f95ee86 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.h @@ -63,7 +63,6 @@ int amdgpu_gem_mmap_ioctl(struct drm_device *dev, void *data, struct drm_file *filp); int amdgpu_gem_wait_idle_ioctl(struct drm_device *dev, void *data, struct drm_file *filp); -uint64_t amdgpu_gem_va_map_flags(struct amdgpu_device *adev, uint32_t flags); int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data, struct drm_file *filp); int amdgpu_gem_op_ioctl(struct drm_device *dev, void *data, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h index 397c6ccdb903..55097ca10738 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h @@ -154,15 +154,15 @@ struct amdgpu_gmc_funcs { unsigned pasid); /* enable/disable PRT support */ void (*set_prt)(struct amdgpu_device *adev, bool enable); - /* map mtype to hardware flags */ - uint64_t (*map_mtype)(struct amdgpu_device *adev, uint32_t flags); /* get the pde for a given mc addr */ void (*get_vm_pde)(struct amdgpu_device *adev, int level, u64 *dst, u64 *flags); - /* get the pte flags to use for a BO VA mapping */ + /* get the pte flags to use for PTEs */ void (*get_vm_pte)(struct amdgpu_device *adev, - struct amdgpu_bo_va_mapping *mapping, - uint64_t *flags); + struct amdgpu_vm *vm, + struct amdgpu_bo *bo, + uint32_t vm_flags, + uint64_t *pte_flags); /* override per-page pte flags */ void (*override_vm_pte_flags)(struct amdgpu_device *dev, struct amdgpu_vm *vm, @@ -356,9 +356,10 @@ struct amdgpu_gmc { #define amdgpu_gmc_emit_flush_gpu_tlb(r, vmid, addr) (r)->adev->gmc.gmc_funcs->emit_flush_gpu_tlb((r), (vmid), (addr)) #define amdgpu_gmc_emit_pasid_mapping(r, vmid, pasid) (r)->adev->gmc.gmc_funcs->emit_pasid_mapping((r), (vmid), (pasid)) -#define amdgpu_gmc_map_mtype(adev, flags) (adev)->gmc.gmc_funcs->map_mtype((adev),(flags)) #define amdgpu_gmc_get_vm_pde(adev, level, dst, flags) (adev)->gmc.gmc_funcs->get_vm_pde((adev), (level), (dst), (flags)) -#define amdgpu_gmc_get_vm_pte(adev, mapping, flags) (adev)->gmc.gmc_funcs->get_vm_pte((adev), (mapping), (flags)) +#define amdgpu_gmc_get_vm_pte(adev, vm, bo, vm_flags, pte_flags) \ + ((adev)->gmc.gmc_funcs->get_vm_pte((adev), (vm), (bo), (vm_flags), \ + (pte_flags))) #define amdgpu_gmc_override_vm_pte_flags(adev, vm, addr, pte_flags) \ (adev)->gmc.gmc_funcs->override_vm_pte_flags \ ((adev), (vm), (addr), (pte_flags)) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c index 4980595bcddd..22da65f45226 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c @@ -121,10 +121,12 @@ static void amdgpu_jpeg_idle_work_handler(struct work_struct *work) fences += amdgpu_fence_count_emitted(&adev->jpeg.inst[i].ring_dec[j]); } - if (!fences && !atomic_read(&adev->jpeg.total_submission_cnt)) + if (!fences && !atomic_read(&adev->jpeg.total_submission_cnt)) { + mutex_lock(&adev->jpeg.jpeg_pg_lock); amdgpu_device_ip_set_powergating_state(adev, AMD_IP_BLOCK_TYPE_JPEG, AMD_PG_STATE_GATE); - else + mutex_unlock(&adev->jpeg.jpeg_pg_lock); + } else schedule_delayed_work(&adev->jpeg.idle_work, JPEG_IDLE_TIMEOUT); } @@ -369,7 +371,7 @@ static int amdgpu_debugfs_jpeg_sched_mask_set(void *data, u64 val) for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) { for (j = 0; j < adev->jpeg.num_jpeg_rings; ++j) { ring = &adev->jpeg.inst[i].ring_dec[j]; - if (val & (1 << ((i * adev->jpeg.num_jpeg_rings) + j))) + if (val & (BIT_ULL(1) << ((i * adev->jpeg.num_jpeg_rings) + j))) ring->sched.ready = true; else ring->sched.ready = false; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h index c0d2c195fe2e..489a4a0f0610 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h @@ -280,6 +280,13 @@ struct mes_reset_queue_input { bool is_kq; }; +struct mes_inv_tlbs_pasid_input { + uint32_t xcc_id; + uint16_t pasid; + uint8_t hub_id; + uint8_t flush_type; +}; + enum mes_misc_opcode { MES_MISC_OP_WRITE_REG, MES_MISC_OP_READ_REG, @@ -367,6 +374,9 @@ struct amdgpu_mes_funcs { int (*reset_hw_queue)(struct amdgpu_mes *mes, struct mes_reset_queue_input *input); + + int (*invalidate_tlbs_pasid)(struct amdgpu_mes *mes, + struct mes_inv_tlbs_pasid_input *input); }; #define amdgpu_mes_kiq_hw_init(adev) (adev)->mes.kiq_hw_init((adev)) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h index c316920f3450..87523fcd4386 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h @@ -69,7 +69,7 @@ struct amdgpu_bo_va_mapping { uint64_t last; uint64_t __subtree_last; uint64_t offset; - uint64_t flags; + uint32_t flags; }; /* User space allocated BO in a VM */ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c index 23484317a5fa..1d6e1d5de8fa 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c @@ -506,8 +506,7 @@ static int psp_sw_init(struct amdgpu_ip_block *ip_block) } ret = amdgpu_bo_create_kernel(adev, PSP_1_MEG, PSP_1_MEG, - (amdgpu_sriov_vf(adev) || adev->debug_use_vram_fw_buf) ? - AMDGPU_GEM_DOMAIN_VRAM : AMDGPU_GEM_DOMAIN_GTT, + AMDGPU_GEM_DOMAIN_VRAM, &psp->fw_pri_bo, &psp->fw_pri_mc_addr, &psp->fw_pri_buf); @@ -666,6 +665,10 @@ static const char *psp_gfx_cmd_name(enum psp_gfx_cmd_id cmd_id) return "FB_FW_RESERV_ADDR"; case GFX_CMD_ID_FB_FW_RESERV_EXT_ADDR: return "FB_FW_RESERV_EXT_ADDR"; + case GFX_CMD_ID_SRIOV_SPATIAL_PART: + return "SPATIAL_PARTITION"; + case GFX_CMD_ID_FB_NPS_MODE: + return "NPS_MODE_CHANGE"; default: return "UNKNOWN CMD"; } @@ -877,9 +880,7 @@ static int psp_tmr_init(struct psp_context *psp) pptr = amdgpu_sriov_vf(psp->adev) ? &tmr_buf : NULL; ret = amdgpu_bo_create_kernel(psp->adev, tmr_size, PSP_TMR_ALIGNMENT, - AMDGPU_HAS_VRAM(psp->adev) ? - AMDGPU_GEM_DOMAIN_VRAM : - AMDGPU_GEM_DOMAIN_GTT, + AMDGPU_GEM_DOMAIN_GTT | AMDGPU_GEM_DOMAIN_VRAM, &psp->tmr_bo, &psp->tmr_mc_addr, pptr); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c index 540817e296da..7fe5b1940df8 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c @@ -122,12 +122,15 @@ const char *get_ras_block_str(struct ras_common_if *ras_block) /* typical ECC bad page rate is 1 bad page per 100MB VRAM */ #define RAS_BAD_PAGE_COVER (100 * 1024 * 1024ULL) -#define MAX_UMC_POISON_POLLING_TIME_ASYNC 300 //ms +#define MAX_UMC_POISON_POLLING_TIME_ASYNC 10 #define AMDGPU_RAS_RETIRE_PAGE_INTERVAL 100 //ms #define MAX_FLUSH_RETIRE_DWORK_TIMES 100 +#define BYPASS_ALLOCATED_ADDRESS 0x0 +#define BYPASS_INITIALIZATION_ADDRESS 0x1 + enum amdgpu_ras_retire_page_reservation { AMDGPU_RAS_RETIRE_PAGE_RESERVED, AMDGPU_RAS_RETIRE_PAGE_PENDING, @@ -136,10 +139,14 @@ enum amdgpu_ras_retire_page_reservation { atomic_t amdgpu_ras_in_intr = ATOMIC_INIT(0); -static bool amdgpu_ras_check_bad_page_unlock(struct amdgpu_ras *con, +static int amdgpu_ras_check_bad_page_unlock(struct amdgpu_ras *con, uint64_t addr); -static bool amdgpu_ras_check_bad_page(struct amdgpu_device *adev, +static int amdgpu_ras_check_bad_page(struct amdgpu_device *adev, uint64_t addr); + +static void amdgpu_ras_critical_region_init(struct amdgpu_device *adev); +static void amdgpu_ras_critical_region_fini(struct amdgpu_device *adev); + #ifdef CONFIG_X86_MCE_AMD static void amdgpu_register_bad_pages_mca_notifier(struct amdgpu_device *adev); struct mce_notifier_adev_list { @@ -169,18 +176,16 @@ static int amdgpu_reserve_page_direct(struct amdgpu_device *adev, uint64_t addre struct eeprom_table_record err_rec; int ret; - if ((address >= adev->gmc.mc_vram_size) || - (address >= RAS_UMC_INJECT_ADDR_LIMIT)) { + ret = amdgpu_ras_check_bad_page(adev, address); + if (ret == -EINVAL) { dev_warn(adev->dev, - "RAS WARN: input address 0x%llx is invalid.\n", - address); + "RAS WARN: input address 0x%llx is invalid.\n", + address); return -EINVAL; - } - - if (amdgpu_ras_check_bad_page(adev, address)) { + } else if (ret == 1) { dev_warn(adev->dev, - "RAS WARN: 0x%llx has already been marked as bad page!\n", - address); + "RAS WARN: 0x%llx has already been marked as bad page!\n", + address); return 0; } @@ -207,6 +212,49 @@ static int amdgpu_reserve_page_direct(struct amdgpu_device *adev, uint64_t addre return 0; } +static int amdgpu_check_address_validity(struct amdgpu_device *adev, + uint64_t address, uint64_t flags) +{ + struct amdgpu_ras *con = amdgpu_ras_get_context(adev); + struct amdgpu_vram_block_info blk_info; + uint64_t page_pfns[32] = {0}; + int i, ret, count; + + if (amdgpu_ip_version(adev, UMC_HWIP, 0) < IP_VERSION(12, 0, 0)) + return 0; + + if ((address >= adev->gmc.mc_vram_size) || + (address >= RAS_UMC_INJECT_ADDR_LIMIT)) + return -EFAULT; + + count = amdgpu_umc_lookup_bad_pages_in_a_row(adev, + address, page_pfns, ARRAY_SIZE(page_pfns)); + if (count <= 0) + return -EPERM; + + for (i = 0; i < count; i++) { + memset(&blk_info, 0, sizeof(blk_info)); + ret = amdgpu_vram_mgr_query_address_block_info(&adev->mman.vram_mgr, + page_pfns[i] << AMDGPU_GPU_PAGE_SHIFT, &blk_info); + if (!ret) { + /* The input address that needs to be checked is allocated by + * current calling process, so it is necessary to exclude + * the calling process. + */ + if ((flags == BYPASS_ALLOCATED_ADDRESS) && + ((blk_info.task.pid != task_pid_nr(current)) || + strncmp(blk_info.task.comm, current->comm, TASK_COMM_LEN))) + return -EACCES; + else if ((flags == BYPASS_INITIALIZATION_ADDRESS) && + (blk_info.task.pid == con->init_task_pid) && + !strncmp(blk_info.task.comm, con->init_task_comm, TASK_COMM_LEN)) + return -EACCES; + } + } + + return 0; +} + static ssize_t amdgpu_ras_debugfs_read(struct file *f, char __user *buf, size_t size, loff_t *pos) { @@ -297,6 +345,8 @@ static int amdgpu_ras_debugfs_ctrl_parse_data(struct file *f, op = 2; else if (strstr(str, "retire_page") != NULL) op = 3; + else if (strstr(str, "check_address") != NULL) + op = 4; else if (str[0] && str[1] && str[2] && str[3]) /* ascii string, but commands are not matched. */ return -EINVAL; @@ -311,6 +361,15 @@ static int amdgpu_ras_debugfs_ctrl_parse_data(struct file *f, data->inject.address = address; return 0; + } else if (op == 4) { + if (sscanf(str, "%*s 0x%llx 0x%llx", &address, &value) != 2 && + sscanf(str, "%*s %llu %llu", &address, &value) != 2) + return -EINVAL; + + data->op = op; + data->inject.address = address; + data->inject.value = value; + return 0; } if (amdgpu_ras_find_block_id_by_name(block_name, &block_id)) @@ -500,6 +559,9 @@ static ssize_t amdgpu_ras_debugfs_ctrl_write(struct file *f, return size; else return ret; + } else if (data.op == 4) { + ret = amdgpu_check_address_validity(adev, data.inject.address, data.inject.value); + return ret ? ret : size; } if (!amdgpu_ras_is_supported(adev, data.head.block)) @@ -513,22 +575,16 @@ static ssize_t amdgpu_ras_debugfs_ctrl_write(struct file *f, ret = amdgpu_ras_feature_enable(adev, &data.head, 1); break; case 2: - if ((data.inject.address >= adev->gmc.mc_vram_size && - adev->gmc.mc_vram_size) || - (data.inject.address >= RAS_UMC_INJECT_ADDR_LIMIT)) { - dev_warn(adev->dev, "RAS WARN: input address " - "0x%llx is invalid.", + /* umc ce/ue error injection for a bad page is not allowed */ + if (data.head.block == AMDGPU_RAS_BLOCK__UMC) + ret = amdgpu_ras_check_bad_page(adev, data.inject.address); + if (ret == -EINVAL) { + dev_warn(adev->dev, "RAS WARN: input address 0x%llx is invalid.", data.inject.address); - ret = -EINVAL; break; - } - - /* umc ce/ue error injection for a bad page is not allowed */ - if ((data.head.block == AMDGPU_RAS_BLOCK__UMC) && - amdgpu_ras_check_bad_page(adev, data.inject.address)) { - dev_warn(adev->dev, "RAS WARN: inject: 0x%llx has " - "already been marked as bad!\n", - data.inject.address); + } else if (ret == 1) { + dev_warn(adev->dev, "RAS WARN: inject: 0x%llx has already been marked as bad!\n", + data.inject.address); break; } @@ -2566,18 +2622,26 @@ static int amdgpu_ras_badpages_read(struct amdgpu_device *adev, goto out; } - *bps = kmalloc(sizeof(struct ras_badpage) * data->count, GFP_KERNEL); + *bps = kmalloc_array(data->count, sizeof(struct ras_badpage), GFP_KERNEL); if (!*bps) { ret = -ENOMEM; goto out; } for (; i < data->count; i++) { + if (!data->bps[i].ts) + continue; + (*bps)[i] = (struct ras_badpage){ .bp = data->bps[i].retired_page, .size = AMDGPU_GPU_PAGE_SIZE, .flags = AMDGPU_RAS_RETIRE_PAGE_RESERVED, }; + + if (amdgpu_ras_check_critical_address(adev, + data->bps[i].retired_page << AMDGPU_GPU_PAGE_SHIFT)) + continue; + status = amdgpu_vram_mgr_query_page_status(&adev->mman.vram_mgr, data->bps[i].retired_page << AMDGPU_GPU_PAGE_SHIFT); if (status == -EBUSY) @@ -2586,7 +2650,7 @@ static int amdgpu_ras_badpages_read(struct amdgpu_device *adev, (*bps)[i].flags = AMDGPU_RAS_RETIRE_PAGE_FAULT; } - *count = data->count; + *count = con->bad_page_num; out: mutex_unlock(&con->recovery_lock); return ret; @@ -2722,7 +2786,7 @@ static int amdgpu_ras_realloc_eh_data_space(struct amdgpu_device *adev, unsigned int old_space = data->count + data->space_left; unsigned int new_space = old_space + pages; unsigned int align_space = ALIGN(new_space, 512); - void *bps = kmalloc(align_space * sizeof(*data->bps), GFP_KERNEL); + void *bps = kmalloc_array(align_space, sizeof(*data->bps), GFP_KERNEL); if (!bps) { return -ENOMEM; @@ -2814,8 +2878,11 @@ static int __amdgpu_ras_restore_bad_pages(struct amdgpu_device *adev, for (j = 0; j < count; j++) { if (amdgpu_ras_check_bad_page_unlock(con, - bps[j].retired_page << AMDGPU_GPU_PAGE_SHIFT)) + bps[j].retired_page << AMDGPU_GPU_PAGE_SHIFT)) { + data->count++; + data->space_left--; continue; + } if (!data->space_left && amdgpu_ras_realloc_eh_data_space(adev, data, 256)) { @@ -2828,6 +2895,7 @@ static int __amdgpu_ras_restore_bad_pages(struct amdgpu_device *adev, sizeof(struct eeprom_table_record)); data->count++; data->space_left--; + con->bad_page_num++; } return 0; @@ -2974,7 +3042,7 @@ int amdgpu_ras_add_bad_pages(struct amdgpu_device *adev, ret = __amdgpu_ras_convert_rec_array_from_rom(adev, &bps[i], &err_data, nps); if (ret) - control->ras_num_bad_pages -= adev->umc.retire_unit; + con->bad_page_num -= adev->umc.retire_unit; i += (adev->umc.retire_unit - 1); } else { break; @@ -2988,8 +3056,10 @@ int amdgpu_ras_add_bad_pages(struct amdgpu_device *adev, ret = __amdgpu_ras_convert_rec_from_rom(adev, &bps[i], &err_data, nps); if (ret) - control->ras_num_bad_pages -= adev->umc.retire_unit; + con->bad_page_num -= adev->umc.retire_unit; } + + con->eh_data->count_saved = con->eh_data->count; } else { ret = __amdgpu_ras_restore_bad_pages(adev, bps, pages); } @@ -3012,7 +3082,7 @@ int amdgpu_ras_save_bad_pages(struct amdgpu_device *adev, struct amdgpu_ras *con = amdgpu_ras_get_context(adev); struct ras_err_handler_data *data; struct amdgpu_ras_eeprom_control *control; - int save_count, unit_num, bad_page_num, i; + int save_count, unit_num, i; if (!con || !con->eh_data) { if (new_cnt) @@ -3033,27 +3103,26 @@ int amdgpu_ras_save_bad_pages(struct amdgpu_device *adev, mutex_lock(&con->recovery_lock); control = &con->eeprom_control; data = con->eh_data; - bad_page_num = control->ras_num_bad_pages; - save_count = data->count - bad_page_num; + unit_num = data->count / adev->umc.retire_unit - control->ras_num_recs; + save_count = con->bad_page_num - control->ras_num_bad_pages; mutex_unlock(&con->recovery_lock); - unit_num = save_count / adev->umc.retire_unit; if (new_cnt) *new_cnt = unit_num; /* only new entries are saved */ - if (save_count > 0) { + if (unit_num > 0) { /*old asics only save pa to eeprom like before*/ if (IP_VERSION_MAJ(amdgpu_ip_version(adev, UMC_HWIP, 0)) < 12) { if (amdgpu_ras_eeprom_append(control, - &data->bps[bad_page_num], save_count)) { + &data->bps[data->count_saved], unit_num)) { dev_err(adev->dev, "Failed to save EEPROM table data!"); return -EIO; } } else { for (i = 0; i < unit_num; i++) { if (amdgpu_ras_eeprom_append(control, - &data->bps[bad_page_num + + &data->bps[data->count_saved + i * adev->umc.retire_unit], 1)) { dev_err(adev->dev, "Failed to save EEPROM table data!"); return -EIO; @@ -3062,6 +3131,7 @@ int amdgpu_ras_save_bad_pages(struct amdgpu_device *adev, } dev_info(adev->dev, "Saved %d pages to EEPROM table.\n", save_count); + data->count_saved = data->count; } return 0; @@ -3116,17 +3186,17 @@ static int amdgpu_ras_load_bad_pages(struct amdgpu_device *adev) } } + ret = amdgpu_ras_add_bad_pages(adev, bps, control->ras_num_recs, true); + if (ret) + goto out; + ret = amdgpu_ras_eeprom_check(control); if (ret) goto out; /* HW not usable */ - if (amdgpu_ras_is_rma(adev)) { + if (amdgpu_ras_is_rma(adev)) ret = -EHWPOISON; - goto out; - } - - ret = amdgpu_ras_add_bad_pages(adev, bps, control->ras_num_recs, true); } out: @@ -3134,18 +3204,24 @@ out: return ret; } -static bool amdgpu_ras_check_bad_page_unlock(struct amdgpu_ras *con, +static int amdgpu_ras_check_bad_page_unlock(struct amdgpu_ras *con, uint64_t addr) { struct ras_err_handler_data *data = con->eh_data; + struct amdgpu_device *adev = con->adev; int i; + if ((addr >= adev->gmc.mc_vram_size && + adev->gmc.mc_vram_size) || + (addr >= RAS_UMC_INJECT_ADDR_LIMIT)) + return -EINVAL; + addr >>= AMDGPU_GPU_PAGE_SHIFT; for (i = 0; i < data->count; i++) if (addr == data->bps[i].retired_page) - return true; + return 1; - return false; + return 0; } /* @@ -3153,11 +3229,11 @@ static bool amdgpu_ras_check_bad_page_unlock(struct amdgpu_ras *con, * * Note: this check is only for umc block */ -static bool amdgpu_ras_check_bad_page(struct amdgpu_device *adev, +static int amdgpu_ras_check_bad_page(struct amdgpu_device *adev, uint64_t addr) { struct amdgpu_ras *con = amdgpu_ras_get_context(adev); - bool ret = false; + int ret = 0; if (!con || !con->eh_data) return ret; @@ -3241,7 +3317,7 @@ static void amdgpu_ras_ecc_log_init(struct ras_ecc_log_info *ecc_log) INIT_RADIX_TREE(&ecc_log->de_page_tree, GFP_KERNEL); ecc_log->de_queried_count = 0; - ecc_log->prev_de_queried_count = 0; + ecc_log->consumption_q_count = 0; } static void amdgpu_ras_ecc_log_fini(struct ras_ecc_log_info *ecc_log) @@ -3261,7 +3337,7 @@ static void amdgpu_ras_ecc_log_fini(struct ras_ecc_log_info *ecc_log) mutex_destroy(&ecc_log->lock); ecc_log->de_queried_count = 0; - ecc_log->prev_de_queried_count = 0; + ecc_log->consumption_q_count = 0; } static bool amdgpu_ras_schedule_retirement_dwork(struct amdgpu_ras *con, @@ -3287,7 +3363,6 @@ static void amdgpu_ras_do_page_retirement(struct work_struct *work) page_retirement_dwork.work); struct amdgpu_device *adev = con->adev; struct ras_err_data err_data; - unsigned long err_cnt; /* If gpu reset is ongoing, delay retiring the bad pages */ if (amdgpu_in_reset(adev) || amdgpu_ras_in_recovery(adev)) { @@ -3299,13 +3374,9 @@ static void amdgpu_ras_do_page_retirement(struct work_struct *work) amdgpu_ras_error_data_init(&err_data); amdgpu_umc_handle_bad_pages(adev, &err_data); - err_cnt = err_data.err_addr_cnt; amdgpu_ras_error_data_fini(&err_data); - if (err_cnt && amdgpu_ras_is_rma(adev)) - amdgpu_ras_reset_gpu(adev); - amdgpu_ras_schedule_retirement_dwork(con, AMDGPU_RAS_RETIRE_PAGE_INTERVAL); } @@ -3316,49 +3387,39 @@ static int amdgpu_ras_poison_creation_handler(struct amdgpu_device *adev, int ret = 0; struct ras_ecc_log_info *ecc_log; struct ras_query_if info; - uint32_t timeout = 0; + u32 timeout = MAX_UMC_POISON_POLLING_TIME_ASYNC; struct amdgpu_ras *ras = amdgpu_ras_get_context(adev); - uint64_t de_queried_count; - uint32_t new_detect_count, total_detect_count; - uint32_t need_query_count = poison_creation_count; + u64 de_queried_count; + u64 consumption_q_count; enum ras_event_type type = RAS_EVENT_TYPE_POISON_CREATION; memset(&info, 0, sizeof(info)); info.head.block = AMDGPU_RAS_BLOCK__UMC; ecc_log = &ras->umc_ecc_log; - total_detect_count = 0; + ecc_log->de_queried_count = 0; + ecc_log->consumption_q_count = 0; + do { ret = amdgpu_ras_query_error_status_with_event(adev, &info, type); if (ret) return ret; de_queried_count = ecc_log->de_queried_count; - if (de_queried_count > ecc_log->prev_de_queried_count) { - new_detect_count = de_queried_count - ecc_log->prev_de_queried_count; - ecc_log->prev_de_queried_count = de_queried_count; - timeout = 0; - } else { - new_detect_count = 0; - } + consumption_q_count = ecc_log->consumption_q_count; - if (new_detect_count) { - total_detect_count += new_detect_count; - } else { - if (!timeout && need_query_count) - timeout = MAX_UMC_POISON_POLLING_TIME_ASYNC; + if (de_queried_count && consumption_q_count) + break; - if (timeout) { - if (!--timeout) - break; - msleep(1); - } - } - } while (total_detect_count < need_query_count); + msleep(100); + } while (--timeout); - if (total_detect_count) + if (de_queried_count) schedule_delayed_work(&ras->page_retirement_dwork, 0); + if (amdgpu_ras_is_rma(adev) && atomic_cmpxchg(&ras->rma_in_recovery, 0, 1) == 0) + amdgpu_ras_reset_gpu(adev); + return 0; } @@ -3394,6 +3455,12 @@ static int amdgpu_ras_poison_consumption_handler(struct amdgpu_device *adev, reset_flags |= msg.reset; } + /* + * Try to ensure poison creation handler is completed first + * to set rma if bad page exceed threshold. + */ + flush_delayed_work(&con->page_retirement_dwork); + /* for RMA, amdgpu_ras_poison_creation_handler will trigger gpu reset */ if (reset_flags && !amdgpu_ras_is_rma(adev)) { if (reset_flags & AMDGPU_RAS_GPU_RESET_MODE1_RESET) @@ -3403,8 +3470,6 @@ static int amdgpu_ras_poison_consumption_handler(struct amdgpu_device *adev, else reset = reset_flags; - flush_delayed_work(&con->page_retirement_dwork); - con->gpu_reset_flags |= reset; amdgpu_ras_reset_gpu(adev); @@ -3434,6 +3499,7 @@ static int amdgpu_ras_page_retirement_thread(void *param) if (kthread_should_stop()) break; + mutex_lock(&con->poison_lock); gpu_reset = 0; do { @@ -3446,7 +3512,8 @@ static int amdgpu_ras_page_retirement_thread(void *param) atomic_sub(poison_creation_count, &con->poison_creation_count); atomic_sub(poison_creation_count, &con->page_retirement_req_cnt); } - } while (atomic_read(&con->poison_creation_count)); + } while (atomic_read(&con->poison_creation_count) && + !atomic_read(&con->poison_consumption_count)); if (ret != -EIO) { msg_count = kfifo_len(&con->poison_fifo); @@ -3463,6 +3530,7 @@ static int amdgpu_ras_page_retirement_thread(void *param) /* gpu mode-1 reset is ongoing or just completed ras mode-1 reset */ /* Clear poison creation request */ atomic_set(&con->poison_creation_count, 0); + atomic_set(&con->poison_consumption_count, 0); /* Clear poison fifo */ amdgpu_ras_clear_poison_fifo(adev); @@ -3487,9 +3555,12 @@ static int amdgpu_ras_page_retirement_thread(void *param) atomic_sub(msg_count, &con->page_retirement_req_cnt); } + atomic_set(&con->poison_consumption_count, 0); + /* Wake up work to save bad pages to eeprom */ schedule_delayed_work(&con->page_retirement_dwork, 0); } + mutex_unlock(&con->poison_lock); } return 0; @@ -3570,8 +3641,10 @@ int amdgpu_ras_recovery_init(struct amdgpu_device *adev, bool init_bp_info) } mutex_init(&con->recovery_lock); + mutex_init(&con->poison_lock); INIT_WORK(&con->recovery_work, amdgpu_ras_do_recovery); atomic_set(&con->in_recovery, 0); + atomic_set(&con->rma_in_recovery, 0); con->eeprom_control.bad_channel_bitmap = 0; max_eeprom_records_count = amdgpu_ras_eeprom_max_record_count(&con->eeprom_control); @@ -3589,6 +3662,7 @@ int amdgpu_ras_recovery_init(struct amdgpu_device *adev, bool init_bp_info) init_waitqueue_head(&con->page_retirement_wq); atomic_set(&con->page_retirement_req_cnt, 0); atomic_set(&con->poison_creation_count, 0); + atomic_set(&con->poison_consumption_count, 0); con->page_retirement_thread = kthread_run(amdgpu_ras_page_retirement_thread, adev, "umc_page_retirement"); if (IS_ERR(con->page_retirement_thread)) { @@ -3661,6 +3735,8 @@ static int amdgpu_ras_recovery_fini(struct amdgpu_device *adev) kfree(data); mutex_unlock(&con->recovery_lock); + amdgpu_ras_critical_region_init(adev); + return 0; } /* recovery end */ @@ -4087,6 +4163,12 @@ int amdgpu_ras_init(struct amdgpu_device *adev) goto release_con; } + con->init_task_pid = task_pid_nr(current); + get_task_comm(con->init_task_comm, current); + + mutex_init(&con->critical_region_lock); + INIT_LIST_HEAD(&con->critical_region_head); + dev_info(adev->dev, "RAS INFO: ras initialized successfully, " "hardware ability[%x] ras_mask[%x]\n", adev->ras_hw_enabled, adev->ras_enabled); @@ -4366,6 +4448,9 @@ int amdgpu_ras_fini(struct amdgpu_device *adev) if (!adev->ras_enabled || !con) return 0; + amdgpu_ras_critical_region_fini(adev); + mutex_destroy(&con->critical_region_lock); + list_for_each_entry_safe(ras_node, tmp, &adev->ras_list, node) { if (ras_node->ras_obj) { obj = ras_node->ras_obj; @@ -5274,6 +5359,9 @@ int amdgpu_ras_reserve_page(struct amdgpu_device *adev, uint64_t pfn) uint64_t start = pfn << AMDGPU_GPU_PAGE_SHIFT; int ret = 0; + if (amdgpu_ras_check_critical_address(adev, start)) + return 0; + mutex_lock(&con->page_rsv_lock); ret = amdgpu_vram_mgr_query_page_status(mgr, start); if (ret == -ENOENT) @@ -5310,3 +5398,80 @@ bool amdgpu_ras_is_rma(struct amdgpu_device *adev) return con->is_rma; } + +int amdgpu_ras_add_critical_region(struct amdgpu_device *adev, + struct amdgpu_bo *bo) +{ + struct amdgpu_ras *con = amdgpu_ras_get_context(adev); + struct amdgpu_vram_mgr_resource *vres; + struct ras_critical_region *region; + struct drm_buddy_block *block; + int ret = 0; + + if (!bo || !bo->tbo.resource) + return -EINVAL; + + vres = to_amdgpu_vram_mgr_resource(bo->tbo.resource); + + mutex_lock(&con->critical_region_lock); + + /* Check if the bo had been recorded */ + list_for_each_entry(region, &con->critical_region_head, node) + if (region->bo == bo) + goto out; + + /* Record new critical amdgpu bo */ + list_for_each_entry(block, &vres->blocks, link) { + region = kzalloc(sizeof(*region), GFP_KERNEL); + if (!region) { + ret = -ENOMEM; + goto out; + } + region->bo = bo; + region->start = amdgpu_vram_mgr_block_start(block); + region->size = amdgpu_vram_mgr_block_size(block); + list_add_tail(®ion->node, &con->critical_region_head); + } + +out: + mutex_unlock(&con->critical_region_lock); + + return ret; +} + +static void amdgpu_ras_critical_region_init(struct amdgpu_device *adev) +{ + amdgpu_ras_add_critical_region(adev, adev->mman.fw_reserved_memory); +} + +static void amdgpu_ras_critical_region_fini(struct amdgpu_device *adev) +{ + struct amdgpu_ras *con = amdgpu_ras_get_context(adev); + struct ras_critical_region *region, *tmp; + + mutex_lock(&con->critical_region_lock); + list_for_each_entry_safe(region, tmp, &con->critical_region_head, node) { + list_del(®ion->node); + kfree(region); + } + mutex_unlock(&con->critical_region_lock); +} + +bool amdgpu_ras_check_critical_address(struct amdgpu_device *adev, uint64_t addr) +{ + struct amdgpu_ras *con = amdgpu_ras_get_context(adev); + struct ras_critical_region *region; + bool ret = false; + + mutex_lock(&con->critical_region_lock); + list_for_each_entry(region, &con->critical_region_head, node) { + if ((region->start <= addr) && + (addr < (region->start + region->size))) { + ret = true; + break; + } + } + mutex_unlock(&con->critical_region_lock); + + return ret; +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h index 927d6bff734a..6cf0dfd38be8 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h @@ -492,8 +492,15 @@ struct ras_ecc_err { struct ras_ecc_log_info { struct mutex lock; struct radix_tree_root de_page_tree; - uint64_t de_queried_count; - uint64_t prev_de_queried_count; + uint64_t de_queried_count; + uint64_t consumption_q_count; +}; + +struct ras_critical_region { + struct list_head node; + struct amdgpu_bo *bo; + uint64_t start; + uint64_t size; }; struct amdgpu_ras { @@ -515,6 +522,7 @@ struct amdgpu_ras { /* gpu recovery */ struct work_struct recovery_work; atomic_t in_recovery; + atomic_t rma_in_recovery; struct amdgpu_device *adev; /* error handler data */ struct ras_err_handler_data *eh_data; @@ -557,6 +565,7 @@ struct amdgpu_ras { struct mutex page_retirement_lock; atomic_t page_retirement_req_cnt; atomic_t poison_creation_count; + atomic_t poison_consumption_count; struct mutex page_rsv_lock; DECLARE_KFIFO(poison_fifo, struct ras_poison_msg, 128); struct ras_ecc_log_info umc_ecc_log; @@ -570,6 +579,17 @@ struct amdgpu_ras { struct ras_event_manager *event_mgr; uint64_t reserved_pages_in_bytes; + + pid_t init_task_pid; + char init_task_comm[TASK_COMM_LEN]; + + int bad_page_num; + + struct list_head critical_region_head; + struct mutex critical_region_lock; + + /* Protect poison injection */ + struct mutex poison_lock; }; struct ras_fs_data { @@ -608,6 +628,7 @@ struct ras_err_handler_data { struct eeprom_table_record *bps; /* the count of entries */ int count; + int count_saved; /* the space can place new entries */ int space_left; }; @@ -973,6 +994,9 @@ int amdgpu_ras_mark_ras_event_caller(struct amdgpu_device *adev, enum ras_event_ int amdgpu_ras_reserve_page(struct amdgpu_device *adev, uint64_t pfn); +int amdgpu_ras_add_critical_region(struct amdgpu_device *adev, struct amdgpu_bo *bo); +bool amdgpu_ras_check_critical_address(struct amdgpu_device *adev, uint64_t addr); + int amdgpu_ras_put_poison_req(struct amdgpu_device *adev, enum amdgpu_ras_block block, uint16_t pasid, pasid_notify pasid_fn, void *data, uint32_t reset); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c index 9bda9ad13f88..3eb3fb55ccb0 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c @@ -743,8 +743,7 @@ amdgpu_ras_eeprom_append_table(struct amdgpu_ras_eeprom_control *control, else control->ras_num_mca_recs += num; - control->ras_num_bad_pages = control->ras_num_pa_recs + - control->ras_num_mca_recs * adev->umc.retire_unit; + control->ras_num_bad_pages = con->bad_page_num; Out: kfree(buf); return res; @@ -766,6 +765,10 @@ amdgpu_ras_eeprom_update_header(struct amdgpu_ras_eeprom_control *control) dev_warn(adev->dev, "Saved bad pages %d reaches threshold value %d\n", control->ras_num_bad_pages, ras->bad_page_cnt_threshold); + + if (adev->cper.enabled && amdgpu_cper_generate_bp_threshold_record(adev)) + dev_warn(adev->dev, "fail to generate bad page threshold cper records\n"); + if ((amdgpu_bad_page_threshold != -1) && (amdgpu_bad_page_threshold != -2)) { control->tbl_hdr.header = RAS_TABLE_HDR_BAD; @@ -774,9 +777,10 @@ amdgpu_ras_eeprom_update_header(struct amdgpu_ras_eeprom_control *control) control->tbl_rai.health_percent = 0; } ras->is_rma = true; - /* ignore the -ENOTSUPP return value */ - amdgpu_dpm_send_rma_reason(adev); } + + /* ignore the -ENOTSUPP return value */ + amdgpu_dpm_send_rma_reason(adev); } if (control->tbl_hdr.version >= RAS_TABLE_VER_V2_1) @@ -1457,8 +1461,7 @@ int amdgpu_ras_eeprom_check(struct amdgpu_ras_eeprom_control *control) if (!__get_eeprom_i2c_addr(adev, control)) return -EINVAL; - control->ras_num_bad_pages = control->ras_num_pa_recs + - control->ras_num_mca_recs * adev->umc.retire_unit; + control->ras_num_bad_pages = ras->bad_page_num; if (hdr->header == RAS_TABLE_HDR_VAL) { dev_dbg(adev->dev, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.h index 4d9b9701139b..2f92b3be40f5 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.h @@ -160,4 +160,16 @@ int amdgpu_reset_do_xgmi_reset_on_init( bool amdgpu_reset_in_recovery(struct amdgpu_device *adev); +static inline void amdgpu_reset_set_dpc_status(struct amdgpu_device *adev, + bool status) +{ + adev->pcie_reset_ctx.occurs_dpc = status; + adev->no_hw_access = status; +} + +static inline bool amdgpu_reset_in_dpc(struct amdgpu_device *adev) +{ + return adev->pcie_reset_ctx.occurs_dpc; +} + #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_seq64.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_seq64.c index d45ebfb642ca..a0b479d5fff1 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_seq64.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_seq64.c @@ -67,9 +67,9 @@ static inline u64 amdgpu_seq64_get_va_base(struct amdgpu_device *adev) int amdgpu_seq64_map(struct amdgpu_device *adev, struct amdgpu_vm *vm, struct amdgpu_bo_va **bo_va) { - u64 seq64_addr, va_flags; struct amdgpu_bo *bo; struct drm_exec exec; + u64 seq64_addr; int r; bo = adev->seq64.sbo; @@ -94,9 +94,9 @@ int amdgpu_seq64_map(struct amdgpu_device *adev, struct amdgpu_vm *vm, seq64_addr = amdgpu_seq64_get_va_base(adev) & AMDGPU_GMC_HOLE_MASK; - va_flags = amdgpu_gem_va_map_flags(adev, AMDGPU_VM_PAGE_READABLE | AMDGPU_VM_MTYPE_UC); - r = amdgpu_vm_bo_map(adev, *bo_va, seq64_addr, 0, AMDGPU_VA_RESERVED_SEQ64_SIZE, - va_flags); + r = amdgpu_vm_bo_map(adev, *bo_va, seq64_addr, 0, + AMDGPU_VA_RESERVED_SEQ64_SIZE, + AMDGPU_VM_PAGE_READABLE | AMDGPU_VM_MTYPE_UC); if (r) { DRM_ERROR("failed to do bo_map on userq sem, err=%d\n", r); amdgpu_vm_bo_del(adev, *bo_va); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c index c92b8794aa73..2e039fb778ea 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c @@ -252,6 +252,7 @@ int amdgpu_umc_pasid_poison_handler(struct amdgpu_device *adev, block, pasid, pasid_fn, data, reset); if (!ret) { atomic_inc(&con->page_retirement_req_cnt); + atomic_inc(&con->poison_consumption_count); wake_up(&con->page_retirement_wq); } } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c index c3ace8030530..f72de06a4ac8 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c @@ -471,6 +471,7 @@ amdgpu_userq_create(struct drm_file *filp, union drm_amdgpu_userq *args) if (index == (uint64_t)-EINVAL) { drm_file_err(uq_mgr->file, "Failed to get doorbell for queue\n"); kfree(queue); + r = -EINVAL; goto unlock; } @@ -567,7 +568,6 @@ int amdgpu_userq_ioctl(struct drm_device *dev, void *data, args->in.queue_size || args->in.rptr_va || args->in.wptr_va || - args->in.wptr_va || args->in.mqd || args->in.mqd_size) return -EINVAL; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c index d287c44ddbc1..595f0df17bcc 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c @@ -92,6 +92,7 @@ MODULE_FIRMWARE(FIRMWARE_VCN5_0_0); MODULE_FIRMWARE(FIRMWARE_VCN5_0_1); static void amdgpu_vcn_idle_work_handler(struct work_struct *work); +static void amdgpu_vcn_reg_dump_fini(struct amdgpu_device *adev); int amdgpu_vcn_early_init(struct amdgpu_device *adev, int i) { @@ -184,16 +185,16 @@ int amdgpu_vcn_sw_init(struct amdgpu_device *adev, int i) dec_ver = (le32_to_cpu(hdr->ucode_version) >> 24) & 0xf; vep = (le32_to_cpu(hdr->ucode_version) >> 28) & 0xf; dev_info(adev->dev, - "Found VCN firmware Version ENC: %u.%u DEC: %u VEP: %u Revision: %u\n", - enc_major, enc_minor, dec_ver, vep, fw_rev); + "[VCN instance %d] Found VCN firmware Version ENC: %u.%u DEC: %u VEP: %u Revision: %u\n", + i, enc_major, enc_minor, dec_ver, vep, fw_rev); } else { unsigned int version_major, version_minor, family_id; family_id = le32_to_cpu(hdr->ucode_version) & 0xff; version_major = (le32_to_cpu(hdr->ucode_version) >> 24) & 0xff; version_minor = (le32_to_cpu(hdr->ucode_version) >> 8) & 0xff; - dev_info(adev->dev, "Found VCN firmware Version: %u.%u Family ID: %u\n", - version_major, version_minor, family_id); + dev_info(adev->dev, "[VCN instance %d] Found VCN firmware Version: %u.%u Family ID: %u\n", + i, version_major, version_minor, family_id); } bo_size = AMDGPU_VCN_STACK_SIZE + AMDGPU_VCN_CONTEXT_SIZE; @@ -285,6 +286,10 @@ int amdgpu_vcn_sw_fini(struct amdgpu_device *adev, int i) amdgpu_ucode_release(&adev->vcn.inst[0].fw); adev->vcn.inst[i].fw = NULL; } + + if (adev->vcn.reg_list) + amdgpu_vcn_reg_dump_fini(adev); + mutex_destroy(&adev->vcn.inst[i].vcn_pg_lock); mutex_destroy(&adev->vcn.inst[i].vcn1_jpeg1_workaround); @@ -352,8 +357,6 @@ int amdgpu_vcn_suspend(struct amdgpu_device *adev, int i) if (adev->vcn.harvest_config & (1 << i)) return 0; - cancel_delayed_work_sync(&adev->vcn.inst[i].idle_work); - /* err_event_athub and dpc recovery will corrupt VCPU buffer, so we need to * restore fw data and clear buffer in amdgpu_vcn_resume() */ if (in_ras_intr || adev->pcie_reset_ctx.in_link_reset) @@ -405,6 +408,54 @@ int amdgpu_vcn_resume(struct amdgpu_device *adev, int i) return 0; } +void amdgpu_vcn_get_profile(struct amdgpu_device *adev) +{ + int r; + + mutex_lock(&adev->vcn.workload_profile_mutex); + + if (adev->vcn.workload_profile_active) { + mutex_unlock(&adev->vcn.workload_profile_mutex); + return; + } + r = amdgpu_dpm_switch_power_profile(adev, PP_SMC_POWER_PROFILE_VIDEO, + true); + if (r) + dev_warn(adev->dev, + "(%d) failed to enable video power profile mode\n", r); + else + adev->vcn.workload_profile_active = true; + mutex_unlock(&adev->vcn.workload_profile_mutex); +} + +void amdgpu_vcn_put_profile(struct amdgpu_device *adev) +{ + bool pg = true; + int r, i; + + mutex_lock(&adev->vcn.workload_profile_mutex); + for (i = 0; i < adev->vcn.num_vcn_inst; i++) { + if (adev->vcn.inst[i].cur_state != AMD_PG_STATE_GATE) { + pg = false; + break; + } + } + + if (pg) { + r = amdgpu_dpm_switch_power_profile( + adev, PP_SMC_POWER_PROFILE_VIDEO, false); + if (r) + dev_warn( + adev->dev, + "(%d) failed to disable video power profile mode\n", + r); + else + adev->vcn.workload_profile_active = false; + } + + mutex_unlock(&adev->vcn.workload_profile_mutex); +} + static void amdgpu_vcn_idle_work_handler(struct work_struct *work) { struct amdgpu_vcn_inst *vcn_inst = @@ -412,7 +463,6 @@ static void amdgpu_vcn_idle_work_handler(struct work_struct *work) struct amdgpu_device *adev = vcn_inst->adev; unsigned int fences = 0, fence[AMDGPU_MAX_VCN_INSTANCES] = {0}; unsigned int i = vcn_inst->inst, j; - int r = 0; if (adev->vcn.harvest_config & (1 << i)) return; @@ -438,16 +488,11 @@ static void amdgpu_vcn_idle_work_handler(struct work_struct *work) fences += fence[i]; if (!fences && !atomic_read(&vcn_inst->total_submission_cnt)) { + mutex_lock(&vcn_inst->vcn_pg_lock); vcn_inst->set_pg_state(vcn_inst, AMD_PG_STATE_GATE); - mutex_lock(&adev->vcn.workload_profile_mutex); - if (adev->vcn.workload_profile_active) { - r = amdgpu_dpm_switch_power_profile(adev, PP_SMC_POWER_PROFILE_VIDEO, - false); - if (r) - dev_warn(adev->dev, "(%d) failed to disable video power profile mode\n", r); - adev->vcn.workload_profile_active = false; - } - mutex_unlock(&adev->vcn.workload_profile_mutex); + mutex_unlock(&vcn_inst->vcn_pg_lock); + amdgpu_vcn_put_profile(adev); + } else { schedule_delayed_work(&vcn_inst->idle_work, VCN_IDLE_TIMEOUT); } @@ -457,30 +502,11 @@ void amdgpu_vcn_ring_begin_use(struct amdgpu_ring *ring) { struct amdgpu_device *adev = ring->adev; struct amdgpu_vcn_inst *vcn_inst = &adev->vcn.inst[ring->me]; - int r = 0; atomic_inc(&vcn_inst->total_submission_cnt); cancel_delayed_work_sync(&vcn_inst->idle_work); - /* We can safely return early here because we've cancelled the - * the delayed work so there is no one else to set it to false - * and we don't care if someone else sets it to true. - */ - if (adev->vcn.workload_profile_active) - goto pg_lock; - - mutex_lock(&adev->vcn.workload_profile_mutex); - if (!adev->vcn.workload_profile_active) { - r = amdgpu_dpm_switch_power_profile(adev, PP_SMC_POWER_PROFILE_VIDEO, - true); - if (r) - dev_warn(adev->dev, "(%d) failed to switch to video power profile mode\n", r); - adev->vcn.workload_profile_active = true; - } - mutex_unlock(&adev->vcn.workload_profile_mutex); - -pg_lock: mutex_lock(&vcn_inst->vcn_pg_lock); vcn_inst->set_pg_state(vcn_inst, AMD_PG_STATE_UNGATE); @@ -508,6 +534,7 @@ pg_lock: vcn_inst->pause_dpg_mode(vcn_inst, &new_state); } mutex_unlock(&vcn_inst->vcn_pg_lock); + amdgpu_vcn_get_profile(adev); } void amdgpu_vcn_ring_end_use(struct amdgpu_ring *ring) @@ -1527,3 +1554,86 @@ int amdgpu_vcn_ring_reset(struct amdgpu_ring *ring, return amdgpu_vcn_reset_engine(adev, ring->me); } + +int amdgpu_vcn_reg_dump_init(struct amdgpu_device *adev, + const struct amdgpu_hwip_reg_entry *reg, u32 count) +{ + adev->vcn.ip_dump = kcalloc(adev->vcn.num_vcn_inst * count, + sizeof(uint32_t), GFP_KERNEL); + if (!adev->vcn.ip_dump) + return -ENOMEM; + adev->vcn.reg_list = reg; + adev->vcn.reg_count = count; + + return 0; +} + +static void amdgpu_vcn_reg_dump_fini(struct amdgpu_device *adev) +{ + kfree(adev->vcn.ip_dump); + adev->vcn.ip_dump = NULL; + adev->vcn.reg_list = NULL; + adev->vcn.reg_count = 0; +} + +void amdgpu_vcn_dump_ip_state(struct amdgpu_ip_block *ip_block) +{ + struct amdgpu_device *adev = ip_block->adev; + int i, j; + bool is_powered; + u32 inst_off; + + if (!adev->vcn.ip_dump) + return; + + for (i = 0; i < adev->vcn.num_vcn_inst; i++) { + if (adev->vcn.harvest_config & (1 << i)) + continue; + + inst_off = i * adev->vcn.reg_count; + /* mmUVD_POWER_STATUS is always readable and is the first in reg_list */ + adev->vcn.ip_dump[inst_off] = + RREG32(SOC15_REG_ENTRY_OFFSET_INST(adev->vcn.reg_list[0], i)); + is_powered = (adev->vcn.ip_dump[inst_off] & + UVD_POWER_STATUS__UVD_POWER_STATUS_TILES_OFF) != + UVD_POWER_STATUS__UVD_POWER_STATUS_TILES_OFF; + + if (is_powered) + for (j = 1; j < adev->vcn.reg_count; j++) + adev->vcn.ip_dump[inst_off + j] = + RREG32(SOC15_REG_ENTRY_OFFSET_INST(adev->vcn.reg_list[j], i)); + } +} + +void amdgpu_vcn_print_ip_state(struct amdgpu_ip_block *ip_block, struct drm_printer *p) +{ + struct amdgpu_device *adev = ip_block->adev; + int i, j; + bool is_powered; + u32 inst_off; + + if (!adev->vcn.ip_dump) + return; + + drm_printf(p, "num_instances:%d\n", adev->vcn.num_vcn_inst); + for (i = 0; i < adev->vcn.num_vcn_inst; i++) { + if (adev->vcn.harvest_config & (1 << i)) { + drm_printf(p, "\nHarvested Instance:VCN%d Skipping dump\n", i); + continue; + } + + inst_off = i * adev->vcn.reg_count; + is_powered = (adev->vcn.ip_dump[inst_off] & + UVD_POWER_STATUS__UVD_POWER_STATUS_TILES_OFF) != + UVD_POWER_STATUS__UVD_POWER_STATUS_TILES_OFF; + + if (is_powered) { + drm_printf(p, "\nActive Instance:VCN%d\n", i); + for (j = 0; j < adev->vcn.reg_count; j++) + drm_printf(p, "%-50s \t 0x%08x\n", adev->vcn.reg_list[j].reg_name, + adev->vcn.ip_dump[inst_off + j]); + } else { + drm_printf(p, "\nInactive Instance:VCN%d\n", i); + } + } +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h index 0bc0a94d7cf0..6d9acd36041d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h @@ -237,6 +237,8 @@ #define AMDGPU_DRM_KEY_INJECT_WORKAROUND_VCNFW_ASD_HANDSHAKING 2 +struct amdgpu_hwip_reg_entry; + enum amdgpu_vcn_caps { AMDGPU_VCN_RRMT_ENABLED, }; @@ -362,6 +364,8 @@ struct amdgpu_vcn { bool workload_profile_active; struct mutex workload_profile_mutex; + u32 reg_count; + const struct amdgpu_hwip_reg_entry *reg_list; }; struct amdgpu_fw_shared_rb_ptrs_struct { @@ -557,4 +561,11 @@ int vcn_set_powergating_state(struct amdgpu_ip_block *ip_block, int amdgpu_vcn_ring_reset(struct amdgpu_ring *ring, unsigned int vmid, struct amdgpu_fence *guilty_fence); +int amdgpu_vcn_reg_dump_init(struct amdgpu_device *adev, + const struct amdgpu_hwip_reg_entry *reg, u32 count); +void amdgpu_vcn_dump_ip_state(struct amdgpu_ip_block *ip_block); +void amdgpu_vcn_print_ip_state(struct amdgpu_ip_block *ip_block, struct drm_printer *p); +void amdgpu_vcn_get_profile(struct amdgpu_device *adev); +void amdgpu_vcn_put_profile(struct amdgpu_device *adev); + #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h index 3da3ebb1d9a1..58accf2259b3 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h @@ -267,7 +267,8 @@ struct amdgpu_virt { struct amdgpu_irq_src rcv_irq; struct work_struct flr_work; - struct work_struct bad_pages_work; + struct work_struct req_bad_pages_work; + struct work_struct handle_bad_pages_work; struct amdgpu_mm_table mm_table; const struct amdgpu_virt_ops *ops; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c index 314f11e6c78b..cc74a9827ac9 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c @@ -1341,13 +1341,14 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev, struct amdgpu_bo_va *bo_va, /* normally,bo_va->flags only contians READABLE and WIRTEABLE bit go here * but in case of something, we filter the flags in first place */ - if (!(mapping->flags & AMDGPU_PTE_READABLE)) + if (!(mapping->flags & AMDGPU_VM_PAGE_READABLE)) update_flags &= ~AMDGPU_PTE_READABLE; - if (!(mapping->flags & AMDGPU_PTE_WRITEABLE)) + if (!(mapping->flags & AMDGPU_VM_PAGE_WRITEABLE)) update_flags &= ~AMDGPU_PTE_WRITEABLE; /* Apply ASIC specific mapping flags */ - amdgpu_gmc_get_vm_pte(adev, mapping, &update_flags); + amdgpu_gmc_get_vm_pte(adev, vm, bo, mapping->flags, + &update_flags); trace_amdgpu_vm_bo_update(mapping); @@ -1488,7 +1489,7 @@ static void amdgpu_vm_free_mapping(struct amdgpu_device *adev, struct amdgpu_bo_va_mapping *mapping, struct dma_fence *fence) { - if (mapping->flags & AMDGPU_PTE_PRT_FLAG(adev)) + if (mapping->flags & AMDGPU_VM_PAGE_PRT) amdgpu_vm_add_prt_cb(adev, fence); kfree(mapping); } @@ -1767,7 +1768,7 @@ static void amdgpu_vm_bo_insert_map(struct amdgpu_device *adev, list_add(&mapping->list, &bo_va->invalids); amdgpu_vm_it_insert(mapping, &vm->va); - if (mapping->flags & AMDGPU_PTE_PRT_FLAG(adev)) + if (mapping->flags & AMDGPU_VM_PAGE_PRT) amdgpu_vm_prt_get(adev); if (amdgpu_vm_is_bo_always_valid(vm, bo) && !bo_va->base.moved) @@ -1827,7 +1828,7 @@ static int amdgpu_vm_verify_parameters(struct amdgpu_device *adev, int amdgpu_vm_bo_map(struct amdgpu_device *adev, struct amdgpu_bo_va *bo_va, uint64_t saddr, uint64_t offset, - uint64_t size, uint64_t flags) + uint64_t size, uint32_t flags) { struct amdgpu_bo_va_mapping *mapping, *tmp; struct amdgpu_bo *bo = bo_va->base.bo; @@ -1886,7 +1887,7 @@ int amdgpu_vm_bo_map(struct amdgpu_device *adev, int amdgpu_vm_bo_replace_map(struct amdgpu_device *adev, struct amdgpu_bo_va *bo_va, uint64_t saddr, uint64_t offset, - uint64_t size, uint64_t flags) + uint64_t size, uint32_t flags) { struct amdgpu_bo_va_mapping *mapping; struct amdgpu_bo *bo = bo_va->base.bo; @@ -2743,7 +2744,7 @@ void amdgpu_vm_fini(struct amdgpu_device *adev, struct amdgpu_vm *vm) dma_fence_put(vm->last_tlb_flush); list_for_each_entry_safe(mapping, tmp, &vm->freed, list) { - if (mapping->flags & AMDGPU_PTE_PRT_FLAG(adev) && prt_fini_needed) { + if (mapping->flags & AMDGPU_VM_PAGE_PRT && prt_fini_needed) { amdgpu_vm_prt_fini(adev, vm); prt_fini_needed = false; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h index ff2c8c6f4d7c..6cc6ea4b681e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h @@ -538,11 +538,11 @@ struct amdgpu_bo_va *amdgpu_vm_bo_add(struct amdgpu_device *adev, int amdgpu_vm_bo_map(struct amdgpu_device *adev, struct amdgpu_bo_va *bo_va, uint64_t addr, uint64_t offset, - uint64_t size, uint64_t flags); + uint64_t size, uint32_t flags); int amdgpu_vm_bo_replace_map(struct amdgpu_device *adev, struct amdgpu_bo_va *bo_va, uint64_t addr, uint64_t offset, - uint64_t size, uint64_t flags); + uint64_t size, uint32_t flags); int amdgpu_vm_bo_unmap(struct amdgpu_device *adev, struct amdgpu_bo_va *bo_va, uint64_t addr); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vpe.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vpe.c index 121ee17b522b..474bfe36c0c2 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vpe.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vpe.c @@ -379,9 +379,10 @@ static int vpe_sw_init(struct amdgpu_ip_block *ip_block) if (ret) goto out; - /* TODO: Add queue reset mask when FW fully supports it */ adev->vpe.supported_reset = amdgpu_get_soft_full_reset_mask(&adev->vpe.ring); + if (!amdgpu_sriov_vf(adev)) + adev->vpe.supported_reset |= AMDGPU_RESET_TYPE_PER_QUEUE; ret = amdgpu_vpe_sysfs_reset_mask_init(adev); if (ret) goto out; @@ -435,6 +436,8 @@ static int vpe_hw_fini(struct amdgpu_ip_block *ip_block) struct amdgpu_device *adev = ip_block->adev; struct amdgpu_vpe *vpe = &adev->vpe; + cancel_delayed_work_sync(&adev->vpe.idle_work); + vpe_ring_stop(vpe); /* Power off VPE */ @@ -445,10 +448,6 @@ static int vpe_hw_fini(struct amdgpu_ip_block *ip_block) static int vpe_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = ip_block->adev; - - cancel_delayed_work_sync(&adev->vpe.idle_work); - return vpe_hw_fini(ip_block); } @@ -874,6 +873,27 @@ static void vpe_ring_end_use(struct amdgpu_ring *ring) schedule_delayed_work(&adev->vpe.idle_work, VPE_IDLE_TIMEOUT); } +static int vpe_ring_reset(struct amdgpu_ring *ring, + unsigned int vmid, + struct amdgpu_fence *timedout_fence) +{ + struct amdgpu_device *adev = ring->adev; + int r; + + amdgpu_ring_reset_helper_begin(ring, timedout_fence); + + r = amdgpu_device_ip_set_powergating_state(adev, AMD_IP_BLOCK_TYPE_VPE, + AMD_PG_STATE_GATE); + if (r) + return r; + r = amdgpu_device_ip_set_powergating_state(adev, AMD_IP_BLOCK_TYPE_VPE, + AMD_PG_STATE_UNGATE); + if (r) + return r; + + return amdgpu_ring_reset_helper_end(ring, timedout_fence); +} + static ssize_t amdgpu_get_vpe_reset_mask(struct device *dev, struct device_attribute *attr, char *buf) @@ -942,6 +962,7 @@ static const struct amdgpu_ring_funcs vpe_ring_funcs = { .preempt_ib = vpe_ring_preempt_ib, .begin_use = vpe_ring_begin_use, .end_use = vpe_ring_end_use, + .reset = vpe_ring_reset, }; static void vpe_set_ring_funcs(struct amdgpu_device *adev) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c index 78f9e86ccc09..e69db0a93378 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c @@ -396,6 +396,35 @@ out: return ret; } +int amdgpu_vram_mgr_query_address_block_info(struct amdgpu_vram_mgr *mgr, + uint64_t address, struct amdgpu_vram_block_info *info) +{ + struct amdgpu_vram_mgr_resource *vres; + struct drm_buddy_block *block; + u64 start, size; + int ret = -ENOENT; + + mutex_lock(&mgr->lock); + list_for_each_entry(vres, &mgr->allocated_vres_list, vres_node) { + list_for_each_entry(block, &vres->blocks, link) { + start = amdgpu_vram_mgr_block_start(block); + size = amdgpu_vram_mgr_block_size(block); + if ((start <= address) && (address < (start + size))) { + info->start = start; + info->size = size; + memcpy(&info->task, &vres->task, sizeof(vres->task)); + ret = 0; + goto out; + } + } + } + +out: + mutex_unlock(&mgr->lock); + + return ret; +} + static void amdgpu_dummy_vram_mgr_debug(struct ttm_resource_manager *man, struct drm_printer *printer) { @@ -568,6 +597,10 @@ static int amdgpu_vram_mgr_new(struct ttm_resource_manager *man, remaining_size -= size; } + vres->task.pid = task_pid_nr(current); + get_task_comm(vres->task.comm, current); + list_add_tail(&vres->vres_node, &mgr->allocated_vres_list); + if (bo->flags & AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS && adjust_dcc_size) { struct drm_buddy_block *dcc_block; unsigned long dcc_start; @@ -645,6 +678,10 @@ static void amdgpu_vram_mgr_del(struct ttm_resource_manager *man, uint64_t vis_usage = 0; mutex_lock(&mgr->lock); + + list_del(&vres->vres_node); + memset(&vres->task, 0, sizeof(vres->task)); + list_for_each_entry(block, &vres->blocks, link) vis_usage += amdgpu_vram_mgr_vis_size(adev, block); @@ -933,6 +970,7 @@ int amdgpu_vram_mgr_init(struct amdgpu_device *adev) mutex_init(&mgr->lock); INIT_LIST_HEAD(&mgr->reservations_pending); INIT_LIST_HEAD(&mgr->reserved_pages); + INIT_LIST_HEAD(&mgr->allocated_vres_list); mgr->default_page_size = PAGE_SIZE; if (!adev->gmc.is_app_apu) { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.h index 2c88d5fd87da..5f5fd9a911c2 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.h @@ -35,12 +35,26 @@ struct amdgpu_vram_mgr { struct list_head reserved_pages; atomic64_t vis_usage; u64 default_page_size; + struct list_head allocated_vres_list; +}; + +struct amdgpu_vres_task { + pid_t pid; + char comm[TASK_COMM_LEN]; +}; + +struct amdgpu_vram_block_info { + u64 start; + u64 size; + struct amdgpu_vres_task task; }; struct amdgpu_vram_mgr_resource { struct ttm_resource base; struct list_head blocks; unsigned long flags; + struct list_head vres_node; + struct amdgpu_vres_task task; }; static inline u64 amdgpu_vram_mgr_block_start(struct drm_buddy_block *block) @@ -72,4 +86,7 @@ static inline void amdgpu_vram_mgr_set_cleared(struct ttm_resource *res) ares->flags |= DRM_BUDDY_CLEARED; } +int amdgpu_vram_mgr_query_address_block_info(struct amdgpu_vram_mgr *mgr, + uint64_t address, struct amdgpu_vram_block_info *info); + #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c index c417f8689220..1083db8cea2e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c @@ -120,6 +120,25 @@ static void __amdgpu_xcp_add_block(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id, xcp->valid = true; } +static void __amdgpu_xcp_set_unique_id(struct amdgpu_xcp_mgr *xcp_mgr, + int xcp_id) +{ + struct amdgpu_xcp *xcp = &xcp_mgr->xcp[xcp_id]; + struct amdgpu_device *adev = xcp_mgr->adev; + uint32_t inst_mask; + uint64_t uid; + int i; + + if (!amdgpu_xcp_get_inst_details(xcp, AMDGPU_XCP_GFX, &inst_mask) && + inst_mask) { + i = GET_INST(GC, (ffs(inst_mask) - 1)); + uid = amdgpu_device_get_uid(xcp_mgr->adev->uid_info, + AMDGPU_UID_TYPE_XCD, i); + if (uid) + xcp->unique_id = uid; + } +} + int amdgpu_xcp_init(struct amdgpu_xcp_mgr *xcp_mgr, int num_xcps, int mode) { struct amdgpu_device *adev = xcp_mgr->adev; @@ -158,6 +177,7 @@ int amdgpu_xcp_init(struct amdgpu_xcp_mgr *xcp_mgr, int num_xcps, int mode) else xcp_mgr->xcp[i].mem_id = mem_id; } + __amdgpu_xcp_set_unique_id(xcp_mgr, i); } xcp_mgr->num_xcps = num_xcps; @@ -406,6 +426,7 @@ void amdgpu_xcp_dev_unplug(struct amdgpu_device *adev) p_ddev->primary->dev = adev->xcp_mgr->xcp[i].pdev; p_ddev->driver = adev->xcp_mgr->xcp[i].driver; p_ddev->vma_offset_manager = adev->xcp_mgr->xcp[i].vma_offset_manager; + amdgpu_xcp_drm_dev_free(p_ddev); } } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h index 70a0f8400b57..1928d9e224fc 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h @@ -112,6 +112,7 @@ struct amdgpu_xcp { struct amdgpu_sched gpu_sched[AMDGPU_HW_IP_NUM][AMDGPU_RING_PRIO_MAX]; struct amdgpu_xcp_mgr *xcp_mgr; struct kobject kobj; + uint64_t unique_id; }; struct amdgpu_xcp_mgr { diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c index bf7c22f81cda..ba73518f5cdf 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c @@ -1462,17 +1462,12 @@ static int dce_v10_0_audio_init(struct amdgpu_device *adev) static void dce_v10_0_audio_fini(struct amdgpu_device *adev) { - int i; - if (!amdgpu_audio) return; if (!adev->mode_info.audio.enabled) return; - for (i = 0; i < adev->mode_info.audio.num_pins; i++) - dce_v10_0_audio_enable(adev, &adev->mode_info.audio.pin[i], false); - adev->mode_info.audio.enabled = false; } diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c index 47e05783c4a0..b01d88d078fa 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c @@ -1511,17 +1511,12 @@ static int dce_v11_0_audio_init(struct amdgpu_device *adev) static void dce_v11_0_audio_fini(struct amdgpu_device *adev) { - int i; - if (!amdgpu_audio) return; if (!adev->mode_info.audio.enabled) return; - for (i = 0; i < adev->mode_info.audio.num_pins; i++) - dce_v11_0_audio_enable(adev, &adev->mode_info.audio.pin[i], false); - adev->mode_info.audio.enabled = false; } diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c index 276c025c4c03..81760a26f2ff 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c @@ -1451,17 +1451,12 @@ static int dce_v6_0_audio_init(struct amdgpu_device *adev) static void dce_v6_0_audio_fini(struct amdgpu_device *adev) { - int i; - if (!amdgpu_audio) return; if (!adev->mode_info.audio.enabled) return; - for (i = 0; i < adev->mode_info.audio.num_pins; i++) - dce_v6_0_audio_enable(adev, &adev->mode_info.audio.pin[i], false); - adev->mode_info.audio.enabled = false; } diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c index e62ccf9eb73d..19a265bd4d19 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c @@ -1443,17 +1443,12 @@ static int dce_v8_0_audio_init(struct amdgpu_device *adev) static void dce_v8_0_audio_fini(struct amdgpu_device *adev) { - int i; - if (!amdgpu_audio) return; if (!adev->mode_info.audio.enabled) return; - for (i = 0; i < adev->mode_info.audio.num_pins; i++) - dce_v8_0_audio_enable(adev, &adev->mode_info.audio.pin[i], false); - adev->mode_info.audio.enabled = false; } diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c index 7bd506f06eb1..264183ab24ec 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c @@ -7668,19 +7668,17 @@ static int gfx_v10_0_soft_reset(struct amdgpu_ip_block *ip_block) /* Disable MEC parsing/prefetching */ gfx_v10_0_cp_compute_enable(adev, false); - if (grbm_soft_reset) { - tmp = RREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET); - tmp |= grbm_soft_reset; - dev_info(adev->dev, "GRBM_SOFT_RESET=0x%08X\n", tmp); - WREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET, tmp); - tmp = RREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET); - - udelay(50); - - tmp &= ~grbm_soft_reset; - WREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET, tmp); - tmp = RREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET); - } + tmp = RREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET); + tmp |= grbm_soft_reset; + dev_info(adev->dev, "GRBM_SOFT_RESET=0x%08X\n", tmp); + WREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET, tmp); + tmp = RREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET); + + udelay(50); + + tmp &= ~grbm_soft_reset; + WREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET, tmp); + tmp = RREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET); /* Wait a little for things to settle down */ udelay(50); diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c index c01c241a1b06..c85de8c8f6f5 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c @@ -1612,9 +1612,9 @@ static int gfx_v11_0_sw_init(struct amdgpu_ip_block *ip_block) case IP_VERSION(11, 0, 2): case IP_VERSION(11, 0, 3): if (!adev->gfx.disable_uq && - adev->gfx.me_fw_version >= 2390 && - adev->gfx.pfp_fw_version >= 2530 && - adev->gfx.mec_fw_version >= 2600 && + adev->gfx.me_fw_version >= 2420 && + adev->gfx.pfp_fw_version >= 2580 && + adev->gfx.mec_fw_version >= 2650 && adev->mes.fw_version[0] >= 120) { adev->userq_funcs[AMDGPU_HW_IP_GFX] = &userq_mes_funcs; adev->userq_funcs[AMDGPU_HW_IP_COMPUTE] = &userq_mes_funcs; @@ -4129,6 +4129,8 @@ static int gfx_v11_0_gfx_mqd_init(struct amdgpu_device *adev, void *m, #endif if (prop->tmz_queue) tmp = REG_SET_FIELD(tmp, CP_GFX_HQD_CNTL, TMZ_MATCH, 1); + if (!prop->kernel_queue) + tmp = REG_SET_FIELD(tmp, CP_GFX_HQD_CNTL, RB_NON_PRIV, 1); mqd->cp_gfx_hqd_cntl = tmp; /* set up cp_doorbell_control */ @@ -4281,8 +4283,10 @@ static int gfx_v11_0_compute_mqd_init(struct amdgpu_device *adev, void *m, tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, UNORD_DISPATCH, 1); tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, TUNNEL_DISPATCH, prop->allow_tunneling); - tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, PRIV_STATE, 1); - tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, KMD_QUEUE, 1); + if (prop->kernel_queue) { + tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, PRIV_STATE, 1); + tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, KMD_QUEUE, 1); + } if (prop->tmz_queue) tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, TMZ, 1); mqd->cp_hqd_pq_control = tmp; diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c index 3e138527d534..fd44d5503e28 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c @@ -3026,6 +3026,8 @@ static int gfx_v12_0_gfx_mqd_init(struct amdgpu_device *adev, void *m, #endif if (prop->tmz_queue) tmp = REG_SET_FIELD(tmp, CP_GFX_HQD_CNTL, TMZ_MATCH, 1); + if (!prop->kernel_queue) + tmp = REG_SET_FIELD(tmp, CP_GFX_HQD_CNTL, RB_NON_PRIV, 1); mqd->cp_gfx_hqd_cntl = tmp; /* set up cp_doorbell_control */ @@ -3175,8 +3177,10 @@ static int gfx_v12_0_compute_mqd_init(struct amdgpu_device *adev, void *m, (order_base_2(AMDGPU_GPU_PAGE_SIZE / 4) - 1)); tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, UNORD_DISPATCH, 1); tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, TUNNEL_DISPATCH, 0); - tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, PRIV_STATE, 1); - tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, KMD_QUEUE, 1); + if (prop->kernel_queue) { + tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, PRIV_STATE, 1); + tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, KMD_QUEUE, 1); + } if (prop->tmz_queue) tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, TMZ, 1); mqd->cp_hqd_pq_control = tmp; diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c index 20b30f4b3c7d..a6ff9a137a83 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c @@ -2650,6 +2650,9 @@ static void gfx_v9_0_init_sq_config(struct amdgpu_device *adev) !READ_ONCE(adev->barrier_has_auto_waitcnt)); WREG32_SOC15(GC, 0, mmSQ_CONFIG, tmp); break; + case IP_VERSION(9, 4, 2): + gfx_v9_4_2_init_sq(adev); + break; default: break; } @@ -4172,19 +4175,17 @@ static int gfx_v9_0_soft_reset(struct amdgpu_ip_block *ip_block) /* Disable MEC parsing/prefetching */ gfx_v9_0_cp_compute_enable(adev, false); - if (grbm_soft_reset) { - tmp = RREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET); - tmp |= grbm_soft_reset; - dev_info(adev->dev, "GRBM_SOFT_RESET=0x%08X\n", tmp); - WREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET, tmp); - tmp = RREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET); + tmp = RREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET); + tmp |= grbm_soft_reset; + dev_info(adev->dev, "GRBM_SOFT_RESET=0x%08X\n", tmp); + WREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET, tmp); + tmp = RREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET); - udelay(50); + udelay(50); - tmp &= ~grbm_soft_reset; - WREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET, tmp); - tmp = RREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET); - } + tmp &= ~grbm_soft_reset; + WREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET, tmp); + tmp = RREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET); /* Wait a little for things to settle down */ udelay(50); diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c index c48cd47b531f..8058ea91ecaf 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c @@ -748,6 +748,18 @@ void gfx_v9_4_2_init_golden_registers(struct amdgpu_device *adev, } } +void gfx_v9_4_2_init_sq(struct amdgpu_device *adev) +{ + uint32_t data; + + if (adev->gfx.mec_fw_version >= 98) { + adev->gmc.xnack_flags |= AMDGPU_GMC_XNACK_FLAG_CHAIN; + data = RREG32_SOC15(GC, 0, regSQ_CONFIG1); + data = REG_SET_FIELD(data, SQ_CONFIG1, DISABLE_XNACK_CHECK_IN_RETRY_DISABLE, 1); + WREG32_SOC15(GC, 0, regSQ_CONFIG1, data); + } +} + void gfx_v9_4_2_debug_trap_config_init(struct amdgpu_device *adev, uint32_t first_vmid, uint32_t last_vmid) diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.h b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.h index 7584624b641c..a603724c1dfc 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.h +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.h @@ -28,6 +28,7 @@ void gfx_v9_4_2_debug_trap_config_init(struct amdgpu_device *adev, uint32_t first_vmid, uint32_t last_vmid); void gfx_v9_4_2_init_golden_registers(struct amdgpu_device *adev, uint32_t die_id); +void gfx_v9_4_2_init_sq(struct amdgpu_device *adev); void gfx_v9_4_2_set_power_brake_sequence(struct amdgpu_device *adev); int gfx_v9_4_2_do_edc_gpr_workarounds(struct amdgpu_device *adev); diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c index 51babf5c78c8..8ba66d4dfe86 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c @@ -2461,19 +2461,17 @@ static int gfx_v9_4_3_soft_reset(struct amdgpu_ip_block *ip_block) /* Disable MEC parsing/prefetching */ gfx_v9_4_3_xcc_cp_compute_enable(adev, false, 0); - if (grbm_soft_reset) { - tmp = RREG32_SOC15(GC, GET_INST(GC, 0), regGRBM_SOFT_RESET); - tmp |= grbm_soft_reset; - dev_info(adev->dev, "GRBM_SOFT_RESET=0x%08X\n", tmp); - WREG32_SOC15(GC, GET_INST(GC, 0), regGRBM_SOFT_RESET, tmp); - tmp = RREG32_SOC15(GC, GET_INST(GC, 0), regGRBM_SOFT_RESET); - - udelay(50); - - tmp &= ~grbm_soft_reset; - WREG32_SOC15(GC, GET_INST(GC, 0), regGRBM_SOFT_RESET, tmp); - tmp = RREG32_SOC15(GC, GET_INST(GC, 0), regGRBM_SOFT_RESET); - } + tmp = RREG32_SOC15(GC, GET_INST(GC, 0), regGRBM_SOFT_RESET); + tmp |= grbm_soft_reset; + dev_info(adev->dev, "GRBM_SOFT_RESET=0x%08X\n", tmp); + WREG32_SOC15(GC, GET_INST(GC, 0), regGRBM_SOFT_RESET, tmp); + tmp = RREG32_SOC15(GC, GET_INST(GC, 0), regGRBM_SOFT_RESET); + + udelay(50); + + tmp &= ~grbm_soft_reset; + WREG32_SOC15(GC, GET_INST(GC, 0), regGRBM_SOFT_RESET, tmp); + tmp = RREG32_SOC15(GC, GET_INST(GC, 0), regGRBM_SOFT_RESET); /* Wait a little for things to settle down */ udelay(50); diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c index 7923f491cf73..7031dd8c3c5e 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c @@ -466,24 +466,6 @@ static void gmc_v10_0_emit_pasid_mapping(struct amdgpu_ring *ring, unsigned int * 0 valid */ -static uint64_t gmc_v10_0_map_mtype(struct amdgpu_device *adev, uint32_t flags) -{ - switch (flags) { - case AMDGPU_VM_MTYPE_DEFAULT: - return AMDGPU_PTE_MTYPE_NV10(0ULL, MTYPE_NC); - case AMDGPU_VM_MTYPE_NC: - return AMDGPU_PTE_MTYPE_NV10(0ULL, MTYPE_NC); - case AMDGPU_VM_MTYPE_WC: - return AMDGPU_PTE_MTYPE_NV10(0ULL, MTYPE_WC); - case AMDGPU_VM_MTYPE_CC: - return AMDGPU_PTE_MTYPE_NV10(0ULL, MTYPE_CC); - case AMDGPU_VM_MTYPE_UC: - return AMDGPU_PTE_MTYPE_NV10(0ULL, MTYPE_UC); - default: - return AMDGPU_PTE_MTYPE_NV10(0ULL, MTYPE_NC); - } -} - static void gmc_v10_0_get_vm_pde(struct amdgpu_device *adev, int level, uint64_t *addr, uint64_t *flags) { @@ -508,21 +490,39 @@ static void gmc_v10_0_get_vm_pde(struct amdgpu_device *adev, int level, } static void gmc_v10_0_get_vm_pte(struct amdgpu_device *adev, - struct amdgpu_bo_va_mapping *mapping, + struct amdgpu_vm *vm, + struct amdgpu_bo *bo, + uint32_t vm_flags, uint64_t *flags) { - struct amdgpu_bo *bo = mapping->bo_va->base.bo; - - *flags &= ~AMDGPU_PTE_EXECUTABLE; - *flags |= mapping->flags & AMDGPU_PTE_EXECUTABLE; + if (vm_flags & AMDGPU_VM_PAGE_EXECUTABLE) + *flags |= AMDGPU_PTE_EXECUTABLE; + else + *flags &= ~AMDGPU_PTE_EXECUTABLE; - *flags &= ~AMDGPU_PTE_MTYPE_NV10_MASK; - *flags |= (mapping->flags & AMDGPU_PTE_MTYPE_NV10_MASK); + switch (vm_flags & AMDGPU_VM_MTYPE_MASK) { + case AMDGPU_VM_MTYPE_DEFAULT: + case AMDGPU_VM_MTYPE_NC: + default: + *flags = AMDGPU_PTE_MTYPE_NV10(*flags, MTYPE_NC); + break; + case AMDGPU_VM_MTYPE_WC: + *flags = AMDGPU_PTE_MTYPE_NV10(*flags, MTYPE_WC); + break; + case AMDGPU_VM_MTYPE_CC: + *flags = AMDGPU_PTE_MTYPE_NV10(*flags, MTYPE_CC); + break; + case AMDGPU_VM_MTYPE_UC: + *flags = AMDGPU_PTE_MTYPE_NV10(*flags, MTYPE_UC); + break; + } - *flags &= ~AMDGPU_PTE_NOALLOC; - *flags |= (mapping->flags & AMDGPU_PTE_NOALLOC); + if (vm_flags & AMDGPU_VM_PAGE_NOALLOC) + *flags |= AMDGPU_PTE_NOALLOC; + else + *flags &= ~AMDGPU_PTE_NOALLOC; - if (mapping->flags & AMDGPU_PTE_PRT) { + if (vm_flags & AMDGPU_VM_PAGE_PRT) { *flags |= AMDGPU_PTE_PRT; *flags |= AMDGPU_PTE_SNOOPED; *flags |= AMDGPU_PTE_LOG; @@ -563,7 +563,6 @@ static const struct amdgpu_gmc_funcs gmc_v10_0_gmc_funcs = { .flush_gpu_tlb_pasid = gmc_v10_0_flush_gpu_tlb_pasid, .emit_flush_gpu_tlb = gmc_v10_0_emit_flush_gpu_tlb, .emit_pasid_mapping = gmc_v10_0_emit_pasid_mapping, - .map_mtype = gmc_v10_0_map_mtype, .get_vm_pde = gmc_v10_0_get_vm_pde, .get_vm_pte = gmc_v10_0_get_vm_pte, .get_vbios_fb_size = gmc_v10_0_get_vbios_fb_size, diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c index f15d691e9a20..93d2b0bbe641 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c @@ -430,24 +430,6 @@ static void gmc_v11_0_emit_pasid_mapping(struct amdgpu_ring *ring, unsigned int * 0 valid */ -static uint64_t gmc_v11_0_map_mtype(struct amdgpu_device *adev, uint32_t flags) -{ - switch (flags) { - case AMDGPU_VM_MTYPE_DEFAULT: - return AMDGPU_PTE_MTYPE_NV10(0ULL, MTYPE_NC); - case AMDGPU_VM_MTYPE_NC: - return AMDGPU_PTE_MTYPE_NV10(0ULL, MTYPE_NC); - case AMDGPU_VM_MTYPE_WC: - return AMDGPU_PTE_MTYPE_NV10(0ULL, MTYPE_WC); - case AMDGPU_VM_MTYPE_CC: - return AMDGPU_PTE_MTYPE_NV10(0ULL, MTYPE_CC); - case AMDGPU_VM_MTYPE_UC: - return AMDGPU_PTE_MTYPE_NV10(0ULL, MTYPE_UC); - default: - return AMDGPU_PTE_MTYPE_NV10(0ULL, MTYPE_NC); - } -} - static void gmc_v11_0_get_vm_pde(struct amdgpu_device *adev, int level, uint64_t *addr, uint64_t *flags) { @@ -472,21 +454,39 @@ static void gmc_v11_0_get_vm_pde(struct amdgpu_device *adev, int level, } static void gmc_v11_0_get_vm_pte(struct amdgpu_device *adev, - struct amdgpu_bo_va_mapping *mapping, + struct amdgpu_vm *vm, + struct amdgpu_bo *bo, + uint32_t vm_flags, uint64_t *flags) { - struct amdgpu_bo *bo = mapping->bo_va->base.bo; - - *flags &= ~AMDGPU_PTE_EXECUTABLE; - *flags |= mapping->flags & AMDGPU_PTE_EXECUTABLE; + if (vm_flags & AMDGPU_VM_PAGE_EXECUTABLE) + *flags |= AMDGPU_PTE_EXECUTABLE; + else + *flags &= ~AMDGPU_PTE_EXECUTABLE; - *flags &= ~AMDGPU_PTE_MTYPE_NV10_MASK; - *flags |= (mapping->flags & AMDGPU_PTE_MTYPE_NV10_MASK); + switch (vm_flags & AMDGPU_VM_MTYPE_MASK) { + case AMDGPU_VM_MTYPE_DEFAULT: + case AMDGPU_VM_MTYPE_NC: + default: + *flags = AMDGPU_PTE_MTYPE_NV10(*flags, MTYPE_NC); + break; + case AMDGPU_VM_MTYPE_WC: + *flags = AMDGPU_PTE_MTYPE_NV10(*flags, MTYPE_WC); + break; + case AMDGPU_VM_MTYPE_CC: + *flags = AMDGPU_PTE_MTYPE_NV10(*flags, MTYPE_CC); + break; + case AMDGPU_VM_MTYPE_UC: + *flags = AMDGPU_PTE_MTYPE_NV10(*flags, MTYPE_UC); + break; + } - *flags &= ~AMDGPU_PTE_NOALLOC; - *flags |= (mapping->flags & AMDGPU_PTE_NOALLOC); + if (vm_flags & AMDGPU_VM_PAGE_NOALLOC) + *flags |= AMDGPU_PTE_NOALLOC; + else + *flags &= ~AMDGPU_PTE_NOALLOC; - if (mapping->flags & AMDGPU_PTE_PRT) { + if (vm_flags & AMDGPU_VM_PAGE_PRT) { *flags |= AMDGPU_PTE_PRT; *flags |= AMDGPU_PTE_SNOOPED; *flags |= AMDGPU_PTE_LOG; @@ -527,7 +527,6 @@ static const struct amdgpu_gmc_funcs gmc_v11_0_gmc_funcs = { .flush_gpu_tlb_pasid = gmc_v11_0_flush_gpu_tlb_pasid, .emit_flush_gpu_tlb = gmc_v11_0_emit_flush_gpu_tlb, .emit_pasid_mapping = gmc_v11_0_emit_pasid_mapping, - .map_mtype = gmc_v11_0_map_mtype, .get_vm_pde = gmc_v11_0_get_vm_pde, .get_vm_pte = gmc_v11_0_get_vm_pte, .get_vbios_fb_size = gmc_v11_0_get_vbios_fb_size, diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c index de763105fdfd..9ba055ddc00f 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c @@ -336,6 +336,22 @@ static void gmc_v12_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev, uint16_t queried; int vmid, i; + if (adev->enable_uni_mes && adev->mes.ring[AMDGPU_MES_SCHED_PIPE].sched.ready && + (adev->mes.sched_version & AMDGPU_MES_VERSION_MASK) >= 0x81) { + struct mes_inv_tlbs_pasid_input input = {0}; + input.pasid = pasid; + input.flush_type = flush_type; + input.hub_id = AMDGPU_GFXHUB(0); + /* MES will invalidate all gc_hub for the device from master */ + adev->mes.funcs->invalidate_tlbs_pasid(&adev->mes, &input); + if (all_hub) { + /* Only need to invalidate mm_hub now, gfx12 only support one mmhub */ + input.hub_id = AMDGPU_MMHUB0(0); + adev->mes.funcs->invalidate_tlbs_pasid(&adev->mes, &input); + } + return; + } + for (vmid = 1; vmid < 16; vmid++) { bool valid; @@ -453,20 +469,6 @@ static void gmc_v12_0_emit_pasid_mapping(struct amdgpu_ring *ring, unsigned vmid * 0 valid */ -static uint64_t gmc_v12_0_map_mtype(struct amdgpu_device *adev, uint32_t flags) -{ - switch (flags) { - case AMDGPU_VM_MTYPE_DEFAULT: - return AMDGPU_PTE_MTYPE_GFX12(0ULL, MTYPE_NC); - case AMDGPU_VM_MTYPE_NC: - return AMDGPU_PTE_MTYPE_GFX12(0ULL, MTYPE_NC); - case AMDGPU_VM_MTYPE_UC: - return AMDGPU_PTE_MTYPE_GFX12(0ULL, MTYPE_UC); - default: - return AMDGPU_PTE_MTYPE_GFX12(0ULL, MTYPE_NC); - } -} - static void gmc_v12_0_get_vm_pde(struct amdgpu_device *adev, int level, uint64_t *addr, uint64_t *flags) { @@ -490,19 +492,35 @@ static void gmc_v12_0_get_vm_pde(struct amdgpu_device *adev, int level, } static void gmc_v12_0_get_vm_pte(struct amdgpu_device *adev, - struct amdgpu_bo_va_mapping *mapping, + struct amdgpu_vm *vm, + struct amdgpu_bo *bo, + uint32_t vm_flags, uint64_t *flags) { - struct amdgpu_bo *bo = mapping->bo_va->base.bo; + if (vm_flags & AMDGPU_VM_PAGE_EXECUTABLE) + *flags |= AMDGPU_PTE_EXECUTABLE; + else + *flags &= ~AMDGPU_PTE_EXECUTABLE; - *flags &= ~AMDGPU_PTE_EXECUTABLE; - *flags |= mapping->flags & AMDGPU_PTE_EXECUTABLE; + switch (vm_flags & AMDGPU_VM_MTYPE_MASK) { + case AMDGPU_VM_MTYPE_DEFAULT: + *flags = AMDGPU_PTE_MTYPE_GFX12(*flags, MTYPE_NC); + break; + case AMDGPU_VM_MTYPE_NC: + default: + *flags = AMDGPU_PTE_MTYPE_GFX12(*flags, MTYPE_NC); + break; + case AMDGPU_VM_MTYPE_UC: + *flags = AMDGPU_PTE_MTYPE_GFX12(*flags, MTYPE_UC); + break; + } - *flags &= ~AMDGPU_PTE_MTYPE_GFX12_MASK; - *flags |= (mapping->flags & AMDGPU_PTE_MTYPE_GFX12_MASK); + if (vm_flags & AMDGPU_VM_PAGE_NOALLOC) + *flags |= AMDGPU_PTE_NOALLOC; + else + *flags &= ~AMDGPU_PTE_NOALLOC; - if (mapping->flags & AMDGPU_PTE_PRT_GFX12) { - *flags |= AMDGPU_PTE_PRT_GFX12; + if (vm_flags & AMDGPU_VM_PAGE_PRT) { *flags |= AMDGPU_PTE_SNOOPED; *flags |= AMDGPU_PTE_SYSTEM; *flags |= AMDGPU_PTE_IS_PTE; @@ -543,7 +561,6 @@ static const struct amdgpu_gmc_funcs gmc_v12_0_gmc_funcs = { .flush_gpu_tlb_pasid = gmc_v12_0_flush_gpu_tlb_pasid, .emit_flush_gpu_tlb = gmc_v12_0_emit_flush_gpu_tlb, .emit_pasid_mapping = gmc_v12_0_emit_pasid_mapping, - .map_mtype = gmc_v12_0_map_mtype, .get_vm_pde = gmc_v12_0_get_vm_pde, .get_vm_pte = gmc_v12_0_get_vm_pte, .get_vbios_fb_size = gmc_v12_0_get_vbios_fb_size, diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c index 8030fcd64210..f6ad7911f1e6 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c @@ -382,7 +382,9 @@ static void gmc_v6_0_get_vm_pde(struct amdgpu_device *adev, int level, } static void gmc_v6_0_get_vm_pte(struct amdgpu_device *adev, - struct amdgpu_bo_va_mapping *mapping, + struct amdgpu_vm *vm, + struct amdgpu_bo *bo, + uint32_t vm_flags, uint64_t *flags) { *flags &= ~AMDGPU_PTE_EXECUTABLE; diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c index a8d5795084fc..93d7ccb7d013 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c @@ -504,7 +504,9 @@ static void gmc_v7_0_get_vm_pde(struct amdgpu_device *adev, int level, } static void gmc_v7_0_get_vm_pte(struct amdgpu_device *adev, - struct amdgpu_bo_va_mapping *mapping, + struct amdgpu_vm *vm, + struct amdgpu_bo *bo, + uint32_t vm_flags, uint64_t *flags) { *flags &= ~AMDGPU_PTE_EXECUTABLE; diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c index b45fa0cea9d2..c5e2a2c41e06 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c @@ -716,11 +716,15 @@ static void gmc_v8_0_get_vm_pde(struct amdgpu_device *adev, int level, } static void gmc_v8_0_get_vm_pte(struct amdgpu_device *adev, - struct amdgpu_bo_va_mapping *mapping, + struct amdgpu_vm *vm, + struct amdgpu_bo *bo, + uint32_t vm_flags, uint64_t *flags) { - *flags &= ~AMDGPU_PTE_EXECUTABLE; - *flags |= mapping->flags & AMDGPU_PTE_EXECUTABLE; + if (vm_flags & AMDGPU_VM_PAGE_EXECUTABLE) + *flags |= AMDGPU_PTE_EXECUTABLE; + else + *flags &= ~AMDGPU_PTE_EXECUTABLE; *flags &= ~AMDGPU_PTE_PRT; } diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c index c4d69cf4e06c..8404695eb13f 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c @@ -1073,27 +1073,6 @@ static void gmc_v9_0_emit_pasid_mapping(struct amdgpu_ring *ring, unsigned int v * 0 valid */ -static uint64_t gmc_v9_0_map_mtype(struct amdgpu_device *adev, uint32_t flags) - -{ - switch (flags) { - case AMDGPU_VM_MTYPE_DEFAULT: - return AMDGPU_PTE_MTYPE_VG10(0ULL, MTYPE_NC); - case AMDGPU_VM_MTYPE_NC: - return AMDGPU_PTE_MTYPE_VG10(0ULL, MTYPE_NC); - case AMDGPU_VM_MTYPE_WC: - return AMDGPU_PTE_MTYPE_VG10(0ULL, MTYPE_WC); - case AMDGPU_VM_MTYPE_RW: - return AMDGPU_PTE_MTYPE_VG10(0ULL, MTYPE_RW); - case AMDGPU_VM_MTYPE_CC: - return AMDGPU_PTE_MTYPE_VG10(0ULL, MTYPE_CC); - case AMDGPU_VM_MTYPE_UC: - return AMDGPU_PTE_MTYPE_VG10(0ULL, MTYPE_UC); - default: - return AMDGPU_PTE_MTYPE_VG10(0ULL, MTYPE_NC); - } -} - static void gmc_v9_0_get_vm_pde(struct amdgpu_device *adev, int level, uint64_t *addr, uint64_t *flags) { @@ -1123,6 +1102,7 @@ static void gmc_v9_0_get_vm_pde(struct amdgpu_device *adev, int level, static void gmc_v9_0_get_coherence_flags(struct amdgpu_device *adev, struct amdgpu_vm *vm, struct amdgpu_bo *bo, + uint32_t vm_flags, uint64_t *flags) { struct amdgpu_device *bo_adev = amdgpu_ttm_adev(bo->tbo.bdev); @@ -1236,25 +1216,43 @@ static void gmc_v9_0_get_coherence_flags(struct amdgpu_device *adev, } static void gmc_v9_0_get_vm_pte(struct amdgpu_device *adev, - struct amdgpu_bo_va_mapping *mapping, + struct amdgpu_vm *vm, + struct amdgpu_bo *bo, + uint32_t vm_flags, uint64_t *flags) { - struct amdgpu_bo *bo = mapping->bo_va->base.bo; - - *flags &= ~AMDGPU_PTE_EXECUTABLE; - *flags |= mapping->flags & AMDGPU_PTE_EXECUTABLE; + if (vm_flags & AMDGPU_VM_PAGE_EXECUTABLE) + *flags |= AMDGPU_PTE_EXECUTABLE; + else + *flags &= ~AMDGPU_PTE_EXECUTABLE; - *flags &= ~AMDGPU_PTE_MTYPE_VG10_MASK; - *flags |= mapping->flags & AMDGPU_PTE_MTYPE_VG10_MASK; + switch (vm_flags & AMDGPU_VM_MTYPE_MASK) { + case AMDGPU_VM_MTYPE_DEFAULT: + case AMDGPU_VM_MTYPE_NC: + default: + *flags = AMDGPU_PTE_MTYPE_VG10(*flags, MTYPE_NC); + break; + case AMDGPU_VM_MTYPE_WC: + *flags |= AMDGPU_PTE_MTYPE_VG10(*flags, MTYPE_WC); + break; + case AMDGPU_VM_MTYPE_RW: + *flags |= AMDGPU_PTE_MTYPE_VG10(*flags, MTYPE_RW); + break; + case AMDGPU_VM_MTYPE_CC: + *flags |= AMDGPU_PTE_MTYPE_VG10(*flags, MTYPE_CC); + break; + case AMDGPU_VM_MTYPE_UC: + *flags |= AMDGPU_PTE_MTYPE_VG10(*flags, MTYPE_UC); + break; + } - if (mapping->flags & AMDGPU_PTE_PRT) { + if (vm_flags & AMDGPU_VM_PAGE_PRT) { *flags |= AMDGPU_PTE_PRT; *flags &= ~AMDGPU_PTE_VALID; } if ((*flags & AMDGPU_PTE_VALID) && bo) - gmc_v9_0_get_coherence_flags(adev, mapping->bo_va->base.vm, bo, - flags); + gmc_v9_0_get_coherence_flags(adev, vm, bo, vm_flags, flags); } static void gmc_v9_0_override_vm_pte_flags(struct amdgpu_device *adev, @@ -1391,7 +1389,6 @@ static const struct amdgpu_gmc_funcs gmc_v9_0_gmc_funcs = { .flush_gpu_tlb_pasid = gmc_v9_0_flush_gpu_tlb_pasid, .emit_flush_gpu_tlb = gmc_v9_0_emit_flush_gpu_tlb, .emit_pasid_mapping = gmc_v9_0_emit_pasid_mapping, - .map_mtype = gmc_v9_0_map_mtype, .get_vm_pde = gmc_v9_0_get_vm_pde, .get_vm_pte = gmc_v9_0_get_vm_pte, .override_vm_pte_flags = gmc_v9_0_override_vm_pte_flags, diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c index b86288a69e7b..a78144773fab 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c @@ -444,7 +444,7 @@ static int jpeg_v4_0_3_hw_fini(struct amdgpu_ip_block *ip_block) ret = jpeg_v4_0_3_set_powergating_state(ip_block, AMD_PG_STATE_GATE); } - if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__JPEG)) + if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__JPEG) && !amdgpu_sriov_vf(adev)) amdgpu_irq_put(adev, &adev->jpeg.inst->ras_poison_irq, 0); return ret; diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_1.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_1.c index 54523dc1f702..47bde62b3909 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_1.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_1.c @@ -196,6 +196,14 @@ static int jpeg_v5_0_1_sw_init(struct amdgpu_ip_block *ip_block) } } + if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__JPEG)) { + r = amdgpu_jpeg_ras_sw_init(adev); + if (r) { + dev_err(adev->dev, "Failed to initialize jpeg ras block!\n"); + return r; + } + } + r = amdgpu_jpeg_reg_dump_init(adev, jpeg_reg_list_5_0_1, ARRAY_SIZE(jpeg_reg_list_5_0_1)); if (r) return r; @@ -307,7 +315,7 @@ static int jpeg_v5_0_1_hw_fini(struct amdgpu_ip_block *ip_block) ret = jpeg_v5_0_1_set_powergating_state(ip_block, AMD_PG_STATE_GATE); } - if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__JPEG)) + if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__JPEG) && !amdgpu_sriov_vf(adev)) amdgpu_irq_put(adev, &adev->jpeg.inst->ras_poison_irq, 0); return ret; @@ -1016,8 +1024,9 @@ static int jpeg_v5_0_1_aca_bank_parser(struct aca_handle *handle, struct aca_ban /* reference to smu driver if header file */ static int jpeg_v5_0_1_err_codes[] = { - 16, 17, 18, 19, 20, 21, 22, 23, /* JPEG[0-7][S|D] */ - 24, 25, 26, 27, 28, 29, 30, 31 + 16, 17, 18, 19, 20, 21, 22, 23, /* JPEG[0-9][S|D] */ + 24, 25, 26, 27, 28, 29, 30, 31, + 48, 49, 50, 51, }; static bool jpeg_v5_0_1_aca_bank_is_valid(struct aca_handle *handle, struct aca_bank *bank, @@ -1058,6 +1067,11 @@ static int jpeg_v5_0_1_ras_late_init(struct amdgpu_device *adev, struct ras_comm if (r) return r; + r = amdgpu_ras_bind_aca(adev, AMDGPU_RAS_BLOCK__JPEG, + &jpeg_v5_0_1_aca_info, NULL); + if (r) + goto late_fini; + if (amdgpu_ras_is_supported(adev, ras_block->block) && adev->jpeg.inst->ras_poison_irq.funcs) { r = amdgpu_irq_get(adev, &adev->jpeg.inst->ras_poison_irq, 0); @@ -1065,11 +1079,6 @@ static int jpeg_v5_0_1_ras_late_init(struct amdgpu_device *adev, struct ras_comm goto late_fini; } - r = amdgpu_ras_bind_aca(adev, AMDGPU_RAS_BLOCK__JPEG, - &jpeg_v5_0_1_aca_info, NULL); - if (r) - goto late_fini; - return 0; late_fini: diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c index 28eb846280dd..3f6a828cad8a 100644 --- a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c @@ -641,8 +641,9 @@ static int mes_v11_0_misc_op(struct amdgpu_mes *mes, break; case MES_MISC_OP_CHANGE_CONFIG: if ((mes->adev->mes.sched_version & AMDGPU_MES_VERSION_MASK) < 0x63) { - dev_err(mes->adev->dev, "MES FW version must be larger than 0x63 to support limit single process feature.\n"); - return -EINVAL; + dev_warn_once(mes->adev->dev, + "MES FW version must be larger than 0x63 to support limit single process feature.\n"); + return 0; } misc_pkt.opcode = MESAPI_MISC__CHANGE_CONFIG; misc_pkt.change_config.opcode = diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c b/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c index 6b222630f3fa..cd5c966cee95 100644 --- a/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c @@ -108,6 +108,7 @@ static const char *mes_v12_0_opcodes[] = { "SET_SE_MODE", "SET_GANG_SUBMIT", "SET_HW_RSRC_1", + "INVALIDATE_TLBS", }; static const char *mes_v12_0_misc_opcodes[] = { @@ -879,6 +880,46 @@ static int mes_v12_0_reset_hw_queue(struct amdgpu_mes *mes, offsetof(union MESAPI__RESET, api_status)); } +static int mes_v12_inv_tlb_convert_hub_id(uint8_t id) +{ + /* + * MES doesn't support invalidate gc_hub on slave xcc individually + * master xcc will invalidate all gc_hub for the partition + */ + if (AMDGPU_IS_GFXHUB(id)) + return 0; + else if (AMDGPU_IS_MMHUB0(id)) + return 1; + else + return -EINVAL; + +} + +static int mes_v12_0_inv_tlbs_pasid(struct amdgpu_mes *mes, + struct mes_inv_tlbs_pasid_input *input) +{ + union MESAPI__INV_TLBS mes_inv_tlbs; + + memset(&mes_inv_tlbs, 0, sizeof(mes_inv_tlbs)); + + mes_inv_tlbs.header.type = MES_API_TYPE_SCHEDULER; + mes_inv_tlbs.header.opcode = MES_SCH_API_INV_TLBS; + mes_inv_tlbs.header.dwsize = API_FRAME_SIZE_IN_DWORDS; + + mes_inv_tlbs.invalidate_tlbs.inv_sel = 0; + mes_inv_tlbs.invalidate_tlbs.flush_type = input->flush_type; + mes_inv_tlbs.invalidate_tlbs.inv_sel_id = input->pasid; + + /*convert amdgpu_mes_hub_id to mes expected hub_id */ + mes_inv_tlbs.invalidate_tlbs.hub_id = mes_v12_inv_tlb_convert_hub_id(input->hub_id); + if (mes_inv_tlbs.invalidate_tlbs.hub_id < 0) + return -EINVAL; + return mes_v12_0_submit_pkt_and_poll_completion(mes, AMDGPU_MES_KIQ_PIPE, + &mes_inv_tlbs, sizeof(mes_inv_tlbs), + offsetof(union MESAPI__INV_TLBS, api_status)); + +} + static const struct amdgpu_mes_funcs mes_v12_0_funcs = { .add_hw_queue = mes_v12_0_add_hw_queue, .remove_hw_queue = mes_v12_0_remove_hw_queue, @@ -888,6 +929,7 @@ static const struct amdgpu_mes_funcs mes_v12_0_funcs = { .resume_gang = mes_v12_0_resume_gang, .misc_op = mes_v12_0_misc_op, .reset_hw_queue = mes_v12_0_reset_hw_queue, + .invalidate_tlbs_pasid = mes_v12_0_inv_tlbs_pasid, }; static int mes_v12_0_allocate_ucode_buffer(struct amdgpu_device *adev, diff --git a/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c b/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c index 48101a34e049..9a40107a0869 100644 --- a/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c +++ b/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c @@ -292,14 +292,32 @@ static void xgpu_ai_mailbox_flr_work(struct work_struct *work) } } -static void xgpu_ai_mailbox_bad_pages_work(struct work_struct *work) +static void xgpu_ai_mailbox_req_bad_pages_work(struct work_struct *work) { - struct amdgpu_virt *virt = container_of(work, struct amdgpu_virt, bad_pages_work); + struct amdgpu_virt *virt = container_of(work, struct amdgpu_virt, req_bad_pages_work); struct amdgpu_device *adev = container_of(virt, struct amdgpu_device, virt); if (down_read_trylock(&adev->reset_domain->sem)) { amdgpu_virt_fini_data_exchange(adev); amdgpu_virt_request_bad_pages(adev); + up_read(&adev->reset_domain->sem); + } +} + +/** + * xgpu_ai_mailbox_handle_bad_pages_work - Reinitialize the data exchange region to get fresh bad page information + * @work: pointer to the work_struct + * + * This work handler is triggered when bad pages are ready, and it reinitializes + * the data exchange region to retrieve updated bad page information from the host. + */ +static void xgpu_ai_mailbox_handle_bad_pages_work(struct work_struct *work) +{ + struct amdgpu_virt *virt = container_of(work, struct amdgpu_virt, handle_bad_pages_work); + struct amdgpu_device *adev = container_of(virt, struct amdgpu_device, virt); + + if (down_read_trylock(&adev->reset_domain->sem)) { + amdgpu_virt_fini_data_exchange(adev); amdgpu_virt_init_data_exchange(adev); up_read(&adev->reset_domain->sem); } @@ -327,10 +345,15 @@ static int xgpu_ai_mailbox_rcv_irq(struct amdgpu_device *adev, struct amdgpu_ras *ras = amdgpu_ras_get_context(adev); switch (event) { + case IDH_RAS_BAD_PAGES_READY: + xgpu_ai_mailbox_send_ack(adev); + if (amdgpu_sriov_runtime(adev)) + schedule_work(&adev->virt.handle_bad_pages_work); + break; case IDH_RAS_BAD_PAGES_NOTIFICATION: xgpu_ai_mailbox_send_ack(adev); if (amdgpu_sriov_runtime(adev)) - schedule_work(&adev->virt.bad_pages_work); + schedule_work(&adev->virt.req_bad_pages_work); break; case IDH_UNRECOV_ERR_NOTIFICATION: xgpu_ai_mailbox_send_ack(adev); @@ -415,7 +438,8 @@ int xgpu_ai_mailbox_get_irq(struct amdgpu_device *adev) } INIT_WORK(&adev->virt.flr_work, xgpu_ai_mailbox_flr_work); - INIT_WORK(&adev->virt.bad_pages_work, xgpu_ai_mailbox_bad_pages_work); + INIT_WORK(&adev->virt.req_bad_pages_work, xgpu_ai_mailbox_req_bad_pages_work); + INIT_WORK(&adev->virt.handle_bad_pages_work, xgpu_ai_mailbox_handle_bad_pages_work); return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c b/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c index f6d8597452ed..457972aa5632 100644 --- a/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c +++ b/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c @@ -202,9 +202,6 @@ send_request: case IDH_REQ_RAS_CPER_DUMP: event = IDH_RAS_CPER_DUMP_READY; break; - case IDH_REQ_RAS_BAD_PAGES: - event = IDH_RAS_BAD_PAGES_READY; - break; default: break; } @@ -359,14 +356,32 @@ static void xgpu_nv_mailbox_flr_work(struct work_struct *work) } } -static void xgpu_nv_mailbox_bad_pages_work(struct work_struct *work) +static void xgpu_nv_mailbox_req_bad_pages_work(struct work_struct *work) { - struct amdgpu_virt *virt = container_of(work, struct amdgpu_virt, bad_pages_work); + struct amdgpu_virt *virt = container_of(work, struct amdgpu_virt, req_bad_pages_work); struct amdgpu_device *adev = container_of(virt, struct amdgpu_device, virt); if (down_read_trylock(&adev->reset_domain->sem)) { amdgpu_virt_fini_data_exchange(adev); amdgpu_virt_request_bad_pages(adev); + up_read(&adev->reset_domain->sem); + } +} + +/** + * xgpu_nv_mailbox_handle_bad_pages_work - Reinitialize the data exchange region to get fresh bad page information + * @work: pointer to the work_struct + * + * This work handler is triggered when bad pages are ready, and it reinitializes + * the data exchange region to retrieve updated bad page information from the host. + */ +static void xgpu_nv_mailbox_handle_bad_pages_work(struct work_struct *work) +{ + struct amdgpu_virt *virt = container_of(work, struct amdgpu_virt, handle_bad_pages_work); + struct amdgpu_device *adev = container_of(virt, struct amdgpu_device, virt); + + if (down_read_trylock(&adev->reset_domain->sem)) { + amdgpu_virt_fini_data_exchange(adev); amdgpu_virt_init_data_exchange(adev); up_read(&adev->reset_domain->sem); } @@ -397,10 +412,15 @@ static int xgpu_nv_mailbox_rcv_irq(struct amdgpu_device *adev, struct amdgpu_ras *ras = amdgpu_ras_get_context(adev); switch (event) { + case IDH_RAS_BAD_PAGES_READY: + xgpu_nv_mailbox_send_ack(adev); + if (amdgpu_sriov_runtime(adev)) + schedule_work(&adev->virt.handle_bad_pages_work); + break; case IDH_RAS_BAD_PAGES_NOTIFICATION: xgpu_nv_mailbox_send_ack(adev); if (amdgpu_sriov_runtime(adev)) - schedule_work(&adev->virt.bad_pages_work); + schedule_work(&adev->virt.req_bad_pages_work); break; case IDH_UNRECOV_ERR_NOTIFICATION: xgpu_nv_mailbox_send_ack(adev); @@ -485,7 +505,8 @@ int xgpu_nv_mailbox_get_irq(struct amdgpu_device *adev) } INIT_WORK(&adev->virt.flr_work, xgpu_nv_mailbox_flr_work); - INIT_WORK(&adev->virt.bad_pages_work, xgpu_nv_mailbox_bad_pages_work); + INIT_WORK(&adev->virt.req_bad_pages_work, xgpu_nv_mailbox_req_bad_pages_work); + INIT_WORK(&adev->virt.handle_bad_pages_work, xgpu_nv_mailbox_handle_bad_pages_work); return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c index d5002ff931d8..860bc5cb03c8 100644 --- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c +++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c @@ -151,9 +151,9 @@ static void nbio_v7_4_sdma_doorbell_range(struct amdgpu_device *adev, int instan * BIF_SDMA0_DOORBELL_RANGE: 0x3bc0 * BIF_SDMA1_DOORBELL_RANGE: 0x3bc4 * BIF_SDMA2_DOORBELL_RANGE: 0x3bd8 -+ * BIF_SDMA4_DOORBELL_RANGE: -+ * ARCTURUS: 0x3be0 -+ * ALDEBARAN: 0x3be4 + * BIF_SDMA4_DOORBELL_RANGE: + * ARCTURUS: 0x3be0 + * ALDEBARAN: 0x3be4 */ if (adev->asic_type == CHIP_ALDEBARAN && instance == 4) reg = instance + 0x4 + 0x1 + diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c index 6cc05d36e359..64b240b51f1a 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c @@ -149,12 +149,12 @@ static int psp_v11_0_wait_for_bootloader(struct psp_context *psp) int ret; int retry_loop; - for (retry_loop = 0; retry_loop < 10; retry_loop++) { + for (retry_loop = 0; retry_loop < 20; retry_loop++) { /* Wait for bootloader to signify that is ready having bit 31 of C2PMSG_35 set to 1 */ ret = psp_wait_for( psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_35), - 0x80000000, 0x80000000, PSP_WAITREG_NOVERBOSE); + 0x80000000, 0x8000FFFF, PSP_WAITREG_NOVERBOSE); if (ret == 0) return 0; @@ -397,18 +397,6 @@ static int psp_v11_0_mode1_reset(struct psp_context *psp) msleep(500); - offset = SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_33); - - ret = psp_wait_for(psp, offset, MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, - 0); - - if (ret) { - DRM_INFO("psp mode 1 reset failed!\n"); - return -EINVAL; - } - - DRM_INFO("psp mode1 reset succeed \n"); - return 0; } @@ -665,7 +653,8 @@ static const struct psp_funcs psp_v11_0_funcs = { .ring_get_wptr = psp_v11_0_ring_get_wptr, .ring_set_wptr = psp_v11_0_ring_set_wptr, .load_usbc_pd_fw = psp_v11_0_load_usbc_pd_fw, - .read_usbc_pd_fw = psp_v11_0_read_usbc_pd_fw + .read_usbc_pd_fw = psp_v11_0_read_usbc_pd_fw, + .wait_for_bootloader = psp_v11_0_wait_for_bootloader }; void psp_v11_0_set_psp_funcs(struct psp_context *psp) diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c index e6d8eddda2bf..db6e41967f12 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c @@ -1377,7 +1377,7 @@ static int sdma_v6_0_sw_init(struct amdgpu_ip_block *ip_block) switch (amdgpu_ip_version(adev, SDMA0_HWIP, 0)) { case IP_VERSION(6, 0, 0): - if ((adev->sdma.instance[0].fw_version >= 24) && !adev->sdma.disable_uq) + if ((adev->sdma.instance[0].fw_version >= 27) && !adev->sdma.disable_uq) adev->userq_funcs[AMDGPU_HW_IP_DMA] = &userq_mes_funcs; break; case IP_VERSION(6, 0, 1): @@ -1385,11 +1385,11 @@ static int sdma_v6_0_sw_init(struct amdgpu_ip_block *ip_block) adev->userq_funcs[AMDGPU_HW_IP_DMA] = &userq_mes_funcs; break; case IP_VERSION(6, 0, 2): - if ((adev->sdma.instance[0].fw_version >= 21) && !adev->sdma.disable_uq) + if ((adev->sdma.instance[0].fw_version >= 23) && !adev->sdma.disable_uq) adev->userq_funcs[AMDGPU_HW_IP_DMA] = &userq_mes_funcs; break; case IP_VERSION(6, 0, 3): - if ((adev->sdma.instance[0].fw_version >= 25) && !adev->sdma.disable_uq) + if ((adev->sdma.instance[0].fw_version >= 27) && !adev->sdma.disable_uq) adev->userq_funcs[AMDGPU_HW_IP_DMA] = &userq_mes_funcs; break; case IP_VERSION(6, 1, 0): diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c index 9e74c9822e62..9785fada4fa7 100644 --- a/drivers/gpu/drm/amd/amdgpu/soc15.c +++ b/drivers/gpu/drm/amd/amdgpu/soc15.c @@ -741,7 +741,6 @@ static void soc15_reg_base_init(struct amdgpu_device *adev) void soc15_set_virt_ops(struct amdgpu_device *adev) { adev->virt.ops = &xgpu_ai_virt_ops; - /* init soc15 reg base early enough so we can * request request full access for sriov before * set_ip_blocks. */ diff --git a/drivers/gpu/drm/amd/amdgpu/umc_v12_0.c b/drivers/gpu/drm/amd/amdgpu/umc_v12_0.c index e590cbdd8de9..8dc32787d625 100644 --- a/drivers/gpu/drm/amd/amdgpu/umc_v12_0.c +++ b/drivers/gpu/drm/amd/amdgpu/umc_v12_0.c @@ -536,8 +536,11 @@ static int umc_v12_0_update_ecc_status(struct amdgpu_device *adev, hwid = REG_GET_FIELD(ipid, MCMP1_IPIDT0, HardwareID); mcatype = REG_GET_FIELD(ipid, MCMP1_IPIDT0, McaType); - if ((hwid != MCA_UMC_HWID_V12_0) || (mcatype != MCA_UMC_MCATYPE_V12_0)) + /* The IP block decode of consumption is SMU */ + if (hwid != MCA_UMC_HWID_V12_0 || mcatype != MCA_UMC_MCATYPE_V12_0) { + con->umc_ecc_log.consumption_q_count++; return 0; + } if (!status) return 0; diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c index c74947705d77..1e89ba153d9d 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c @@ -1338,7 +1338,6 @@ static int vcn_v1_0_pause_dpg_mode(struct amdgpu_vcn_inst *vinst, WREG32_SOC15(UVD, 0, mmUVD_RB_RPTR2, lower_32_bits(ring->wptr)); WREG32_SOC15(UVD, 0, mmUVD_RB_WPTR2, lower_32_bits(ring->wptr)); - ring = &adev->vcn.inst->ring_dec; WREG32_SOC15(UVD, 0, mmUVD_RBC_RB_WPTR, RREG32_SOC15(UVD, 0, mmUVD_SCRATCH2) & 0x7FFFFFFF); SOC15_WAIT_ON_RREG(UVD, 0, mmUVD_POWER_STATUS, @@ -1399,7 +1398,6 @@ static int vcn_v1_0_pause_dpg_mode(struct amdgpu_vcn_inst *vinst, WREG32_SOC15(UVD, 0, mmUVD_JRBC_RB_CNTL, UVD_JRBC_RB_CNTL__RB_RPTR_WR_EN_MASK); - ring = &adev->vcn.inst->ring_dec; WREG32_SOC15(UVD, 0, mmUVD_RBC_RB_WPTR, RREG32_SOC15(UVD, 0, mmUVD_SCRATCH2) & 0x7FFFFFFF); SOC15_WAIT_ON_RREG(UVD, 0, mmUVD_POWER_STATUS, diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c index 68b4371df0f1..b115137ab2d6 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c @@ -136,8 +136,6 @@ static int vcn_v2_0_sw_init(struct amdgpu_ip_block *ip_block) { struct amdgpu_ring *ring; int i, r; - uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_2_0); - uint32_t *ptr; struct amdgpu_device *adev = ip_block->adev; volatile struct amdgpu_fw_shared *fw_shared; @@ -232,14 +230,9 @@ static int vcn_v2_0_sw_init(struct amdgpu_ip_block *ip_block) if (amdgpu_vcnfw_log) amdgpu_vcn_fwlog_init(adev->vcn.inst); - /* Allocate memory for VCN IP Dump buffer */ - ptr = kcalloc(adev->vcn.num_vcn_inst * reg_count, sizeof(uint32_t), GFP_KERNEL); - if (!ptr) { - DRM_ERROR("Failed to allocate memory for VCN IP Dump\n"); - adev->vcn.ip_dump = NULL; - } else { - adev->vcn.ip_dump = ptr; - } + r = amdgpu_vcn_reg_dump_init(adev, vcn_reg_list_2_0, ARRAY_SIZE(vcn_reg_list_2_0)); + if (r) + return r; r = amdgpu_vcn_sysfs_reset_mask_init(adev); if (r) @@ -276,8 +269,6 @@ static int vcn_v2_0_sw_fini(struct amdgpu_ip_block *ip_block) r = amdgpu_vcn_sw_fini(adev, 0); - kfree(adev->vcn.ip_dump); - return r; } @@ -865,6 +856,7 @@ static int vcn_v2_0_start_dpg_mode(struct amdgpu_vcn_inst *vinst, bool indirect) volatile struct amdgpu_fw_shared *fw_shared = adev->vcn.inst->fw_shared.cpu_addr; struct amdgpu_ring *ring = &adev->vcn.inst->ring_dec; uint32_t rb_bufsz, tmp; + int ret; vcn_v2_0_enable_static_power_gating(vinst); @@ -948,8 +940,13 @@ static int vcn_v2_0_start_dpg_mode(struct amdgpu_vcn_inst *vinst, bool indirect) UVD, 0, mmUVD_MASTINT_EN), UVD_MASTINT_EN__VCPU_EN_MASK, 0, indirect); - if (indirect) - amdgpu_vcn_psp_update_sram(adev, 0, 0); + if (indirect) { + ret = amdgpu_vcn_psp_update_sram(adev, 0, 0); + if (ret) { + dev_err(adev->dev, "vcn sram load failed %d\n", ret); + return ret; + } + } /* force RBC into idle state */ rb_bufsz = order_base_2(ring->ring_size); @@ -2095,66 +2092,6 @@ static int vcn_v2_0_start_sriov(struct amdgpu_device *adev) return vcn_v2_0_start_mmsch(adev, &adev->virt.mm_table); } -static void vcn_v2_0_print_ip_state(struct amdgpu_ip_block *ip_block, struct drm_printer *p) -{ - struct amdgpu_device *adev = ip_block->adev; - int i, j; - uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_2_0); - uint32_t inst_off, is_powered; - - if (!adev->vcn.ip_dump) - return; - - drm_printf(p, "num_instances:%d\n", adev->vcn.num_vcn_inst); - for (i = 0; i < adev->vcn.num_vcn_inst; i++) { - if (adev->vcn.harvest_config & (1 << i)) { - drm_printf(p, "\nHarvested Instance:VCN%d Skipping dump\n", i); - continue; - } - - inst_off = i * reg_count; - is_powered = (adev->vcn.ip_dump[inst_off] & - UVD_POWER_STATUS__UVD_POWER_STATUS_MASK) != 1; - - if (is_powered) { - drm_printf(p, "\nActive Instance:VCN%d\n", i); - for (j = 0; j < reg_count; j++) - drm_printf(p, "%-50s \t 0x%08x\n", vcn_reg_list_2_0[j].reg_name, - adev->vcn.ip_dump[inst_off + j]); - } else { - drm_printf(p, "\nInactive Instance:VCN%d\n", i); - } - } -} - -static void vcn_v2_0_dump_ip_state(struct amdgpu_ip_block *ip_block) -{ - struct amdgpu_device *adev = ip_block->adev; - int i, j; - bool is_powered; - uint32_t inst_off; - uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_2_0); - - if (!adev->vcn.ip_dump) - return; - - for (i = 0; i < adev->vcn.num_vcn_inst; i++) { - if (adev->vcn.harvest_config & (1 << i)) - continue; - - inst_off = i * reg_count; - /* mmUVD_POWER_STATUS is always readable and is first element of the array */ - adev->vcn.ip_dump[inst_off] = RREG32_SOC15(VCN, i, mmUVD_POWER_STATUS); - is_powered = (adev->vcn.ip_dump[inst_off] & - UVD_POWER_STATUS__UVD_POWER_STATUS_MASK) != 1; - - if (is_powered) - for (j = 1; j < reg_count; j++) - adev->vcn.ip_dump[inst_off + j] = - RREG32(SOC15_REG_ENTRY_OFFSET_INST(vcn_reg_list_2_0[j], i)); - } -} - static const struct amd_ip_funcs vcn_v2_0_ip_funcs = { .name = "vcn_v2_0", .early_init = vcn_v2_0_early_init, @@ -2168,8 +2105,8 @@ static const struct amd_ip_funcs vcn_v2_0_ip_funcs = { .wait_for_idle = vcn_v2_0_wait_for_idle, .set_clockgating_state = vcn_v2_0_set_clockgating_state, .set_powergating_state = vcn_set_powergating_state, - .dump_ip_state = vcn_v2_0_dump_ip_state, - .print_ip_state = vcn_v2_0_print_ip_state, + .dump_ip_state = amdgpu_vcn_dump_ip_state, + .print_ip_state = amdgpu_vcn_print_ip_state, }; static const struct amdgpu_ring_funcs vcn_v2_0_dec_ring_vm_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c b/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c index bc30a5326866..904b94bc8693 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c @@ -116,7 +116,6 @@ static void vcn_v2_5_idle_work_handler(struct work_struct *work) struct amdgpu_device *adev = vcn_inst->adev; unsigned int fences = 0, fence[AMDGPU_MAX_VCN_INSTANCES] = {0}; unsigned int i, j; - int r = 0; for (i = 0; i < adev->vcn.num_vcn_inst; ++i) { struct amdgpu_vcn_inst *v = &adev->vcn.inst[i]; @@ -149,15 +148,7 @@ static void vcn_v2_5_idle_work_handler(struct work_struct *work) if (!fences && !atomic_read(&adev->vcn.inst[0].total_submission_cnt)) { amdgpu_device_ip_set_powergating_state(adev, AMD_IP_BLOCK_TYPE_VCN, AMD_PG_STATE_GATE); - mutex_lock(&adev->vcn.workload_profile_mutex); - if (adev->vcn.workload_profile_active) { - r = amdgpu_dpm_switch_power_profile(adev, PP_SMC_POWER_PROFILE_VIDEO, - false); - if (r) - dev_warn(adev->dev, "(%d) failed to disable video power profile mode\n", r); - adev->vcn.workload_profile_active = false; - } - mutex_unlock(&adev->vcn.workload_profile_mutex); + amdgpu_vcn_put_profile(adev); } else { schedule_delayed_work(&adev->vcn.inst[0].idle_work, VCN_IDLE_TIMEOUT); } @@ -167,7 +158,6 @@ static void vcn_v2_5_ring_begin_use(struct amdgpu_ring *ring) { struct amdgpu_device *adev = ring->adev; struct amdgpu_vcn_inst *v = &adev->vcn.inst[ring->me]; - int r = 0; atomic_inc(&adev->vcn.inst[0].total_submission_cnt); @@ -177,20 +167,6 @@ static void vcn_v2_5_ring_begin_use(struct amdgpu_ring *ring) * the delayed work so there is no one else to set it to false * and we don't care if someone else sets it to true. */ - if (adev->vcn.workload_profile_active) - goto pg_lock; - - mutex_lock(&adev->vcn.workload_profile_mutex); - if (!adev->vcn.workload_profile_active) { - r = amdgpu_dpm_switch_power_profile(adev, PP_SMC_POWER_PROFILE_VIDEO, - true); - if (r) - dev_warn(adev->dev, "(%d) failed to switch to video power profile mode\n", r); - adev->vcn.workload_profile_active = true; - } - mutex_unlock(&adev->vcn.workload_profile_mutex); - -pg_lock: mutex_lock(&adev->vcn.inst[0].vcn_pg_lock); amdgpu_device_ip_set_powergating_state(adev, AMD_IP_BLOCK_TYPE_VCN, AMD_PG_STATE_UNGATE); @@ -218,6 +194,7 @@ pg_lock: v->pause_dpg_mode(v, &new_state); } mutex_unlock(&adev->vcn.inst[0].vcn_pg_lock); + amdgpu_vcn_get_profile(adev); } static void vcn_v2_5_ring_end_use(struct amdgpu_ring *ring) @@ -297,8 +274,6 @@ static int vcn_v2_5_sw_init(struct amdgpu_ip_block *ip_block) { struct amdgpu_ring *ring; int i, j, r; - uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_2_5); - uint32_t *ptr; struct amdgpu_device *adev = ip_block->adev; for (j = 0; j < adev->vcn.num_vcn_inst; j++) { @@ -423,14 +398,9 @@ static int vcn_v2_5_sw_init(struct amdgpu_ip_block *ip_block) if (r) return r; - /* Allocate memory for VCN IP Dump buffer */ - ptr = kcalloc(adev->vcn.num_vcn_inst * reg_count, sizeof(uint32_t), GFP_KERNEL); - if (!ptr) { - DRM_ERROR("Failed to allocate memory for VCN IP Dump\n"); - adev->vcn.ip_dump = NULL; - } else { - adev->vcn.ip_dump = ptr; - } + r = amdgpu_vcn_reg_dump_init(adev, vcn_reg_list_2_5, ARRAY_SIZE(vcn_reg_list_2_5)); + if (r) + return r; r = amdgpu_vcn_sysfs_reset_mask_init(adev); if (r) @@ -477,8 +447,6 @@ static int vcn_v2_5_sw_fini(struct amdgpu_ip_block *ip_block) return r; } - kfree(adev->vcn.ip_dump); - return 0; } @@ -1035,6 +1003,7 @@ static int vcn_v2_5_start_dpg_mode(struct amdgpu_vcn_inst *vinst, bool indirect) volatile struct amdgpu_fw_shared *fw_shared = adev->vcn.inst[inst_idx].fw_shared.cpu_addr; struct amdgpu_ring *ring; uint32_t rb_bufsz, tmp; + int ret; /* disable register anti-hang mechanism */ WREG32_P(SOC15_REG_OFFSET(VCN, inst_idx, mmUVD_POWER_STATUS), 1, @@ -1125,8 +1094,13 @@ static int vcn_v2_5_start_dpg_mode(struct amdgpu_vcn_inst *vinst, bool indirect) VCN, 0, mmUVD_MASTINT_EN), UVD_MASTINT_EN__VCPU_EN_MASK, 0, indirect); - if (indirect) - amdgpu_vcn_psp_update_sram(adev, inst_idx, 0); + if (indirect) { + ret = amdgpu_vcn_psp_update_sram(adev, inst_idx, 0); + if (ret) { + dev_err(adev->dev, "vcn sram load failed %d\n", ret); + return ret; + } + } ring = &adev->vcn.inst[inst_idx].ring_dec; /* force RBC into idle state */ @@ -2127,66 +2101,6 @@ static void vcn_v2_5_set_irq_funcs(struct amdgpu_device *adev) } } -static void vcn_v2_5_print_ip_state(struct amdgpu_ip_block *ip_block, struct drm_printer *p) -{ - struct amdgpu_device *adev = ip_block->adev; - int i, j; - uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_2_5); - uint32_t inst_off, is_powered; - - if (!adev->vcn.ip_dump) - return; - - drm_printf(p, "num_instances:%d\n", adev->vcn.num_vcn_inst); - for (i = 0; i < adev->vcn.num_vcn_inst; i++) { - if (adev->vcn.harvest_config & (1 << i)) { - drm_printf(p, "\nHarvested Instance:VCN%d Skipping dump\n", i); - continue; - } - - inst_off = i * reg_count; - is_powered = (adev->vcn.ip_dump[inst_off] & - UVD_POWER_STATUS__UVD_POWER_STATUS_MASK) != 1; - - if (is_powered) { - drm_printf(p, "\nActive Instance:VCN%d\n", i); - for (j = 0; j < reg_count; j++) - drm_printf(p, "%-50s \t 0x%08x\n", vcn_reg_list_2_5[j].reg_name, - adev->vcn.ip_dump[inst_off + j]); - } else { - drm_printf(p, "\nInactive Instance:VCN%d\n", i); - } - } -} - -static void vcn_v2_5_dump_ip_state(struct amdgpu_ip_block *ip_block) -{ - struct amdgpu_device *adev = ip_block->adev; - int i, j; - bool is_powered; - uint32_t inst_off; - uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_2_5); - - if (!adev->vcn.ip_dump) - return; - - for (i = 0; i < adev->vcn.num_vcn_inst; i++) { - if (adev->vcn.harvest_config & (1 << i)) - continue; - - inst_off = i * reg_count; - /* mmUVD_POWER_STATUS is always readable and is first element of the array */ - adev->vcn.ip_dump[inst_off] = RREG32_SOC15(VCN, i, mmUVD_POWER_STATUS); - is_powered = (adev->vcn.ip_dump[inst_off] & - UVD_POWER_STATUS__UVD_POWER_STATUS_MASK) != 1; - - if (is_powered) - for (j = 1; j < reg_count; j++) - adev->vcn.ip_dump[inst_off + j] = - RREG32(SOC15_REG_ENTRY_OFFSET_INST(vcn_reg_list_2_5[j], i)); - } -} - static const struct amd_ip_funcs vcn_v2_5_ip_funcs = { .name = "vcn_v2_5", .early_init = vcn_v2_5_early_init, @@ -2200,8 +2114,8 @@ static const struct amd_ip_funcs vcn_v2_5_ip_funcs = { .wait_for_idle = vcn_v2_5_wait_for_idle, .set_clockgating_state = vcn_v2_5_set_clockgating_state, .set_powergating_state = vcn_set_powergating_state, - .dump_ip_state = vcn_v2_5_dump_ip_state, - .print_ip_state = vcn_v2_5_print_ip_state, + .dump_ip_state = amdgpu_vcn_dump_ip_state, + .print_ip_state = amdgpu_vcn_print_ip_state, }; static const struct amd_ip_funcs vcn_v2_6_ip_funcs = { @@ -2217,8 +2131,8 @@ static const struct amd_ip_funcs vcn_v2_6_ip_funcs = { .wait_for_idle = vcn_v2_5_wait_for_idle, .set_clockgating_state = vcn_v2_5_set_clockgating_state, .set_powergating_state = vcn_set_powergating_state, - .dump_ip_state = vcn_v2_5_dump_ip_state, - .print_ip_state = vcn_v2_5_print_ip_state, + .dump_ip_state = amdgpu_vcn_dump_ip_state, + .print_ip_state = amdgpu_vcn_print_ip_state, }; const struct amdgpu_ip_block_version vcn_v2_5_ip_block = diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c index 4b8f4407047f..95173156f956 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c @@ -175,8 +175,6 @@ static int vcn_v3_0_sw_init(struct amdgpu_ip_block *ip_block) struct amdgpu_ring *ring; int i, j, r; int vcn_doorbell_index = 0; - uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_3_0); - uint32_t *ptr; struct amdgpu_device *adev = ip_block->adev; /* @@ -304,14 +302,9 @@ static int vcn_v3_0_sw_init(struct amdgpu_ip_block *ip_block) return r; } - /* Allocate memory for VCN IP Dump buffer */ - ptr = kcalloc(adev->vcn.num_vcn_inst * reg_count, sizeof(uint32_t), GFP_KERNEL); - if (ptr == NULL) { - DRM_ERROR("Failed to allocate memory for VCN IP Dump\n"); - adev->vcn.ip_dump = NULL; - } else { - adev->vcn.ip_dump = ptr; - } + r = amdgpu_vcn_reg_dump_init(adev, vcn_reg_list_3_0, ARRAY_SIZE(vcn_reg_list_3_0)); + if (r) + return r; r = amdgpu_vcn_sysfs_reset_mask_init(adev); if (r) @@ -361,7 +354,6 @@ static int vcn_v3_0_sw_fini(struct amdgpu_ip_block *ip_block) return r; } - kfree(adev->vcn.ip_dump); return 0; } @@ -1042,6 +1034,7 @@ static int vcn_v3_0_start_dpg_mode(struct amdgpu_vcn_inst *vinst, bool indirect) volatile struct amdgpu_fw_shared *fw_shared = adev->vcn.inst[inst_idx].fw_shared.cpu_addr; struct amdgpu_ring *ring; uint32_t rb_bufsz, tmp; + int ret; /* disable register anti-hang mechanism */ WREG32_P(SOC15_REG_OFFSET(VCN, inst_idx, mmUVD_POWER_STATUS), 1, @@ -1134,8 +1127,13 @@ static int vcn_v3_0_start_dpg_mode(struct amdgpu_vcn_inst *vinst, bool indirect) WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( VCN, inst_idx, mmUVD_VCPU_CNTL), tmp, 0, indirect); - if (indirect) - amdgpu_vcn_psp_update_sram(adev, inst_idx, 0); + if (indirect) { + ret = amdgpu_vcn_psp_update_sram(adev, inst_idx, 0); + if (ret) { + dev_err(adev->dev, "vcn sram load failed %d\n", ret); + return ret; + } + } ring = &adev->vcn.inst[inst_idx].ring_dec; /* force RBC into idle state */ @@ -2342,67 +2340,6 @@ static void vcn_v3_0_set_irq_funcs(struct amdgpu_device *adev) } } -static void vcn_v3_0_print_ip_state(struct amdgpu_ip_block *ip_block, struct drm_printer *p) -{ - struct amdgpu_device *adev = ip_block->adev; - int i, j; - uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_3_0); - uint32_t inst_off; - bool is_powered; - - if (!adev->vcn.ip_dump) - return; - - drm_printf(p, "num_instances:%d\n", adev->vcn.num_vcn_inst); - for (i = 0; i < adev->vcn.num_vcn_inst; i++) { - if (adev->vcn.harvest_config & (1 << i)) { - drm_printf(p, "\nHarvested Instance:VCN%d Skipping dump\n", i); - continue; - } - - inst_off = i * reg_count; - is_powered = (adev->vcn.ip_dump[inst_off] & - UVD_POWER_STATUS__UVD_POWER_STATUS_MASK) != 1; - - if (is_powered) { - drm_printf(p, "\nActive Instance:VCN%d\n", i); - for (j = 0; j < reg_count; j++) - drm_printf(p, "%-50s \t 0x%08x\n", vcn_reg_list_3_0[j].reg_name, - adev->vcn.ip_dump[inst_off + j]); - } else { - drm_printf(p, "\nInactive Instance:VCN%d\n", i); - } - } -} - -static void vcn_v3_0_dump_ip_state(struct amdgpu_ip_block *ip_block) -{ - struct amdgpu_device *adev = ip_block->adev; - int i, j; - bool is_powered; - uint32_t inst_off; - uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_3_0); - - if (!adev->vcn.ip_dump) - return; - - for (i = 0; i < adev->vcn.num_vcn_inst; i++) { - if (adev->vcn.harvest_config & (1 << i)) - continue; - - inst_off = i * reg_count; - /* mmUVD_POWER_STATUS is always readable and is first element of the array */ - adev->vcn.ip_dump[inst_off] = RREG32_SOC15(VCN, i, mmUVD_POWER_STATUS); - is_powered = (adev->vcn.ip_dump[inst_off] & - UVD_POWER_STATUS__UVD_POWER_STATUS_MASK) != 1; - - if (is_powered) - for (j = 1; j < reg_count; j++) - adev->vcn.ip_dump[inst_off + j] = - RREG32(SOC15_REG_ENTRY_OFFSET_INST(vcn_reg_list_3_0[j], i)); - } -} - static const struct amd_ip_funcs vcn_v3_0_ip_funcs = { .name = "vcn_v3_0", .early_init = vcn_v3_0_early_init, @@ -2416,8 +2353,8 @@ static const struct amd_ip_funcs vcn_v3_0_ip_funcs = { .wait_for_idle = vcn_v3_0_wait_for_idle, .set_clockgating_state = vcn_v3_0_set_clockgating_state, .set_powergating_state = vcn_set_powergating_state, - .dump_ip_state = vcn_v3_0_dump_ip_state, - .print_ip_state = vcn_v3_0_print_ip_state, + .dump_ip_state = amdgpu_vcn_dump_ip_state, + .print_ip_state = amdgpu_vcn_print_ip_state, }; const struct amdgpu_ip_block_version vcn_v3_0_ip_block = { diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c index 1924e075b66f..d0d27790b73b 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c @@ -183,8 +183,6 @@ static int vcn_v4_0_sw_init(struct amdgpu_ip_block *ip_block) struct amdgpu_ring *ring; struct amdgpu_device *adev = ip_block->adev; int i, r; - uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_4_0); - uint32_t *ptr; for (i = 0; i < adev->vcn.num_vcn_inst; i++) { if (adev->vcn.harvest_config & (1 << i)) @@ -255,14 +253,9 @@ static int vcn_v4_0_sw_init(struct amdgpu_ip_block *ip_block) if (r) return r; - /* Allocate memory for VCN IP Dump buffer */ - ptr = kcalloc(adev->vcn.num_vcn_inst * reg_count, sizeof(uint32_t), GFP_KERNEL); - if (!ptr) { - DRM_ERROR("Failed to allocate memory for VCN IP Dump\n"); - adev->vcn.ip_dump = NULL; - } else { - adev->vcn.ip_dump = ptr; - } + r = amdgpu_vcn_reg_dump_init(adev, vcn_reg_list_4_0, ARRAY_SIZE(vcn_reg_list_4_0)); + if (r) + return r; r = amdgpu_vcn_sysfs_reset_mask_init(adev); if (r) @@ -315,8 +308,6 @@ static int vcn_v4_0_sw_fini(struct amdgpu_ip_block *ip_block) return r; } - kfree(adev->vcn.ip_dump); - return 0; } @@ -1012,6 +1003,7 @@ static int vcn_v4_0_start_dpg_mode(struct amdgpu_vcn_inst *vinst, bool indirect) volatile struct amdgpu_vcn4_fw_shared *fw_shared = adev->vcn.inst[inst_idx].fw_shared.cpu_addr; struct amdgpu_ring *ring; uint32_t tmp; + int ret; /* disable register anti-hang mechanism */ WREG32_P(SOC15_REG_OFFSET(VCN, inst_idx, regUVD_POWER_STATUS), 1, @@ -1094,8 +1086,13 @@ static int vcn_v4_0_start_dpg_mode(struct amdgpu_vcn_inst *vinst, bool indirect) UVD_MASTINT_EN__VCPU_EN_MASK, 0, indirect); - if (indirect) - amdgpu_vcn_psp_update_sram(adev, inst_idx, 0); + if (indirect) { + ret = amdgpu_vcn_psp_update_sram(adev, inst_idx, 0); + if (ret) { + dev_err(adev->dev, "vcn sram load failed %d\n", ret); + return ret; + } + } ring = &adev->vcn.inst[inst_idx].ring_enc[0]; @@ -1624,7 +1621,6 @@ static int vcn_v4_0_stop(struct amdgpu_vcn_inst *vinst) if (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG) { vcn_v4_0_stop_dpg_mode(vinst); - r = 0; goto done; } @@ -2246,67 +2242,6 @@ static void vcn_v4_0_set_irq_funcs(struct amdgpu_device *adev) } } -static void vcn_v4_0_print_ip_state(struct amdgpu_ip_block *ip_block, struct drm_printer *p) -{ - struct amdgpu_device *adev = ip_block->adev; - int i, j; - uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_4_0); - uint32_t inst_off, is_powered; - - if (!adev->vcn.ip_dump) - return; - - drm_printf(p, "num_instances:%d\n", adev->vcn.num_vcn_inst); - for (i = 0; i < adev->vcn.num_vcn_inst; i++) { - if (adev->vcn.harvest_config & (1 << i)) { - drm_printf(p, "\nHarvested Instance:VCN%d Skipping dump\n", i); - continue; - } - - inst_off = i * reg_count; - is_powered = (adev->vcn.ip_dump[inst_off] & - UVD_POWER_STATUS__UVD_POWER_STATUS_MASK) != 1; - - if (is_powered) { - drm_printf(p, "\nActive Instance:VCN%d\n", i); - for (j = 0; j < reg_count; j++) - drm_printf(p, "%-50s \t 0x%08x\n", vcn_reg_list_4_0[j].reg_name, - adev->vcn.ip_dump[inst_off + j]); - } else { - drm_printf(p, "\nInactive Instance:VCN%d\n", i); - } - } -} - -static void vcn_v4_0_dump_ip_state(struct amdgpu_ip_block *ip_block) -{ - struct amdgpu_device *adev = ip_block->adev; - int i, j; - bool is_powered; - uint32_t inst_off; - uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_4_0); - - if (!adev->vcn.ip_dump) - return; - - for (i = 0; i < adev->vcn.num_vcn_inst; i++) { - if (adev->vcn.harvest_config & (1 << i)) - continue; - - inst_off = i * reg_count; - /* mmUVD_POWER_STATUS is always readable and is first element of the array */ - adev->vcn.ip_dump[inst_off] = RREG32_SOC15(VCN, i, regUVD_POWER_STATUS); - is_powered = (adev->vcn.ip_dump[inst_off] & - UVD_POWER_STATUS__UVD_POWER_STATUS_MASK) != 1; - - if (is_powered) - for (j = 1; j < reg_count; j++) - adev->vcn.ip_dump[inst_off + j] = - RREG32(SOC15_REG_ENTRY_OFFSET_INST(vcn_reg_list_4_0[j], - i)); - } -} - static const struct amd_ip_funcs vcn_v4_0_ip_funcs = { .name = "vcn_v4_0", .early_init = vcn_v4_0_early_init, @@ -2320,8 +2255,8 @@ static const struct amd_ip_funcs vcn_v4_0_ip_funcs = { .wait_for_idle = vcn_v4_0_wait_for_idle, .set_clockgating_state = vcn_v4_0_set_clockgating_state, .set_powergating_state = vcn_set_powergating_state, - .dump_ip_state = vcn_v4_0_dump_ip_state, - .print_ip_state = vcn_v4_0_print_ip_state, + .dump_ip_state = amdgpu_vcn_dump_ip_state, + .print_ip_state = amdgpu_vcn_print_ip_state, }; const struct amdgpu_ip_block_version vcn_v4_0_ip_block = { diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c index 2a3663b551af..7b93a275ec4f 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c @@ -134,6 +134,19 @@ static int vcn_v4_0_3_early_init(struct amdgpu_ip_block *ip_block) return 0; } +static int vcn_v4_0_3_late_init(struct amdgpu_ip_block *ip_block) +{ + struct amdgpu_device *adev = ip_block->adev; + + adev->vcn.supported_reset = + amdgpu_get_soft_full_reset_mask(&adev->vcn.inst[0].ring_enc[0]); + + if (amdgpu_dpm_reset_vcn_is_supported(adev)) + adev->vcn.supported_reset |= AMDGPU_RESET_TYPE_PER_QUEUE; + + return 0; +} + static int vcn_v4_0_3_fw_shared_init(struct amdgpu_device *adev, int inst_idx) { struct amdgpu_vcn4_fw_shared *fw_shared; @@ -160,8 +173,6 @@ static int vcn_v4_0_3_sw_init(struct amdgpu_ip_block *ip_block) struct amdgpu_device *adev = ip_block->adev; struct amdgpu_ring *ring; int i, r, vcn_inst; - uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_4_0_3); - uint32_t *ptr; /* VCN DEC TRAP */ r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_VCN, @@ -213,10 +224,6 @@ static int vcn_v4_0_3_sw_init(struct amdgpu_ip_block *ip_block) adev->vcn.inst[i].pause_dpg_mode = vcn_v4_0_3_pause_dpg_mode; } - /* TODO: Add queue reset mask when FW fully supports it */ - adev->vcn.supported_reset = - amdgpu_get_soft_full_reset_mask(&adev->vcn.inst[0].ring_enc[0]); - if (amdgpu_sriov_vf(adev)) { r = amdgpu_virt_alloc_mm_table(adev); if (r) @@ -231,20 +238,11 @@ static int vcn_v4_0_3_sw_init(struct amdgpu_ip_block *ip_block) } } - /* Allocate memory for VCN IP Dump buffer */ - ptr = kcalloc(adev->vcn.num_vcn_inst * reg_count, sizeof(uint32_t), GFP_KERNEL); - if (!ptr) { - DRM_ERROR("Failed to allocate memory for VCN IP Dump\n"); - adev->vcn.ip_dump = NULL; - } else { - adev->vcn.ip_dump = ptr; - } - - r = amdgpu_vcn_sysfs_reset_mask_init(adev); + r = amdgpu_vcn_reg_dump_init(adev, vcn_reg_list_4_0_3, ARRAY_SIZE(vcn_reg_list_4_0_3)); if (r) return r; - return 0; + return amdgpu_vcn_sysfs_reset_mask_init(adev); } /** @@ -287,8 +285,6 @@ static int vcn_v4_0_3_sw_fini(struct amdgpu_ip_block *ip_block) return r; } - kfree(adev->vcn.ip_dump); - return 0; } @@ -391,7 +387,7 @@ static int vcn_v4_0_3_hw_fini(struct amdgpu_ip_block *ip_block) vinst->set_pg_state(vinst, AMD_PG_STATE_GATE); } - if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__VCN)) + if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__VCN) && !amdgpu_sriov_vf(adev)) amdgpu_irq_put(adev, &adev->vcn.inst->ras_poison_irq, 0); return 0; @@ -851,7 +847,7 @@ static int vcn_v4_0_3_start_dpg_mode(struct amdgpu_vcn_inst *vinst, volatile struct amdgpu_vcn4_fw_shared *fw_shared = adev->vcn.inst[inst_idx].fw_shared.cpu_addr; struct amdgpu_ring *ring; - int vcn_inst; + int vcn_inst, ret; uint32_t tmp; vcn_inst = GET_INST(VCN, inst_idx); @@ -944,8 +940,13 @@ static int vcn_v4_0_3_start_dpg_mode(struct amdgpu_vcn_inst *vinst, VCN, 0, regUVD_MASTINT_EN), UVD_MASTINT_EN__VCPU_EN_MASK, 0, indirect); - if (indirect) - amdgpu_vcn_psp_update_sram(adev, inst_idx, AMDGPU_UCODE_ID_VCN0_RAM); + if (indirect) { + ret = amdgpu_vcn_psp_update_sram(adev, inst_idx, AMDGPU_UCODE_ID_VCN0_RAM); + if (ret) { + dev_err(adev->dev, "vcn sram load failed %d\n", ret); + return ret; + } + } ring = &adev->vcn.inst[inst_idx].ring_enc[0]; @@ -1872,71 +1873,10 @@ static void vcn_v4_0_3_set_irq_funcs(struct amdgpu_device *adev) adev->vcn.inst->ras_poison_irq.funcs = &vcn_v4_0_3_ras_irq_funcs; } -static void vcn_v4_0_3_print_ip_state(struct amdgpu_ip_block *ip_block, struct drm_printer *p) -{ - struct amdgpu_device *adev = ip_block->adev; - int i, j; - uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_4_0_3); - uint32_t inst_off, is_powered; - - if (!adev->vcn.ip_dump) - return; - - drm_printf(p, "num_instances:%d\n", adev->vcn.num_vcn_inst); - for (i = 0; i < adev->vcn.num_vcn_inst; i++) { - if (adev->vcn.harvest_config & (1 << i)) { - drm_printf(p, "\nHarvested Instance:VCN%d Skipping dump\n", i); - continue; - } - - inst_off = i * reg_count; - is_powered = (adev->vcn.ip_dump[inst_off] & - UVD_POWER_STATUS__UVD_POWER_STATUS_MASK) != 1; - - if (is_powered) { - drm_printf(p, "\nActive Instance:VCN%d\n", i); - for (j = 0; j < reg_count; j++) - drm_printf(p, "%-50s \t 0x%08x\n", vcn_reg_list_4_0_3[j].reg_name, - adev->vcn.ip_dump[inst_off + j]); - } else { - drm_printf(p, "\nInactive Instance:VCN%d\n", i); - } - } -} - -static void vcn_v4_0_3_dump_ip_state(struct amdgpu_ip_block *ip_block) -{ - struct amdgpu_device *adev = ip_block->adev; - int i, j; - bool is_powered; - uint32_t inst_off, inst_id; - uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_4_0_3); - - if (!adev->vcn.ip_dump) - return; - - for (i = 0; i < adev->vcn.num_vcn_inst; i++) { - if (adev->vcn.harvest_config & (1 << i)) - continue; - - inst_id = GET_INST(VCN, i); - inst_off = i * reg_count; - /* mmUVD_POWER_STATUS is always readable and is first element of the array */ - adev->vcn.ip_dump[inst_off] = RREG32_SOC15(VCN, inst_id, regUVD_POWER_STATUS); - is_powered = (adev->vcn.ip_dump[inst_off] & - UVD_POWER_STATUS__UVD_POWER_STATUS_MASK) != 1; - - if (is_powered) - for (j = 1; j < reg_count; j++) - adev->vcn.ip_dump[inst_off + j] = - RREG32(SOC15_REG_ENTRY_OFFSET_INST(vcn_reg_list_4_0_3[j], - inst_id)); - } -} - static const struct amd_ip_funcs vcn_v4_0_3_ip_funcs = { .name = "vcn_v4_0_3", .early_init = vcn_v4_0_3_early_init, + .late_init = vcn_v4_0_3_late_init, .sw_init = vcn_v4_0_3_sw_init, .sw_fini = vcn_v4_0_3_sw_fini, .hw_init = vcn_v4_0_3_hw_init, @@ -1947,8 +1887,8 @@ static const struct amd_ip_funcs vcn_v4_0_3_ip_funcs = { .wait_for_idle = vcn_v4_0_3_wait_for_idle, .set_clockgating_state = vcn_v4_0_3_set_clockgating_state, .set_powergating_state = vcn_set_powergating_state, - .dump_ip_state = vcn_v4_0_3_dump_ip_state, - .print_ip_state = vcn_v4_0_3_print_ip_state, + .dump_ip_state = amdgpu_vcn_dump_ip_state, + .print_ip_state = amdgpu_vcn_print_ip_state, }; const struct amdgpu_ip_block_version vcn_v4_0_3_ip_block = { diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c index caf2d95a85d4..75c884a8f556 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c @@ -147,9 +147,6 @@ static int vcn_v4_0_5_sw_init(struct amdgpu_ip_block *ip_block) struct amdgpu_ring *ring; struct amdgpu_device *adev = ip_block->adev; int i, r; - uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_4_0_5); - uint32_t *ptr; - for (i = 0; i < adev->vcn.num_vcn_inst; i++) { volatile struct amdgpu_vcn4_fw_shared *fw_shared; @@ -233,15 +230,9 @@ static int vcn_v4_0_5_sw_init(struct amdgpu_ip_block *ip_block) return r; } - /* Allocate memory for VCN IP Dump buffer */ - ptr = kcalloc(adev->vcn.num_vcn_inst * reg_count, sizeof(uint32_t), GFP_KERNEL); - if (!ptr) { - DRM_ERROR("Failed to allocate memory for VCN IP Dump\n"); - adev->vcn.ip_dump = NULL; - } else { - adev->vcn.ip_dump = ptr; - } - return 0; + r = amdgpu_vcn_reg_dump_init(adev, vcn_reg_list_4_0_5, ARRAY_SIZE(vcn_reg_list_4_0_5)); + + return r; } /** @@ -284,8 +275,6 @@ static int vcn_v4_0_5_sw_fini(struct amdgpu_ip_block *ip_block) return r; } - kfree(adev->vcn.ip_dump); - return 0; } @@ -926,6 +915,7 @@ static int vcn_v4_0_5_start_dpg_mode(struct amdgpu_vcn_inst *vinst, volatile struct amdgpu_vcn4_fw_shared *fw_shared = adev->vcn.inst[inst_idx].fw_shared.cpu_addr; struct amdgpu_ring *ring; uint32_t tmp; + int ret; /* disable register anti-hang mechanism */ WREG32_P(SOC15_REG_OFFSET(VCN, inst_idx, regUVD_POWER_STATUS), 1, @@ -1006,8 +996,13 @@ static int vcn_v4_0_5_start_dpg_mode(struct amdgpu_vcn_inst *vinst, VCN, inst_idx, regUVD_MASTINT_EN), UVD_MASTINT_EN__VCPU_EN_MASK, 0, indirect); - if (indirect) - amdgpu_vcn_psp_update_sram(adev, inst_idx, 0); + if (indirect) { + ret = amdgpu_vcn_psp_update_sram(adev, inst_idx, 0); + if (ret) { + dev_err(adev->dev, "vcn sram load failed %d\n", ret); + return ret; + } + } ring = &adev->vcn.inst[inst_idx].ring_enc[0]; @@ -1704,67 +1699,6 @@ static void vcn_v4_0_5_set_irq_funcs(struct amdgpu_device *adev) } } -static void vcn_v4_0_5_print_ip_state(struct amdgpu_ip_block *ip_block, struct drm_printer *p) -{ - struct amdgpu_device *adev = ip_block->adev; - int i, j; - uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_4_0_5); - uint32_t inst_off, is_powered; - - if (!adev->vcn.ip_dump) - return; - - drm_printf(p, "num_instances:%d\n", adev->vcn.num_vcn_inst); - for (i = 0; i < adev->vcn.num_vcn_inst; i++) { - if (adev->vcn.harvest_config & (1 << i)) { - drm_printf(p, "\nHarvested Instance:VCN%d Skipping dump\n", i); - continue; - } - - inst_off = i * reg_count; - is_powered = (adev->vcn.ip_dump[inst_off] & - UVD_POWER_STATUS__UVD_POWER_STATUS_MASK) != 1; - - if (is_powered) { - drm_printf(p, "\nActive Instance:VCN%d\n", i); - for (j = 0; j < reg_count; j++) - drm_printf(p, "%-50s \t 0x%08x\n", vcn_reg_list_4_0_5[j].reg_name, - adev->vcn.ip_dump[inst_off + j]); - } else { - drm_printf(p, "\nInactive Instance:VCN%d\n", i); - } - } -} - -static void vcn_v4_0_5_dump_ip_state(struct amdgpu_ip_block *ip_block) -{ - struct amdgpu_device *adev = ip_block->adev; - int i, j; - bool is_powered; - uint32_t inst_off; - uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_4_0_5); - - if (!adev->vcn.ip_dump) - return; - - for (i = 0; i < adev->vcn.num_vcn_inst; i++) { - if (adev->vcn.harvest_config & (1 << i)) - continue; - - inst_off = i * reg_count; - /* mmUVD_POWER_STATUS is always readable and is first element of the array */ - adev->vcn.ip_dump[inst_off] = RREG32_SOC15(VCN, i, regUVD_POWER_STATUS); - is_powered = (adev->vcn.ip_dump[inst_off] & - UVD_POWER_STATUS__UVD_POWER_STATUS_MASK) != 1; - - if (is_powered) - for (j = 1; j < reg_count; j++) - adev->vcn.ip_dump[inst_off + j] = - RREG32(SOC15_REG_ENTRY_OFFSET_INST(vcn_reg_list_4_0_5[j], - i)); - } -} - static const struct amd_ip_funcs vcn_v4_0_5_ip_funcs = { .name = "vcn_v4_0_5", .early_init = vcn_v4_0_5_early_init, @@ -1778,8 +1712,8 @@ static const struct amd_ip_funcs vcn_v4_0_5_ip_funcs = { .wait_for_idle = vcn_v4_0_5_wait_for_idle, .set_clockgating_state = vcn_v4_0_5_set_clockgating_state, .set_powergating_state = vcn_set_powergating_state, - .dump_ip_state = vcn_v4_0_5_dump_ip_state, - .print_ip_state = vcn_v4_0_5_print_ip_state, + .dump_ip_state = amdgpu_vcn_dump_ip_state, + .print_ip_state = amdgpu_vcn_print_ip_state, }; const struct amdgpu_ip_block_version vcn_v4_0_5_ip_block = { diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c index 07a6e9582880..455f829b8bb9 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c @@ -115,21 +115,6 @@ static int vcn_v5_0_0_early_init(struct amdgpu_ip_block *ip_block) return 0; } -void vcn_v5_0_0_alloc_ip_dump(struct amdgpu_device *adev) -{ - uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_5_0); - uint32_t *ptr; - - /* Allocate memory for VCN IP Dump buffer */ - ptr = kcalloc(adev->vcn.num_vcn_inst * reg_count, sizeof(uint32_t), GFP_KERNEL); - if (!ptr) { - DRM_ERROR("Failed to allocate memory for VCN IP Dump\n"); - adev->vcn.ip_dump = NULL; - } else { - adev->vcn.ip_dump = ptr; - } -} - /** * vcn_v5_0_0_sw_init - sw init for VCN block * @@ -201,7 +186,9 @@ static int vcn_v5_0_0_sw_init(struct amdgpu_ip_block *ip_block) if (!amdgpu_sriov_vf(adev)) adev->vcn.supported_reset |= AMDGPU_RESET_TYPE_PER_QUEUE; - vcn_v5_0_0_alloc_ip_dump(adev); + r = amdgpu_vcn_reg_dump_init(adev, vcn_reg_list_5_0, ARRAY_SIZE(vcn_reg_list_5_0)); + if (r) + return r; r = amdgpu_vcn_sysfs_reset_mask_init(adev); if (r) @@ -251,8 +238,6 @@ static int vcn_v5_0_0_sw_fini(struct amdgpu_ip_block *ip_block) return r; } - kfree(adev->vcn.ip_dump); - return 0; } @@ -713,6 +698,7 @@ static int vcn_v5_0_0_start_dpg_mode(struct amdgpu_vcn_inst *vinst, volatile struct amdgpu_vcn5_fw_shared *fw_shared = adev->vcn.inst[inst_idx].fw_shared.cpu_addr; struct amdgpu_ring *ring; uint32_t tmp; + int ret; /* disable register anti-hang mechanism */ WREG32_P(SOC15_REG_OFFSET(VCN, inst_idx, regUVD_POWER_STATUS), 1, @@ -766,8 +752,13 @@ static int vcn_v5_0_0_start_dpg_mode(struct amdgpu_vcn_inst *vinst, VCN, inst_idx, regUVD_MASTINT_EN), UVD_MASTINT_EN__VCPU_EN_MASK, 0, indirect); - if (indirect) - amdgpu_vcn_psp_update_sram(adev, inst_idx, 0); + if (indirect) { + ret = amdgpu_vcn_psp_update_sram(adev, inst_idx, 0); + if (ret) { + dev_err(adev->dev, "%s: vcn sram load failed %d\n", __func__, ret); + return ret; + } + } ring = &adev->vcn.inst[inst_idx].ring_enc[0]; @@ -1428,67 +1419,6 @@ static void vcn_v5_0_0_set_irq_funcs(struct amdgpu_device *adev) } } -void vcn_v5_0_0_print_ip_state(struct amdgpu_ip_block *ip_block, - struct drm_printer *p) -{ - struct amdgpu_device *adev = ip_block->adev; - int i, j; - uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_5_0); - uint32_t inst_off, is_powered; - - if (!adev->vcn.ip_dump) - return; - - drm_printf(p, "num_instances:%d\n", adev->vcn.num_vcn_inst); - for (i = 0; i < adev->vcn.num_vcn_inst; i++) { - if (adev->vcn.harvest_config & (1 << i)) { - drm_printf(p, "\nHarvested Instance:VCN%d Skipping dump\n", i); - continue; - } - - inst_off = i * reg_count; - is_powered = (adev->vcn.ip_dump[inst_off] & - UVD_POWER_STATUS__UVD_POWER_STATUS_MASK) != 1; - - if (is_powered) { - drm_printf(p, "\nActive Instance:VCN%d\n", i); - for (j = 0; j < reg_count; j++) - drm_printf(p, "%-50s \t 0x%08x\n", vcn_reg_list_5_0[j].reg_name, - adev->vcn.ip_dump[inst_off + j]); - } else { - drm_printf(p, "\nInactive Instance:VCN%d\n", i); - } - } -} - -void vcn_v5_0_0_dump_ip_state(struct amdgpu_ip_block *ip_block) -{ - struct amdgpu_device *adev = ip_block->adev; - int i, j; - bool is_powered; - uint32_t inst_off; - uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_5_0); - - if (!adev->vcn.ip_dump) - return; - - for (i = 0; i < adev->vcn.num_vcn_inst; i++) { - if (adev->vcn.harvest_config & (1 << i)) - continue; - - inst_off = i * reg_count; - /* mmUVD_POWER_STATUS is always readable and is first element of the array */ - adev->vcn.ip_dump[inst_off] = RREG32_SOC15(VCN, i, regUVD_POWER_STATUS); - is_powered = (adev->vcn.ip_dump[inst_off] & - UVD_POWER_STATUS__UVD_POWER_STATUS_MASK) != 1; - - if (is_powered) - for (j = 1; j < reg_count; j++) - adev->vcn.ip_dump[inst_off + j] = - RREG32(SOC15_REG_ENTRY_OFFSET_INST(vcn_reg_list_5_0[j], i)); - } -} - static const struct amd_ip_funcs vcn_v5_0_0_ip_funcs = { .name = "vcn_v5_0_0", .early_init = vcn_v5_0_0_early_init, @@ -1502,8 +1432,8 @@ static const struct amd_ip_funcs vcn_v5_0_0_ip_funcs = { .wait_for_idle = vcn_v5_0_0_wait_for_idle, .set_clockgating_state = vcn_v5_0_0_set_clockgating_state, .set_powergating_state = vcn_set_powergating_state, - .dump_ip_state = vcn_v5_0_0_dump_ip_state, - .print_ip_state = vcn_v5_0_0_print_ip_state, + .dump_ip_state = amdgpu_vcn_dump_ip_state, + .print_ip_state = amdgpu_vcn_print_ip_state, }; const struct amdgpu_ip_block_version vcn_v5_0_0_ip_block = { diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.h b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.h index b8927652bc50..51bbccd4360f 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.h +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.h @@ -32,11 +32,6 @@ #define VCN_VID_IP_ADDRESS 0x0 #define VCN_AON_IP_ADDRESS 0x30000 -void vcn_v5_0_0_alloc_ip_dump(struct amdgpu_device *adev); -void vcn_v5_0_0_print_ip_state(struct amdgpu_ip_block *ip_block, - struct drm_printer *p); -void vcn_v5_0_0_dump_ip_state(struct amdgpu_ip_block *ip_block); - extern const struct amdgpu_ip_block_version vcn_v5_0_0_ip_block; #endif /* __VCN_V5_0_0_H__ */ diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c index cdefd7fcb0da..4b01e35ad7ef 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c @@ -40,6 +40,40 @@ #include <drm/drm_drv.h> +static const struct amdgpu_hwip_reg_entry vcn_reg_list_5_0_1[] = { + SOC15_REG_ENTRY_STR(VCN, 0, regUVD_POWER_STATUS), + SOC15_REG_ENTRY_STR(VCN, 0, regUVD_STATUS), + SOC15_REG_ENTRY_STR(VCN, 0, regUVD_CONTEXT_ID), + SOC15_REG_ENTRY_STR(VCN, 0, regUVD_CONTEXT_ID2), + SOC15_REG_ENTRY_STR(VCN, 0, regUVD_GPCOM_VCPU_DATA0), + SOC15_REG_ENTRY_STR(VCN, 0, regUVD_GPCOM_VCPU_DATA1), + SOC15_REG_ENTRY_STR(VCN, 0, regUVD_GPCOM_VCPU_CMD), + SOC15_REG_ENTRY_STR(VCN, 0, regUVD_RB_BASE_HI), + SOC15_REG_ENTRY_STR(VCN, 0, regUVD_RB_BASE_LO), + SOC15_REG_ENTRY_STR(VCN, 0, regUVD_RB_BASE_HI2), + SOC15_REG_ENTRY_STR(VCN, 0, regUVD_RB_BASE_LO2), + SOC15_REG_ENTRY_STR(VCN, 0, regUVD_RB_BASE_HI3), + SOC15_REG_ENTRY_STR(VCN, 0, regUVD_RB_BASE_LO3), + SOC15_REG_ENTRY_STR(VCN, 0, regUVD_RB_BASE_HI4), + SOC15_REG_ENTRY_STR(VCN, 0, regUVD_RB_BASE_LO4), + SOC15_REG_ENTRY_STR(VCN, 0, regUVD_RB_RPTR), + SOC15_REG_ENTRY_STR(VCN, 0, regUVD_RB_WPTR), + SOC15_REG_ENTRY_STR(VCN, 0, regUVD_RB_RPTR2), + SOC15_REG_ENTRY_STR(VCN, 0, regUVD_RB_WPTR2), + SOC15_REG_ENTRY_STR(VCN, 0, regUVD_RB_RPTR3), + SOC15_REG_ENTRY_STR(VCN, 0, regUVD_RB_WPTR3), + SOC15_REG_ENTRY_STR(VCN, 0, regUVD_RB_RPTR4), + SOC15_REG_ENTRY_STR(VCN, 0, regUVD_RB_WPTR4), + SOC15_REG_ENTRY_STR(VCN, 0, regUVD_RB_SIZE), + SOC15_REG_ENTRY_STR(VCN, 0, regUVD_RB_SIZE2), + SOC15_REG_ENTRY_STR(VCN, 0, regUVD_RB_SIZE3), + SOC15_REG_ENTRY_STR(VCN, 0, regUVD_RB_SIZE4), + SOC15_REG_ENTRY_STR(VCN, 0, regUVD_DPG_LMA_CTL), + SOC15_REG_ENTRY_STR(VCN, 0, regUVD_DPG_LMA_DATA), + SOC15_REG_ENTRY_STR(VCN, 0, regUVD_DPG_LMA_MASK), + SOC15_REG_ENTRY_STR(VCN, 0, regUVD_DPG_PAUSE) +}; + static int vcn_v5_0_1_start_sriov(struct amdgpu_device *adev); static void vcn_v5_0_1_set_unified_ring_funcs(struct amdgpu_device *adev); static void vcn_v5_0_1_set_irq_funcs(struct amdgpu_device *adev); @@ -163,7 +197,17 @@ static int vcn_v5_0_1_sw_init(struct amdgpu_ip_block *ip_block) return r; } - vcn_v5_0_0_alloc_ip_dump(adev); + if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__VCN)) { + r = amdgpu_vcn_ras_sw_init(adev); + if (r) { + dev_err(adev->dev, "Failed to initialize vcn ras block!\n"); + return r; + } + } + + r = amdgpu_vcn_reg_dump_init(adev, vcn_reg_list_5_0_1, ARRAY_SIZE(vcn_reg_list_5_0_1)); + if (r) + return r; return amdgpu_vcn_sysfs_reset_mask_init(adev); } @@ -209,8 +253,6 @@ static int vcn_v5_0_1_sw_fini(struct amdgpu_ip_block *ip_block) amdgpu_vcn_sysfs_reset_mask_fini(adev); - kfree(adev->vcn.ip_dump); - return 0; } @@ -284,7 +326,7 @@ static int vcn_v5_0_1_hw_fini(struct amdgpu_ip_block *ip_block) vinst->set_pg_state(vinst, AMD_PG_STATE_GATE); } - if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__VCN)) + if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__VCN) && !amdgpu_sriov_vf(adev)) amdgpu_irq_put(adev, &adev->vcn.inst->ras_poison_irq, 0); return 0; @@ -605,7 +647,7 @@ static int vcn_v5_0_1_start_dpg_mode(struct amdgpu_vcn_inst *vinst, adev->vcn.inst[inst_idx].fw_shared.cpu_addr; struct amdgpu_ring *ring; struct dpg_pause_state state = {.fw_based = VCN_DPG_STATE__PAUSE}; - int vcn_inst; + int vcn_inst, ret; uint32_t tmp; vcn_inst = GET_INST(VCN, inst_idx); @@ -666,8 +708,13 @@ static int vcn_v5_0_1_start_dpg_mode(struct amdgpu_vcn_inst *vinst, VCN, 0, regUVD_MASTINT_EN), UVD_MASTINT_EN__VCPU_EN_MASK, 0, indirect); - if (indirect) - amdgpu_vcn_psp_update_sram(adev, inst_idx, AMDGPU_UCODE_ID_VCN0_RAM); + if (indirect) { + ret = amdgpu_vcn_psp_update_sram(adev, inst_idx, AMDGPU_UCODE_ID_VCN0_RAM); + if (ret) { + dev_err(adev->dev, "vcn sram load failed %d\n", ret); + return ret; + } + } /* resetting ring, fw should not check RB ring */ fw_shared->sq.queue_mode |= FW_QUEUE_RING_RESET; @@ -1475,8 +1522,8 @@ static const struct amd_ip_funcs vcn_v5_0_1_ip_funcs = { .post_soft_reset = NULL, .set_clockgating_state = vcn_v5_0_1_set_clockgating_state, .set_powergating_state = vcn_set_powergating_state, - .dump_ip_state = vcn_v5_0_0_dump_ip_state, - .print_ip_state = vcn_v5_0_0_print_ip_state, + .dump_ip_state = amdgpu_vcn_dump_ip_state, + .print_ip_state = amdgpu_vcn_print_ip_state, }; const struct amdgpu_ip_block_version vcn_v5_0_1_ip_block = { @@ -1557,7 +1604,7 @@ static int vcn_v5_0_1_aca_bank_parser(struct aca_handle *handle, struct aca_bank /* reference to smu driver if header file */ static int vcn_v5_0_1_err_codes[] = { - 14, 15, /* VCN */ + 14, 15, 47, /* VCN [D|V|S] */ }; static bool vcn_v5_0_1_aca_bank_is_valid(struct aca_handle *handle, struct aca_bank *bank, @@ -1603,6 +1650,13 @@ static int vcn_v5_0_1_ras_late_init(struct amdgpu_device *adev, struct ras_commo if (r) goto late_fini; + if (amdgpu_ras_is_supported(adev, ras_block->block) && + adev->vcn.inst->ras_poison_irq.funcs) { + r = amdgpu_irq_get(adev, &adev->vcn.inst->ras_poison_irq, 0); + if (r) + goto late_fini; + } + return 0; late_fini: diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c index 828a9ceef1e7..43115a374469 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c @@ -1070,7 +1070,12 @@ static int kfd_ioctl_alloc_memory_of_gpu(struct file *filep, svm_range_list_lock_and_flush_work(&p->svms, current->mm); mutex_lock(&p->svms.lock); mmap_write_unlock(current->mm); - if (interval_tree_iter_first(&p->svms.objects, + + /* Skip a special case that allocates VRAM without VA, + * VA will be invalid of 0. + */ + if (!(!args->va_addr && (flags & KFD_IOC_ALLOC_MEM_FLAGS_VRAM)) && + interval_tree_iter_first(&p->svms.objects, args->va_addr >> PAGE_SHIFT, (args->va_addr + args->size - 1) >> PAGE_SHIFT)) { pr_err("Address: 0x%llx already allocated by SVM\n", @@ -3252,8 +3257,10 @@ static long kfd_ioctl(struct file *filep, unsigned int cmd, unsigned long arg) int retcode = -EINVAL; bool ptrace_attached = false; - if (nr >= AMDKFD_CORE_IOCTL_COUNT) + if (nr >= AMDKFD_CORE_IOCTL_COUNT) { + retcode = -ENOTTY; goto err_i1; + } if ((nr >= AMDKFD_COMMAND_START) && (nr < AMDKFD_COMMAND_END)) { u32 amdkfd_size; @@ -3266,8 +3273,10 @@ static long kfd_ioctl(struct file *filep, unsigned int cmd, unsigned long arg) asize = amdkfd_size; cmd = ioctl->cmd; - } else + } else { + retcode = -ENOTTY; goto err_i1; + } dev_dbg(kfd_device, "ioctl cmd 0x%x (#0x%x), arg 0x%lx\n", cmd, nr, arg); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c b/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c index 2b0a830f5b29..fb3129883a4c 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c @@ -46,11 +46,7 @@ static bool kq_initialize(struct kernel_queue *kq, struct kfd_node *dev, int retval; union PM4_MES_TYPE_3_HEADER nop; - if (WARN_ON(type != KFD_QUEUE_TYPE_DIQ && type != KFD_QUEUE_TYPE_HIQ)) - return false; - - pr_debug("Initializing queue type %d size %d\n", KFD_QUEUE_TYPE_HIQ, - queue_size); + pr_debug("Initializing queue type %d size %d\n", type, queue_size); memset(&prop, 0, sizeof(prop)); memset(&nop, 0, sizeof(nop)); @@ -69,6 +65,7 @@ static bool kq_initialize(struct kernel_queue *kq, struct kfd_node *dev, kq->mqd_mgr = dev->dqm->mqd_mgrs[KFD_MQD_TYPE_HIQ]; break; default: + WARN(1, "Invalid queue type %d\n", type); dev_err(dev->adev->dev, "Invalid queue type %d\n", type); return false; } diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c index 683ff02c45af..86315ecb6f1d 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c @@ -261,20 +261,7 @@ static void svm_migrate_put_sys_page(unsigned long addr) put_page(page); } -static unsigned long svm_migrate_unsuccessful_pages(struct migrate_vma *migrate) -{ - unsigned long upages = 0; - unsigned long i; - - for (i = 0; i < migrate->npages; i++) { - if (migrate->src[i] & MIGRATE_PFN_VALID && - !(migrate->src[i] & MIGRATE_PFN_MIGRATE)) - upages++; - } - return upages; -} - -static int +static long svm_migrate_copy_to_vram(struct kfd_node *node, struct svm_range *prange, struct migrate_vma *migrate, struct dma_fence **mfence, dma_addr_t *scratch, uint64_t ttm_res_offset) @@ -283,7 +270,7 @@ svm_migrate_copy_to_vram(struct kfd_node *node, struct svm_range *prange, struct amdgpu_device *adev = node->adev; struct device *dev = adev->dev; struct amdgpu_res_cursor cursor; - uint64_t mpages = 0; + long mpages; dma_addr_t *src; uint64_t *dst; uint64_t i, j; @@ -297,6 +284,7 @@ svm_migrate_copy_to_vram(struct kfd_node *node, struct svm_range *prange, amdgpu_res_first(prange->ttm_res, ttm_res_offset, npages << PAGE_SHIFT, &cursor); + mpages = 0; for (i = j = 0; (i < npages) && (mpages < migrate->cpages); i++) { struct page *spage; @@ -357,13 +345,14 @@ svm_migrate_copy_to_vram(struct kfd_node *node, struct svm_range *prange, out_free_vram_pages: if (r) { pr_debug("failed %d to copy memory to vram\n", r); - for (i = 0; i < npages && mpages; i++) { + while (i-- && mpages) { if (!dst[i]) continue; svm_migrate_put_vram_page(adev, dst[i]); migrate->dst[i] = 0; mpages--; } + mpages = r; } #ifdef DEBUG_FORCE_MIXED_DOMAINS @@ -381,7 +370,7 @@ out_free_vram_pages: } #endif - return r; + return mpages; } static long @@ -396,7 +385,7 @@ svm_migrate_vma_to_vram(struct kfd_node *node, struct svm_range *prange, struct dma_fence *mfence = NULL; struct migrate_vma migrate = { 0 }; unsigned long cpages = 0; - unsigned long mpages = 0; + long mpages = 0; dma_addr_t *scratch; void *buf; int r = -ENOMEM; @@ -442,15 +431,17 @@ svm_migrate_vma_to_vram(struct kfd_node *node, struct svm_range *prange, else pr_debug("0x%lx pages collected\n", cpages); - r = svm_migrate_copy_to_vram(node, prange, &migrate, &mfence, scratch, ttm_res_offset); + mpages = svm_migrate_copy_to_vram(node, prange, &migrate, &mfence, scratch, ttm_res_offset); migrate_vma_pages(&migrate); svm_migrate_copy_done(adev, mfence); migrate_vma_finalize(&migrate); - mpages = cpages - svm_migrate_unsuccessful_pages(&migrate); - pr_debug("successful/cpages/npages 0x%lx/0x%lx/0x%lx\n", + if (mpages >= 0) + pr_debug("migrated/collected/requested 0x%lx/0x%lx/0x%lx\n", mpages, cpages, migrate.npages); + else + r = mpages; svm_range_dma_unmap_dev(adev->dev, scratch, 0, npages); @@ -460,14 +451,13 @@ out_free: start >> PAGE_SHIFT, end >> PAGE_SHIFT, 0, node->id, trigger, r); out: - if (!r && mpages) { + if (!r && mpages > 0) { pdd = svm_range_get_pdd_by_node(prange, node); if (pdd) WRITE_ONCE(pdd->page_in, pdd->page_in + mpages); - - return mpages; } - return r; + + return r ? r : mpages; } /** @@ -578,7 +568,7 @@ static void svm_migrate_page_free(struct page *page) } } -static int +static long svm_migrate_copy_to_ram(struct amdgpu_device *adev, struct svm_range *prange, struct migrate_vma *migrate, struct dma_fence **mfence, dma_addr_t *scratch, uint64_t npages) @@ -587,6 +577,7 @@ svm_migrate_copy_to_ram(struct amdgpu_device *adev, struct svm_range *prange, uint64_t *src; dma_addr_t *dst; struct page *dpage; + long mpages; uint64_t i = 0, j; uint64_t addr; int r = 0; @@ -599,6 +590,7 @@ svm_migrate_copy_to_ram(struct amdgpu_device *adev, struct svm_range *prange, src = (uint64_t *)(scratch + npages); dst = scratch; + mpages = 0; for (i = 0, j = 0; i < npages; i++, addr += PAGE_SIZE) { struct page *spage; @@ -647,6 +639,7 @@ svm_migrate_copy_to_ram(struct amdgpu_device *adev, struct svm_range *prange, dst[i] >> PAGE_SHIFT, page_to_pfn(dpage)); migrate->dst[i] = migrate_pfn(page_to_pfn(dpage)); + mpages++; j++; } @@ -656,13 +649,17 @@ svm_migrate_copy_to_ram(struct amdgpu_device *adev, struct svm_range *prange, out_oom: if (r) { pr_debug("failed %d copy to ram\n", r); - while (i--) { + while (i-- && mpages) { + if (!migrate->dst[i]) + continue; svm_migrate_put_sys_page(dst[i]); migrate->dst[i] = 0; + mpages--; } + mpages = r; } - return r; + return mpages; } /** @@ -689,9 +686,8 @@ svm_migrate_vma_to_ram(struct kfd_node *node, struct svm_range *prange, { struct kfd_process *p = container_of(prange->svms, struct kfd_process, svms); uint64_t npages = (end - start) >> PAGE_SHIFT; - unsigned long upages = npages; unsigned long cpages = 0; - unsigned long mpages = 0; + long mpages = 0; struct amdgpu_device *adev = node->adev; struct kfd_process_device *pdd; struct dma_fence *mfence = NULL; @@ -737,7 +733,6 @@ svm_migrate_vma_to_ram(struct kfd_node *node, struct svm_range *prange, if (!cpages) { pr_debug("failed collect migrate device pages [0x%lx 0x%lx]\n", prange->start, prange->last); - upages = svm_migrate_unsuccessful_pages(&migrate); goto out_free; } if (cpages != npages) @@ -746,13 +741,15 @@ svm_migrate_vma_to_ram(struct kfd_node *node, struct svm_range *prange, else pr_debug("0x%lx pages collected\n", cpages); - r = svm_migrate_copy_to_ram(adev, prange, &migrate, &mfence, + mpages = svm_migrate_copy_to_ram(adev, prange, &migrate, &mfence, scratch, npages); migrate_vma_pages(&migrate); - upages = svm_migrate_unsuccessful_pages(&migrate); - pr_debug("unsuccessful/cpages/npages 0x%lx/0x%lx/0x%lx\n", - upages, cpages, migrate.npages); + if (mpages >= 0) + pr_debug("migrated/collected/requested 0x%lx/0x%lx/0x%lx\n", + mpages, cpages, migrate.npages); + else + r = mpages; svm_migrate_copy_done(adev, mfence); migrate_vma_finalize(&migrate); @@ -765,8 +762,7 @@ out_free: start >> PAGE_SHIFT, end >> PAGE_SHIFT, node->id, 0, trigger, r); out: - if (!r && cpages) { - mpages = cpages - upages; + if (!r && mpages > 0) { pdd = svm_range_get_pdd_by_node(prange, node); if (pdd) WRITE_ONCE(pdd->page_out, pdd->page_out + mpages); @@ -849,6 +845,9 @@ int svm_migrate_vram_to_ram(struct svm_range *prange, struct mm_struct *mm, } if (r >= 0) { + WARN_ONCE(prange->vram_pages < mpages, + "Recorded vram pages(0x%llx) should not be less than migration pages(0x%lx).", + prange->vram_pages, mpages); prange->vram_pages -= mpages; /* prange does not have vram page set its actual_loc to system diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h index 67694bcd9464..d01ef5ac0766 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h +++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h @@ -111,7 +111,14 @@ #define KFD_KERNEL_QUEUE_SIZE 2048 -#define KFD_UNMAP_LATENCY_MS (4000) +/* KFD_UNMAP_LATENCY_MS is the timeout CP waiting for SDMA preemption. One XCC + * can be associated to 2 SDMA engines. queue_preemption_timeout_ms is the time + * driver waiting for CP returning the UNMAP_QUEUE fence. Thus the math is + * queue_preemption_timeout_ms = sdma_preemption_time * 2 + cp workload + * The format here makes CP workload 10% of total timeout + */ +#define KFD_UNMAP_LATENCY_MS \ + ((queue_preemption_timeout_ms - queue_preemption_timeout_ms / 10) >> 1) #define KFD_MAX_SDMA_QUEUES 128 diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c index a0f22ea6d15a..521c14c7a789 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c @@ -1189,7 +1189,7 @@ svm_nodes_in_same_hive(struct kfd_node *node_a, struct kfd_node *node_b) } static uint64_t -svm_range_get_pte_flags(struct kfd_node *node, +svm_range_get_pte_flags(struct kfd_node *node, struct amdgpu_vm *vm, struct svm_range *prange, int domain) { struct kfd_node *bo_node; @@ -1292,10 +1292,6 @@ svm_range_get_pte_flags(struct kfd_node *node, AMDGPU_VM_MTYPE_UC : AMDGPU_VM_MTYPE_NC; } - mapping_flags |= AMDGPU_VM_PAGE_READABLE | AMDGPU_VM_PAGE_WRITEABLE; - - if (flags & KFD_IOCTL_SVM_FLAG_GPU_RO) - mapping_flags &= ~AMDGPU_VM_PAGE_WRITEABLE; if (flags & KFD_IOCTL_SVM_FLAG_GPU_EXEC) mapping_flags |= AMDGPU_VM_PAGE_EXECUTABLE; @@ -1305,7 +1301,10 @@ svm_range_get_pte_flags(struct kfd_node *node, if (gc_ip_version >= IP_VERSION(12, 0, 0)) pte_flags |= AMDGPU_PTE_IS_PTE; - pte_flags |= amdgpu_gem_va_map_flags(node->adev, mapping_flags); + amdgpu_gmc_get_vm_pte(node->adev, vm, NULL, mapping_flags, &pte_flags); + pte_flags |= AMDGPU_PTE_READABLE; + if (!(flags & KFD_IOCTL_SVM_FLAG_GPU_RO)) + pte_flags |= AMDGPU_PTE_WRITEABLE; return pte_flags; } @@ -1412,7 +1411,7 @@ svm_range_map_to_gpu(struct kfd_process_device *pdd, struct svm_range *prange, pr_debug("Mapping range [0x%lx 0x%llx] on domain: %s\n", last_start, prange->start + i, last_domain ? "GPU" : "CPU"); - pte_flags = svm_range_get_pte_flags(pdd->dev, prange, last_domain); + pte_flags = svm_range_get_pte_flags(pdd->dev, vm, prange, last_domain); if (readonly) pte_flags &= ~AMDGPU_PTE_WRITEABLE; @@ -1714,6 +1713,29 @@ static int svm_range_validate_and_map(struct mm_struct *mm, next = min(vma->vm_end, end); npages = (next - addr) >> PAGE_SHIFT; + /* HMM requires at least READ permissions. If provided with PROT_NONE, + * unmap the memory. If it's not already mapped, this is a no-op + * If PROT_WRITE is provided without READ, warn first then unmap + */ + if (!(vma->vm_flags & VM_READ)) { + unsigned long e, s; + + svm_range_lock(prange); + if (vma->vm_flags & VM_WRITE) + pr_debug("VM_WRITE without VM_READ is not supported"); + s = max(start, prange->start); + e = min(end, prange->last); + if (e >= s) + r = svm_range_unmap_from_gpus(prange, s, e, + KFD_SVM_UNMAP_TRIGGER_UNMAP_FROM_CPU); + svm_range_unlock(prange); + /* If unmap returns non-zero, we'll bail on the next for loop + * iteration, so just leave r and continue + */ + addr = next; + continue; + } + WRITE_ONCE(p->svms.faulting_task, current); r = amdgpu_hmm_range_get_pages(&prange->notifier, addr, npages, readonly, owner, NULL, diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c index 4ec73f33535e..82dbd68d8c99 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c @@ -530,6 +530,8 @@ static ssize_t node_show(struct kobject *kobj, struct attribute *attr, sysfs_show_32bit_prop(buffer, offs, "sdma_fw_version", dev->gpu->kfd->sdma_fw_version); sysfs_show_64bit_prop(buffer, offs, "unique_id", + dev->gpu->xcp ? + dev->gpu->xcp->unique_id : dev->gpu->adev->unique_id); sysfs_show_32bit_prop(buffer, offs, "num_xcc", NUM_XCC(dev->gpu->xcc_mask)); diff --git a/drivers/gpu/drm/amd/amdxcp/amdgpu_xcp_drv.c b/drivers/gpu/drm/amd/amdxcp/amdgpu_xcp_drv.c index 8bc36f04b1b7..44009aa8216e 100644 --- a/drivers/gpu/drm/amd/amdxcp/amdgpu_xcp_drv.c +++ b/drivers/gpu/drm/amd/amdxcp/amdgpu_xcp_drv.c @@ -46,18 +46,29 @@ static const struct drm_driver amdgpu_xcp_driver = { static int8_t pdev_num; static struct xcp_device *xcp_dev[MAX_XCP_PLATFORM_DEVICE]; +static DEFINE_MUTEX(xcp_mutex); int amdgpu_xcp_drm_dev_alloc(struct drm_device **ddev) { struct platform_device *pdev; struct xcp_device *pxcp_dev; char dev_name[20]; - int ret; + int ret, i; + + guard(mutex)(&xcp_mutex); if (pdev_num >= MAX_XCP_PLATFORM_DEVICE) return -ENODEV; - snprintf(dev_name, sizeof(dev_name), "amdgpu_xcp_%d", pdev_num); + for (i = 0; i < MAX_XCP_PLATFORM_DEVICE; i++) { + if (!xcp_dev[i]) + break; + } + + if (i >= MAX_XCP_PLATFORM_DEVICE) + return -ENODEV; + + snprintf(dev_name, sizeof(dev_name), "amdgpu_xcp_%d", i); pdev = platform_device_register_simple(dev_name, -1, NULL, 0); if (IS_ERR(pdev)) return PTR_ERR(pdev); @@ -73,8 +84,8 @@ int amdgpu_xcp_drm_dev_alloc(struct drm_device **ddev) goto out_devres; } - xcp_dev[pdev_num] = pxcp_dev; - xcp_dev[pdev_num]->pdev = pdev; + xcp_dev[i] = pxcp_dev; + xcp_dev[i]->pdev = pdev; *ddev = &pxcp_dev->drm; pdev_num++; @@ -89,16 +100,43 @@ out_unregister: } EXPORT_SYMBOL(amdgpu_xcp_drm_dev_alloc); -void amdgpu_xcp_drv_release(void) +static void free_xcp_dev(int8_t index) { - for (--pdev_num; pdev_num >= 0; --pdev_num) { - struct platform_device *pdev = xcp_dev[pdev_num]->pdev; + if ((index < MAX_XCP_PLATFORM_DEVICE) && (xcp_dev[index])) { + struct platform_device *pdev = xcp_dev[index]->pdev; devres_release_group(&pdev->dev, NULL); platform_device_unregister(pdev); - xcp_dev[pdev_num] = NULL; + + xcp_dev[index] = NULL; + pdev_num--; + } +} + +void amdgpu_xcp_drm_dev_free(struct drm_device *ddev) +{ + int8_t i; + + guard(mutex)(&xcp_mutex); + + for (i = 0; i < MAX_XCP_PLATFORM_DEVICE; i++) { + if ((xcp_dev[i]) && (&xcp_dev[i]->drm == ddev)) { + free_xcp_dev(i); + break; + } + } +} +EXPORT_SYMBOL(amdgpu_xcp_drm_dev_free); + +void amdgpu_xcp_drv_release(void) +{ + int8_t i; + + guard(mutex)(&xcp_mutex); + + for (i = 0; pdev_num && i < MAX_XCP_PLATFORM_DEVICE; i++) { + free_xcp_dev(i); } - pdev_num = 0; } EXPORT_SYMBOL(amdgpu_xcp_drv_release); diff --git a/drivers/gpu/drm/amd/amdxcp/amdgpu_xcp_drv.h b/drivers/gpu/drm/amd/amdxcp/amdgpu_xcp_drv.h index c1c4b679bf95..580a1602c8e3 100644 --- a/drivers/gpu/drm/amd/amdxcp/amdgpu_xcp_drv.h +++ b/drivers/gpu/drm/amd/amdxcp/amdgpu_xcp_drv.h @@ -25,5 +25,6 @@ #define _AMDGPU_XCP_DRV_H_ int amdgpu_xcp_drm_dev_alloc(struct drm_device **ddev); +void amdgpu_xcp_drm_dev_free(struct drm_device *ddev); void amdgpu_xcp_drv_release(void); #endif /* _AMDGPU_XCP_DRV_H_ */ diff --git a/drivers/gpu/drm/amd/display/Makefile b/drivers/gpu/drm/amd/display/Makefile index 89d605de0595..0084a8d55254 100644 --- a/drivers/gpu/drm/amd/display/Makefile +++ b/drivers/gpu/drm/amd/display/Makefile @@ -44,6 +44,7 @@ subdir-ccflags-y += -I$(FULL_AMD_DISPLAY_PATH)/dc/mmhubbub subdir-ccflags-y += -I$(FULL_AMD_DISPLAY_PATH)/dc/mpc subdir-ccflags-y += -I$(FULL_AMD_DISPLAY_PATH)/dc/opp subdir-ccflags-y += -I$(FULL_AMD_DISPLAY_PATH)/dc/pg +subdir-ccflags-y += -I$(FULL_AMD_DISPLAY_PATH)/dc/soc_and_ip_translator subdir-ccflags-y += -I$(FULL_AMD_DISPLAY_PATH)/modules/inc subdir-ccflags-y += -I$(FULL_AMD_DISPLAY_PATH)/modules/freesync subdir-ccflags-y += -I$(FULL_AMD_DISPLAY_PATH)/modules/color diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c index 18f0e0912ea6..1cdb195deb63 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -1,3 +1,4 @@ +// SPDX-License-Identifier: MIT /* * Copyright 2015 Advanced Micro Devices, Inc. * @@ -39,13 +40,11 @@ #include "dc/dc_stat.h" #include "dc/dc_state.h" #include "amdgpu_dm_trace.h" -#include "dpcd_defs.h" #include "link/protocols/link_dpcd.h" #include "link_service_types.h" #include "link/protocols/link_dp_capability.h" #include "link/protocols/link_ddc.h" -#include "vid.h" #include "amdgpu.h" #include "amdgpu_display.h" #include "amdgpu_ucode.h" @@ -56,7 +55,6 @@ #include "amdgpu_dm_hdcp.h" #include <drm/display/drm_hdcp_helper.h> #include "amdgpu_dm_wb.h" -#include "amdgpu_pm.h" #include "amdgpu_atombios.h" #include "amd_shared.h" @@ -82,6 +80,7 @@ #include <linux/component.h> #include <linux/sort.h> +#include <drm/drm_privacy_screen_consumer.h> #include <drm/display/drm_dp_mst_helper.h> #include <drm/display/drm_hdmi_helper.h> #include <drm/drm_atomic.h> @@ -102,15 +101,6 @@ #include "ivsrcid/dcn/irqsrcs_dcn_1_0.h" -#include "dcn/dcn_1_0_offset.h" -#include "dcn/dcn_1_0_sh_mask.h" -#include "soc15_hw_ip.h" -#include "soc15_common.h" -#include "vega10_ip_offset.h" - -#include "gc/gc_11_0_0_offset.h" -#include "gc/gc_11_0_0_sh_mask.h" - #include "modules/inc/mod_freesync.h" #include "modules/power/power_helpers.h" @@ -541,6 +531,50 @@ static void dm_pflip_high_irq(void *interrupt_params) amdgpu_crtc->crtc_id, amdgpu_crtc, vrr_active, (int)!e); } +static void dm_handle_vmin_vmax_update(struct work_struct *offload_work) +{ + struct vupdate_offload_work *work = container_of(offload_work, struct vupdate_offload_work, work); + struct amdgpu_device *adev = work->adev; + struct dc_stream_state *stream = work->stream; + struct dc_crtc_timing_adjust *adjust = work->adjust; + + mutex_lock(&adev->dm.dc_lock); + dc_stream_adjust_vmin_vmax(adev->dm.dc, stream, adjust); + mutex_unlock(&adev->dm.dc_lock); + + dc_stream_release(stream); + kfree(work->adjust); + kfree(work); +} + +static void schedule_dc_vmin_vmax(struct amdgpu_device *adev, + struct dc_stream_state *stream, + struct dc_crtc_timing_adjust *adjust) +{ + struct vupdate_offload_work *offload_work = kzalloc(sizeof(*offload_work), GFP_KERNEL); + if (!offload_work) { + drm_dbg_driver(adev_to_drm(adev), "Failed to allocate vupdate_offload_work\n"); + return; + } + + struct dc_crtc_timing_adjust *adjust_copy = kzalloc(sizeof(*adjust_copy), GFP_KERNEL); + if (!adjust_copy) { + drm_dbg_driver(adev_to_drm(adev), "Failed to allocate adjust_copy\n"); + kfree(offload_work); + return; + } + + dc_stream_retain(stream); + memcpy(adjust_copy, adjust, sizeof(*adjust_copy)); + + INIT_WORK(&offload_work->work, dm_handle_vmin_vmax_update); + offload_work->adev = adev; + offload_work->stream = stream; + offload_work->adjust = adjust_copy; + + queue_work(system_wq, &offload_work->work); +} + static void dm_vupdate_high_irq(void *interrupt_params) { struct common_irq_params *irq_params = interrupt_params; @@ -578,22 +612,27 @@ static void dm_vupdate_high_irq(void *interrupt_params) * page-flip completion events that have been queued to us * if a pageflip happened inside front-porch. */ - if (vrr_active) { + if (vrr_active && acrtc->dm_irq_params.stream) { + bool replay_en = acrtc->dm_irq_params.stream->link->replay_settings.replay_feature_enabled; + bool psr_en = acrtc->dm_irq_params.stream->link->psr_settings.psr_feature_enabled; + bool fs_active_var_en = acrtc->dm_irq_params.freesync_config.state + == VRR_STATE_ACTIVE_VARIABLE; + amdgpu_dm_crtc_handle_vblank(acrtc); /* BTR processing for pre-DCE12 ASICs */ - if (acrtc->dm_irq_params.stream && - adev->family < AMDGPU_FAMILY_AI) { + if (adev->family < AMDGPU_FAMILY_AI) { spin_lock_irqsave(&adev_to_drm(adev)->event_lock, flags); mod_freesync_handle_v_update( adev->dm.freesync_module, acrtc->dm_irq_params.stream, &acrtc->dm_irq_params.vrr_params); - dc_stream_adjust_vmin_vmax( - adev->dm.dc, - acrtc->dm_irq_params.stream, - &acrtc->dm_irq_params.vrr_params.adjust); + if (fs_active_var_en || (!fs_active_var_en && !replay_en && !psr_en)) { + schedule_dc_vmin_vmax(adev, + acrtc->dm_irq_params.stream, + &acrtc->dm_irq_params.vrr_params.adjust); + } spin_unlock_irqrestore(&adev_to_drm(adev)->event_lock, flags); } } @@ -676,15 +715,20 @@ static void dm_crtc_high_irq(void *interrupt_params) spin_lock_irqsave(&adev_to_drm(adev)->event_lock, flags); if (acrtc->dm_irq_params.stream && - acrtc->dm_irq_params.vrr_params.supported && - acrtc->dm_irq_params.freesync_config.state == - VRR_STATE_ACTIVE_VARIABLE) { + acrtc->dm_irq_params.vrr_params.supported) { + bool replay_en = acrtc->dm_irq_params.stream->link->replay_settings.replay_feature_enabled; + bool psr_en = acrtc->dm_irq_params.stream->link->psr_settings.psr_feature_enabled; + bool fs_active_var_en = acrtc->dm_irq_params.freesync_config.state == VRR_STATE_ACTIVE_VARIABLE; + mod_freesync_handle_v_update(adev->dm.freesync_module, acrtc->dm_irq_params.stream, &acrtc->dm_irq_params.vrr_params); - dc_stream_adjust_vmin_vmax(adev->dm.dc, acrtc->dm_irq_params.stream, - &acrtc->dm_irq_params.vrr_params.adjust); + /* update vmin_vmax only if freesync is enabled, or only if PSR and REPLAY are disabled */ + if (fs_active_var_en || (!fs_active_var_en && !replay_en && !psr_en)) { + schedule_dc_vmin_vmax(adev, acrtc->dm_irq_params.stream, + &acrtc->dm_irq_params.vrr_params.adjust); + } } /* @@ -4790,8 +4834,8 @@ static void convert_custom_brightness(const struct amdgpu_dm_backlight_caps *cap uint32_t *user_brightness) { u32 brightness = scale_input_to_fw(min, max, *user_brightness); - u8 prev_signal = 0, prev_lum = 0; - int i = 0; + u8 lower_signal, upper_signal, upper_lum, lower_lum, lum; + int left, right; if (amdgpu_dc_debug_mask & DC_DISABLE_CUSTOM_BRIGHTNESS_CURVE) return; @@ -4799,32 +4843,44 @@ static void convert_custom_brightness(const struct amdgpu_dm_backlight_caps *cap if (!caps->data_points) return; - /* choose start to run less interpolation steps */ - if (caps->luminance_data[caps->data_points/2].input_signal > brightness) - i = caps->data_points/2; - do { - u8 signal = caps->luminance_data[i].input_signal; - u8 lum = caps->luminance_data[i].luminance; + left = 0; + right = caps->data_points - 1; + while (left <= right) { + int mid = left + (right - left) / 2; + u8 signal = caps->luminance_data[mid].input_signal; - /* - * brightness == signal: luminance is percent numerator - * brightness < signal: interpolate between previous and current luminance numerator - * brightness > signal: find next data point - */ - if (brightness > signal) { - prev_signal = signal; - prev_lum = lum; - i++; - continue; + /* Exact match found */ + if (signal == brightness) { + lum = caps->luminance_data[mid].luminance; + goto scale; } - if (brightness < signal) - lum = prev_lum + DIV_ROUND_CLOSEST((lum - prev_lum) * - (brightness - prev_signal), - signal - prev_signal); - *user_brightness = scale_fw_to_input(min, max, - DIV_ROUND_CLOSEST(lum * brightness, 101)); - return; - } while (i < caps->data_points); + + if (signal < brightness) + left = mid + 1; + else + right = mid - 1; + } + + /* verify bound */ + if (left >= caps->data_points) + left = caps->data_points - 1; + + /* At this point, left > right */ + lower_signal = caps->luminance_data[right].input_signal; + upper_signal = caps->luminance_data[left].input_signal; + lower_lum = caps->luminance_data[right].luminance; + upper_lum = caps->luminance_data[left].luminance; + + /* interpolate */ + if (right == left || !lower_lum) + lum = upper_lum; + else + lum = lower_lum + DIV_ROUND_CLOSEST((upper_lum - lower_lum) * + (brightness - lower_signal), + upper_signal - lower_signal); +scale: + *user_brightness = scale_fw_to_input(min, max, + DIV_ROUND_CLOSEST(lum * brightness, 101)); } static u32 convert_brightness_from_user(const struct amdgpu_dm_backlight_caps *caps, @@ -4947,10 +5003,8 @@ static u32 amdgpu_dm_backlight_get_level(struct amdgpu_display_manager *dm, if (caps.aux_support) { u32 avg, peak; - bool rc; - rc = dc_link_get_backlight_level_nits(link, &avg, &peak); - if (!rc) + if (!dc_link_get_backlight_level_nits(link, &avg, &peak)) return dm->brightness[bl_idx]; return convert_brightness_to_user(&caps, avg); } @@ -6411,13 +6465,15 @@ static void fill_stream_properties_from_drm_display_mode( (struct drm_connector *)connector, mode_in); if (err < 0) - drm_warn_once(connector->dev, "Failed to setup avi infoframe on connector %s: %zd \n", connector->name, err); + drm_warn_once(connector->dev, "Failed to setup avi infoframe on connector %s: %zd\n", + connector->name, err); timing_out->vic = avi_frame.video_code; err = drm_hdmi_vendor_infoframe_from_display_mode(&hv_frame, (struct drm_connector *)connector, mode_in); if (err < 0) - drm_warn_once(connector->dev, "Failed to setup vendor infoframe on connector %s: %zd \n", connector->name, err); + drm_warn_once(connector->dev, "Failed to setup vendor infoframe on connector %s: %zd\n", + connector->name, err); timing_out->hdmi_vic = hv_frame.vic; } @@ -7813,6 +7869,9 @@ amdgpu_dm_connector_atomic_check(struct drm_connector *conn, struct amdgpu_dm_connector *aconn = to_amdgpu_dm_connector(conn); int ret; + if (WARN_ON(unlikely(!old_con_state || !new_con_state))) + return -EINVAL; + trace_amdgpu_dm_connector_atomic_check(new_con_state); if (conn->connector_type == DRM_MODE_CONNECTOR_DisplayPort) { @@ -7824,6 +7883,14 @@ amdgpu_dm_connector_atomic_check(struct drm_connector *conn, if (!crtc) return 0; + if (new_con_state->privacy_screen_sw_state != old_con_state->privacy_screen_sw_state) { + new_crtc_state = drm_atomic_get_crtc_state(state, crtc); + if (IS_ERR(new_crtc_state)) + return PTR_ERR(new_crtc_state); + + new_crtc_state->mode_changed = true; + } + if (new_con_state->colorspace != old_con_state->colorspace) { new_crtc_state = drm_atomic_get_crtc_state(state, crtc); if (IS_ERR(new_crtc_state)) @@ -8399,8 +8466,7 @@ static int amdgpu_dm_connector_get_modes(struct drm_connector *connector) drm_add_modes_noedid(connector, 1920, 1080); } else { amdgpu_dm_connector_ddc_get_modes(connector, drm_edid); - if (encoder && (connector->connector_type != DRM_MODE_CONNECTOR_eDP) && - (connector->connector_type != DRM_MODE_CONNECTOR_LVDS)) + if (encoder) amdgpu_dm_connector_add_common_modes(encoder, connector); amdgpu_dm_connector_add_freesync_modes(connector, drm_edid); } @@ -8519,6 +8585,18 @@ void amdgpu_dm_connector_init_helper(struct amdgpu_display_manager *dm, if (adev->dm.hdcp_workqueue) drm_connector_attach_content_protection_property(&aconnector->base, true); } + + if (connector_type == DRM_MODE_CONNECTOR_eDP) { + struct drm_privacy_screen *privacy_screen; + + privacy_screen = drm_privacy_screen_get(adev_to_drm(adev)->dev, NULL); + if (!IS_ERR(privacy_screen)) { + drm_connector_attach_privacy_screen_provider(&aconnector->base, + privacy_screen); + } else if (PTR_ERR(privacy_screen) != -ENODEV) { + drm_warn(adev_to_drm(adev), "Error getting privacy-screen\n"); + } + } } static int amdgpu_dm_i2c_xfer(struct i2c_adapter *i2c_adap, @@ -10107,69 +10185,40 @@ static void dm_set_writeback(struct amdgpu_display_manager *dm, drm_writeback_queue_job(wb_conn, new_con_state); } -/** - * amdgpu_dm_atomic_commit_tail() - AMDgpu DM's commit tail implementation. - * @state: The atomic state to commit - * - * This will tell DC to commit the constructed DC state from atomic_check, - * programming the hardware. Any failures here implies a hardware failure, since - * atomic check should have filtered anything non-kosher. - */ -static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state) +static void amdgpu_dm_update_hdcp(struct drm_atomic_state *state) { + struct drm_connector_state *old_con_state, *new_con_state; struct drm_device *dev = state->dev; - struct amdgpu_device *adev = drm_to_adev(dev); - struct amdgpu_display_manager *dm = &adev->dm; - struct dm_atomic_state *dm_state; - struct dc_state *dc_state = NULL; - u32 i, j; - struct drm_crtc *crtc; - struct drm_crtc_state *old_crtc_state, *new_crtc_state; - unsigned long flags; - bool wait_for_vblank = true; struct drm_connector *connector; - struct drm_connector_state *old_con_state, *new_con_state; - struct dm_crtc_state *dm_old_crtc_state, *dm_new_crtc_state; - int crtc_disable_count = 0; - - trace_amdgpu_dm_atomic_commit_tail_begin(state); - - drm_atomic_helper_update_legacy_modeset_state(dev, state); - drm_dp_mst_atomic_wait_for_dependencies(state); + struct amdgpu_device *adev = drm_to_adev(dev); + int i; - dm_state = dm_atomic_get_new_state(state); - if (dm_state && dm_state->context) { - dc_state = dm_state->context; - amdgpu_dm_commit_streams(state, dc_state); - } + if (!adev->dm.hdcp_workqueue) + return; for_each_oldnew_connector_in_state(state, connector, old_con_state, new_con_state, i) { struct dm_connector_state *dm_new_con_state = to_dm_connector_state(new_con_state); struct amdgpu_crtc *acrtc = to_amdgpu_crtc(dm_new_con_state->base.crtc); + struct drm_crtc_state *old_crtc_state, *new_crtc_state; + struct dm_crtc_state *dm_new_crtc_state; struct amdgpu_dm_connector *aconnector; - if (connector->connector_type == DRM_MODE_CONNECTOR_WRITEBACK) + if (!connector || connector->connector_type == DRM_MODE_CONNECTOR_WRITEBACK) continue; aconnector = to_amdgpu_dm_connector(connector); - if (!adev->dm.hdcp_workqueue) - continue; - - pr_debug("[HDCP_DM] -------------- i : %x ----------\n", i); - - if (!connector) - continue; + drm_dbg(dev, "[HDCP_DM] -------------- i : %x ----------\n", i); - pr_debug("[HDCP_DM] connector->index: %x connect_status: %x dpms: %x\n", + drm_dbg(dev, "[HDCP_DM] connector->index: %x connect_status: %x dpms: %x\n", connector->index, connector->status, connector->dpms); - pr_debug("[HDCP_DM] state protection old: %x new: %x\n", + drm_dbg(dev, "[HDCP_DM] state protection old: %x new: %x\n", old_con_state->content_protection, new_con_state->content_protection); if (aconnector->dc_sink) { if (aconnector->dc_sink->sink_signal != SIGNAL_TYPE_VIRTUAL && aconnector->dc_sink->sink_signal != SIGNAL_TYPE_NONE) { - pr_debug("[HDCP_DM] pipe_ctx dispname=%s\n", + drm_dbg(dev, "[HDCP_DM] pipe_ctx dispname=%s\n", aconnector->dc_sink->edid_caps.display_name); } } @@ -10183,7 +10232,7 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state) } if (old_crtc_state) - pr_debug("old crtc en: %x a: %x m: %x a-chg: %x c-chg: %x\n", + drm_dbg(dev, "old crtc en: %x a: %x m: %x a-chg: %x c-chg: %x\n", old_crtc_state->enable, old_crtc_state->active, old_crtc_state->mode_changed, @@ -10191,29 +10240,13 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state) old_crtc_state->connectors_changed); if (new_crtc_state) - pr_debug("NEW crtc en: %x a: %x m: %x a-chg: %x c-chg: %x\n", + drm_dbg(dev, "NEW crtc en: %x a: %x m: %x a-chg: %x c-chg: %x\n", new_crtc_state->enable, new_crtc_state->active, new_crtc_state->mode_changed, new_crtc_state->active_changed, new_crtc_state->connectors_changed); - } - - for_each_oldnew_connector_in_state(state, connector, old_con_state, new_con_state, i) { - struct dm_connector_state *dm_new_con_state = to_dm_connector_state(new_con_state); - struct amdgpu_crtc *acrtc = to_amdgpu_crtc(dm_new_con_state->base.crtc); - struct amdgpu_dm_connector *aconnector = to_amdgpu_dm_connector(connector); - if (!adev->dm.hdcp_workqueue) - continue; - - new_crtc_state = NULL; - old_crtc_state = NULL; - - if (acrtc) { - new_crtc_state = drm_atomic_get_new_crtc_state(state, &acrtc->base); - old_crtc_state = drm_atomic_get_old_crtc_state(state, &acrtc->base); - } dm_new_crtc_state = to_dm_crtc_state(new_crtc_state); @@ -10257,7 +10290,7 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state) new_con_state->content_protection >= DRM_MODE_CONTENT_PROTECTION_DESIRED) enable_encryption = true; - drm_info(adev_to_drm(adev), "[HDCP_DM] hdcp_update_display enable_encryption = %x\n", enable_encryption); + drm_info(dev, "[HDCP_DM] hdcp_update_display enable_encryption = %x\n", enable_encryption); if (aconnector->dc_link) hdcp_update_display( @@ -10265,6 +10298,45 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state) new_con_state->hdcp_content_type, enable_encryption); } } +} + +/** + * amdgpu_dm_atomic_commit_tail() - AMDgpu DM's commit tail implementation. + * @state: The atomic state to commit + * + * This will tell DC to commit the constructed DC state from atomic_check, + * programming the hardware. Any failures here implies a hardware failure, since + * atomic check should have filtered anything non-kosher. + */ +static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state) +{ + struct drm_device *dev = state->dev; + struct amdgpu_device *adev = drm_to_adev(dev); + struct amdgpu_display_manager *dm = &adev->dm; + struct dm_atomic_state *dm_state; + struct dc_state *dc_state = NULL; + u32 i, j; + struct drm_crtc *crtc; + struct drm_crtc_state *old_crtc_state, *new_crtc_state; + unsigned long flags; + bool wait_for_vblank = true; + struct drm_connector *connector; + struct drm_connector_state *old_con_state = NULL, *new_con_state = NULL; + struct dm_crtc_state *dm_old_crtc_state, *dm_new_crtc_state; + int crtc_disable_count = 0; + + trace_amdgpu_dm_atomic_commit_tail_begin(state); + + drm_atomic_helper_update_legacy_modeset_state(dev, state); + drm_dp_mst_atomic_wait_for_dependencies(state); + + dm_state = dm_atomic_get_new_state(state); + if (dm_state && dm_state->context) { + dc_state = dm_state->context; + amdgpu_dm_commit_streams(state, dc_state); + } + + amdgpu_dm_update_hdcp(state); /* Handle connector state changes */ for_each_oldnew_connector_in_state(state, connector, old_con_state, new_con_state, i) { @@ -10367,6 +10439,8 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state) &stream_update); mutex_unlock(&dm->dc_lock); kfree(dummy_updates); + + drm_connector_update_privacy_screen(new_con_state); } /** @@ -10418,6 +10492,7 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state) #if defined(CONFIG_DRM_AMD_SECURE_DISPLAY) if (amdgpu_dm_crc_window_is_activated(crtc)) { uint8_t cnt; + spin_lock_irqsave(&adev_to_drm(adev)->event_lock, flags); for (cnt = 0; cnt < MAX_CRC_WINDOW_NUM; cnt++) { if (acrtc->dm_irq_params.window_param[cnt].enable) { diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h index 60ce2ceb653a..159f8ded0439 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h @@ -1,3 +1,4 @@ +/* SPDX-License-Identifier: MIT */ /* * Copyright (C) 2015-2020 Advanced Micro Devices, Inc. All rights reserved. * @@ -152,6 +153,20 @@ struct idle_workqueue { bool running; }; +/** + * struct vupdate_offload_work - Work data for offloading task from vupdate handler + * @work: Kernel work data for the work event + * @adev: amdgpu_device back pointer + * @stream: DC stream associated with the crtc + * @adjust: DC CRTC timing adjust to be applied to the crtc + */ +struct vupdate_offload_work { + struct work_struct work; + struct amdgpu_device *adev; + struct dc_stream_state *stream; + struct dc_crtc_timing_adjust *adjust; +}; + #define MAX_LUMINANCE_DATA_POINTS 99 /** @@ -758,6 +773,9 @@ struct amdgpu_dm_connector { uint16_t vc_full_pbn; struct mutex handle_mst_msg_ready; + /* branch device specific data */ + uint32_t branch_ieee_oui; + /* TODO see if we can merge with ddc_bus or make a dm_connector */ struct amdgpu_i2c_adapter *i2c; diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c index ebabfe3a512f..c7387af725d6 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c @@ -1,3 +1,4 @@ +// SPDX-License-Identifier: MIT /* * Copyright 2018 Advanced Micro Devices, Inc. * @@ -27,7 +28,6 @@ #include "amdgpu_dm.h" #include "dc.h" #include "modules/color/color_gamma.h" -#include "basics/conversion.h" /** * DOC: overview diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c index 033bd817d871..e20aa7438066 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c @@ -1,3 +1,4 @@ +// SPDX-License-Identifier: MIT /* * Copyright 2015 Advanced Micro Devices, Inc. * diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.h index 3da056c8d20b..95bdb8699d7f 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.h +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.h @@ -1,3 +1,4 @@ +/* SPDX-License-Identifier: MIT */ /* * Copyright 2019 Advanced Micro Devices, Inc. * diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c index 010172f930ae..45feb404b097 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c @@ -299,6 +299,25 @@ static inline int amdgpu_dm_crtc_set_vblank(struct drm_crtc *crtc, bool enable) irq_type = amdgpu_display_crtc_idx_to_irq_type(adev, acrtc->crtc_id); if (enable) { + struct dc *dc = adev->dm.dc; + struct drm_vblank_crtc *vblank = drm_crtc_vblank_crtc(crtc); + struct psr_settings *psr = &acrtc_state->stream->link->psr_settings; + struct replay_settings *pr = &acrtc_state->stream->link->replay_settings; + bool sr_supported = (psr->psr_version != DC_PSR_VERSION_UNSUPPORTED) || + pr->config.replay_supported; + + /* + * IPS & self-refresh feature can cause vblank counter resets between + * vblank disable and enable. + * It may cause system stuck due to waiting for the vblank counter. + * Call this function to estimate missed vblanks by using timestamps and + * update the vblank counter in DRM. + */ + if (dc->caps.ips_support && + dc->config.disable_ips != DMUB_IPS_DISABLE_ALL && + sr_supported && vblank->config.disable_immediate) + drm_crtc_vblank_restore(crtc); + /* vblank irq on -> Only need vupdate irq in vrr mode */ if (amdgpu_dm_crtc_vrr_active(acrtc_state)) rc = amdgpu_dm_crtc_set_vupdate_irq(crtc, true); diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c index b726bcd18e29..f263e1a4537e 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c @@ -1,3 +1,4 @@ +// SPDX-License-Identifier: MIT /* * Copyright 2018 Advanced Micro Devices, Inc. * @@ -3106,6 +3107,35 @@ static int replay_get_state(void *data, u64 *val) } /* + * Start / Stop capture Replay residency + */ +static int replay_set_residency(void *data, u64 val) +{ + struct amdgpu_dm_connector *connector = data; + struct dc_link *link = connector->dc_link; + bool is_start = (val != 0); + u32 residency = 0; + + link->dc->link_srv->edp_replay_residency(link, &residency, is_start, PR_RESIDENCY_MODE_PHY); + return 0; +} + +/* + * Read Replay residency + */ +static int replay_get_residency(void *data, u64 *val) +{ + struct amdgpu_dm_connector *connector = data; + struct dc_link *link = connector->dc_link; + u32 residency = 0; + + link->dc->link_srv->edp_replay_residency(link, &residency, false, PR_RESIDENCY_MODE_PHY); + *val = (u64)residency; + + return 0; +} + +/* * Read PSR state */ static int psr_get(void *data, u64 *val) @@ -3324,7 +3354,8 @@ DEFINE_DEBUGFS_ATTRIBUTE(dmcub_trace_event_state_fops, dmcub_trace_event_state_g dmcub_trace_event_state_set, "%llu\n"); DEFINE_DEBUGFS_ATTRIBUTE(replay_state_fops, replay_get_state, NULL, "%llu\n"); - +DEFINE_DEBUGFS_ATTRIBUTE(replay_residency_fops, replay_get_residency, replay_set_residency, + "%llu\n"); DEFINE_DEBUGFS_ATTRIBUTE(psr_fops, psr_get, NULL, "%llu\n"); DEFINE_DEBUGFS_ATTRIBUTE(psr_residency_fops, psr_read_residency, NULL, "%llu\n"); @@ -3502,6 +3533,8 @@ void connector_debugfs_init(struct amdgpu_dm_connector *connector) debugfs_create_file("replay_capability", 0444, dir, connector, &replay_capability_fops); debugfs_create_file("replay_state", 0444, dir, connector, &replay_state_fops); + debugfs_create_file_unsafe("replay_residency", 0444, dir, + connector, &replay_residency_fops); debugfs_create_file_unsafe("psr_capability", 0444, dir, connector, &psr_capability_fops); debugfs_create_file_unsafe("psr_state", 0444, dir, connector, &psr_fops); debugfs_create_file_unsafe("psr_residency", 0444, dir, diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.h index 071200473c27..122cdc124b3b 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.h +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.h @@ -1,3 +1,4 @@ +/* SPDX-License-Identifier: MIT */ /* * Copyright 2018 Advanced Micro Devices, Inc. * diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c index b1d1897f5eaf..437174d4fed5 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c @@ -1,3 +1,4 @@ +// SPDX-License-Identifier: MIT /* * Copyright 2019 Advanced Micro Devices, Inc. * diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.h index 69b445b011c8..4faa344f196e 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.h +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.h @@ -1,3 +1,4 @@ +/* SPDX-License-Identifier: MIT */ /* * Copyright 2019 Advanced Micro Devices, Inc. * diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c index 9e3e51a2dc49..fe100e4c9801 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c @@ -1,3 +1,4 @@ +// SPDX-License-Identifier: MIT /* * Copyright 2015 Advanced Micro Devices, Inc. * diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.c index b61e210f6246..a1c722112c22 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.c @@ -1,3 +1,4 @@ +// SPDX-License-Identifier: MIT /* * Copyright 2015 Advanced Micro Devices, Inc. * diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.h index ba17c23b2706..4f6b58f4f90d 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.h +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.h @@ -1,3 +1,4 @@ +/* SPDX-License-Identifier: MIT */ /* * Copyright 2015 Advanced Micro Devices, Inc. * diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq_params.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq_params.h index 6c9de834455b..3c9995275cbd 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq_params.h +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq_params.h @@ -1,3 +1,4 @@ +/* SPDX-License-Identifier: MIT */ /* * Copyright 2020 Advanced Micro Devices, Inc. * diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c index 7187d5aedf0a..6a817508c826 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c @@ -1,3 +1,4 @@ +// SPDX-License-Identifier: MIT /* * Copyright 2012-15 Advanced Micro Devices, Inc. * @@ -329,6 +330,34 @@ static bool retrieve_downstream_port_device(struct amdgpu_dm_connector *aconnect return true; } +static bool retrieve_branch_specific_data(struct amdgpu_dm_connector *aconnector) +{ + struct drm_connector *connector = &aconnector->base; + struct drm_dp_mst_port *port = aconnector->mst_output_port; + struct drm_dp_mst_port *port_parent; + struct drm_dp_aux *immediate_upstream_aux; + struct drm_dp_desc branch_desc; + + if (!port->parent) + return false; + + port_parent = port->parent->port_parent; + + immediate_upstream_aux = port_parent ? &port_parent->aux : port->mgr->aux; + + if (drm_dp_read_desc(immediate_upstream_aux, &branch_desc, true)) + return false; + + aconnector->branch_ieee_oui = (branch_desc.ident.oui[0] << 16) + + (branch_desc.ident.oui[1] << 8) + + (branch_desc.ident.oui[2]); + + drm_dbg_dp(port->aux.drm_dev, "MST branch oui 0x%x detected at %s\n", + aconnector->branch_ieee_oui, connector->name); + + return true; +} + static int dm_dp_mst_get_modes(struct drm_connector *connector) { struct amdgpu_dm_connector *aconnector = to_amdgpu_dm_connector(connector); @@ -668,6 +697,9 @@ dm_dp_add_mst_connector(struct drm_dp_mst_topology_mgr *mgr, drm_connector_set_path_property(connector, pathprop); + if (!retrieve_branch_specific_data(aconnector)) + aconnector->branch_ieee_oui = 0; + /* * Initialize connector state before adding the connectror to drm and * framebuffer lists @@ -1762,14 +1794,20 @@ static bool dp_get_link_current_set_bw(struct drm_dp_aux *aux, uint32_t *cur_lin union lane_count_set lane_count; u8 dp_link_encoding; u8 link_bw_set = 0; + u8 data[16] = {0}; *cur_link_bw = 0; - if (drm_dp_dpcd_read(aux, DP_MAIN_LINK_CHANNEL_CODING_SET, &dp_link_encoding, 1) != 1 || - drm_dp_dpcd_read(aux, DP_LANE_COUNT_SET, &lane_count.raw, 1) != 1 || - drm_dp_dpcd_read(aux, DP_LINK_BW_SET, &link_bw_set, 1) != 1) + if (drm_dp_dpcd_read(aux, DP_LINK_BW_SET, data, 16) != 16) return false; + dp_link_encoding = data[DP_MAIN_LINK_CHANNEL_CODING_SET - DP_LINK_BW_SET]; + link_bw_set = data[DP_LINK_BW_SET - DP_LINK_BW_SET]; + lane_count.raw = data[DP_LANE_COUNT_SET - DP_LINK_BW_SET]; + + drm_dbg_dp(aux->drm_dev, "MST_DSC downlink setting: %d, 0x%x x %d\n", + dp_link_encoding, link_bw_set, lane_count.bits.LANE_COUNT_SET); + switch (dp_link_encoding) { case DP_8b_10b_ENCODING: link_rate = link_bw_set; @@ -1866,8 +1904,10 @@ enum dc_status dm_dp_mst_is_port_support_mode( end_link_bw = aconnector->mst_local_bw; } - if (end_link_bw > 0 && stream_kbps > end_link_bw) { - DRM_DEBUG_DRIVER("MST_DSC dsc decode at last link." + if (end_link_bw > 0 && + stream_kbps > end_link_bw && + aconnector->branch_ieee_oui != DP_BRANCH_DEVICE_ID_90CC24) { + DRM_DEBUG_DRIVER("MST_DSC dsc decode at last link. " "Mode required bw can't fit into last link\n"); return DC_FAIL_BANDWIDTH_VALIDATE; } diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.h index 600d6e221011..65f76a7d00db 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.h +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.h @@ -1,3 +1,4 @@ +/* SPDX-License-Identifier: MIT */ /* * Copyright 2012-15 Advanced Micro Devices, Inc. * diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_pp_smu.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_pp_smu.c index 848c5b4bb301..e5771f490f2e 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_pp_smu.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_pp_smu.c @@ -1,3 +1,4 @@ +// SPDX-License-Identifier: MIT /* * Copyright 2018 Advanced Micro Devices, Inc. * diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.c index ff7b867ae98b..fd491b7a3cd7 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.c @@ -1,3 +1,4 @@ +// SPDX-License-Identifier: MIT /* * Copyright 2021 Advanced Micro Devices, Inc. * @@ -26,7 +27,6 @@ #include "amdgpu_dm_psr.h" #include "dc_dmub_srv.h" #include "dc.h" -#include "dm_helpers.h" #include "amdgpu_dm.h" #include "modules/power/power_helpers.h" diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.h index e2366321a3c1..4fb8626913cf 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.h +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.h @@ -1,3 +1,4 @@ +/* SPDX-License-Identifier: MIT */ /* * Copyright 2021 Advanced Micro Devices, Inc. * diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_replay.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_replay.c index 41f07f13a7b5..82ea3fe5e764 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_replay.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_replay.c @@ -1,3 +1,4 @@ +// SPDX-License-Identifier: MIT /* * Copyright 2023 Advanced Micro Devices, Inc. * diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_replay.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_replay.h index 8126bdb1eb6b..73b6c67ae5e7 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_replay.h +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_replay.h @@ -1,3 +1,4 @@ +/* SPDX-License-Identifier: MIT */ /* * Copyright 2021 Advanced Micro Devices, Inc. * diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_services.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_services.c index 0005f5f8f34f..132de4071efd 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_services.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_services.c @@ -1,3 +1,4 @@ +//SPDX-License-Identifier: MIT /* * Copyright 2015 Advanced Micro Devices, Inc. * diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_trace.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_trace.h index 95f890fda8aa..aa56fd6d56c3 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_trace.h +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_trace.h @@ -1,3 +1,4 @@ +//SPDX-License-Identifier: MIT /* * Copyright 2018 Advanced Micro Devices, Inc. * diff --git a/drivers/gpu/drm/amd/display/dc/Makefile b/drivers/gpu/drm/amd/display/dc/Makefile index 3c9ecea7eebc..dc943abd6dba 100644 --- a/drivers/gpu/drm/amd/display/dc/Makefile +++ b/drivers/gpu/drm/amd/display/dc/Makefile @@ -37,6 +37,7 @@ DC_LIBS += dcn301 DC_LIBS += dcn31 DC_LIBS += dml DC_LIBS += dml2 +DC_LIBS += soc_and_ip_translator endif DC_LIBS += dce120 diff --git a/drivers/gpu/drm/amd/display/dc/basics/dce_calcs.c b/drivers/gpu/drm/amd/display/dc/basics/dce_calcs.c index d897f8a30ede..4da5adab799c 100644 --- a/drivers/gpu/drm/amd/display/dc/basics/dce_calcs.c +++ b/drivers/gpu/drm/amd/display/dc/basics/dce_calcs.c @@ -1136,7 +1136,7 @@ static void calculate_bandwidth( } } } - data->total_dmifmc_urgent_trips = bw_ceil2(bw_div(data->total_requests_for_adjusted_dmif_size, (bw_add(dceip->dmif_request_buffer_size, bw_int_to_fixed(vbios->number_of_request_slots_gmc_reserves_for_dmif_per_channel * data->number_of_dram_channels)))), bw_int_to_fixed(1)); + data->total_dmifmc_urgent_trips = bw_ceil2(bw_div(data->total_requests_for_adjusted_dmif_size, (bw_add(dceip->dmif_request_buffer_size, bw_int_to_fixed((uint64_t)vbios->number_of_request_slots_gmc_reserves_for_dmif_per_channel * data->number_of_dram_channels)))), bw_int_to_fixed(1)); data->total_dmifmc_urgent_latency = bw_mul(vbios->dmifmc_urgent_latency, data->total_dmifmc_urgent_trips); data->total_display_reads_required_data = bw_int_to_fixed(0); data->total_display_reads_required_dram_access_data = bw_int_to_fixed(0); diff --git a/drivers/gpu/drm/amd/display/dc/basics/fixpt31_32.c b/drivers/gpu/drm/amd/display/dc/basics/fixpt31_32.c index 452206b5095e..6073cadde76c 100644 --- a/drivers/gpu/drm/amd/display/dc/basics/fixpt31_32.c +++ b/drivers/gpu/drm/amd/display/dc/basics/fixpt31_32.c @@ -284,7 +284,7 @@ struct fixed31_32 dc_fixpt_cos(struct fixed31_32 arg) dc_fixpt_mul( square, res), - n * (n - 1))); + (long long)n * (n - 1))); n -= 2; } while (n != 0); diff --git a/drivers/gpu/drm/amd/display/dc/basics/vector.c b/drivers/gpu/drm/amd/display/dc/basics/vector.c index 6d2924114a3e..b413a672c2c0 100644 --- a/drivers/gpu/drm/amd/display/dc/basics/vector.c +++ b/drivers/gpu/drm/amd/display/dc/basics/vector.c @@ -170,7 +170,7 @@ bool dal_vector_remove_at_index( memmove( vector->container + (index * vector->struct_size), vector->container + ((index + 1) * vector->struct_size), - (vector->count - index - 1) * vector->struct_size); + (size_t)(vector->count - index - 1) * vector->struct_size); vector->count -= 1; return true; @@ -219,7 +219,7 @@ bool dal_vector_insert_at( memmove( insert_address + vector->struct_size, insert_address, - vector->struct_size * (vector->count - position)); + (size_t)vector->struct_size * (vector->count - position)); memmove( insert_address, @@ -271,7 +271,7 @@ struct vector *dal_vector_clone( /* copy vector's data */ memmove(vec_cloned->container, vector->container, - vec_cloned->struct_size * vec_cloned->capacity); + (size_t)vec_cloned->struct_size * vec_cloned->capacity); return vec_cloned; } diff --git a/drivers/gpu/drm/amd/display/dc/bios/bios_parser.c b/drivers/gpu/drm/amd/display/dc/bios/bios_parser.c index 67f08495b7e6..154fd2c18e88 100644 --- a/drivers/gpu/drm/amd/display/dc/bios/bios_parser.c +++ b/drivers/gpu/drm/amd/display/dc/bios/bios_parser.c @@ -174,11 +174,8 @@ static struct graphics_object_id bios_parser_get_connector_id( return object_id; } - if (tbl->ucNumberOfObjects <= i) { - dm_error("Can't find connector id %d in connector table of size %d.\n", - i, tbl->ucNumberOfObjects); + if (tbl->ucNumberOfObjects <= i) return object_id; - } id = le16_to_cpu(tbl->asObjects[i].usObjectID); object_id = object_id_from_bios_object_id(id); diff --git a/drivers/gpu/drm/amd/display/dc/bios/command_table.c b/drivers/gpu/drm/amd/display/dc/bios/command_table.c index 2bcae0643e61..58e88778da7f 100644 --- a/drivers/gpu/drm/amd/display/dc/bios/command_table.c +++ b/drivers/gpu/drm/amd/display/dc/bios/command_table.c @@ -993,7 +993,7 @@ static enum bp_result set_pixel_clock_v3( allocation.sPCLKInput.usFbDiv = cpu_to_le16((uint16_t)bp_params->feedback_divider); allocation.sPCLKInput.ucFracFbDiv = - (uint8_t)bp_params->fractional_feedback_divider; + (uint8_t)(bp_params->fractional_feedback_divider / 100000); allocation.sPCLKInput.ucPostDiv = (uint8_t)bp_params->pixel_clock_post_divider; diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce100/dce_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce100/dce_clk_mgr.c index f5ad0a177038..dbd6ef1b60a0 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce100/dce_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce100/dce_clk_mgr.c @@ -72,9 +72,9 @@ static const struct state_dependent_clocks dce80_max_clks_by_state[] = { /* ClocksStateLow */ { .display_clk_khz = 352000, .pixel_clk_khz = 330000}, /* ClocksStateNominal */ -{ .display_clk_khz = 600000, .pixel_clk_khz = 400000 }, +{ .display_clk_khz = 625000, .pixel_clk_khz = 400000 }, /* ClocksStatePerformance */ -{ .display_clk_khz = 600000, .pixel_clk_khz = 400000 } }; +{ .display_clk_khz = 625000, .pixel_clk_khz = 400000 } }; int dentist_get_divider_from_did(int did) { @@ -391,8 +391,6 @@ static void dce_pplib_apply_display_requirements( { struct dm_pp_display_configuration *pp_display_cfg = &context->pp_display_cfg; - pp_display_cfg->avail_mclk_switch_time_us = dce110_get_min_vblank_time_us(context); - dce110_fill_display_configs(context, pp_display_cfg); if (memcmp(&dc->current_state->pp_display_cfg, pp_display_cfg, sizeof(*pp_display_cfg)) != 0) @@ -405,11 +403,9 @@ static void dce_update_clocks(struct clk_mgr *clk_mgr_base, { struct clk_mgr_internal *clk_mgr_dce = TO_CLK_MGR_INTERNAL(clk_mgr_base); struct dm_pp_power_level_change_request level_change_req; - int patched_disp_clk = context->bw_ctx.bw.dce.dispclk_khz; - - /*TODO: W/A for dal3 linux, investigate why this works */ - if (!clk_mgr_dce->dfs_bypass_active) - patched_disp_clk = patched_disp_clk * 115 / 100; + const int max_disp_clk = + clk_mgr_dce->max_clks_by_state[DM_PP_CLOCKS_STATE_PERFORMANCE].display_clk_khz; + int patched_disp_clk = MIN(max_disp_clk, context->bw_ctx.bw.dce.dispclk_khz); level_change_req.power_level = dce_get_required_clocks_state(clk_mgr_base, context); /* get max clock state from PPLIB */ diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce110/dce110_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce110/dce110_clk_mgr.c index f8409453434c..13cf415e38e5 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce110/dce110_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce110/dce110_clk_mgr.c @@ -120,9 +120,15 @@ void dce110_fill_display_configs( const struct dc_state *context, struct dm_pp_display_configuration *pp_display_cfg) { + struct dc *dc = context->clk_mgr->ctx->dc; int j; int num_cfgs = 0; + pp_display_cfg->avail_mclk_switch_time_us = dce110_get_min_vblank_time_us(context); + pp_display_cfg->disp_clk_khz = dc->clk_mgr->clks.dispclk_khz; + pp_display_cfg->avail_mclk_switch_time_in_disp_active_us = 0; + pp_display_cfg->crtc_index = dc->res_pool->res_cap->num_timing_generator; + for (j = 0; j < context->stream_count; j++) { int k; @@ -164,6 +170,23 @@ void dce110_fill_display_configs( cfg->v_refresh /= stream->timing.h_total; cfg->v_refresh = (cfg->v_refresh + stream->timing.v_total / 2) / stream->timing.v_total; + + /* Find first CRTC index and calculate its line time. + * This is necessary for DPM on SI GPUs. + */ + if (cfg->pipe_idx < pp_display_cfg->crtc_index) { + const struct dc_crtc_timing *timing = + &context->streams[0]->timing; + + pp_display_cfg->crtc_index = cfg->pipe_idx; + pp_display_cfg->line_time_in_us = + timing->h_total * 10000 / timing->pix_clk_100hz; + } + } + + if (!num_cfgs) { + pp_display_cfg->crtc_index = 0; + pp_display_cfg->line_time_in_us = 0; } pp_display_cfg->display_count = num_cfgs; @@ -223,25 +246,8 @@ void dce11_pplib_apply_display_requirements( pp_display_cfg->min_engine_clock_deep_sleep_khz = context->bw_ctx.bw.dce.sclk_deep_sleep_khz; - pp_display_cfg->avail_mclk_switch_time_us = - dce110_get_min_vblank_time_us(context); - /* TODO: dce11.2*/ - pp_display_cfg->avail_mclk_switch_time_in_disp_active_us = 0; - - pp_display_cfg->disp_clk_khz = dc->clk_mgr->clks.dispclk_khz; - dce110_fill_display_configs(context, pp_display_cfg); - /* TODO: is this still applicable?*/ - if (pp_display_cfg->display_count == 1) { - const struct dc_crtc_timing *timing = - &context->streams[0]->timing; - - pp_display_cfg->crtc_index = - pp_display_cfg->disp_configs[0].pipe_idx; - pp_display_cfg->line_time_in_us = timing->h_total * 10000 / timing->pix_clk_100hz; - } - if (memcmp(&dc->current_state->pp_display_cfg, pp_display_cfg, sizeof(*pp_display_cfg)) != 0) dm_pp_apply_display_requirements(dc->ctx, pp_display_cfg); } diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce60/dce60_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce60/dce60_clk_mgr.c index 0267644717b2..a39641a0ff09 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce60/dce60_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce60/dce60_clk_mgr.c @@ -83,22 +83,13 @@ static const struct state_dependent_clocks dce60_max_clks_by_state[] = { static int dce60_get_dp_ref_freq_khz(struct clk_mgr *clk_mgr_base) { struct clk_mgr_internal *clk_mgr = TO_CLK_MGR_INTERNAL(clk_mgr_base); - int dprefclk_wdivider; - int dp_ref_clk_khz; - int target_div; + struct dc_context *ctx = clk_mgr_base->ctx; + int dp_ref_clk_khz = 0; - /* DCE6 has no DPREFCLK_CNTL to read DP Reference Clock source */ - - /* Read the mmDENTIST_DISPCLK_CNTL to get the currently - * programmed DID DENTIST_DPREFCLK_WDIVIDER*/ - REG_GET(DENTIST_DISPCLK_CNTL, DENTIST_DPREFCLK_WDIVIDER, &dprefclk_wdivider); - - /* Convert DENTIST_DPREFCLK_WDIVIDERto actual divider*/ - target_div = dentist_get_divider_from_did(dprefclk_wdivider); - - /* Calculate the current DFS clock, in kHz.*/ - dp_ref_clk_khz = (DENTIST_DIVIDER_RANGE_SCALE_FACTOR - * clk_mgr->base.dentist_vco_freq_khz) / target_div; + if (ASIC_REV_IS_TAHITI_P(ctx->asic_id.hw_internal_rev)) + dp_ref_clk_khz = ctx->dc_bios->fw_info.default_display_engine_pll_frequency; + else + dp_ref_clk_khz = clk_mgr_base->clks.dispclk_khz; return dce_adjust_dp_ref_freq_for_ss(clk_mgr, dp_ref_clk_khz); } @@ -109,8 +100,6 @@ static void dce60_pplib_apply_display_requirements( { struct dm_pp_display_configuration *pp_display_cfg = &context->pp_display_cfg; - pp_display_cfg->avail_mclk_switch_time_us = dce110_get_min_vblank_time_us(context); - dce110_fill_display_configs(context, pp_display_cfg); if (memcmp(&dc->current_state->pp_display_cfg, pp_display_cfg, sizeof(*pp_display_cfg)) != 0) @@ -123,11 +112,9 @@ static void dce60_update_clocks(struct clk_mgr *clk_mgr_base, { struct clk_mgr_internal *clk_mgr_dce = TO_CLK_MGR_INTERNAL(clk_mgr_base); struct dm_pp_power_level_change_request level_change_req; - int patched_disp_clk = context->bw_ctx.bw.dce.dispclk_khz; - - /*TODO: W/A for dal3 linux, investigate why this works */ - if (!clk_mgr_dce->dfs_bypass_active) - patched_disp_clk = patched_disp_clk * 115 / 100; + const int max_disp_clk = + clk_mgr_dce->max_clks_by_state[DM_PP_CLOCKS_STATE_PERFORMANCE].display_clk_khz; + int patched_disp_clk = MIN(max_disp_clk, context->bw_ctx.bw.dce.dispclk_khz); level_change_req.power_level = dce_get_required_clocks_state(clk_mgr_base, context); /* get max clock state from PPLIB */ diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn301/vg_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn301/vg_clk_mgr.c index 9e2ef0e724fc..7aee02d56292 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn301/vg_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn301/vg_clk_mgr.c @@ -563,6 +563,7 @@ static void vg_clk_mgr_helper_populate_bw_params( { int i, j; struct clk_bw_params *bw_params = clk_mgr->base.bw_params; + uint32_t max_dispclk = 0, max_dppclk = 0; j = -1; @@ -584,6 +585,15 @@ static void vg_clk_mgr_helper_populate_bw_params( return; } + /* dispclk and dppclk can be max at any voltage, same number of levels for both */ + if (clock_table->NumDispClkLevelsEnabled <= VG_NUM_DISPCLK_DPM_LEVELS && + clock_table->NumDispClkLevelsEnabled <= VG_NUM_DPPCLK_DPM_LEVELS) { + max_dispclk = find_max_clk_value(clock_table->DispClocks, clock_table->NumDispClkLevelsEnabled); + max_dppclk = find_max_clk_value(clock_table->DppClocks, clock_table->NumDispClkLevelsEnabled); + } else { + ASSERT(0); + } + bw_params->clk_table.num_entries = j + 1; for (i = 0; i < bw_params->clk_table.num_entries - 1; i++, j--) { @@ -591,11 +601,17 @@ static void vg_clk_mgr_helper_populate_bw_params( bw_params->clk_table.entries[i].memclk_mhz = clock_table->DfPstateTable[j].memclk; bw_params->clk_table.entries[i].voltage = clock_table->DfPstateTable[j].voltage; bw_params->clk_table.entries[i].dcfclk_mhz = find_dcfclk_for_voltage(clock_table, clock_table->DfPstateTable[j].voltage); + + /* Now update clocks we do read */ + bw_params->clk_table.entries[i].dispclk_mhz = max_dispclk; + bw_params->clk_table.entries[i].dppclk_mhz = max_dppclk; } bw_params->clk_table.entries[i].fclk_mhz = clock_table->DfPstateTable[j].fclk; bw_params->clk_table.entries[i].memclk_mhz = clock_table->DfPstateTable[j].memclk; bw_params->clk_table.entries[i].voltage = clock_table->DfPstateTable[j].voltage; bw_params->clk_table.entries[i].dcfclk_mhz = find_max_clk_value(clock_table->DcfClocks, VG_NUM_DCFCLK_DPM_LEVELS); + bw_params->clk_table.entries[i].dispclk_mhz = find_max_clk_value(clock_table->DispClocks, VG_NUM_DISPCLK_DPM_LEVELS); + bw_params->clk_table.entries[i].dppclk_mhz = find_max_clk_value(clock_table->DppClocks, VG_NUM_DPPCLK_DPM_LEVELS); bw_params->vram_type = bios_info->memory_type; bw_params->num_channels = bios_info->ma_channel_number; diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c index 084994c650c4..8376e2b0e73d 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c @@ -1047,11 +1047,8 @@ static void dcn32_get_memclk_states_from_smu(struct clk_mgr *clk_mgr_base) &num_entries_per_clk->num_fclk_levels); clk_mgr_base->bw_params->dc_mode_limit.fclk_mhz = dcn30_smu_get_dc_mode_max_dpm_freq(clk_mgr, PPCLK_FCLK); - if (num_entries_per_clk->num_memclk_levels >= num_entries_per_clk->num_fclk_levels) { - num_levels = num_entries_per_clk->num_memclk_levels; - } else { - num_levels = num_entries_per_clk->num_fclk_levels; - } + num_levels = max(num_entries_per_clk->num_memclk_levels, num_entries_per_clk->num_fclk_levels); + clk_mgr_base->bw_params->max_memclk_mhz = clk_mgr_base->bw_params->clk_table.entries[num_entries_per_clk->num_memclk_levels - 1].memclk_mhz; clk_mgr_base->bw_params->clk_table.num_entries = num_levels ? num_levels : 1; diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr.c index b59703467128..47ff4c965d76 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr.c @@ -1404,11 +1404,7 @@ static void dcn401_get_memclk_states_from_smu(struct clk_mgr *clk_mgr_base) clk_mgr_base->bw_params->clk_table.entries[num_entries_per_clk->num_fclk_levels - 1].fclk_mhz) clk_mgr_base->bw_params->dc_mode_limit.fclk_mhz = 0; - if (num_entries_per_clk->num_memclk_levels >= num_entries_per_clk->num_fclk_levels) { - num_levels = num_entries_per_clk->num_memclk_levels; - } else { - num_levels = num_entries_per_clk->num_fclk_levels; - } + num_levels = max(num_entries_per_clk->num_memclk_levels, num_entries_per_clk->num_fclk_levels); clk_mgr_base->bw_params->clk_table.num_entries = num_levels ? num_levels : 1; diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c index 9ab0ee20ca6f..261b7d43e91d 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc.c @@ -84,6 +84,7 @@ #if defined(CONFIG_DRM_AMD_DC_FP) #include "dml2/dml2_internal_types.h" +#include "soc_and_ip_translator.h" #endif #include "dce/dmub_outbox.h" @@ -217,11 +218,24 @@ static bool create_links( connectors_num, num_virtual_links); - // condition loop on link_count to allow skipping invalid indices + /* When getting the number of connectors, the VBIOS reports the number of valid indices, + * but it doesn't say which indices are valid, and not every index has an actual connector. + * So, if we don't find a connector on an index, that is not an error. + * + * - There is no guarantee that the first N indices will be valid + * - VBIOS may report a higher amount of valid indices than there are actual connectors + * - Some VBIOS have valid configurations for more connectors than there actually are + * on the card. This may be because the manufacturer used the same VBIOS for different + * variants of the same card. + */ for (i = 0; dc->link_count < connectors_num && i < MAX_LINKS; i++) { + struct graphics_object_id connector_id = bios->funcs->get_connector_id(bios, i); struct link_init_data link_init_params = {0}; struct dc_link *link; + if (connector_id.id == CONNECTOR_ID_UNKNOWN) + continue; + DC_LOG_DC("BIOS object table - printing link object info for connector number: %d, link_index: %d", i, dc->link_count); link_init_params.ctx = dc->ctx; @@ -446,7 +460,9 @@ bool dc_stream_adjust_vmin_vmax(struct dc *dc, * avoid conflicting with firmware updates. */ if (dc->ctx->dce_version > DCE_VERSION_MAX) { - if (dc->optimized_required || dc->wm_optimized_required) { + if ((dc->optimized_required || dc->wm_optimized_required) && + (stream->adjust.v_total_max != adjust->v_total_max || + stream->adjust.v_total_min != adjust->v_total_min)) { stream->adjust.timing_adjust_pending = true; return false; } @@ -934,7 +950,9 @@ static void dc_destruct(struct dc *dc) } dc_destroy_resource_pool(dc); - +#ifdef CONFIG_DRM_AMD_DC_FP + dc_destroy_soc_and_ip_translator(&dc->soc_and_ip_translator); +#endif if (dc->link_srv) link_destroy_link_service(&dc->link_srv); @@ -1138,6 +1156,9 @@ static bool dc_construct(struct dc *dc, dc->res_pool->funcs->update_bw_bounding_box(dc, dc->clk_mgr->bw_params); DC_FP_END(); } + dc->soc_and_ip_translator = dc_create_soc_and_ip_translator(dc_ctx->dce_version); + if (!dc->soc_and_ip_translator) + goto fail; #endif if (!create_links(dc, init_params->num_virtual_links)) @@ -2398,6 +2419,18 @@ enum dc_status dc_commit_streams(struct dc *dc, struct dc_commit_streams_params goto fail; } + /* + * If not already seamless, make transition seamless by inserting intermediate minimal transition + */ + if (dc->hwss.is_pipe_topology_transition_seamless && + !dc->hwss.is_pipe_topology_transition_seamless(dc, dc->current_state, context)) { + res = commit_minimal_transition_state(dc, context); + if (res != DC_OK) { + BREAK_TO_DEBUGGER(); + goto fail; + } + } + res = dc_commit_state_no_check(dc, context); for (i = 0; i < params->stream_count; i++) { @@ -3377,7 +3410,7 @@ static void update_seamless_boot_flags(struct dc *dc, int surface_count, struct dc_stream_state *stream) { - if (get_seamless_boot_stream_count(context) > 0 && surface_count > 0) { + if (get_seamless_boot_stream_count(context) > 0 && (surface_count > 0 || stream->dpms_off)) { /* Optimize seamless boot flag keeps clocks and watermarks high until * first flip. After first flip, optimization is required to lower * bandwidth. Important to note that it is expected UEFI will @@ -5089,129 +5122,6 @@ static bool fast_update_only(struct dc *dc, && !full_update_required(dc, srf_updates, surface_count, stream_update, stream); } -static bool update_planes_and_stream_v1(struct dc *dc, - struct dc_surface_update *srf_updates, int surface_count, - struct dc_stream_state *stream, - struct dc_stream_update *stream_update, - struct dc_state *state) -{ - const struct dc_stream_status *stream_status; - enum surface_update_type update_type; - struct dc_state *context; - struct dc_context *dc_ctx = dc->ctx; - int i, j; - struct dc_fast_update fast_update[MAX_SURFACES] = {0}; - - dc_exit_ips_for_hw_access(dc); - - populate_fast_updates(fast_update, srf_updates, surface_count, stream_update); - stream_status = dc_stream_get_status(stream); - context = dc->current_state; - - update_type = dc_check_update_surfaces_for_stream( - dc, srf_updates, surface_count, stream_update, stream_status); - /* It is possible to receive a flip for one plane while there are multiple flip_immediate planes in the same stream. - * E.g. Desktop and MPO plane are flip_immediate but only the MPO plane received a flip - * Force the other flip_immediate planes to flip so GSL doesn't wait for a flip that won't come. - */ - force_immediate_gsl_plane_flip(dc, srf_updates, surface_count); - - if (update_type >= UPDATE_TYPE_FULL) { - - /* initialize scratch memory for building context */ - context = dc_state_create_copy(state); - if (context == NULL) { - DC_ERROR("Failed to allocate new validate context!\n"); - return false; - } - - for (i = 0; i < dc->res_pool->pipe_count; i++) { - struct pipe_ctx *new_pipe = &context->res_ctx.pipe_ctx[i]; - struct pipe_ctx *old_pipe = &dc->current_state->res_ctx.pipe_ctx[i]; - - if (new_pipe->plane_state && new_pipe->plane_state != old_pipe->plane_state) - new_pipe->plane_state->force_full_update = true; - } - } else if (update_type == UPDATE_TYPE_FAST) { - /* - * Previous frame finished and HW is ready for optimization. - */ - dc_post_update_surfaces_to_stream(dc); - } - - for (i = 0; i < surface_count; i++) { - struct dc_plane_state *surface = srf_updates[i].surface; - - copy_surface_update_to_plane(surface, &srf_updates[i]); - - if (update_type >= UPDATE_TYPE_MED) { - for (j = 0; j < dc->res_pool->pipe_count; j++) { - struct pipe_ctx *pipe_ctx = - &context->res_ctx.pipe_ctx[j]; - - if (pipe_ctx->plane_state != surface) - continue; - - resource_build_scaling_params(pipe_ctx); - } - } - } - - copy_stream_update_to_stream(dc, context, stream, stream_update); - - if (update_type >= UPDATE_TYPE_FULL) { - if (dc->res_pool->funcs->validate_bandwidth(dc, context, DC_VALIDATE_MODE_AND_PROGRAMMING) != DC_OK) { - DC_ERROR("Mode validation failed for stream update!\n"); - dc_state_release(context); - return false; - } - } - - TRACE_DC_PIPE_STATE(pipe_ctx, i, MAX_PIPES); - - if (fast_update_only(dc, fast_update, srf_updates, surface_count, stream_update, stream) && - !dc->debug.enable_legacy_fast_update) { - commit_planes_for_stream_fast(dc, - srf_updates, - surface_count, - stream, - stream_update, - update_type, - context); - } else { - commit_planes_for_stream( - dc, - srf_updates, - surface_count, - stream, - stream_update, - update_type, - context); - } - /*update current_State*/ - if (dc->current_state != context) { - - struct dc_state *old = dc->current_state; - - dc->current_state = context; - dc_state_release(old); - - for (i = 0; i < dc->res_pool->pipe_count; i++) { - struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; - - if (pipe_ctx->plane_state && pipe_ctx->stream == stream) - pipe_ctx->plane_state->force_full_update = false; - } - } - - /* Legacy optimization path for DCE. */ - if (update_type >= UPDATE_TYPE_FULL && dc_ctx->dce_version < DCE_VERSION_MAX) { - dc_post_update_surfaces_to_stream(dc); - TRACE_DCE_CLOCK_STATE(&context->bw_ctx.bw.dce); - } - return true; -} - static bool update_planes_and_stream_v2(struct dc *dc, struct dc_surface_update *srf_updates, int surface_count, struct dc_stream_state *stream, @@ -5469,12 +5379,10 @@ void dc_commit_updates_for_stream(struct dc *dc, if (dc->ctx->dce_version >= DCN_VERSION_4_01) { ret = update_planes_and_stream_v3(dc, srf_updates, surface_count, stream, stream_update); - } else if (dc->ctx->dce_version >= DCN_VERSION_3_2) { + } else { ret = update_planes_and_stream_v2(dc, srf_updates, surface_count, stream, stream_update); - } else - ret = update_planes_and_stream_v1(dc, srf_updates, surface_count, stream, - stream_update, state); + } if (ret && dc->ctx->dce_version >= DCN_VERSION_3_2) clear_update_flags(srf_updates, surface_count, stream); @@ -6450,3 +6358,21 @@ bool dc_can_clear_cursor_limit(struct dc *dc) return false; } + +void dc_get_underflow_debug_data_for_otg(struct dc *dc, int primary_otg_inst, + struct dc_underflow_debug_data *out_data) +{ + struct timing_generator *tg = NULL; + + for (int i = 0; i < MAX_PIPES; i++) { + if (dc->res_pool->timing_generators[i] && + dc->res_pool->timing_generators[i]->inst == primary_otg_inst) { + tg = dc->res_pool->timing_generators[i]; + break; + } + } + + dc_exit_ips_for_hw_access(dc); + if (dc->hwss.get_underflow_debug_data) + dc->hwss.get_underflow_debug_data(dc, tg, out_data); +} diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c index ec4e80e5b6eb..d82b1cb467f4 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c @@ -1177,6 +1177,8 @@ void hwss_wait_for_odm_update_pending_complete(struct dc *dc, struct dc_state *c tg = otg_master->stream_res.tg; if (tg->funcs->wait_odm_doublebuffer_pending_clear) tg->funcs->wait_odm_doublebuffer_pending_clear(tg); + if (tg->funcs->wait_otg_disable) + tg->funcs->wait_otg_disable(tg); } /* ODM update may require to reprogram blank pattern for each OPP */ diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_exports.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_exports.c index 130455f2802a..b7a5de4ecb61 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link_exports.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_exports.c @@ -520,3 +520,10 @@ enum dc_status dc_link_validate_dp_tunneling_bandwidth(const struct dc *dc, cons return dc->link_srv->validate_dp_tunnel_bandwidth(dc, new_ctx); } +void dc_link_get_alpm_support(struct dc_link *link, + bool *auxless_support, + bool *auxwake_support) +{ + link->dc->link_srv->edp_get_alpm_support(link, auxless_support, auxwake_support); +} + diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c index 4d6181e7c612..d712548b1927 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c @@ -165,7 +165,13 @@ enum dce_version resource_parse_asic_id(struct hw_asic_id asic_id) case FAMILY_NV: dc_version = DCN_VERSION_2_0; - if (asic_id.chip_id == DEVICE_ID_NV_13FE || asic_id.chip_id == DEVICE_ID_NV_143F) { + if (asic_id.chip_id == DEVICE_ID_NV_13FE || + asic_id.chip_id == DEVICE_ID_NV_143F || + asic_id.chip_id == DEVICE_ID_NV_13F9 || + asic_id.chip_id == DEVICE_ID_NV_13FA || + asic_id.chip_id == DEVICE_ID_NV_13FB || + asic_id.chip_id == DEVICE_ID_NV_13FC || + asic_id.chip_id == DEVICE_ID_NV_13DB) { dc_version = DCN_VERSION_2_01; break; } diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_stat.c b/drivers/gpu/drm/amd/display/dc/core/dc_stat.c index fe9f99f1bdf9..f976ffd6d466 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_stat.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_stat.c @@ -65,7 +65,7 @@ void dc_stat_get_dmub_notification(const struct dc *dc, struct dmub_notification notify->type == DMUB_NOTIFICATION_DPIA_NOTIFICATION || notify->type == DMUB_NOTIFICATION_SET_CONFIG_REPLY) { notify->link_index = - get_link_index_from_dpia_port_index(dc, notify->link_index); + get_link_index_from_dpia_port_index(dc, notify->instance); } } diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_stream.c b/drivers/gpu/drm/amd/display/dc/core/dc_stream.c index 4d6bc9fd4faa..9ac2d41f8fca 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_stream.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_stream.c @@ -316,6 +316,9 @@ bool dc_stream_set_cursor_attributes( { bool result = false; + if (!stream) + return false; + if (dc_stream_check_cursor_attributes(stream, stream->ctx->dc->current_state, attributes)) { stream->cursor_attributes = *attributes; result = true; @@ -331,7 +334,10 @@ bool dc_stream_program_cursor_attributes( struct dc *dc; bool reset_idle_optimizations = false; - dc = stream ? stream->ctx->dc : NULL; + if (!stream) + return false; + + dc = stream->ctx->dc; if (dc_stream_set_cursor_attributes(stream, attributes)) { dc_z10_restore(dc); diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h index 59c07756130d..b41e66c31e6a 100644 --- a/drivers/gpu/drm/amd/display/dc/dc.h +++ b/drivers/gpu/drm/amd/display/dc/dc.h @@ -55,7 +55,7 @@ struct aux_payload; struct set_config_cmd_payload; struct dmub_notification; -#define DC_VER "3.2.340" +#define DC_VER "3.2.348" /** * MAX_SURFACES - representative of the upper bound of surfaces that can be piped to a single CRTC @@ -234,6 +234,7 @@ struct lut3d_caps { * @ogam_ram: programmable out gamma LUT * @ocsc: output color space conversion matrix * @num_3dluts: MPC 3D LUT; always assumes a preceding shaper LUT + * @num_rmcm_3dluts: number of RMCM 3D LUTS; always assumes a preceding shaper LUT * @shared_3d_lut: shared 3D LUT flag. Can be either DPP or MPC, but single * instance * @ogam_rom_caps: pre-definied curve caps for regamma 1D LUT @@ -694,6 +695,15 @@ struct dc_clocks { int idle_fclk_khz; int subvp_prefetch_dramclk_khz; int subvp_prefetch_fclk_khz; + + /* Stutter efficiency is technically not clock values + * but stored here so the values are part of the update_clocks call similar to num_ways + * Efficiencies are stored as percentage (0-100) + */ + struct { + uint8_t base_efficiency; //LP1 + uint8_t low_power_efficiency; //LP2 + } stutter_efficiency; }; struct dc_bw_validation_profile { @@ -839,7 +849,8 @@ union dpia_debug_options { uint32_t enable_force_tbt3_work_around:1; /* bit 4 */ uint32_t disable_usb4_pm_support:1; /* bit 5 */ uint32_t enable_usb4_bw_zero_alloc_patch:1; /* bit 6 */ - uint32_t reserved:25; + uint32_t enable_bw_allocation_mode:1; /* bit 7 */ + uint32_t reserved:24; } bits; uint32_t raw; }; @@ -1072,6 +1083,7 @@ struct dc_debug_options { unsigned int force_mall_ss_num_ways; bool alloc_extra_way_for_cursor; uint32_t subvp_extra_lines; + bool disable_force_pstate_allow_on_hw_release; bool force_usr_allow; /* uses value at boot and disables switch */ bool disable_dtb_ref_clk_switch; @@ -1145,6 +1157,11 @@ struct dc_debug_options { bool enable_hblank_borrow; bool force_subvp_df_throttle; uint32_t acpi_transition_bitmasks[MAX_PIPES]; + unsigned int auxless_alpm_lfps_setup_ns; + unsigned int auxless_alpm_lfps_period_ns; + unsigned int auxless_alpm_lfps_silence_ns; + unsigned int auxless_alpm_lfps_t1t2_us; + short auxless_alpm_lfps_t1t2_offset_us; }; @@ -1305,6 +1322,32 @@ union dc_3dlut_state { }; +#define MATRIX_9C__DIM_128_ALIGNED_LEN 16 // 9+8 : 9 * 8 + 7 * 8 = 72 + 56 = 128 % 128 = 0 +#define MATRIX_17C__DIM_128_ALIGNED_LEN 32 //17+15: 17 * 8 + 15 * 8 = 136 + 120 = 256 % 128 = 0 +#define MATRIX_33C__DIM_128_ALIGNED_LEN 64 //17+47: 17 * 8 + 47 * 8 = 136 + 376 = 512 % 128 = 0 + +struct lut_rgb { + uint16_t b; + uint16_t g; + uint16_t r; + uint16_t padding; +}; + +//this structure maps directly to how the lut will read it from memory +struct lut_mem_mapping { + union { + //NATIVE MODE 1, 2 + //RGB layout [b][g][r] //red is 128 byte aligned + //BGR layout [r][g][b] //blue is 128 byte aligned + struct lut_rgb rgb_17c[17][17][MATRIX_17C__DIM_128_ALIGNED_LEN]; + struct lut_rgb rgb_33c[33][33][MATRIX_33C__DIM_128_ALIGNED_LEN]; + + //TRANSFORMED + uint16_t linear_rgb[(33*33*33*4/128+1)*128]; + }; + uint16_t size; +}; + struct dc_rmcm_3dlut { bool isInUse; const struct dc_stream_state *stream; @@ -1733,7 +1776,7 @@ struct dc { struct dml2_configuration_options dml2_options; struct dml2_configuration_options dml2_dc_power_options; enum dc_acpi_cm_power_state power_state; - + struct soc_and_ip_translator *soc_and_ip_translator; }; struct dc_scaling_info { @@ -1786,6 +1829,23 @@ struct dc_surface_update { struct dc_bias_and_scale bias_and_scale; }; +struct dc_underflow_debug_data { + uint32_t otg_inst; + uint32_t otg_underflow; + uint32_t h_position; + uint32_t v_position; + uint32_t otg_frame_count; + struct dc_underflow_per_hubp_debug_data { + uint32_t hubp_underflow; + uint32_t hubp_in_blank; + uint32_t hubp_readline; + uint32_t det_config_error; + } hubps[MAX_PIPES]; + uint32_t curr_det_sizes[MAX_PIPES]; + uint32_t target_det_sizes[MAX_PIPES]; + uint32_t compbuf_config_error; +}; + /* * Create a new surface with default parameters; */ @@ -1804,8 +1864,6 @@ void dc_3dlut_func_retain(struct dc_3dlut *lut); void dc_post_update_surfaces_to_stream( struct dc *dc); -#include "dc_stream.h" - /** * struct dc_validation_set - Struct to store surface/stream associations for validation */ @@ -2447,6 +2505,12 @@ void dc_link_dp_dpia_handle_usb4_bandwidth_allocation_for_link( */ enum dc_status dc_link_validate_dp_tunneling_bandwidth(const struct dc *dc, const struct dc_state *new_ctx); +/* + * Get if ALPM is supported by the link + */ +void dc_link_get_alpm_support(struct dc_link *link, bool *auxless_support, + bool *auxwake_support); + /* Sink Interfaces - A sink corresponds to a display output device */ struct dc_container_id { @@ -2674,4 +2738,17 @@ bool dc_is_timing_changed(struct dc_stream_state *cur_stream, bool dc_is_cursor_limit_pending(struct dc *dc); bool dc_can_clear_cursor_limit(struct dc *dc); +/** + * dc_get_underflow_debug_data_for_otg() - Retrieve underflow debug data. + * + * @dc: Pointer to the display core context. + * @primary_otg_inst: Instance index of the primary OTG that underflowed. + * @out_data: Pointer to a dc_underflow_debug_data struct to be filled with debug information. + * + * This function collects and logs underflow-related HW states when underflow happens, + * including OTG underflow status, current read positions, frame count, and per-HUBP debug data. + * The results are stored in the provided out_data structure for further analysis or logging. + */ +void dc_get_underflow_debug_data_for_otg(struct dc *dc, int primary_otg_inst, struct dc_underflow_debug_data *out_data); + #endif /* DC_INTERFACE_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c index f5ef1a07078e..53a088ebddef 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c +++ b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c @@ -2010,11 +2010,12 @@ bool dmub_lsdma_init(struct dc_dmub_srv *dc_dmub_srv) return result; } -bool dmub_lsdma_send_linear_copy_packet( +bool dmub_lsdma_send_linear_copy_command( struct dc_dmub_srv *dc_dmub_srv, uint64_t src_addr, uint64_t dst_addr, - uint32_t count) + uint32_t count +) { struct dc_context *dc_ctx = dc_dmub_srv->ctx; union dmub_rb_cmd cmd; @@ -2042,9 +2043,54 @@ bool dmub_lsdma_send_linear_copy_packet( return result; } +bool dmub_lsdma_send_linear_sub_window_copy_command( + struct dc_dmub_srv *dc_dmub_srv, + struct lsdma_linear_sub_window_copy_params copy_data +) +{ + struct dc_context *dc_ctx = dc_dmub_srv->ctx; + union dmub_rb_cmd cmd; + enum dm_dmub_wait_type wait_type; + struct dmub_cmd_lsdma_data *lsdma_data = &cmd.lsdma.lsdma_data; + bool result; + + memset(&cmd, 0, sizeof(cmd)); + + cmd.cmd_common.header.type = DMUB_CMD__LSDMA; + cmd.cmd_common.header.sub_type = DMUB_CMD__LSDMA_LINEAR_SUB_WINDOW_COPY; + wait_type = DM_DMUB_WAIT_TYPE_NO_WAIT; + + lsdma_data->u.linear_sub_window_copy_data.tmz = copy_data.tmz; + lsdma_data->u.linear_sub_window_copy_data.element_size = copy_data.element_size; + lsdma_data->u.linear_sub_window_copy_data.src_lo = copy_data.src_lo; + lsdma_data->u.linear_sub_window_copy_data.src_hi = copy_data.src_hi; + lsdma_data->u.linear_sub_window_copy_data.src_x = copy_data.src_x; + lsdma_data->u.linear_sub_window_copy_data.src_y = copy_data.src_y; + lsdma_data->u.linear_sub_window_copy_data.src_pitch = copy_data.src_pitch; + lsdma_data->u.linear_sub_window_copy_data.src_slice_pitch = copy_data.src_slice_pitch; + lsdma_data->u.linear_sub_window_copy_data.dst_lo = copy_data.dst_lo; + lsdma_data->u.linear_sub_window_copy_data.dst_hi = copy_data.dst_hi; + lsdma_data->u.linear_sub_window_copy_data.dst_x = copy_data.dst_x; + lsdma_data->u.linear_sub_window_copy_data.dst_y = copy_data.dst_y; + lsdma_data->u.linear_sub_window_copy_data.dst_pitch = copy_data.dst_pitch; + lsdma_data->u.linear_sub_window_copy_data.dst_slice_pitch = copy_data.dst_slice_pitch; + lsdma_data->u.linear_sub_window_copy_data.rect_x = copy_data.rect_x; + lsdma_data->u.linear_sub_window_copy_data.rect_y = copy_data.rect_y; + lsdma_data->u.linear_sub_window_copy_data.src_cache_policy = copy_data.src_cache_policy; + lsdma_data->u.linear_sub_window_copy_data.dst_cache_policy = copy_data.dst_cache_policy; + + result = dc_wake_and_execute_dmub_cmd(dc_ctx, &cmd, wait_type); + + if (!result) + DC_ERROR("LSDMA Linear Sub Window Copy failed in DMUB"); + + return result; +} + bool dmub_lsdma_send_tiled_to_tiled_copy_command( struct dc_dmub_srv *dc_dmub_srv, - struct lsdma_send_tiled_to_tiled_copy_command_params params) + struct lsdma_send_tiled_to_tiled_copy_command_params params +) { struct dc_context *dc_ctx = dc_dmub_srv->ctx; union dmub_rb_cmd cmd; @@ -2066,8 +2112,8 @@ bool dmub_lsdma_send_tiled_to_tiled_copy_command( lsdma_data->u.tiled_copy_data.src_y = params.src_y; lsdma_data->u.tiled_copy_data.dst_x = params.dst_x; lsdma_data->u.tiled_copy_data.dst_y = params.dst_y; - lsdma_data->u.tiled_copy_data.src_width = params.src_width - 1; // LSDMA controller expects width -1 - lsdma_data->u.tiled_copy_data.dst_width = params.dst_width - 1; // LSDMA controller expects width -1 + lsdma_data->u.tiled_copy_data.src_width = params.src_width; + lsdma_data->u.tiled_copy_data.dst_width = params.dst_width; lsdma_data->u.tiled_copy_data.src_swizzle_mode = params.swizzle_mode; lsdma_data->u.tiled_copy_data.dst_swizzle_mode = params.swizzle_mode; lsdma_data->u.tiled_copy_data.src_element_size = params.element_size; @@ -2078,8 +2124,8 @@ bool dmub_lsdma_send_tiled_to_tiled_copy_command( lsdma_data->u.tiled_copy_data.tmz = params.tmz; lsdma_data->u.tiled_copy_data.read_compress = params.read_compress; lsdma_data->u.tiled_copy_data.write_compress = params.write_compress; - lsdma_data->u.tiled_copy_data.src_height = params.src_height - 1; // LSDMA controller expects height -1 - lsdma_data->u.tiled_copy_data.dst_height = params.dst_height - 1; // LSDMA controller expects height -1 + lsdma_data->u.tiled_copy_data.src_height = params.src_height; + lsdma_data->u.tiled_copy_data.dst_height = params.dst_height; lsdma_data->u.tiled_copy_data.data_format = params.data_format; lsdma_data->u.tiled_copy_data.max_com = params.max_com; lsdma_data->u.tiled_copy_data.max_uncom = params.max_uncom; @@ -2097,7 +2143,8 @@ bool dmub_lsdma_send_pio_copy_command( uint64_t src_addr, uint64_t dst_addr, uint32_t byte_count, - uint32_t overlap_disable) + uint32_t overlap_disable +) { struct dc_context *dc_ctx = dc_dmub_srv->ctx; union dmub_rb_cmd cmd; @@ -2130,7 +2177,8 @@ bool dmub_lsdma_send_pio_constfill_command( struct dc_dmub_srv *dc_dmub_srv, uint64_t dst_addr, uint32_t byte_count, - uint32_t data) + uint32_t data +) { struct dc_context *dc_ctx = dc_dmub_srv->ctx; union dmub_rb_cmd cmd; diff --git a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h index 8ea320f21269..7ef93444ef3c 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h +++ b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h @@ -211,11 +211,45 @@ void dc_dmub_srv_fams2_passthrough_flip( int surface_count); bool dmub_lsdma_init(struct dc_dmub_srv *dc_dmub_srv); -bool dmub_lsdma_send_linear_copy_packet( +bool dmub_lsdma_send_linear_copy_command( struct dc_dmub_srv *dc_dmub_srv, uint64_t src_addr, uint64_t dst_addr, uint32_t count); + +struct lsdma_linear_sub_window_copy_params { + uint32_t src_lo; + uint32_t src_hi; + + uint32_t dst_lo; + uint32_t dst_hi; + + uint32_t src_x : 16; + uint32_t src_y : 16; + + uint32_t dst_x : 16; + uint32_t dst_y : 16; + + uint32_t rect_x : 16; + uint32_t rect_y : 16; + + uint32_t src_pitch : 16; + uint32_t dst_pitch : 16; + + uint32_t src_slice_pitch; + uint32_t dst_slice_pitch; + + uint32_t tmz : 1; + uint32_t element_size : 3; + uint32_t src_cache_policy : 3; + uint32_t dst_cache_policy : 3; + uint32_t padding : 22; +}; + +bool dmub_lsdma_send_linear_sub_window_copy_command( + struct dc_dmub_srv *dc_dmub_srv, + struct lsdma_linear_sub_window_copy_params copy_data +); bool dmub_lsdma_send_pio_copy_command( struct dc_dmub_srv *dc_dmub_srv, uint64_t src_addr, diff --git a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h index 5ce1be362534..3a3ec38cdf8b 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h +++ b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h @@ -1021,7 +1021,8 @@ union dp_128b_132b_supported_lttpr_link_rates { union dp_alpm_lttpr_cap { struct { uint8_t AUX_LESS_ALPM_SUPPORTED :1; - uint8_t RESERVED :7; + uint8_t ASSR_SUPPORTED :1; + uint8_t RESERVED :6; } bits; uint8_t raw; }; @@ -1119,10 +1120,11 @@ union dp_128b_132b_training_aux_rd_interval { union edp_alpm_caps { struct { - uint8_t AUX_WAKE_ALPM_CAP :1; - uint8_t PM_STATE_2A_SUPPORT :1; - uint8_t AUX_LESS_ALPM_CAP :1; - uint8_t RESERVED :5; + uint8_t AUX_WAKE_ALPM_CAP :1; + uint8_t PM_STATE_2A_SUPPORT :1; + uint8_t AUX_LESS_ALPM_CAP :1; + uint8_t AUX_LESS_ALPM_ML_PHY_SLEEP_STATUS_SUPPORTED :1; + uint8_t RESERVED :4; } bits; uint8_t raw; }; @@ -1347,7 +1349,9 @@ union dpcd_alpm_configuration { struct { unsigned char ENABLE : 1; unsigned char IRQ_HPD_ENABLE : 1; - unsigned char RESERVED : 6; + unsigned char ALPM_MODE_SEL : 1; + unsigned char ACDS_PERIOD_DURATION : 1; + unsigned char RESERVED : 4; } bits; unsigned char raw; }; diff --git a/drivers/gpu/drm/amd/display/dc/dc_helper.c b/drivers/gpu/drm/amd/display/dc/dc_helper.c index 7217de258851..51e41aed7316 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_helper.c +++ b/drivers/gpu/drm/amd/display/dc/dc_helper.c @@ -732,7 +732,7 @@ char *dce_version_to_string(const int version) case DCN_VERSION_3_03: return "DCN 3.0.3"; case DCN_VERSION_3_1: - return "DCN 3.1"; + return "DCN 3.1.2"; case DCN_VERSION_3_14: return "DCN 3.1.4"; case DCN_VERSION_3_15: diff --git a/drivers/gpu/drm/amd/display/dc/dc_types.h b/drivers/gpu/drm/amd/display/dc/dc_types.h index 375ca2f13b7a..619834a328a3 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_types.h +++ b/drivers/gpu/drm/amd/display/dc/dc_types.h @@ -563,6 +563,12 @@ struct dc_info_packet_128 { uint8_t sb[128]; }; +struct dc_edid_read_policy { + uint32_t max_retry_count; + uint32_t delay_time_ms; + uint32_t ignore_checksum; +}; + #define DC_PLANE_UPDATE_TIMES_MAX 10 struct dc_plane_flip_time { @@ -571,6 +577,12 @@ struct dc_plane_flip_time { unsigned int prev_update_time_in_us; }; +enum dc_alpm_mode { + DC_ALPM_AUXWAKE = 0, + DC_ALPM_AUXLESS = 1, + DC_ALPM_UNSUPPORTED = 0xF, +}; + enum dc_psr_state { PSR_STATE0 = 0x0, PSR_STATE1, @@ -616,6 +628,7 @@ struct psr_config { unsigned int line_time_in_us; uint8_t rate_control_caps; uint16_t dsc_slice_height; + bool os_request_force_ffu; }; union dmcu_psr_level { @@ -728,6 +741,7 @@ struct psr_context { unsigned int line_time_in_us; uint8_t rate_control_caps; uint16_t dsc_slice_height; + bool os_request_force_ffu; }; struct colorspace_transform { @@ -1137,6 +1151,10 @@ struct replay_config { bool low_rr_supported; /* Replay Video Conferencing Optimization Enabled */ bool replay_video_conferencing_optimization_enabled; + /* Replay alpm mode */ + enum dc_alpm_mode alpm_mode; + /* Replay full screen only */ + bool os_request_force_ffu; }; /* Replay feature flags*/ diff --git a/drivers/gpu/drm/amd/display/dc/dccg/dcn401/dcn401_dccg.c b/drivers/gpu/drm/amd/display/dc/dccg/dcn401/dcn401_dccg.c index 668ee2d405fd..0b8ed9b94d3c 100644 --- a/drivers/gpu/drm/amd/display/dc/dccg/dcn401/dcn401_dccg.c +++ b/drivers/gpu/drm/amd/display/dc/dccg/dcn401/dcn401_dccg.c @@ -619,7 +619,7 @@ void dccg401_set_dp_dto( dto_integer = div_u64(params->pixclk_hz, dto_modulo_hz); dto_phase_hz = params->pixclk_hz - dto_integer * dto_modulo_hz; - if (dto_phase_hz <= 0) { + if (dto_phase_hz <= 0 && dto_integer <= 0) { /* negative pixel rate should never happen */ BREAK_TO_DEBUGGER(); return; diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_hw.c b/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_hw.c index 0421b267a0b5..365dd2e37aea 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_hw.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_hw.c @@ -591,7 +591,7 @@ static bool dce_i2c_hw_engine_submit_payload(struct dce_i2c_hw *dce_i2c_hw, DCE_I2C_TRANSACTION_ACTION_I2C_WRITE; - request.address = (uint8_t) ((payload->address << 1) | !payload->write); + request.address = (uint8_t) ((payload->address << 1) | (payload->write ? 0 : 1)); request.length = payload->length; request.data = payload->data; diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_sw.c b/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_sw.c index e188447c8156..2d73b94c515c 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_sw.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_sw.c @@ -451,7 +451,7 @@ static bool dce_i2c_sw_engine_submit_payload(struct dce_i2c_sw *engine, DCE_I2C_TRANSACTION_ACTION_I2C_WRITE_MOT : DCE_I2C_TRANSACTION_ACTION_I2C_WRITE; - request.address = (uint8_t) ((payload->address << 1) | !payload->write); + request.address = (uint8_t) ((payload->address << 1) | (payload->write ? 0 : 1)); request.length = payload->length; request.data = payload->data; diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_link_encoder.c b/drivers/gpu/drm/amd/display/dc/dce/dce_link_encoder.c index 4a9d07c31bc5..0c50fe266c8a 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_link_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_link_encoder.c @@ -896,13 +896,13 @@ void dce110_link_encoder_construct( enc110->base.id, &bp_cap_info); /* Override features with DCE-specific values */ - if (BP_RESULT_OK == result) { + if (result == BP_RESULT_OK) { enc110->base.features.flags.bits.IS_HBR2_CAPABLE = bp_cap_info.DP_HBR2_EN; enc110->base.features.flags.bits.IS_HBR3_CAPABLE = bp_cap_info.DP_HBR3_EN; enc110->base.features.flags.bits.HDMI_6GB_EN = bp_cap_info.HDMI_6GB_EN; - } else { + } else if (result != BP_RESULT_NORECORD) { DC_LOG_WARNING("%s: Failed to get encoder_cap_info from VBIOS with error code %d!\n", __func__, result); @@ -1798,13 +1798,13 @@ void dce60_link_encoder_construct( enc110->base.id, &bp_cap_info); /* Override features with DCE-specific values */ - if (BP_RESULT_OK == result) { + if (result == BP_RESULT_OK) { enc110->base.features.flags.bits.IS_HBR2_CAPABLE = bp_cap_info.DP_HBR2_EN; enc110->base.features.flags.bits.IS_HBR3_CAPABLE = bp_cap_info.DP_HBR3_EN; enc110->base.features.flags.bits.HDMI_6GB_EN = bp_cap_info.HDMI_6GB_EN; - } else { + } else if (result != BP_RESULT_NORECORD) { DC_LOG_WARNING("%s: Failed to get encoder_cap_info from VBIOS with error code %d!\n", __func__, result); diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c b/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c index ff3b8244ba3d..87af4fdc04a6 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c @@ -391,7 +391,7 @@ static bool dmub_psr_copy_settings(struct dmub_psr *dmub, sizeof(DP_SINK_DEVICE_STR_ID_1))) link->psr_settings.force_ffu_mode = 1; - copy_settings_data->force_ffu_mode = link->psr_settings.force_ffu_mode; + copy_settings_data->force_ffu_mode = link->psr_settings.force_ffu_mode || psr_context->os_request_force_ffu; if (((link->dpcd_caps.fec_cap.bits.FEC_CAPABLE && !link->dc->debug.disable_fec) && diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.c b/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.c index e7a318e26d38..65b979617b0c 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.c @@ -3,8 +3,8 @@ // Copyright 2024 Advanced Micro Devices, Inc. #include "dc.h" +#include "link.h" #include "dc_dmub_srv.h" -#include "dc_dp_types.h" #include "dmub/dmub_srv.h" #include "core_types.h" #include "dmub_replay.h" @@ -44,45 +44,21 @@ static void dmub_replay_get_state(struct dmub_replay *dmub, enum replay_state *s /* * Enable/Disable Replay. */ -static void dmub_replay_enable(struct dmub_replay *dmub, bool enable, bool wait, uint8_t panel_inst, - struct dc_link *link) +static void dmub_replay_enable(struct dmub_replay *dmub, bool enable, bool wait, uint8_t panel_inst) { union dmub_rb_cmd cmd; struct dc_context *dc = dmub->ctx; uint32_t retry_count; enum replay_state state = REPLAY_STATE_0; - struct pipe_ctx *pipe_ctx = NULL; - struct resource_context *res_ctx = &link->ctx->dc->current_state->res_ctx; - uint8_t i; memset(&cmd, 0, sizeof(cmd)); cmd.replay_enable.header.type = DMUB_CMD__REPLAY; cmd.replay_enable.data.panel_inst = panel_inst; cmd.replay_enable.header.sub_type = DMUB_CMD__REPLAY_ENABLE; - if (enable) { + if (enable) cmd.replay_enable.data.enable = REPLAY_ENABLE; - // hpo stream/link encoder assignments are not static, need to update everytime we try to enable replay - if (link->cur_link_settings.link_rate >= LINK_RATE_UHBR10) { - for (i = 0; i < MAX_PIPES; i++) { - if (res_ctx && - res_ctx->pipe_ctx[i].stream && - res_ctx->pipe_ctx[i].stream->link && - res_ctx->pipe_ctx[i].stream->link == link && - res_ctx->pipe_ctx[i].stream->link->connector_signal == SIGNAL_TYPE_EDP) { - pipe_ctx = &res_ctx->pipe_ctx[i]; - //TODO: refactor for multi edp support - break; - } - } - - if (!pipe_ctx) - return; - - cmd.replay_enable.data.hpo_stream_enc_inst = pipe_ctx->stream_res.hpo_dp_stream_enc->inst; - cmd.replay_enable.data.hpo_link_enc_inst = pipe_ctx->link_res.hpo_dp_link_enc->inst; - } - } else + else cmd.replay_enable.data.enable = REPLAY_DISABLE; cmd.replay_enable.header.payload_bytes = sizeof(struct dmub_rb_cmd_replay_enable_data); @@ -174,17 +150,6 @@ static bool dmub_replay_copy_settings(struct dmub_replay *dmub, copy_settings_data->digbe_inst = replay_context->digbe_inst; copy_settings_data->digfe_inst = replay_context->digfe_inst; - if (link->cur_link_settings.link_rate >= LINK_RATE_UHBR10) { - if (pipe_ctx->stream_res.hpo_dp_stream_enc) - copy_settings_data->hpo_stream_enc_inst = pipe_ctx->stream_res.hpo_dp_stream_enc->inst; - else - copy_settings_data->hpo_stream_enc_inst = 0; - if (pipe_ctx->link_res.hpo_dp_link_enc) - copy_settings_data->hpo_link_enc_inst = pipe_ctx->link_res.hpo_dp_link_enc->inst; - else - copy_settings_data->hpo_link_enc_inst = 0; - } - if (pipe_ctx->plane_res.dpp) copy_settings_data->dpp_inst = pipe_ctx->plane_res.dpp->inst; else @@ -225,6 +190,18 @@ static bool dmub_replay_copy_settings(struct dmub_replay *dmub, else copy_settings_data->flags.bitfields.force_wakeup_by_tps3 = 0; + copy_settings_data->flags.bitfields.alpm_mode = (enum dmub_alpm_mode)link->replay_settings.config.alpm_mode; + if (link->replay_settings.config.alpm_mode == DC_ALPM_AUXLESS) { + copy_settings_data->auxless_alpm_data.lfps_setup_ns = dc->dc->debug.auxless_alpm_lfps_setup_ns; + copy_settings_data->auxless_alpm_data.lfps_period_ns = dc->dc->debug.auxless_alpm_lfps_period_ns; + copy_settings_data->auxless_alpm_data.lfps_silence_ns = dc->dc->debug.auxless_alpm_lfps_silence_ns; + copy_settings_data->auxless_alpm_data.lfps_t1_t2_override_us = + dc->dc->debug.auxless_alpm_lfps_t1t2_us; + copy_settings_data->auxless_alpm_data.lfps_t1_t2_offset_us = + dc->dc->debug.auxless_alpm_lfps_t1t2_offset_us; + copy_settings_data->auxless_alpm_data.lttpr_count = link->dc->link_srv->dp_get_lttpr_count(link); + } + dc_wake_and_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); return true; @@ -247,7 +224,6 @@ static void dmub_replay_set_coasting_vtotal(struct dmub_replay *dmub, pCmd->header.type = DMUB_CMD__REPLAY; pCmd->header.sub_type = DMUB_CMD__REPLAY_SET_COASTING_VTOTAL; pCmd->header.payload_bytes = sizeof(struct dmub_cmd_replay_set_coasting_vtotal_data); - pCmd->replay_set_coasting_vtotal_data.panel_inst = panel_inst; pCmd->replay_set_coasting_vtotal_data.coasting_vtotal = (coasting_vtotal & 0xFFFF); pCmd->replay_set_coasting_vtotal_data.coasting_vtotal_high = (coasting_vtotal & 0xFFFF0000) >> 16; diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.h b/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.h index ccbe385e132c..e6346c0ffc0e 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.h +++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.h @@ -19,7 +19,7 @@ struct dmub_replay_funcs { void (*replay_get_state)(struct dmub_replay *dmub, enum replay_state *state, uint8_t panel_inst); void (*replay_enable)(struct dmub_replay *dmub, bool enable, bool wait, - uint8_t panel_inst, struct dc_link *link); + uint8_t panel_inst); bool (*replay_copy_settings)(struct dmub_replay *dmub, struct dc_link *link, struct replay_context *replay_context, uint8_t panel_inst); void (*replay_set_power_opt)(struct dmub_replay *dmub, unsigned int power_opt, diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20.c b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20.c index 390c1a77fda6..9c58ff1069d6 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20.c @@ -646,7 +646,7 @@ static void get_meta_and_pte_attr(struct display_mode_lib *mode_lib, // the dpte_group_bytes is reduced for the specific case of vertical // access of a tile surface that has dpte request of 8x1 ptes. - if (!surf_linear & (log2_dpte_req_height_ptes == 0) & surf_vert) //reduced, in this case, will have page fault within a group + if (!surf_linear && (log2_dpte_req_height_ptes == 0) && surf_vert) //reduced, in this case, will have page fault within a group rq_sizing_param->dpte_group_bytes = 512; else //full size diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20v2.c b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20v2.c index 843d6004258c..570e6e39eb45 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20v2.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20v2.c @@ -646,7 +646,7 @@ static void get_meta_and_pte_attr(struct display_mode_lib *mode_lib, // the dpte_group_bytes is reduced for the specific case of vertical // access of a tile surface that has dpte request of 8x1 ptes. - if (!surf_linear & (log2_dpte_req_height_ptes == 0) & surf_vert) //reduced, in this case, will have page fault within a group + if (!surf_linear && (log2_dpte_req_height_ptes == 0) && surf_vert) //reduced, in this case, will have page fault within a group rq_sizing_param->dpte_group_bytes = 512; else //full size diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_rq_dlg_calc_21.c b/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_rq_dlg_calc_21.c index 5718000627b0..f549da082c01 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_rq_dlg_calc_21.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_rq_dlg_calc_21.c @@ -652,7 +652,7 @@ static void get_meta_and_pte_attr( if (hostvm_enable) rq_sizing_param->dpte_group_bytes = 512; else { - if (!surf_linear & (log2_dpte_req_height_ptes == 0) & surf_vert) //reduced, in this case, will have page fault within a group + if (!surf_linear && (log2_dpte_req_height_ptes == 0) && surf_vert) //reduced, in this case, will have page fault within a group rq_sizing_param->dpte_group_bytes = 512; else //full size diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_rq_dlg_calc_30.c b/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_rq_dlg_calc_30.c index 8d4873f80df0..4fb37df54d59 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_rq_dlg_calc_30.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_rq_dlg_calc_30.c @@ -620,7 +620,7 @@ static void get_meta_and_pte_attr(struct display_mode_lib *mode_lib, if (hostvm_enable) rq_sizing_param->dpte_group_bytes = 512; else { - if (!surf_linear & (log2_dpte_req_height_ptes == 0) & surf_vert) //reduced, in this case, will have page fault within a group + if (!surf_linear && (log2_dpte_req_height_ptes == 0) && surf_vert) //reduced, in this case, will have page fault within a group rq_sizing_param->dpte_group_bytes = 512; else rq_sizing_param->dpte_group_bytes = 2048; diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn301/dcn301_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn301/dcn301_fpu.c index 0c0b2d67c9cd..1aaa77265eed 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn301/dcn301_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn301/dcn301_fpu.c @@ -326,7 +326,7 @@ void dcn301_fpu_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_p struct dcn301_resource_pool *pool = TO_DCN301_RES_POOL(dc->res_pool); struct clk_limit_table *clk_table = &bw_params->clk_table; unsigned int i, closest_clk_lvl; - int j; + int j = 0, max_dispclk_mhz = 0, max_dppclk_mhz = 0; dc_assert_fp_enabled(); @@ -338,6 +338,15 @@ void dcn301_fpu_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_p dcn3_01_soc.num_chans = bw_params->num_channels; ASSERT(clk_table->num_entries); + + /* Prepass to find max clocks independent of voltage level. */ + for (i = 0; i < clk_table->num_entries; ++i) { + if (clk_table->entries[i].dispclk_mhz > max_dispclk_mhz) + max_dispclk_mhz = clk_table->entries[i].dispclk_mhz; + if (clk_table->entries[i].dppclk_mhz > max_dppclk_mhz) + max_dppclk_mhz = clk_table->entries[i].dppclk_mhz; + } + for (i = 0; i < clk_table->num_entries; i++) { /* loop backwards*/ for (closest_clk_lvl = 0, j = dcn3_01_soc.num_states - 1; j >= 0; j--) { @@ -353,8 +362,13 @@ void dcn301_fpu_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_p s[i].socclk_mhz = clk_table->entries[i].socclk_mhz; s[i].dram_speed_mts = clk_table->entries[i].memclk_mhz * 2; - s[i].dispclk_mhz = dcn3_01_soc.clock_limits[closest_clk_lvl].dispclk_mhz; - s[i].dppclk_mhz = dcn3_01_soc.clock_limits[closest_clk_lvl].dppclk_mhz; + /* Clocks independent of voltage level. */ + s[i].dispclk_mhz = max_dispclk_mhz ? max_dispclk_mhz : + dcn3_01_soc.clock_limits[closest_clk_lvl].dispclk_mhz; + + s[i].dppclk_mhz = max_dppclk_mhz ? max_dppclk_mhz : + dcn3_01_soc.clock_limits[closest_clk_lvl].dppclk_mhz; + s[i].dram_bw_per_chan_gbps = dcn3_01_soc.clock_limits[closest_clk_lvl].dram_bw_per_chan_gbps; s[i].dscclk_mhz = dcn3_01_soc.clock_limits[closest_clk_lvl].dscclk_mhz; @@ -435,12 +449,12 @@ void dcn301_fpu_calculate_wm_and_dlg(struct dc *dc, &context->bw_ctx.dml, pipes, pipe_cnt); /* WM Set C */ table_entry = &bw_params->wm_table.entries[WM_C]; - vlevel = min(max(vlevel_req, 2), vlevel_max); + vlevel = clamp(vlevel_req, 2, vlevel_max); calculate_wm_set_for_vlevel(vlevel, table_entry, &context->bw_ctx.bw.dcn.watermarks.c, &context->bw_ctx.dml, pipes, pipe_cnt); /* WM Set B */ table_entry = &bw_params->wm_table.entries[WM_B]; - vlevel = min(max(vlevel_req, 1), vlevel_max); + vlevel = clamp(vlevel_req, 1, vlevel_max); calculate_wm_set_for_vlevel(vlevel, table_entry, &context->bw_ctx.bw.dcn.watermarks.b, &context->bw_ctx.dml, pipes, pipe_cnt); diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn302/dcn302_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn302/dcn302_fpu.c index 8da97a96b1ce..8d7c59ec701d 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn302/dcn302_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn302/dcn302_fpu.c @@ -280,7 +280,7 @@ void dcn302_fpu_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_p j = 0; /* create the final dcfclk and uclk table */ while (i < num_dcfclk_sta_targets && j < num_uclk_states && num_states < DC__VOLTAGE_STATES) { - if (dcfclk_sta_targets[i] < optimal_dcfclk_for_uclk[j] && i < num_dcfclk_sta_targets) { + if (dcfclk_sta_targets[i] < optimal_dcfclk_for_uclk[j]) { dcfclk_mhz[num_states] = dcfclk_sta_targets[i]; dram_speed_mts[num_states++] = optimal_uclk_for_dcfclk_sta_targets[i++]; } else { diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn303/dcn303_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn303/dcn303_fpu.c index e968870a4b81..b5d3fd4c3694 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn303/dcn303_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn303/dcn303_fpu.c @@ -285,7 +285,7 @@ void dcn303_fpu_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_p j = 0; /* create the final dcfclk and uclk table */ while (i < num_dcfclk_sta_targets && j < num_uclk_states && num_states < DC__VOLTAGE_STATES) { - if (dcfclk_sta_targets[i] < optimal_dcfclk_for_uclk[j] && i < num_dcfclk_sta_targets) { + if (dcfclk_sta_targets[i] < optimal_dcfclk_for_uclk[j]) { dcfclk_mhz[num_states] = dcfclk_sta_targets[i]; dram_speed_mts[num_states++] = optimal_uclk_for_dcfclk_sta_targets[i++]; } else { diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_rq_dlg_calc_31.c b/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_rq_dlg_calc_31.c index c46bda2141ac..bfeb01477f0c 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_rq_dlg_calc_31.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_rq_dlg_calc_31.c @@ -615,7 +615,7 @@ static void get_meta_and_pte_attr( if (hostvm_enable) rq_sizing_param->dpte_group_bytes = 512; else { - if (!surf_linear & (log2_dpte_req_height_ptes == 0) & surf_vert) //reduced, in this case, will have page fault within a group + if (!surf_linear && (log2_dpte_req_height_ptes == 0) && surf_vert) //reduced, in this case, will have page fault within a group rq_sizing_param->dpte_group_bytes = 512; else rq_sizing_param->dpte_group_bytes = 2048; diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_rq_dlg_calc_314.c b/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_rq_dlg_calc_314.c index b7d2a0caec11..04df263ff65e 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_rq_dlg_calc_314.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_rq_dlg_calc_314.c @@ -703,7 +703,7 @@ static void get_meta_and_pte_attr( if (hostvm_enable) rq_sizing_param->dpte_group_bytes = 512; else { - if (!surf_linear & (log2_dpte_req_height_ptes == 0) & surf_vert) //reduced, in this case, will have page fault within a group + if (!surf_linear && (log2_dpte_req_height_ptes == 0) && surf_vert) //reduced, in this case, will have page fault within a group rq_sizing_param->dpte_group_bytes = 512; else rq_sizing_param->dpte_group_bytes = 2048; diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c index 6160952245b4..18388fb00be8 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c @@ -3229,7 +3229,7 @@ void dcn32_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_pa j = 0; // create the final dcfclk and uclk table while (i < num_dcfclk_sta_targets && j < num_uclk_states && num_states < DC__VOLTAGE_STATES) { - if (dcfclk_sta_targets[i] < optimal_dcfclk_for_uclk[j] && i < num_dcfclk_sta_targets) { + if (dcfclk_sta_targets[i] < optimal_dcfclk_for_uclk[j]) { dcfclk_mhz[num_states] = dcfclk_sta_targets[i]; dram_speed_mts[num_states++] = optimal_uclk_for_dcfclk_sta_targets[i++]; } else { @@ -3401,7 +3401,7 @@ bool dcn32_allow_subvp_with_active_margin(struct pipe_ctx *pipe) uint32_t height = subvp_active_margin_list.res[i].height; refresh_rate = (pipe->stream->timing.pix_clk_100hz * (uint64_t)100 + - pipe->stream->timing.v_total * pipe->stream->timing.h_total - (uint64_t)1); + (uint64_t)pipe->stream->timing.v_total * pipe->stream->timing.h_total - (uint64_t)1); refresh_rate = div_u64(refresh_rate, pipe->stream->timing.v_total); refresh_rate = div_u64(refresh_rate, pipe->stream->timing.h_total); diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_rq_dlg_calc_32.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_rq_dlg_calc_32.c index 9ba6cb67655f..6c75aa82327a 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_rq_dlg_calc_32.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_rq_dlg_calc_32.c @@ -139,7 +139,6 @@ void dml32_rq_dlg_get_rq_reg(display_rq_regs_st *rq_regs, if (dual_plane) { unsigned int p1_pte_row_height_linear = get_dpte_row_height_linear_c(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); - ; if (src->sw_mode == dm_sw_linear) ASSERT(p1_pte_row_height_linear >= 8); diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.c index 8839faf42207..e0a1dc89ce43 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.c @@ -779,7 +779,7 @@ void dcn321_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_p j = 0; // create the final dcfclk and uclk table while (i < num_dcfclk_sta_targets && j < num_uclk_states && num_states < DC__VOLTAGE_STATES) { - if (dcfclk_sta_targets[i] < optimal_dcfclk_for_uclk[j] && i < num_dcfclk_sta_targets) { + if (dcfclk_sta_targets[i] < optimal_dcfclk_for_uclk[j]) { dcfclk_mhz[num_states] = dcfclk_sta_targets[i]; dram_speed_mts[num_states++] = optimal_uclk_for_dcfclk_sta_targets[i++]; } else { diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.c b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.c index a06217a9eef6..f6879e622271 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.c @@ -8,7 +8,7 @@ #include "dml2_internal_types.h" #include "dml21_utils.h" #include "dml21_translation_helper.h" -#include "bounding_boxes/dcn4_soc_bb.h" +#include "soc_and_ip_translator.h" static void dml21_populate_pmo_options(struct dml2_pmo_options *pmo_options, const struct dc *in_dc, @@ -38,375 +38,37 @@ static void dml21_populate_pmo_options(struct dml2_pmo_options *pmo_options, pmo_options->disable_drr_clamped_when_var_active = in_dc->debug.disable_fams_gaming == INGAME_FAMS_DISABLE; } -/* - * Populate dml_init based on default static values in soc bb. The default - * values are for reference and support at least minimal operation of current - * SoC and DCN hardware. The values could be modifed by subsequent override - * functions to reflect our true hardware capability. - */ -static void populate_default_dml_init_params(struct dml2_initialize_instance_in_out *dml_init, - const struct dml2_configuration_options *config, - const struct dc *in_dc) +static enum dml2_project_id dml21_dcn_revision_to_dml2_project_id(enum dce_version dcn_version) { - switch (in_dc->ctx->dce_version) { + enum dml2_project_id project_id; + switch (dcn_version) { case DCN_VERSION_4_01: - dml_init->options.project_id = dml2_project_dcn4x_stage2_auto_drr_svp; - dml21_populate_pmo_options(&dml_init->options.pmo_options, in_dc, config); - dml_init->soc_bb = dml2_socbb_dcn401; - dml_init->soc_bb.qos_parameters = dml_dcn4_variant_a_soc_qos_params; - dml_init->ip_caps = dml2_dcn401_max_ip_caps; + project_id = dml2_project_dcn4x_stage2_auto_drr_svp; break; default: - memset(dml_init, 0, sizeof(*dml_init)); + project_id = dml2_project_invalid; DC_ERR("unsupported dcn version for DML21!"); - return; - } -} - -static void override_dml_init_with_values_from_hardware_default(struct dml2_initialize_instance_in_out *dml_init, - const struct dml2_configuration_options *config, - const struct dc *in_dc) -{ - dml_init->soc_bb.dchub_refclk_mhz = in_dc->res_pool->ref_clocks.dchub_ref_clock_inKhz / 1000; - dml_init->soc_bb.dprefclk_mhz = in_dc->clk_mgr->dprefclk_khz / 1000; - dml_init->soc_bb.dispclk_dppclk_vco_speed_mhz = in_dc->clk_mgr->dentist_vco_freq_khz / 1000.0; -} - -/* - * SMU stands for System Management Unit. It is a power management processor. - * It owns the initialization of dc's clock table and programming of clock values - * based on dc's requests. - * Our clock values in base soc bb is a dummy placeholder. The real clock values - * are retrieved from SMU firmware to dc clock table at runtime. - * This function overrides our dummy placeholder values with real values in dc - * clock table. - */ -static void override_dml_init_with_values_from_smu( - struct dml2_initialize_instance_in_out *dml_init, - const struct dml2_configuration_options *config, - const struct dc *in_dc) -{ - int i; - const struct clk_bw_params *dc_bw_params = in_dc->clk_mgr->bw_params; - const struct clk_limit_table *dc_clk_table = &dc_bw_params->clk_table; - struct dml2_soc_state_table *dml_clk_table = &dml_init->soc_bb.clk_table; - - if (!in_dc->clk_mgr->funcs->is_smu_present || - !in_dc->clk_mgr->funcs->is_smu_present(in_dc->clk_mgr)) - /* skip if smu is not present */ - return; - - /* dcfclk */ - if (dc_clk_table->num_entries_per_clk.num_dcfclk_levels) { - dml_clk_table->dcfclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_dcfclk_levels; - for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) { - if (i < dml_clk_table->dcfclk.num_clk_values) { - if (config->use_clock_dc_limits && dc_bw_params->dc_mode_limit.dcfclk_mhz && - dc_clk_table->entries[i].dcfclk_mhz > dc_bw_params->dc_mode_limit.dcfclk_mhz) { - if (i == 0 || dc_clk_table->entries[i-1].dcfclk_mhz < dc_bw_params->dc_mode_limit.dcfclk_mhz) { - dml_clk_table->dcfclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.dcfclk_mhz * 1000; - dml_clk_table->dcfclk.num_clk_values = i + 1; - } else { - dml_clk_table->dcfclk.clk_values_khz[i] = 0; - dml_clk_table->dcfclk.num_clk_values = i; - } - } else { - dml_clk_table->dcfclk.clk_values_khz[i] = dc_clk_table->entries[i].dcfclk_mhz * 1000; - } - } else { - dml_clk_table->dcfclk.clk_values_khz[i] = 0; - } - } - } - - /* fclk */ - if (dc_clk_table->num_entries_per_clk.num_fclk_levels) { - dml_clk_table->fclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_fclk_levels; - for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) { - if (i < dml_clk_table->fclk.num_clk_values) { - if (config->use_clock_dc_limits && dc_bw_params->dc_mode_limit.fclk_mhz && - dc_clk_table->entries[i].fclk_mhz > dc_bw_params->dc_mode_limit.fclk_mhz) { - if (i == 0 || dc_clk_table->entries[i-1].fclk_mhz < dc_bw_params->dc_mode_limit.fclk_mhz) { - dml_clk_table->fclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.fclk_mhz * 1000; - dml_clk_table->fclk.num_clk_values = i + 1; - } else { - dml_clk_table->fclk.clk_values_khz[i] = 0; - dml_clk_table->fclk.num_clk_values = i; - } - } else { - dml_clk_table->fclk.clk_values_khz[i] = dc_clk_table->entries[i].fclk_mhz * 1000; - } - } else { - dml_clk_table->fclk.clk_values_khz[i] = 0; - } - } - } - - /* uclk */ - if (dc_clk_table->num_entries_per_clk.num_memclk_levels) { - dml_clk_table->uclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_memclk_levels; - for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) { - if (i < dml_clk_table->uclk.num_clk_values) { - if (config->use_clock_dc_limits && dc_bw_params->dc_mode_limit.memclk_mhz && - dc_clk_table->entries[i].memclk_mhz > dc_bw_params->dc_mode_limit.memclk_mhz) { - if (i == 0 || dc_clk_table->entries[i-1].memclk_mhz < dc_bw_params->dc_mode_limit.memclk_mhz) { - dml_clk_table->uclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.memclk_mhz * 1000; - dml_clk_table->uclk.num_clk_values = i + 1; - } else { - dml_clk_table->uclk.clk_values_khz[i] = 0; - dml_clk_table->uclk.num_clk_values = i; - } - } else { - dml_clk_table->uclk.clk_values_khz[i] = dc_clk_table->entries[i].memclk_mhz * 1000; - } - } else { - dml_clk_table->uclk.clk_values_khz[i] = 0; - } - } - } - - /* dispclk */ - if (dc_clk_table->num_entries_per_clk.num_dispclk_levels) { - dml_clk_table->dispclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_dispclk_levels; - for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) { - if (i < dml_clk_table->dispclk.num_clk_values) { - if (config->use_clock_dc_limits && dc_bw_params->dc_mode_limit.dispclk_mhz && - dc_clk_table->entries[i].dispclk_mhz > dc_bw_params->dc_mode_limit.dispclk_mhz) { - if (i == 0 || dc_clk_table->entries[i-1].dispclk_mhz < dc_bw_params->dc_mode_limit.dispclk_mhz) { - dml_clk_table->dispclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.dispclk_mhz * 1000; - dml_clk_table->dispclk.num_clk_values = i + 1; - } else { - dml_clk_table->dispclk.clk_values_khz[i] = 0; - dml_clk_table->dispclk.num_clk_values = i; - } - } else { - dml_clk_table->dispclk.clk_values_khz[i] = dc_clk_table->entries[i].dispclk_mhz * 1000; - } - } else { - dml_clk_table->dispclk.clk_values_khz[i] = 0; - } - } - } - - /* dppclk */ - if (dc_clk_table->num_entries_per_clk.num_dppclk_levels) { - dml_clk_table->dppclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_dppclk_levels; - for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) { - if (i < dml_clk_table->dppclk.num_clk_values) { - if (config->use_clock_dc_limits && dc_bw_params->dc_mode_limit.dppclk_mhz && - dc_clk_table->entries[i].dppclk_mhz > dc_bw_params->dc_mode_limit.dppclk_mhz) { - if (i == 0 || dc_clk_table->entries[i-1].dppclk_mhz < dc_bw_params->dc_mode_limit.dppclk_mhz) { - dml_clk_table->dppclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.dppclk_mhz * 1000; - dml_clk_table->dppclk.num_clk_values = i + 1; - } else { - dml_clk_table->dppclk.clk_values_khz[i] = 0; - dml_clk_table->dppclk.num_clk_values = i; - } - } else { - dml_clk_table->dppclk.clk_values_khz[i] = dc_clk_table->entries[i].dppclk_mhz * 1000; - } - } else { - dml_clk_table->dppclk.clk_values_khz[i] = 0; - } - } - } - - /* dtbclk */ - if (dc_clk_table->num_entries_per_clk.num_dtbclk_levels) { - dml_clk_table->dtbclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_dtbclk_levels; - for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) { - if (i < dml_clk_table->dtbclk.num_clk_values) { - if (config->use_clock_dc_limits && dc_bw_params->dc_mode_limit.dtbclk_mhz && - dc_clk_table->entries[i].dtbclk_mhz > dc_bw_params->dc_mode_limit.dtbclk_mhz) { - if (i == 0 || dc_clk_table->entries[i-1].dtbclk_mhz < dc_bw_params->dc_mode_limit.dtbclk_mhz) { - dml_clk_table->dtbclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.dtbclk_mhz * 1000; - dml_clk_table->dtbclk.num_clk_values = i + 1; - } else { - dml_clk_table->dtbclk.clk_values_khz[i] = 0; - dml_clk_table->dtbclk.num_clk_values = i; - } - } else { - dml_clk_table->dtbclk.clk_values_khz[i] = dc_clk_table->entries[i].dtbclk_mhz * 1000; - } - } else { - dml_clk_table->dtbclk.clk_values_khz[i] = 0; - } - } - } - - /* socclk */ - if (dc_clk_table->num_entries_per_clk.num_socclk_levels) { - dml_clk_table->socclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_socclk_levels; - for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) { - if (i < dml_clk_table->socclk.num_clk_values) { - if (config->use_clock_dc_limits && dc_bw_params->dc_mode_limit.socclk_mhz && - dc_clk_table->entries[i].socclk_mhz > dc_bw_params->dc_mode_limit.socclk_mhz) { - if (i == 0 || dc_clk_table->entries[i-1].socclk_mhz < dc_bw_params->dc_mode_limit.socclk_mhz) { - dml_clk_table->socclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.socclk_mhz * 1000; - dml_clk_table->socclk.num_clk_values = i + 1; - } else { - dml_clk_table->socclk.clk_values_khz[i] = 0; - dml_clk_table->socclk.num_clk_values = i; - } - } else { - dml_clk_table->socclk.clk_values_khz[i] = dc_clk_table->entries[i].socclk_mhz * 1000; - } - } else { - dml_clk_table->socclk.clk_values_khz[i] = 0; - } - } - } -} - -static void override_dml_init_with_values_from_vbios( - struct dml2_initialize_instance_in_out *dml_init, - const struct dml2_configuration_options *config, - const struct dc *in_dc) -{ - const struct clk_bw_params *dc_bw_params = in_dc->clk_mgr->bw_params; - struct dml2_soc_bb *dml_soc_bb = &dml_init->soc_bb; - struct dml2_soc_state_table *dml_clk_table = &dml_init->soc_bb.clk_table; - - if (in_dc->ctx->dc_bios->bb_info.dram_clock_change_latency_100ns > 0) - dml_soc_bb->power_management_parameters.dram_clk_change_blackout_us = - (in_dc->ctx->dc_bios->bb_info.dram_clock_change_latency_100ns + 9) / 10; - - if (in_dc->ctx->dc_bios->bb_info.dram_sr_enter_exit_latency_100ns > 0) - dml_soc_bb->power_management_parameters.stutter_enter_plus_exit_latency_us = - (in_dc->ctx->dc_bios->bb_info.dram_sr_enter_exit_latency_100ns + 9) / 10; - - if (in_dc->ctx->dc_bios->bb_info.dram_sr_exit_latency_100ns > 0) - dml_soc_bb->power_management_parameters.stutter_exit_latency_us = - (in_dc->ctx->dc_bios->bb_info.dram_sr_exit_latency_100ns + 9) / 10; - - if (dc_bw_params->num_channels) { - dml_clk_table->dram_config.channel_count = dc_bw_params->num_channels; - dml_soc_bb->mall_allocated_for_dcn_mbytes = in_dc->caps.mall_size_total / 1048576; - } else if (in_dc->ctx->dc_bios->vram_info.num_chans) { - dml_clk_table->dram_config.channel_count = in_dc->ctx->dc_bios->vram_info.num_chans; - dml_soc_bb->mall_allocated_for_dcn_mbytes = in_dc->caps.mall_size_total / 1048576; - } - - if (dc_bw_params->dram_channel_width_bytes) { - dml_clk_table->dram_config.channel_width_bytes = dc_bw_params->dram_channel_width_bytes; - } else if (in_dc->ctx->dc_bios->vram_info.dram_channel_width_bytes) { - dml_clk_table->dram_config.channel_width_bytes = in_dc->ctx->dc_bios->vram_info.dram_channel_width_bytes; + break; } - dml_init->soc_bb.xtalclk_mhz = in_dc->ctx->dc_bios->fw_info.pll_info.crystal_frequency / 1000; + return project_id; } - -static void override_dml_init_with_values_from_dmub(struct dml2_initialize_instance_in_out *dml_init, +void dml21_populate_dml_init_params(struct dml2_initialize_instance_in_out *dml_init, const struct dml2_configuration_options *config, const struct dc *in_dc) { - /* - * TODO - There seems to be overlaps between the values overriden from - * dmub and vbios. Investigate and identify the values that DMUB needs - * to own. - */ -// const struct dmub_soc_bb_params *dmub_bb_params = -// (const struct dmub_soc_bb_params *)config->bb_from_dmub; - -// if (dmub_bb_params == NULL) -// return; - -// if (dmub_bb_params->dram_clk_change_blackout_ns > 0) -// dml_init->soc_bb.power_management_parameters.dram_clk_change_blackout_us = -// (double) dmub_bb_params->dram_clk_change_blackout_ns / 1000.0; -// if (dmub_bb_params->dram_clk_change_read_only_ns > 0) -// dml_init->soc_bb.power_management_parameters.dram_clk_change_read_only_us = -// (double) dmub_bb_params->dram_clk_change_read_only_ns / 1000.0; -// if (dmub_bb_params->dram_clk_change_write_only_ns > 0) -// dml_init->soc_bb.power_management_parameters.dram_clk_change_write_only_us = -// (double) dmub_bb_params->dram_clk_change_write_only_ns / 1000.0; -// if (dmub_bb_params->fclk_change_blackout_ns > 0) -// dml_init->soc_bb.power_management_parameters.fclk_change_blackout_us = -// (double) dmub_bb_params->fclk_change_blackout_ns / 1000.0; -// if (dmub_bb_params->g7_ppt_blackout_ns > 0) -// dml_init->soc_bb.power_management_parameters.g7_ppt_blackout_us = -// (double) dmub_bb_params->g7_ppt_blackout_ns / 1000.0; -// if (dmub_bb_params->stutter_enter_plus_exit_latency_ns > 0) -// dml_init->soc_bb.power_management_parameters.stutter_enter_plus_exit_latency_us = -// (double) dmub_bb_params->stutter_enter_plus_exit_latency_ns / 1000.0; -// if (dmub_bb_params->stutter_exit_latency_ns > 0) -// dml_init->soc_bb.power_management_parameters.stutter_exit_latency_us = -// (double) dmub_bb_params->stutter_exit_latency_ns / 1000.0; -// if (dmub_bb_params->z8_stutter_enter_plus_exit_latency_ns > 0) -// dml_init->soc_bb.power_management_parameters.z8_stutter_enter_plus_exit_latency_us = -// (double) dmub_bb_params->z8_stutter_enter_plus_exit_latency_ns / 1000.0; -// if (dmub_bb_params->z8_stutter_exit_latency_ns > 0) -// dml_init->soc_bb.power_management_parameters.z8_stutter_exit_latency_us = -// (double) dmub_bb_params->z8_stutter_exit_latency_ns / 1000.0; -// if (dmub_bb_params->z8_min_idle_time_ns > 0) -// dml_init->soc_bb.power_management_parameters.z8_min_idle_time = -// (double) dmub_bb_params->z8_min_idle_time_ns / 1000.0; -// #ifndef TRIM_DML2_DCN6B_IP_SENSITIVE -// if (dmub_bb_params->type_b_dram_clk_change_blackout_ns > 0) -// dml_init->soc_bb.power_management_parameters.lpddr5_dram_clk_change_blackout_us = -// (double) dmub_bb_params->type_b_dram_clk_change_blackout_ns / 1000.0; -// if (dmub_bb_params->type_b_ppt_blackout_ns > 0) -// dml_init->soc_bb.power_management_parameters.lpddr5_ppt_blackout_us = -// (double) dmub_bb_params->type_b_ppt_blackout_ns / 1000.0; -// #else -// if (dmub_bb_params->type_b_dram_clk_change_blackout_ns > 0) -// dml_init->soc_bb.power_management_parameters.type_b_dram_clk_change_blackout_us = -// (double) dmub_bb_params->type_b_dram_clk_change_blackout_ns / 1000.0; -// if (dmub_bb_params->type_b_ppt_blackout_ns > 0) -// dml_init->soc_bb.power_management_parameters.type_b_ppt_blackout_us = -// (double) dmub_bb_params->type_b_ppt_blackout_ns / 1000.0; -// #endif -// if (dmub_bb_params->vmin_limit_dispclk_khz > 0) -// dml_init->soc_bb.vmin_limit.dispclk_khz = dmub_bb_params->vmin_limit_dispclk_khz; -// if (dmub_bb_params->vmin_limit_dcfclk_khz > 0) -// dml_init->soc_bb.vmin_limit.dcfclk_khz = dmub_bb_params->vmin_limit_dcfclk_khz; -// if (dmub_bb_params->g7_temperature_read_blackout_ns > 0) -// dml_init->soc_bb.power_management_parameters.g7_temperature_read_blackout_us = -// (double) dmub_bb_params->g7_temperature_read_blackout_ns / 1000.0; -} + dml_init->options.project_id = dml21_dcn_revision_to_dml2_project_id(in_dc->ctx->dce_version); -static void override_dml_init_with_values_from_software_policy(struct dml2_initialize_instance_in_out *dml_init, - const struct dml2_configuration_options *config, - const struct dc *in_dc) -{ - if (!config->use_native_soc_bb_construction) { + if (config->use_native_soc_bb_construction) { + in_dc->soc_and_ip_translator->translator_funcs->get_soc_bb(&dml_init->soc_bb, in_dc, config); + in_dc->soc_and_ip_translator->translator_funcs->get_ip_caps(&dml_init->ip_caps); + } else { dml_init->soc_bb = config->external_socbb_ip_params->soc_bb; dml_init->ip_caps = config->external_socbb_ip_params->ip_params; } - if (in_dc->bb_overrides.sr_exit_time_ns) - dml_init->soc_bb.power_management_parameters.stutter_exit_latency_us = - in_dc->bb_overrides.sr_exit_time_ns / 1000.0; - - if (in_dc->bb_overrides.sr_enter_plus_exit_time_ns) - dml_init->soc_bb.power_management_parameters.stutter_enter_plus_exit_latency_us = - in_dc->bb_overrides.sr_enter_plus_exit_time_ns / 1000.0; - - if (in_dc->bb_overrides.dram_clock_change_latency_ns) - dml_init->soc_bb.power_management_parameters.dram_clk_change_blackout_us = - in_dc->bb_overrides.dram_clock_change_latency_ns / 1000.0; - - if (in_dc->bb_overrides.fclk_clock_change_latency_ns) - dml_init->soc_bb.power_management_parameters.fclk_change_blackout_us = - in_dc->bb_overrides.fclk_clock_change_latency_ns / 1000.0; -} - -void dml21_populate_dml_init_params(struct dml2_initialize_instance_in_out *dml_init, - const struct dml2_configuration_options *config, - const struct dc *in_dc) -{ - populate_default_dml_init_params(dml_init, config, in_dc); - - override_dml_init_with_values_from_hardware_default(dml_init, config, in_dc); - - override_dml_init_with_values_from_smu(dml_init, config, in_dc); - - override_dml_init_with_values_from_vbios(dml_init, config, in_dc); - - override_dml_init_with_values_from_dmub(dml_init, config, in_dc); - - override_dml_init_with_values_from_software_policy(dml_init, config, in_dc); + dml21_populate_pmo_options(&dml_init->options.pmo_options, in_dc, config); } static unsigned int calc_max_hardware_v_total(const struct dc_stream_state *stream) @@ -463,11 +125,7 @@ static void populate_dml21_timing_config_from_stream_state(struct dml2_timing_cf (stream->timing.h_total * (long long)calc_max_hardware_v_total(stream))); } - if (stream->timing.min_refresh_in_uhz > min_hardware_refresh_in_uhz) { - timing->drr_config.min_refresh_uhz = stream->timing.min_refresh_in_uhz; - } else { - timing->drr_config.min_refresh_uhz = min_hardware_refresh_in_uhz; - } + timing->drr_config.min_refresh_uhz = max(stream->timing.min_refresh_in_uhz, min_hardware_refresh_in_uhz); if (dml_ctx->config.callbacks.get_max_flickerless_instant_vtotal_increase && stream->ctx->dc->config.enable_fpo_flicker_detection == 1) @@ -1165,6 +823,8 @@ void dml21_copy_clocks_to_dc_state(struct dml2_context *in_ctx, struct dc_state context->bw_ctx.bw.dcn.clk.socclk_khz = in_ctx->v21.mode_programming.programming->min_clocks.dcn4x.socclk_khz; context->bw_ctx.bw.dcn.clk.subvp_prefetch_dramclk_khz = in_ctx->v21.mode_programming.programming->min_clocks.dcn4x.svp_prefetch_no_throttle.uclk_khz; context->bw_ctx.bw.dcn.clk.subvp_prefetch_fclk_khz = in_ctx->v21.mode_programming.programming->min_clocks.dcn4x.svp_prefetch_no_throttle.fclk_khz; + context->bw_ctx.bw.dcn.clk.stutter_efficiency.base_efficiency = in_ctx->v21.mode_programming.programming->stutter.base_percent_efficiency; + context->bw_ctx.bw.dcn.clk.stutter_efficiency.low_power_efficiency = in_ctx->v21.mode_programming.programming->stutter.low_power_percent_efficiency; } static struct dml2_dchub_watermark_regs *wm_set_index_to_dc_wm_set(union dcn_watermark_set *watermarks, const enum dml2_dchub_watermark_reg_set_index wm_index) diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_wrapper.c b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_wrapper.c index 03de3cf06ae5..798abb2b2e67 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_wrapper.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_wrapper.c @@ -60,7 +60,7 @@ static void dml21_init(const struct dc *in_dc, struct dml2_context *dml_ctx, con DC_FP_START(); - dml21_populate_dml_init_params(&dml_ctx->v21.dml_init, config, in_dc); + dml21_populate_dml_init_params(&dml_ctx->v21.dml_init, &dml_ctx->config, in_dc); dml2_initialize_instance(&dml_ctx->v21.dml_init); diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_dchub_registers.h b/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_dchub_registers.h index b05030926ce8..91955bbe24b8 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_dchub_registers.h +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_dchub_registers.h @@ -159,6 +159,8 @@ struct dml2_dchub_watermark_regs { uint32_t sr_exit; uint32_t sr_enter_z8; uint32_t sr_exit_z8; + uint32_t sr_enter_low_power; + uint32_t sr_exit_low_power; uint32_t uclk_pstate; uint32_t fclk_pstate; uint32_t temp_read_or_ppt; diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_soc_parameter_types.h b/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_soc_parameter_types.h index 8c9f414aa6bf..176f55947664 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_soc_parameter_types.h +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_soc_parameter_types.h @@ -96,6 +96,8 @@ struct dml2_soc_power_management_parameters { double g7_temperature_read_blackout_us; double stutter_enter_plus_exit_latency_us; double stutter_exit_latency_us; + double low_power_stutter_enter_plus_exit_latency_us; + double low_power_stutter_exit_latency_us; double z8_stutter_enter_plus_exit_latency_us; double z8_stutter_exit_latency_us; double z8_min_idle_time; diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_types.h b/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_types.h index 98c0234e2f47..7de10a95cfdb 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_types.h +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_types.h @@ -417,6 +417,8 @@ struct dml2_display_cfg_programming { struct { bool supported_in_blank; // Changing to configurations where this is false requires stutter to be disabled during the transition + uint8_t base_percent_efficiency; //LP1 + uint8_t low_power_percent_efficiency; //LP2 } stutter; struct { diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_dcn4_calcs.c b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_dcn4_calcs.c index b9cff2198511..bf62d42b3f78 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_dcn4_calcs.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_dcn4_calcs.c @@ -1238,18 +1238,27 @@ static void CalculateDETBufferSize( static double CalculateRequiredDispclk( enum dml2_odm_mode ODMMode, - double PixelClock) + double PixelClock, + bool isTMDS420) { + double DispClk; if (ODMMode == dml2_odm_mode_combine_4to1) { - return PixelClock / 4.0; + DispClk = PixelClock / 4.0; } else if (ODMMode == dml2_odm_mode_combine_3to1) { - return PixelClock / 3.0; + DispClk = PixelClock / 3.0; } else if (ODMMode == dml2_odm_mode_combine_2to1) { - return PixelClock / 2.0; + DispClk = PixelClock / 2.0; } else { - return PixelClock; + DispClk = PixelClock; + } + + if (isTMDS420) { + double TMDS420MinPixClock = PixelClock / 2.0; + DispClk = math_max2(DispClk, TMDS420MinPixClock); } + + return DispClk; } static double TruncToValidBPP( @@ -4122,11 +4131,12 @@ static noinline_for_stack void CalculateODMMode( bool success; bool UseDSC = DSCEnable && (NumberOfDSCSlices > 0); enum dml2_odm_mode DecidedODMMode; + bool isTMDS420 = (OutFormat == dml2_420 && Output == dml2_hdmi); - SurfaceRequiredDISPCLKWithoutODMCombine = CalculateRequiredDispclk(dml2_odm_mode_bypass, PixelClock); - SurfaceRequiredDISPCLKWithODMCombineTwoToOne = CalculateRequiredDispclk(dml2_odm_mode_combine_2to1, PixelClock); - SurfaceRequiredDISPCLKWithODMCombineThreeToOne = CalculateRequiredDispclk(dml2_odm_mode_combine_3to1, PixelClock); - SurfaceRequiredDISPCLKWithODMCombineFourToOne = CalculateRequiredDispclk(dml2_odm_mode_combine_4to1, PixelClock); + SurfaceRequiredDISPCLKWithoutODMCombine = CalculateRequiredDispclk(dml2_odm_mode_bypass, PixelClock, isTMDS420); + SurfaceRequiredDISPCLKWithODMCombineTwoToOne = CalculateRequiredDispclk(dml2_odm_mode_combine_2to1, PixelClock, isTMDS420); + SurfaceRequiredDISPCLKWithODMCombineThreeToOne = CalculateRequiredDispclk(dml2_odm_mode_combine_3to1, PixelClock, isTMDS420); + SurfaceRequiredDISPCLKWithODMCombineFourToOne = CalculateRequiredDispclk(dml2_odm_mode_combine_4to1, PixelClock, isTMDS420); #ifdef __DML_VBA_DEBUG__ DML_LOG_VERBOSE("DML::%s: ODMUse = %d\n", __func__, ODMUse); DML_LOG_VERBOSE("DML::%s: Output = %d\n", __func__, Output); diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_factory.c b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_factory.c index 28394de02885..640087e862f8 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_factory.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_factory.c @@ -10,7 +10,7 @@ bool dml2_core_create(enum dml2_project_id project_id, struct dml2_core_instance { bool result = false; - if (out == 0) + if (!out) return false; memset(out, 0, sizeof(struct dml2_core_instance)); diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_shared_types.h b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_shared_types.h index 28687565ac22..ffb8c09f37a5 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_shared_types.h +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_shared_types.h @@ -201,6 +201,8 @@ struct dml2_core_internal_watermarks { double WritebackFCLKChangeWatermark; double StutterExitWatermark; double StutterEnterPlusExitWatermark; + double LowPowerStutterExitWatermark; + double LowPowerStutterEnterPlusExitWatermark; double Z8StutterExitWatermark; double Z8StutterEnterPlusExitWatermark; double USRRetrainingWatermark; @@ -877,6 +879,9 @@ struct dml2_core_internal_mode_program { double Z8StutterEfficiency; unsigned int Z8NumberOfStutterBurstsPerFrame; double Z8StutterEfficiencyNotIncludingVBlank; + double LowPowerStutterEfficiency; + double LowPowerStutterEfficiencyNotIncludingVBlank; + unsigned int LowPowerNumberOfStutterBurstsPerFrame; double StutterPeriod; double Z8StutterEfficiencyBestCase; unsigned int Z8NumberOfStutterBurstsPerFrameBestCase; @@ -1016,6 +1021,8 @@ struct dml2_core_internal_SOCParametersList { double FCLKChangeLatency; double SRExitTime; double SREnterPlusExitTime; + double SRExitTimeLowPower; + double SREnterPlusExitTimeLowPower; double SRExitZ8Time; double SREnterPlusExitZ8Time; double USRRetrainingLatency; @@ -1851,9 +1858,11 @@ struct dml2_core_calcs_CalculateStutterEfficiency_params { unsigned int CompbufReservedSpaceZs; bool hw_debug5; double SRExitTime; + double SRExitTimeLowPower; double SRExitZ8Time; bool SynchronizeTimings; double StutterEnterPlusExitWatermark; + double LowPowerStutterEnterPlusExitWatermark; double Z8StutterEnterPlusExitWatermark; bool ProgressiveToInterlaceUnitInOPP; double *MinTTUVBlank; @@ -1879,7 +1888,10 @@ struct dml2_core_calcs_CalculateStutterEfficiency_params { // output double *StutterEfficiencyNotIncludingVBlank; double *StutterEfficiency; + double *LowPowerStutterEfficiencyNotIncludingVBlank; + double *LowPowerStutterEfficiency; unsigned int *NumberOfStutterBurstsPerFrame; + unsigned int *LowPowerNumberOfStutterBurstsPerFrame; double *Z8StutterEfficiencyNotIncludingVBlank; double *Z8StutterEfficiency; unsigned int *Z8NumberOfStutterBurstsPerFrame; diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_dpmm/dml2_dpmm_factory.c b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_dpmm/dml2_dpmm_factory.c index 3861bc6c9621..dfd01440737d 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_dpmm/dml2_dpmm_factory.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_dpmm/dml2_dpmm_factory.c @@ -20,7 +20,7 @@ bool dml2_dpmm_create(enum dml2_project_id project_id, struct dml2_dpmm_instance { bool result = false; - if (out == 0) + if (!out) return false; memset(out, 0, sizeof(struct dml2_dpmm_instance)); diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_mcg/dml2_mcg_factory.c b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_mcg/dml2_mcg_factory.c index cd3fbc0591d8..c60b8fe90819 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_mcg/dml2_mcg_factory.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_mcg/dml2_mcg_factory.c @@ -15,7 +15,7 @@ bool dml2_mcg_create(enum dml2_project_id project_id, struct dml2_mcg_instance * { bool result = false; - if (out == 0) + if (!out) return false; memset(out, 0, sizeof(struct dml2_mcg_instance)); diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_pmo/dml2_pmo_factory.c b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_pmo/dml2_pmo_factory.c index 7ed0242a4b33..55d2464365d0 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_pmo/dml2_pmo_factory.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_pmo/dml2_pmo_factory.c @@ -26,7 +26,7 @@ bool dml2_pmo_create(enum dml2_project_id project_id, struct dml2_pmo_instance * { bool result = false; - if (out == 0) + if (!out) return false; memset(out, 0, sizeof(struct dml2_pmo_instance)); diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml2_dc_resource_mgmt.c b/drivers/gpu/drm/amd/display/dc/dml2/dml2_dc_resource_mgmt.c index 5f1b49a50049..4cfe64aa8492 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml2_dc_resource_mgmt.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml2_dc_resource_mgmt.c @@ -473,7 +473,6 @@ static void sort_pipes_for_splitting(struct dc_plane_pipe_pool *pipes) { bool sorted, swapped; unsigned int cur_index; - unsigned int temp; int odm_slice_index; for (odm_slice_index = 0; odm_slice_index < pipes->num_pipes_assigned_to_plane_for_odm_combine; odm_slice_index++) { @@ -489,9 +488,8 @@ static void sort_pipes_for_splitting(struct dc_plane_pipe_pool *pipes) swapped = false; while (!sorted) { if (pipes->pipes_assigned_to_plane[odm_slice_index][cur_index] > pipes->pipes_assigned_to_plane[odm_slice_index][cur_index + 1]) { - temp = pipes->pipes_assigned_to_plane[odm_slice_index][cur_index]; - pipes->pipes_assigned_to_plane[odm_slice_index][cur_index] = pipes->pipes_assigned_to_plane[odm_slice_index][cur_index + 1]; - pipes->pipes_assigned_to_plane[odm_slice_index][cur_index + 1] = temp; + swap(pipes->pipes_assigned_to_plane[odm_slice_index][cur_index + 1], + pipes->pipes_assigned_to_plane[odm_slice_index][cur_index]); swapped = true; } diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml2_mall_phantom.c b/drivers/gpu/drm/amd/display/dc/dml2/dml2_mall_phantom.c index a56e75cdf712..c59f825cfae9 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml2_mall_phantom.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml2_mall_phantom.c @@ -654,14 +654,14 @@ static void set_phantom_stream_timing(struct dml2_context *ctx, struct dc_state unsigned int svp_height, unsigned int svp_vstartup) { - unsigned int i, pipe_idx; + unsigned int i; double line_time, fp_and_sync_width_time; struct pipe_ctx *pipe; uint32_t phantom_vactive, phantom_bp, pstate_width_fw_delay_lines; static const double cvt_rb_vblank_max = ((double) 460 / (1000 * 1000)); // Find DML pipe index (pipe_idx) using dc_pipe_idx - for (i = 0, pipe_idx = 0; i < ctx->config.dcn_pipe_count; i++) { + for (i = 0; i < ctx->config.dcn_pipe_count; i++) { pipe = &state->res_ctx.pipe_ctx[i]; if (!pipe->stream) @@ -669,8 +669,6 @@ static void set_phantom_stream_timing(struct dml2_context *ctx, struct dc_state if (i == dc_pipe_idx) break; - - pipe_idx++; } // Calculate lines required for pstate allow width and FW processing delays diff --git a/drivers/gpu/drm/amd/display/dc/dpp/dcn10/dcn10_dpp.c b/drivers/gpu/drm/amd/display/dc/dpp/dcn10/dcn10_dpp.c index 75fb77bca83b..01480a04f85e 100644 --- a/drivers/gpu/drm/amd/display/dc/dpp/dcn10/dcn10_dpp.c +++ b/drivers/gpu/drm/amd/display/dc/dpp/dcn10/dcn10_dpp.c @@ -520,6 +520,15 @@ void dpp1_dppclk_control( REG_UPDATE(DPP_CONTROL, DPP_CLOCK_ENABLE, 0); } +void dpp_force_disable_cursor(struct dpp *dpp_base) +{ + struct dcn10_dpp *dpp = TO_DCN10_DPP(dpp_base); + + /* Force disable cursor */ + REG_UPDATE(CURSOR0_CONTROL, CUR0_ENABLE, 0); + dpp_base->pos.cur0_ctl.bits.cur0_enable = 0; +} + static const struct dpp_funcs dcn10_dpp_funcs = { .dpp_read_state = dpp_read_state, .dpp_reset = dpp_reset, diff --git a/drivers/gpu/drm/amd/display/dc/dpp/dcn10/dcn10_dpp.h b/drivers/gpu/drm/amd/display/dc/dpp/dcn10/dcn10_dpp.h index c48139bed11f..f466182963f7 100644 --- a/drivers/gpu/drm/amd/display/dc/dpp/dcn10/dcn10_dpp.h +++ b/drivers/gpu/drm/amd/display/dc/dpp/dcn10/dcn10_dpp.h @@ -1525,4 +1525,6 @@ void dpp1_construct(struct dcn10_dpp *dpp1, void dpp1_cm_get_gamut_remap(struct dpp *dpp_base, struct dpp_grph_csc_adjustment *adjust); +void dpp_force_disable_cursor(struct dpp *dpp_base); + #endif diff --git a/drivers/gpu/drm/amd/display/dc/dpp/dcn30/dcn30_dpp.c b/drivers/gpu/drm/amd/display/dc/dpp/dcn30/dcn30_dpp.c index 2d70586cef40..09be2a90cc79 100644 --- a/drivers/gpu/drm/amd/display/dc/dpp/dcn30/dcn30_dpp.c +++ b/drivers/gpu/drm/amd/display/dc/dpp/dcn30/dcn30_dpp.c @@ -1494,6 +1494,7 @@ static struct dpp_funcs dcn30_dpp_funcs = { .dpp_dppclk_control = dpp1_dppclk_control, .dpp_set_hdr_multiplier = dpp3_set_hdr_multiplier, .dpp_get_gamut_remap = dpp3_cm_get_gamut_remap, + .dpp_force_disable_cursor = dpp_force_disable_cursor, }; diff --git a/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp.h b/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp.h index 5a6a861402b3..5f6b431ec398 100644 --- a/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp.h +++ b/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp.h @@ -673,6 +673,16 @@ struct dcn401_dpp { struct pwl_params pwl_data; }; +enum dcn401_dscl_mode_sel { + DCN401_DSCL_MODE_SCALING_444_BYPASS = 0, + DCN401_DSCL_MODE_SCALING_444_RGB_ENABLE = 1, + DCN401_DSCL_MODE_SCALING_444_YCBCR_ENABLE = 2, + DCN401_DSCL_MODE_SCALING_420_YCBCR_ENABLE = 3, + DCN401_DSCL_MODE_SCALING_420_LUMA_BYPASS = 4, + DCN401_DSCL_MODE_SCALING_420_CHROMA_BYPASS = 5, + DCN401_DSCL_MODE_DSCL_BYPASS = 6 +}; + bool dpp401_construct(struct dcn401_dpp *dpp401, struct dc_context *ctx, uint32_t inst, diff --git a/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp_dscl.c b/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp_dscl.c index 2f92e7d4981b..6df3419f825f 100644 --- a/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp_dscl.c +++ b/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp_dscl.c @@ -78,16 +78,6 @@ enum dscl_autocal_mode { AUTOCAL_MODE_AUTOREPLICATE = 3 }; -enum dscl_mode_sel { - DSCL_MODE_SCALING_444_BYPASS = 0, - DSCL_MODE_SCALING_444_RGB_ENABLE = 1, - DSCL_MODE_SCALING_444_YCBCR_ENABLE = 2, - DSCL_MODE_SCALING_420_YCBCR_ENABLE = 3, - DSCL_MODE_SCALING_420_LUMA_BYPASS = 4, - DSCL_MODE_SCALING_420_CHROMA_BYPASS = 5, - DSCL_MODE_DSCL_BYPASS = 6 -}; - static int dpp401_dscl_get_pixel_depth_val(enum lb_pixel_depth depth) { if (depth == LB_PIXEL_DEPTH_30BPP) @@ -122,7 +112,7 @@ static bool dpp401_dscl_is_420_format(enum pixel_format format) return false; } -static enum dscl_mode_sel dpp401_dscl_get_dscl_mode( +static enum dcn401_dscl_mode_sel dpp401_dscl_get_dscl_mode( struct dpp *dpp_base, const struct scaler_data *data, bool dbg_always_scale) @@ -132,7 +122,7 @@ static enum dscl_mode_sel dpp401_dscl_get_dscl_mode( if (dpp_base->caps->dscl_data_proc_format == DSCL_DATA_PRCESSING_FIXED_FORMAT) { /* DSCL is processing data in fixed format */ if (data->format == PIXEL_FORMAT_FP16) - return DSCL_MODE_DSCL_BYPASS; + return DCN401_DSCL_MODE_DSCL_BYPASS; } if (data->ratios.horz.value == one @@ -140,20 +130,20 @@ static enum dscl_mode_sel dpp401_dscl_get_dscl_mode( && data->ratios.horz_c.value == one && data->ratios.vert_c.value == one && !dbg_always_scale) - return DSCL_MODE_SCALING_444_BYPASS; + return DCN401_DSCL_MODE_SCALING_444_BYPASS; if (!dpp401_dscl_is_420_format(data->format)) { if (dpp401_dscl_is_video_format(data->format)) - return DSCL_MODE_SCALING_444_YCBCR_ENABLE; + return DCN401_DSCL_MODE_SCALING_444_YCBCR_ENABLE; else - return DSCL_MODE_SCALING_444_RGB_ENABLE; + return DCN401_DSCL_MODE_SCALING_444_RGB_ENABLE; } if (data->ratios.horz.value == one && data->ratios.vert.value == one) - return DSCL_MODE_SCALING_420_LUMA_BYPASS; + return DCN401_DSCL_MODE_SCALING_420_LUMA_BYPASS; if (data->ratios.horz_c.value == one && data->ratios.vert_c.value == one) - return DSCL_MODE_SCALING_420_CHROMA_BYPASS; + return DCN401_DSCL_MODE_SCALING_420_CHROMA_BYPASS; - return DSCL_MODE_SCALING_420_YCBCR_ENABLE; + return DCN401_DSCL_MODE_SCALING_420_YCBCR_ENABLE; } static void dpp401_power_on_dscl( @@ -1071,7 +1061,7 @@ void dpp401_dscl_set_scaler_manual_scale(struct dpp *dpp_base, uint32_t v_num_taps_c = scl_data->taps.v_taps_c - 1; uint32_t h_num_taps = scl_data->taps.h_taps - 1; uint32_t h_num_taps_c = scl_data->taps.h_taps_c - 1; - enum dscl_mode_sel dscl_mode = dpp401_dscl_get_dscl_mode( + enum dcn401_dscl_mode_sel dscl_mode = dpp401_dscl_get_dscl_mode( dpp_base, scl_data, dpp_base->ctx->dc->debug.always_scale); bool ycbcr = scl_data->format >= PIXEL_FORMAT_VIDEO_BEGIN && scl_data->format <= PIXEL_FORMAT_VIDEO_END; @@ -1102,7 +1092,7 @@ void dpp401_dscl_set_scaler_manual_scale(struct dpp *dpp_base, dpp->scl_data = *scl_data; if ((dpp->base.ctx->dc->config.use_spl) && (!dpp->base.ctx->dc->debug.disable_spl)) { - dscl_mode = (enum dscl_mode_sel) scl_data->dscl_prog_data.dscl_mode; + dscl_mode = (enum dcn401_dscl_mode_sel) scl_data->dscl_prog_data.dscl_mode; rect = (struct rect *)&scl_data->dscl_prog_data.recout; mpc_width = scl_data->dscl_prog_data.mpc_size.width; mpc_height = scl_data->dscl_prog_data.mpc_size.height; @@ -1112,7 +1102,7 @@ void dpp401_dscl_set_scaler_manual_scale(struct dpp *dpp_base, h_num_taps_c = scl_data->dscl_prog_data.taps.h_taps_c; } if (dpp_base->ctx->dc->debug.enable_mem_low_power.bits.dscl) { - if (dscl_mode != DSCL_MODE_DSCL_BYPASS) + if (dscl_mode != DCN401_DSCL_MODE_DSCL_BYPASS) dpp401_power_on_dscl(dpp_base, true); } @@ -1139,7 +1129,7 @@ void dpp401_dscl_set_scaler_manual_scale(struct dpp *dpp_base, /* SCL mode */ REG_UPDATE(SCL_MODE, DSCL_MODE, dscl_mode); - if (dscl_mode == DSCL_MODE_DSCL_BYPASS) { + if (dscl_mode == DCN401_DSCL_MODE_DSCL_BYPASS) { if (dpp_base->ctx->dc->debug.enable_mem_low_power.bits.dscl) dpp401_power_on_dscl(dpp_base, false); return; @@ -1149,7 +1139,7 @@ void dpp401_dscl_set_scaler_manual_scale(struct dpp *dpp_base, lb_config = dpp401_dscl_find_lb_memory_config(dpp, scl_data); dpp401_dscl_set_lb(dpp, &scl_data->lb_params, lb_config); - if (dscl_mode == DSCL_MODE_SCALING_444_BYPASS) { + if (dscl_mode == DCN401_DSCL_MODE_SCALING_444_BYPASS) { if (dpp->base.ctx->dc->config.prefer_easf) dpp401_dscl_disable_easf(dpp_base, scl_data); dpp401_dscl_program_isharp(dpp_base, scl_data, program_isharp_1dlut, &bs_coeffs_updated); diff --git a/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c b/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c index 1f53a9f0c0ac..e4144b244332 100644 --- a/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c +++ b/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c @@ -1157,6 +1157,11 @@ static bool setup_dsc_config( if (!is_dsc_possible) goto done; + /* increase miniumum slice count to meet sink slice width limitations */ + min_slices_h = dc_fixpt_ceil(dc_fixpt_max( + dc_fixpt_div_int(dc_fixpt_from_int(pic_width), dsc_common_caps.max_slice_width), // sink min + dc_fixpt_from_int(min_slices_h))); // source min + min_slices_h = fit_num_slices_up(dsc_common_caps.slice_caps, min_slices_h); /* increase minimum slice count to meet sink throughput limitations */ diff --git a/drivers/gpu/drm/amd/display/dc/hubbub/dcn30/dcn30_hubbub.c b/drivers/gpu/drm/amd/display/dc/hubbub/dcn30/dcn30_hubbub.c index d347bb06577a..e7e5f6d4778e 100644 --- a/drivers/gpu/drm/amd/display/dc/hubbub/dcn30/dcn30_hubbub.c +++ b/drivers/gpu/drm/amd/display/dc/hubbub/dcn30/dcn30_hubbub.c @@ -440,6 +440,35 @@ void hubbub3_init_watermarks(struct hubbub *hubbub) REG_WRITE(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_D, reg); } +void hubbub3_get_det_sizes(struct hubbub *hubbub, uint32_t *curr_det_sizes, uint32_t *target_det_sizes) +{ + struct dcn20_hubbub *hubbub1 = TO_DCN20_HUBBUB(hubbub); + + REG_GET_2(DCHUBBUB_DET0_CTRL, DET0_SIZE_CURRENT, &curr_det_sizes[0], + DET0_SIZE, &target_det_sizes[0]); + + REG_GET_2(DCHUBBUB_DET1_CTRL, DET1_SIZE_CURRENT, &curr_det_sizes[1], + DET1_SIZE, &target_det_sizes[1]); + + REG_GET_2(DCHUBBUB_DET2_CTRL, DET2_SIZE_CURRENT, &curr_det_sizes[2], + DET2_SIZE, &target_det_sizes[2]); + + REG_GET_2(DCHUBBUB_DET3_CTRL, DET3_SIZE_CURRENT, &curr_det_sizes[3], + DET3_SIZE, &target_det_sizes[3]); + +} + +uint32_t hubbub3_compbuf_config_error(struct hubbub *hubbub) +{ + struct dcn20_hubbub *hubbub1 = TO_DCN20_HUBBUB(hubbub); + uint32_t compbuf_config_error = 0; + + REG_GET(DCHUBBUB_COMPBUF_CTRL, CONFIG_ERROR, + &compbuf_config_error); + + return compbuf_config_error; +} + static const struct hubbub_funcs hubbub30_funcs = { .update_dchub = hubbub2_update_dchub, .init_dchub_sys_ctx = hubbub3_init_dchub_sys_ctx, @@ -457,6 +486,8 @@ static const struct hubbub_funcs hubbub30_funcs = { .force_pstate_change_control = hubbub3_force_pstate_change_control, .init_watermarks = hubbub3_init_watermarks, .hubbub_read_state = hubbub2_read_state, + .get_det_sizes = hubbub3_get_det_sizes, + .compbuf_config_error = hubbub3_compbuf_config_error, }; void hubbub3_construct(struct dcn20_hubbub *hubbub3, diff --git a/drivers/gpu/drm/amd/display/dc/hubbub/dcn30/dcn30_hubbub.h b/drivers/gpu/drm/amd/display/dc/hubbub/dcn30/dcn30_hubbub.h index ca6233e8f1f4..49a469969d36 100644 --- a/drivers/gpu/drm/amd/display/dc/hubbub/dcn30/dcn30_hubbub.h +++ b/drivers/gpu/drm/amd/display/dc/hubbub/dcn30/dcn30_hubbub.h @@ -133,4 +133,10 @@ void hubbub3_force_pstate_change_control(struct hubbub *hubbub, void hubbub3_init_watermarks(struct hubbub *hubbub); +void hubbub3_get_det_sizes(struct hubbub *hubbub, + uint32_t *curr_det_sizes, + uint32_t *target_det_sizes); + +uint32_t hubbub3_compbuf_config_error(struct hubbub *hubbub); + #endif diff --git a/drivers/gpu/drm/amd/display/dc/hubbub/dcn31/dcn31_hubbub.c b/drivers/gpu/drm/amd/display/dc/hubbub/dcn31/dcn31_hubbub.c index b98505b240a7..cdb20251a154 100644 --- a/drivers/gpu/drm/amd/display/dc/hubbub/dcn31/dcn31_hubbub.c +++ b/drivers/gpu/drm/amd/display/dc/hubbub/dcn31/dcn31_hubbub.c @@ -1071,6 +1071,8 @@ static const struct hubbub_funcs hubbub31_funcs = { .program_compbuf_size = dcn31_program_compbuf_size, .init_crb = dcn31_init_crb, .hubbub_read_state = hubbub2_read_state, + .get_det_sizes = hubbub3_get_det_sizes, + .compbuf_config_error = hubbub3_compbuf_config_error, }; void hubbub31_construct(struct dcn20_hubbub *hubbub31, diff --git a/drivers/gpu/drm/amd/display/dc/hubbub/dcn32/dcn32_hubbub.c b/drivers/gpu/drm/amd/display/dc/hubbub/dcn32/dcn32_hubbub.c index 32a6be543105..92957398ac0a 100644 --- a/drivers/gpu/drm/amd/display/dc/hubbub/dcn32/dcn32_hubbub.c +++ b/drivers/gpu/drm/amd/display/dc/hubbub/dcn32/dcn32_hubbub.c @@ -1009,6 +1009,8 @@ static const struct hubbub_funcs hubbub32_funcs = { .force_usr_retraining_allow = hubbub32_force_usr_retraining_allow, .set_request_limit = hubbub32_set_request_limit, .get_mall_en = hubbub32_get_mall_en, + .get_det_sizes = hubbub3_get_det_sizes, + .compbuf_config_error = hubbub3_compbuf_config_error, }; void hubbub32_construct(struct dcn20_hubbub *hubbub2, diff --git a/drivers/gpu/drm/amd/display/dc/hubbub/dcn35/dcn35_hubbub.c b/drivers/gpu/drm/amd/display/dc/hubbub/dcn35/dcn35_hubbub.c index 6d41953011f5..a443722a8632 100644 --- a/drivers/gpu/drm/amd/display/dc/hubbub/dcn35/dcn35_hubbub.c +++ b/drivers/gpu/drm/amd/display/dc/hubbub/dcn35/dcn35_hubbub.c @@ -589,6 +589,8 @@ static const struct hubbub_funcs hubbub35_funcs = { .hubbub_read_state = hubbub2_read_state, .force_usr_retraining_allow = hubbub32_force_usr_retraining_allow, .dchubbub_init = hubbub35_init, + .get_det_sizes = hubbub3_get_det_sizes, + .compbuf_config_error = hubbub3_compbuf_config_error, }; void hubbub35_construct(struct dcn20_hubbub *hubbub2, diff --git a/drivers/gpu/drm/amd/display/dc/hubbub/dcn401/dcn401_hubbub.c b/drivers/gpu/drm/amd/display/dc/hubbub/dcn401/dcn401_hubbub.c index 92fab471b183..a36273a52880 100644 --- a/drivers/gpu/drm/amd/display/dc/hubbub/dcn401/dcn401_hubbub.c +++ b/drivers/gpu/drm/amd/display/dc/hubbub/dcn401/dcn401_hubbub.c @@ -1247,6 +1247,8 @@ static const struct hubbub_funcs hubbub4_01_funcs = { .program_compbuf_segments = dcn401_program_compbuf_segments, .wait_for_det_update = dcn401_wait_for_det_update, .program_arbiter = dcn401_program_arbiter, + .get_det_sizes = hubbub3_get_det_sizes, + .compbuf_config_error = hubbub3_compbuf_config_error, }; void hubbub401_construct(struct dcn20_hubbub *hubbub2, diff --git a/drivers/gpu/drm/amd/display/dc/hubp/dcn10/dcn10_hubp.h b/drivers/gpu/drm/amd/display/dc/hubp/dcn10/dcn10_hubp.h index f8f991785d4f..cf2eb9793008 100644 --- a/drivers/gpu/drm/amd/display/dc/hubp/dcn10/dcn10_hubp.h +++ b/drivers/gpu/drm/amd/display/dc/hubp/dcn10/dcn10_hubp.h @@ -104,7 +104,8 @@ SRI(DCN_SURF1_TTU_CNTL1, HUBPREQ, id),\ SRI(DCN_CUR0_TTU_CNTL0, HUBPREQ, id),\ SRI(DCN_CUR0_TTU_CNTL1, HUBPREQ, id),\ - SRI(HUBP_CLK_CNTL, HUBP, id) + SRI(HUBP_CLK_CNTL, HUBP, id),\ + SRI(HUBPRET_READ_LINE_VALUE, HUBPRET, id) /* Register address initialization macro for ASICs with VM */ #define HUBP_REG_LIST_DCN_VM(id)\ @@ -249,7 +250,8 @@ uint32_t CURSOR_POSITION; \ uint32_t CURSOR_HOT_SPOT; \ uint32_t CURSOR_DST_OFFSET; \ - uint32_t HUBP_CLK_CNTL + uint32_t HUBP_CLK_CNTL; \ + uint32_t HUBPRET_READ_LINE_VALUE #define HUBP_SF(reg_name, field_name, post_fix)\ .field_name = reg_name ## __ ## field_name ## post_fix @@ -622,6 +624,8 @@ type DCN_VM_SYSTEM_APERTURE_DEFAULT_SYSTEM;\ type DCN_VM_SYSTEM_APERTURE_DEFAULT_ADDR_MSB;\ type DCN_VM_SYSTEM_APERTURE_DEFAULT_ADDR_LSB;\ + type PIPE_READ_LINE;\ + type HUBP_SEG_ALLOC_ERR_STATUS;\ /* todo: get these from GVM instead of reading registers ourselves */\ type PAGE_DIRECTORY_ENTRY_HI32;\ type PAGE_DIRECTORY_ENTRY_LO32;\ @@ -671,6 +675,7 @@ struct dcn_fl_regs_st { uint32_t lut_done; uint32_t lut_addr_mode; uint32_t lut_width; + uint32_t lut_mpc_width; uint32_t lut_tmz; uint32_t lut_crossbar_sel_r; uint32_t lut_crossbar_sel_g; diff --git a/drivers/gpu/drm/amd/display/dc/hubp/dcn20/dcn20_hubp.h b/drivers/gpu/drm/amd/display/dc/hubp/dcn20/dcn20_hubp.h index 62369be070ea..f325db555102 100644 --- a/drivers/gpu/drm/amd/display/dc/hubp/dcn20/dcn20_hubp.h +++ b/drivers/gpu/drm/amd/display/dc/hubp/dcn20/dcn20_hubp.h @@ -264,6 +264,7 @@ type HUBP_3DLUT_DONE;\ type HUBP_3DLUT_ADDRESSING_MODE;\ type HUBP_3DLUT_WIDTH;\ + type HUBP_3DLUT_MPC_WIDTH;\ type HUBP_3DLUT_TMZ;\ type HUBP_3DLUT_CROSSBAR_SELECT_Y_G;\ type HUBP_3DLUT_CROSSBAR_SELECT_CB_B;\ diff --git a/drivers/gpu/drm/amd/display/dc/hubp/dcn30/dcn30_hubp.c b/drivers/gpu/drm/amd/display/dc/hubp/dcn30/dcn30_hubp.c index 0da70b50e86d..556214b2227d 100644 --- a/drivers/gpu/drm/amd/display/dc/hubp/dcn30/dcn30_hubp.c +++ b/drivers/gpu/drm/amd/display/dc/hubp/dcn30/dcn30_hubp.c @@ -505,6 +505,30 @@ void hubp3_init(struct hubp *hubp) hubp_reset(hubp); } +uint32_t hubp3_get_current_read_line(struct hubp *hubp) +{ + uint32_t read_line = 0; + struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp); + + REG_GET(HUBPRET_READ_LINE_VALUE, + PIPE_READ_LINE, + &read_line); + + return read_line; +} + +unsigned int hubp3_get_underflow_status(struct hubp *hubp) +{ + uint32_t hubp_underflow = 0; + struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp); + + REG_GET(DCHUBP_CNTL, + HUBP_UNDERFLOW_STATUS, + &hubp_underflow); + + return hubp_underflow; +} + static struct hubp_funcs dcn30_hubp_funcs = { .hubp_enable_tripleBuffer = hubp2_enable_triplebuffer, .hubp_is_triplebuffer_enabled = hubp2_is_triplebuffer_enabled, @@ -534,6 +558,8 @@ static struct hubp_funcs dcn30_hubp_funcs = { .hubp_soft_reset = hubp1_soft_reset, .hubp_set_flip_int = hubp1_set_flip_int, .hubp_clear_tiling = hubp3_clear_tiling, + .hubp_get_underflow_status = hubp3_get_underflow_status, + .hubp_get_current_read_line = hubp3_get_current_read_line, }; bool hubp3_construct( diff --git a/drivers/gpu/drm/amd/display/dc/hubp/dcn30/dcn30_hubp.h b/drivers/gpu/drm/amd/display/dc/hubp/dcn30/dcn30_hubp.h index b7d7adf0b58c..842f4eb72cc8 100644 --- a/drivers/gpu/drm/amd/display/dc/hubp/dcn30/dcn30_hubp.h +++ b/drivers/gpu/drm/amd/display/dc/hubp/dcn30/dcn30_hubp.h @@ -243,7 +243,8 @@ HUBP_SF(HUBPREQ0_FLIP_PARAMETERS_6, REFCYC_PER_META_CHUNK_FLIP_C, mask_sh),\ HUBP_SF(HUBPREQ0_VBLANK_PARAMETERS_5, REFCYC_PER_VM_GROUP_VBLANK, mask_sh),\ HUBP_SF(HUBPREQ0_VBLANK_PARAMETERS_6, REFCYC_PER_VM_REQ_VBLANK, mask_sh),\ - HUBP_SF(HUBP0_DCHUBP_REQ_SIZE_CONFIG, VM_GROUP_SIZE, mask_sh) + HUBP_SF(HUBP0_DCHUBP_REQ_SIZE_CONFIG, VM_GROUP_SIZE, mask_sh),\ + HUBP_SF(HUBPRET0_HUBPRET_READ_LINE_VALUE, PIPE_READ_LINE, mask_sh) bool hubp3_construct( struct dcn20_hubp *hubp2, @@ -299,6 +300,11 @@ void hubp3_init(struct hubp *hubp); void hubp3_clear_tiling(struct hubp *hubp); +uint32_t hubp3_get_current_read_line(struct hubp *hubp); + +uint32_t hubp3_get_underflow_status(struct hubp *hubp); + + #endif /* __DC_HUBP_DCN30_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/hubp/dcn31/dcn31_hubp.c b/drivers/gpu/drm/amd/display/dc/hubp/dcn31/dcn31_hubp.c index 7fd582a8a4ba..47101847c2b7 100644 --- a/drivers/gpu/drm/amd/display/dc/hubp/dcn31/dcn31_hubp.c +++ b/drivers/gpu/drm/amd/display/dc/hubp/dcn31/dcn31_hubp.c @@ -68,6 +68,18 @@ void hubp31_program_extended_blank_value( hubp31_program_extended_blank(hubp, min_dst_y_next_start_optimized); } +uint32_t hubp31_get_det_config_error(struct hubp *hubp) +{ + uint32_t config_error = 0; + struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp); + + REG_GET(DCHUBP_CNTL, + HUBP_SEG_ALLOC_ERR_STATUS, + &config_error); + + return config_error; +} + static struct hubp_funcs dcn31_hubp_funcs = { .hubp_enable_tripleBuffer = hubp2_enable_triplebuffer, .hubp_is_triplebuffer_enabled = hubp2_is_triplebuffer_enabled, @@ -98,6 +110,9 @@ static struct hubp_funcs dcn31_hubp_funcs = { .hubp_in_blank = hubp1_in_blank, .program_extended_blank = hubp31_program_extended_blank, .hubp_clear_tiling = hubp3_clear_tiling, + .hubp_get_underflow_status = hubp3_get_underflow_status, + .hubp_get_current_read_line = hubp3_get_current_read_line, + .hubp_get_det_config_error = hubp31_get_det_config_error, }; bool hubp31_construct( diff --git a/drivers/gpu/drm/amd/display/dc/hubp/dcn31/dcn31_hubp.h b/drivers/gpu/drm/amd/display/dc/hubp/dcn31/dcn31_hubp.h index d688db79b750..5952c4671507 100644 --- a/drivers/gpu/drm/amd/display/dc/hubp/dcn31/dcn31_hubp.h +++ b/drivers/gpu/drm/amd/display/dc/hubp/dcn31/dcn31_hubp.h @@ -228,7 +228,9 @@ HUBP_SF(HUBPREQ0_FLIP_PARAMETERS_6, REFCYC_PER_META_CHUNK_FLIP_C, mask_sh),\ HUBP_SF(HUBPREQ0_VBLANK_PARAMETERS_5, REFCYC_PER_VM_GROUP_VBLANK, mask_sh),\ HUBP_SF(HUBPREQ0_VBLANK_PARAMETERS_6, REFCYC_PER_VM_REQ_VBLANK, mask_sh),\ - HUBP_SF(HUBP0_DCHUBP_REQ_SIZE_CONFIG, VM_GROUP_SIZE, mask_sh) + HUBP_SF(HUBP0_DCHUBP_REQ_SIZE_CONFIG, VM_GROUP_SIZE, mask_sh),\ + HUBP_SF(HUBPRET0_HUBPRET_READ_LINE_VALUE, PIPE_READ_LINE, mask_sh),\ + HUBP_SF(HUBP0_DCHUBP_CNTL, HUBP_SEG_ALLOC_ERR_STATUS, mask_sh) bool hubp31_construct( @@ -246,4 +248,6 @@ void hubp31_set_unbounded_requesting(struct hubp *hubp, bool enable); void hubp31_program_extended_blank_value( struct hubp *hubp, unsigned int min_dst_y_next_start_optimized); +uint32_t hubp31_get_det_config_error(struct hubp *hubp); + #endif /* __DC_HUBP_DCN31_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/hubp/dcn32/dcn32_hubp.c b/drivers/gpu/drm/amd/display/dc/hubp/dcn32/dcn32_hubp.c index f3a21c623f44..a5f23bb2a76a 100644 --- a/drivers/gpu/drm/amd/display/dc/hubp/dcn32/dcn32_hubp.c +++ b/drivers/gpu/drm/amd/display/dc/hubp/dcn32/dcn32_hubp.c @@ -206,6 +206,9 @@ static struct hubp_funcs dcn32_hubp_funcs = { .hubp_update_mall_sel = hubp32_update_mall_sel, .hubp_prepare_subvp_buffering = hubp32_prepare_subvp_buffering, .hubp_clear_tiling = hubp3_clear_tiling, + .hubp_get_underflow_status = hubp3_get_underflow_status, + .hubp_get_current_read_line = hubp3_get_current_read_line, + .hubp_get_det_config_error = hubp31_get_det_config_error, }; bool hubp32_construct( diff --git a/drivers/gpu/drm/amd/display/dc/hubp/dcn35/dcn35_hubp.c b/drivers/gpu/drm/amd/display/dc/hubp/dcn35/dcn35_hubp.c index 6d060ba12da8..b140808f21af 100644 --- a/drivers/gpu/drm/amd/display/dc/hubp/dcn35/dcn35_hubp.c +++ b/drivers/gpu/drm/amd/display/dc/hubp/dcn35/dcn35_hubp.c @@ -218,6 +218,9 @@ static struct hubp_funcs dcn35_hubp_funcs = { .hubp_in_blank = hubp1_in_blank, .program_extended_blank = hubp31_program_extended_blank_value, .hubp_clear_tiling = hubp3_clear_tiling, + .hubp_get_underflow_status = hubp3_get_underflow_status, + .hubp_get_current_read_line = hubp3_get_current_read_line, + .hubp_get_det_config_error = hubp31_get_det_config_error, }; bool hubp35_construct( diff --git a/drivers/gpu/drm/amd/display/dc/hubp/dcn401/dcn401_hubp.c b/drivers/gpu/drm/amd/display/dc/hubp/dcn401/dcn401_hubp.c index 705b98b1b6cc..0fcbc6a35be6 100644 --- a/drivers/gpu/drm/amd/display/dc/hubp/dcn401/dcn401_hubp.c +++ b/drivers/gpu/drm/amd/display/dc/hubp/dcn401/dcn401_hubp.c @@ -127,6 +127,43 @@ void hubp401_program_3dlut_fl_format(struct hubp *hubp, enum hubp_3dlut_fl_forma REG_UPDATE(_3DLUT_FL_CONFIG, HUBP0_3DLUT_FL_FORMAT, format); } +void hubp401_program_3dlut_fl_config( + struct hubp *hubp, + struct hubp_fl_3dlut_config *cfg) +{ + struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp); + + uint32_t mpc_width = {(cfg->width == 17) ? 0 : 1}; + uint32_t width = {cfg->width}; + + if (cfg->layout == DC_CM2_GPU_MEM_LAYOUT_1D_PACKED_LINEAR) + width = (cfg->width == 17) ? 4916 : 35940; + + REG_UPDATE_2(_3DLUT_FL_CONFIG, + HUBP0_3DLUT_FL_MODE, cfg->mode, + HUBP0_3DLUT_FL_FORMAT, cfg->format); + + REG_UPDATE_2(_3DLUT_FL_BIAS_SCALE, + HUBP0_3DLUT_FL_BIAS, cfg->bias, + HUBP0_3DLUT_FL_SCALE, cfg->scale); + + REG_UPDATE(HUBP_3DLUT_ADDRESS_HIGH, + HUBP_3DLUT_ADDRESS_HIGH, cfg->address.lut3d.addr.high_part); + REG_UPDATE(HUBP_3DLUT_ADDRESS_LOW, + HUBP_3DLUT_ADDRESS_LOW, cfg->address.lut3d.addr.low_part); + + //cross bar + REG_UPDATE_8(HUBP_3DLUT_CONTROL, + HUBP_3DLUT_MPC_WIDTH, mpc_width, + HUBP_3DLUT_WIDTH, width, + HUBP_3DLUT_CROSSBAR_SELECT_CR_R, cfg->crossbar_bit_slice_cr_r, + HUBP_3DLUT_CROSSBAR_SELECT_Y_G, cfg->crossbar_bit_slice_y_g, + HUBP_3DLUT_CROSSBAR_SELECT_CB_B, cfg->crossbar_bit_slice_cb_b, + HUBP_3DLUT_ADDRESSING_MODE, cfg->addr_mode, + HUBP_3DLUT_TMZ, cfg->protection_bits, + HUBP_3DLUT_ENABLE, cfg->enabled ? 1 : 0); +} + void hubp401_update_mall_sel(struct hubp *hubp, uint32_t mall_sel, bool c_cursor) { struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp); @@ -1033,6 +1070,10 @@ static struct hubp_funcs dcn401_hubp_funcs = { .hubp_program_3dlut_fl_crossbar = hubp401_program_3dlut_fl_crossbar, .hubp_get_3dlut_fl_done = hubp401_get_3dlut_fl_done, .hubp_clear_tiling = hubp401_clear_tiling, + .hubp_program_3dlut_fl_config = hubp401_program_3dlut_fl_config, + .hubp_get_underflow_status = hubp3_get_underflow_status, + .hubp_get_current_read_line = hubp3_get_current_read_line, + .hubp_get_det_config_error = hubp31_get_det_config_error, }; bool hubp401_construct( diff --git a/drivers/gpu/drm/amd/display/dc/hubp/dcn401/dcn401_hubp.h b/drivers/gpu/drm/amd/display/dc/hubp/dcn401/dcn401_hubp.h index 608e6153fa68..fdabbeec8ffa 100644 --- a/drivers/gpu/drm/amd/display/dc/hubp/dcn401/dcn401_hubp.h +++ b/drivers/gpu/drm/amd/display/dc/hubp/dcn401/dcn401_hubp.h @@ -252,7 +252,9 @@ HUBP_SF(HUBP0_DCHUBP_MCACHEID_CONFIG, MCACHEID_MALL_PREF_1H_P0, mask_sh),\ HUBP_SF(HUBP0_DCHUBP_MCACHEID_CONFIG, MCACHEID_MALL_PREF_2H_P0, mask_sh),\ HUBP_SF(HUBP0_DCHUBP_MCACHEID_CONFIG, MCACHEID_MALL_PREF_1H_P1, mask_sh),\ - HUBP_SF(HUBP0_DCHUBP_MCACHEID_CONFIG, MCACHEID_MALL_PREF_2H_P1, mask_sh) + HUBP_SF(HUBP0_DCHUBP_MCACHEID_CONFIG, MCACHEID_MALL_PREF_2H_P1, mask_sh),\ + HUBP_SF(HUBPRET0_HUBPRET_READ_LINE_VALUE, PIPE_READ_LINE, mask_sh),\ + HUBP_SF(HUBP0_DCHUBP_CNTL, HUBP_SEG_ALLOC_ERR_STATUS, mask_sh) void hubp401_update_mall_sel(struct hubp *hubp, uint32_t mall_sel, bool c_cursor); @@ -349,6 +351,10 @@ void hubp401_program_3dlut_fl_format(struct hubp *hubp, enum hubp_3dlut_fl_forma void hubp401_program_3dlut_fl_mode(struct hubp *hubp, enum hubp_3dlut_fl_mode mode); +void hubp401_program_3dlut_fl_config( + struct hubp *hubp, + struct hubp_fl_3dlut_config *cfg); + void hubp401_clear_tiling(struct hubp *hubp); void hubp401_vready_at_or_After_vsync(struct hubp *hubp, diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c index 4ea13d0bf815..153d68375fa3 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c @@ -671,6 +671,7 @@ void dce110_enable_stream(struct pipe_ctx *pipe_ctx) uint32_t early_control = 0; struct timing_generator *tg = pipe_ctx->stream_res.tg; + link_hwss->setup_stream_attribute(pipe_ctx); link_hwss->setup_stream_encoder(pipe_ctx); dc->hwss.update_info_frame(pipe_ctx); @@ -1269,7 +1270,7 @@ void dce110_set_avmute(struct pipe_ctx *pipe_ctx, bool enable) pipe_ctx->stream_res.stream_enc->funcs->set_avmute(pipe_ctx->stream_res.stream_enc, enable); } -static enum audio_dto_source translate_to_dto_source(enum controller_id crtc_id) +enum audio_dto_source translate_to_dto_source(enum controller_id crtc_id) { switch (crtc_id) { case CONTROLLER_ID_D0: @@ -1289,7 +1290,7 @@ static enum audio_dto_source translate_to_dto_source(enum controller_id crtc_id) } } -static void populate_audio_dp_link_info( +void populate_audio_dp_link_info( const struct pipe_ctx *pipe_ctx, struct audio_dp_link_info *dp_link_info) { @@ -1600,19 +1601,17 @@ enum dc_status dce110_apply_single_controller_ctx_to_hw( } if (pipe_ctx->stream_res.audio != NULL) { - struct audio_output audio_output = {0}; + build_audio_output(context, pipe_ctx, &pipe_ctx->stream_res.audio_output); - build_audio_output(context, pipe_ctx, &audio_output); - - link_hwss->setup_audio_output(pipe_ctx, &audio_output, + link_hwss->setup_audio_output(pipe_ctx, &pipe_ctx->stream_res.audio_output, pipe_ctx->stream_res.audio->inst); pipe_ctx->stream_res.audio->funcs->az_configure( pipe_ctx->stream_res.audio, pipe_ctx->stream->signal, - &audio_output.crtc_info, + &pipe_ctx->stream_res.audio_output.crtc_info, &pipe_ctx->stream->audio_info, - &audio_output.dp_link_info); + &pipe_ctx->stream_res.audio_output.dp_link_info); if (dc->config.disable_hbr_audio_dp2) if (pipe_ctx->stream_res.audio->funcs->az_disable_hbr_audio && @@ -2254,7 +2253,7 @@ static bool should_enable_fbc(struct dc *dc, /* * Enable FBC */ -static void enable_fbc( +void enable_fbc( struct dc *dc, struct dc_state *context) { @@ -2386,9 +2385,7 @@ static void dce110_setup_audio_dto( if (pipe_ctx->stream->signal != SIGNAL_TYPE_HDMI_TYPE_A) continue; if (pipe_ctx->stream_res.audio != NULL) { - struct audio_output audio_output; - - build_audio_output(context, pipe_ctx, &audio_output); + build_audio_output(context, pipe_ctx, &pipe_ctx->stream_res.audio_output); if (dc->res_pool->dccg && dc->res_pool->dccg->funcs->set_audio_dtbclk_dto) { struct dtbclk_dto_params dto_params = {0}; @@ -2399,14 +2396,14 @@ static void dce110_setup_audio_dto( pipe_ctx->stream_res.audio->funcs->wall_dto_setup( pipe_ctx->stream_res.audio, pipe_ctx->stream->signal, - &audio_output.crtc_info, - &audio_output.pll_info); + &pipe_ctx->stream_res.audio_output.crtc_info, + &pipe_ctx->stream_res.audio_output.pll_info); } else pipe_ctx->stream_res.audio->funcs->wall_dto_setup( pipe_ctx->stream_res.audio, pipe_ctx->stream->signal, - &audio_output.crtc_info, - &audio_output.pll_info); + &pipe_ctx->stream_res.audio_output.crtc_info, + &pipe_ctx->stream_res.audio_output.pll_info); break; } } @@ -2426,15 +2423,15 @@ static void dce110_setup_audio_dto( continue; if (pipe_ctx->stream_res.audio != NULL) { - struct audio_output audio_output = {0}; - - build_audio_output(context, pipe_ctx, &audio_output); + build_audio_output(context, + pipe_ctx, + &pipe_ctx->stream_res.audio_output); pipe_ctx->stream_res.audio->funcs->wall_dto_setup( pipe_ctx->stream_res.audio, pipe_ctx->stream->signal, - &audio_output.crtc_info, - &audio_output.pll_info); + &pipe_ctx->stream_res.audio_output.crtc_info, + &pipe_ctx->stream_res.audio_output.pll_info); break; } } diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.h b/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.h index 7cd8c1576988..9c032e449481 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.h @@ -114,5 +114,12 @@ void build_audio_output( struct dc_state *state, const struct pipe_ctx *pipe_ctx, struct audio_output *audio_output); +enum audio_dto_source translate_to_dto_source(enum controller_id crtc_id); +void populate_audio_dp_link_info( + const struct pipe_ctx *pipe_ctx, + struct audio_dp_link_info *dp_link_info); +void enable_fbc( + struct dc *dc, + struct dc_state *context); #endif /* __DC_HWSS_DCE110_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c index 39910f73ecd0..506c3bbbf221 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c @@ -328,19 +328,25 @@ static void dcn10_log_hubp_states(struct dc *dc, void *log_ctx) } DTN_INFO("\n=======HUBP FL======\n"); - DTN_INFO( - "HUBP FL: Enabled Done adr_mode width tmz xbar_sel_R xbar_sel_G xbar_sel_B adr_hi adr_low REFCYC Bias Scale Mode Format\n"); + static const char * const pLabels[] = { + "inst", "Enabled ", "Done ", "adr_mode ", "width ", "mpc_width ", + "tmz", "xbar_sel_R", "xbar_sel_G", "xbar_sel_B", "adr_hi ", + "adr_low", "REFCYC", "Bias", "Scale", "Mode", + "Format", "prefetch"}; + for (i = 0; i < pool->pipe_count; i++) { struct dcn_hubp_state *s = &(TO_DCN10_HUBP(pool->hubps[i])->state); struct dcn_fl_regs_st *fl_regs = &s->fl_regs; + struct _vcs_dpi_display_dlg_regs_st *dlg_regs = &s->dlg_attr; if (!s->blank_en) { - DTN_INFO("[%2d]: %5xh %6xh %5d %6d %8xh %2xh %6xh %6d %8d %8d %7d %8xh %5x %5x %5x", + uint32_t values[] = { pool->hubps[i]->inst, fl_regs->lut_enable, fl_regs->lut_done, fl_regs->lut_addr_mode, fl_regs->lut_width, + fl_regs->lut_mpc_width, fl_regs->lut_tmz, fl_regs->lut_crossbar_sel_r, fl_regs->lut_crossbar_sel_g, @@ -351,8 +357,13 @@ static void dcn10_log_hubp_states(struct dc *dc, void *log_ctx) fl_regs->lut_fl_bias, fl_regs->lut_fl_scale, fl_regs->lut_fl_mode, - fl_regs->lut_fl_format); - DTN_INFO("\n"); + fl_regs->lut_fl_format, + dlg_regs->dst_y_prefetch}; + + int num_elements = 18; + + for (int j = 0; j < num_elements; j++) + DTN_INFO("%s \t %8xh\n", pLabels[j], values[j]); } } @@ -541,19 +552,43 @@ static void dcn10_log_color_state(struct dc *dc, dc->caps.color.mpc.ogam_ram, dc->caps.color.mpc.ocsc); DTN_INFO("===== MPC RMCM 3DLUT =====\n"); - DTN_INFO("MPCC: SIZE MODE MODE_CUR RD_SEL 30BIT_EN WR_EN_MASK RAM_SEL OUT_NORM_FACTOR FL_SEL OUT_OFFSET OUT_SCALE FL_DONE SOFT_UNDERFLOW HARD_UNDERFLOW MEM_PWR_ST FORCE DIS MODE\n"); + static const char * const pLabels[] = { + "MPCC", "SIZE", "MODE", "MODE_CUR", "RD_SEL", + "30BIT_EN", "WR_EN_MASK", "RAM_SEL", "OUT_NORM_FACTOR", "FL_SEL", + "OUT_OFFSET", "OUT_SCALE", "FL_DONE", "SOFT_UNDERFLOW", "HARD_UNDERFLOW", + "MEM_PWR_ST", "FORCE", "DIS", "MODE"}; + for (i = 0; i < pool->mpcc_count; i++) { struct mpcc_state s = {0}; pool->mpc->funcs->read_mpcc_state(pool->mpc, i, &s); - if (s.opp_id != 0xf) - DTN_INFO("[%2d]: %4xh %4xh %6xh %4x %4x %4x %4x %4x %4xh %4xh %6xh %4x %4x %4x %4x %4x %4x %4x\n", - i, s.rmcm_regs.rmcm_3dlut_size, s.rmcm_regs.rmcm_3dlut_mode, s.rmcm_regs.rmcm_3dlut_mode_cur, - s.rmcm_regs.rmcm_3dlut_read_sel, s.rmcm_regs.rmcm_3dlut_30bit_en, s.rmcm_regs.rmcm_3dlut_wr_en_mask, - s.rmcm_regs.rmcm_3dlut_ram_sel, s.rmcm_regs.rmcm_3dlut_out_norm_factor, s.rmcm_regs.rmcm_3dlut_fl_sel, - s.rmcm_regs.rmcm_3dlut_out_offset_r, s.rmcm_regs.rmcm_3dlut_out_scale_r, s.rmcm_regs.rmcm_3dlut_fl_done, - s.rmcm_regs.rmcm_3dlut_fl_soft_underflow, s.rmcm_regs.rmcm_3dlut_fl_hard_underflow, s.rmcm_regs.rmcm_3dlut_mem_pwr_state, - s.rmcm_regs.rmcm_3dlut_mem_pwr_force, s.rmcm_regs.rmcm_3dlut_mem_pwr_dis, s.rmcm_regs.rmcm_3dlut_mem_pwr_mode); + if (s.opp_id != 0xf) { + uint32_t values[] = { + i, + s.rmcm_regs.rmcm_3dlut_size, + s.rmcm_regs.rmcm_3dlut_mode, + s.rmcm_regs.rmcm_3dlut_mode_cur, + s.rmcm_regs.rmcm_3dlut_read_sel, + s.rmcm_regs.rmcm_3dlut_30bit_en, + s.rmcm_regs.rmcm_3dlut_wr_en_mask, + s.rmcm_regs.rmcm_3dlut_ram_sel, + s.rmcm_regs.rmcm_3dlut_out_norm_factor, + s.rmcm_regs.rmcm_3dlut_fl_sel, + s.rmcm_regs.rmcm_3dlut_out_offset_r, + s.rmcm_regs.rmcm_3dlut_out_scale_r, + s.rmcm_regs.rmcm_3dlut_fl_done, + s.rmcm_regs.rmcm_3dlut_fl_soft_underflow, + s.rmcm_regs.rmcm_3dlut_fl_hard_underflow, + s.rmcm_regs.rmcm_3dlut_mem_pwr_state, + s.rmcm_regs.rmcm_3dlut_mem_pwr_force, + s.rmcm_regs.rmcm_3dlut_mem_pwr_dis, + s.rmcm_regs.rmcm_3dlut_mem_pwr_mode}; + + int num_elements = 19; + + for (int j = 0; j < num_elements; j++) + DTN_INFO("%s \t %8xh\n", pLabels[j], values[j]); + } } DTN_INFO("\n"); DTN_INFO("===== MPC RMCM Shaper =====\n"); @@ -3628,6 +3663,8 @@ void dcn10_set_cursor_position(struct pipe_ctx *pipe_ctx) int y_plane = pipe_ctx->plane_state->dst_rect.y; int x_pos = pos_cpy.x; int y_pos = pos_cpy.y; + int clip_x = pipe_ctx->plane_state->clip_rect.x; + int clip_width = pipe_ctx->plane_state->clip_rect.width; if ((pipe_ctx->top_pipe != NULL) || (pipe_ctx->bottom_pipe != NULL)) { if ((pipe_ctx->plane_state->src_rect.width != pipe_ctx->plane_res.scl_data.viewport.width) || @@ -3646,7 +3683,7 @@ void dcn10_set_cursor_position(struct pipe_ctx *pipe_ctx) */ /** - * Translate cursor from stream space to plane space. + * Translate cursor and clip offset from stream space to plane space. * * If the cursor is scaled then we need to scale the position * to be in the approximately correct place. We can't do anything @@ -3663,6 +3700,10 @@ void dcn10_set_cursor_position(struct pipe_ctx *pipe_ctx) pipe_ctx->plane_state->dst_rect.width; y_pos = (y_pos - y_plane) * pipe_ctx->plane_state->src_rect.height / pipe_ctx->plane_state->dst_rect.height; + clip_x = (clip_x - x_plane) * pipe_ctx->plane_state->src_rect.width / + pipe_ctx->plane_state->dst_rect.width; + clip_width = clip_width * pipe_ctx->plane_state->src_rect.width / + pipe_ctx->plane_state->dst_rect.width; } /** @@ -3709,30 +3750,18 @@ void dcn10_set_cursor_position(struct pipe_ctx *pipe_ctx) if (param.rotation == ROTATION_ANGLE_0) { - int viewport_width = - pipe_ctx->plane_res.scl_data.viewport.width; - int viewport_x = - pipe_ctx->plane_res.scl_data.viewport.x; if (param.mirror) { - if (pipe_split_on || odm_combine_on) { - if (pos_cpy.x >= viewport_width + viewport_x) { - pos_cpy.x = 2 * viewport_width - - pos_cpy.x + 2 * viewport_x; - } else { - uint32_t temp_x = pos_cpy.x; - - pos_cpy.x = 2 * viewport_x - pos_cpy.x; - if (temp_x >= viewport_x + - (int)hubp->curs_attr.width || pos_cpy.x - <= (int)hubp->curs_attr.width + - pipe_ctx->plane_state->src_rect.x) { - pos_cpy.x = 2 * viewport_width - temp_x; - } - } - } else { - pos_cpy.x = viewport_width - pos_cpy.x + 2 * viewport_x; - } + /* + * The plane is split into multiple viewports. + * The combination of all viewports span the + * entirety of the clip rect. + * + * For no pipe_split, viewport_width is represents + * the full width of the clip_rect, so we can just + * mirror it. + */ + pos_cpy.x = clip_width - pos_cpy.x + 2 * clip_x; } } // Swap axis and mirror horizontally @@ -3802,30 +3831,17 @@ void dcn10_set_cursor_position(struct pipe_ctx *pipe_ctx) } // Mirror horizontally and vertically else if (param.rotation == ROTATION_ANGLE_180) { - int viewport_width = - pipe_ctx->plane_res.scl_data.viewport.width; - int viewport_x = - pipe_ctx->plane_res.scl_data.viewport.x; - if (!param.mirror) { - if (pipe_split_on || odm_combine_on) { - if (pos_cpy.x >= viewport_width + viewport_x) { - pos_cpy.x = 2 * viewport_width - - pos_cpy.x + 2 * viewport_x; - } else { - uint32_t temp_x = pos_cpy.x; - - pos_cpy.x = 2 * viewport_x - pos_cpy.x; - if (temp_x >= viewport_x + - (int)hubp->curs_attr.width || pos_cpy.x - <= (int)hubp->curs_attr.width + - pipe_ctx->plane_state->src_rect.x) { - pos_cpy.x = temp_x + viewport_width; - } - } - } else { - pos_cpy.x = viewport_width - pos_cpy.x + 2 * viewport_x; - } + /* + * The plane is split into multiple viewports. + * The combination of all viewports span the + * entirety of the clip rect. + * + * For no pipe_split, viewport_width is represents + * the full width of the clip_rect, so we can just + * mirror it. + */ + pos_cpy.x = clip_width - pos_cpy.x + 2 * clip_x; } /** diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c index 3207addbd4eb..7d24fa1517bf 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c @@ -3054,6 +3054,8 @@ void dcn20_enable_stream(struct pipe_ctx *pipe_ctx) link_enc->transmitter - TRANSMITTER_UNIPHY_A); } + link_hwss->setup_stream_attribute(pipe_ctx); + if (dc->res_pool->dccg->funcs->set_pixel_rate_div) dc->res_pool->dccg->funcs->set_pixel_rate_div( dc->res_pool->dccg, diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_hwseq.c index 37a239219dfe..139a63101488 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_hwseq.c @@ -1228,3 +1228,51 @@ void dcn30_wait_for_all_pending_updates(const struct pipe_ctx *pipe_ctx) } } } + +void dcn30_get_underflow_debug_data(const struct dc *dc, + struct timing_generator *tg, + struct dc_underflow_debug_data *out_data) +{ + struct hubbub *hubbub = dc->res_pool->hubbub; + + if (tg) { + uint32_t v_blank_start = 0, v_blank_end = 0; + + out_data->otg_inst = tg->inst; + + tg->funcs->get_scanoutpos(tg, + &v_blank_start, + &v_blank_end, + &out_data->h_position, + &out_data->v_position); + + out_data->otg_frame_count = tg->funcs->get_frame_count(tg); + + out_data->otg_underflow = tg->funcs->is_optc_underflow_occurred(tg); + } + + for (int i = 0; i < MAX_PIPES; i++) { + struct hubp *hubp = dc->res_pool->hubps[i]; + + if (hubp) { + if (hubp->funcs->hubp_get_underflow_status) + out_data->hubps[i].hubp_underflow = hubp->funcs->hubp_get_underflow_status(hubp); + + if (hubp->funcs->hubp_in_blank) + out_data->hubps[i].hubp_in_blank = hubp->funcs->hubp_in_blank(hubp); + + if (hubp->funcs->hubp_get_current_read_line) + out_data->hubps[i].hubp_readline = hubp->funcs->hubp_get_current_read_line(hubp); + + if (hubp->funcs->hubp_get_det_config_error) + out_data->hubps[i].det_config_error = hubp->funcs->hubp_get_det_config_error(hubp); + } + } + + if (hubbub->funcs->get_det_sizes) + hubbub->funcs->get_det_sizes(hubbub, out_data->curr_det_sizes, out_data->target_det_sizes); + + if (hubbub->funcs->compbuf_config_error) + out_data->compbuf_config_error = hubbub->funcs->compbuf_config_error(hubbub); + +} diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_hwseq.h b/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_hwseq.h index 4b90b781c4f2..40afbbfb5b9c 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_hwseq.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_hwseq.h @@ -29,6 +29,7 @@ #include "hw_sequencer_private.h" struct dc; +struct dc_underflow_debug_data; void dcn30_init_hw(struct dc *dc); void dcn30_program_all_writeback_pipes_in_tree( @@ -98,4 +99,8 @@ void dcn30_prepare_bandwidth(struct dc *dc, void dcn30_wait_for_all_pending_updates(const struct pipe_ctx *pipe_ctx); +void dcn30_get_underflow_debug_data(const struct dc *dc, + struct timing_generator *tg, + struct dc_underflow_debug_data *out_data); + #endif /* __DC_HWSS_DCN30_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_init.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_init.c index 2ac5d54d1626..d7ff55669bac 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_init.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_init.c @@ -110,6 +110,7 @@ static const struct hw_sequencer_funcs dcn30_funcs = { .update_visual_confirm_color = dcn10_update_visual_confirm_color, .is_abm_supported = dcn21_is_abm_supported, .wait_for_all_pending_updates = dcn30_wait_for_all_pending_updates, + .get_underflow_debug_data = dcn30_get_underflow_debug_data, }; static const struct hwseq_private_funcs dcn30_private_funcs = { diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn31/dcn31_init.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn31/dcn31_init.c index 556f4fe57eda..5a6a459da224 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn31/dcn31_init.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn31/dcn31_init.c @@ -112,6 +112,7 @@ static const struct hw_sequencer_funcs dcn31_funcs = { .exit_optimized_pwr_state = dcn21_exit_optimized_pwr_state, .update_visual_confirm_color = dcn10_update_visual_confirm_color, .setup_hpo_hw_control = dcn31_setup_hpo_hw_control, + .get_underflow_debug_data = dcn30_get_underflow_debug_data, }; static const struct hwseq_private_funcs dcn31_private_funcs = { diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_hwseq.c index e68f21fd5f0f..560984533950 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_hwseq.c @@ -528,3 +528,75 @@ void dcn314_disable_link_output(struct dc_link *link, apply_symclk_on_tx_off_wa(link); } + +/** + * dcn314_dpp_pg_control - DPP power gate control. + * + * @hws: dce_hwseq reference. + * @dpp_inst: DPP instance reference. + * @power_on: true if we want to enable power gate, false otherwise. + * + * Enable or disable power gate in the specific DPP instance. + * If power gating is disabled, will force disable cursor in the DPP instance. + */ +void dcn314_dpp_pg_control( + struct dce_hwseq *hws, + unsigned int dpp_inst, + bool power_on) +{ + uint32_t power_gate = power_on ? 0 : 1; + uint32_t pwr_status = power_on ? 0 : 2; + + + if (hws->ctx->dc->debug.disable_dpp_power_gate) { + /* Workaround for DCN314 with disabled power gating */ + if (!power_on) { + + /* Force disable cursor if power gating is disabled */ + struct dpp *dpp = hws->ctx->dc->res_pool->dpps[dpp_inst]; + if (dpp && dpp->funcs->dpp_force_disable_cursor) + dpp->funcs->dpp_force_disable_cursor(dpp); + } + return; + } + if (REG(DOMAIN1_PG_CONFIG) == 0) + return; + + switch (dpp_inst) { + case 0: /* DPP0 */ + REG_UPDATE(DOMAIN1_PG_CONFIG, + DOMAIN1_POWER_GATE, power_gate); + + REG_WAIT(DOMAIN1_PG_STATUS, + DOMAIN1_PGFSM_PWR_STATUS, pwr_status, + 1, 1000); + break; + case 1: /* DPP1 */ + REG_UPDATE(DOMAIN3_PG_CONFIG, + DOMAIN3_POWER_GATE, power_gate); + + REG_WAIT(DOMAIN3_PG_STATUS, + DOMAIN3_PGFSM_PWR_STATUS, pwr_status, + 1, 1000); + break; + case 2: /* DPP2 */ + REG_UPDATE(DOMAIN5_PG_CONFIG, + DOMAIN5_POWER_GATE, power_gate); + + REG_WAIT(DOMAIN5_PG_STATUS, + DOMAIN5_PGFSM_PWR_STATUS, pwr_status, + 1, 1000); + break; + case 3: /* DPP3 */ + REG_UPDATE(DOMAIN7_PG_CONFIG, + DOMAIN7_POWER_GATE, power_gate); + + REG_WAIT(DOMAIN7_PG_STATUS, + DOMAIN7_PGFSM_PWR_STATUS, pwr_status, + 1, 1000); + break; + default: + BREAK_TO_DEBUGGER(); + break; + } +} diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_hwseq.h b/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_hwseq.h index 2305ad282f21..6c072d0274ea 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_hwseq.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_hwseq.h @@ -47,4 +47,6 @@ void dcn314_dpp_root_clock_control(struct dce_hwseq *hws, unsigned int dpp_inst, void dcn314_disable_link_output(struct dc_link *link, const struct link_resource *link_res, enum signal_type signal); +void dcn314_dpp_pg_control(struct dce_hwseq *hws, unsigned int dpp_inst, bool power_on); + #endif /* __DC_HWSS_DCN314_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_init.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_init.c index f5112742edf9..79faab1125d4 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_init.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_init.c @@ -115,6 +115,7 @@ static const struct hw_sequencer_funcs dcn314_funcs = { .update_visual_confirm_color = dcn10_update_visual_confirm_color, .calculate_pix_rate_divider = dcn314_calculate_pix_rate_divider, .setup_hpo_hw_control = dcn31_setup_hpo_hw_control, + .get_underflow_debug_data = dcn30_get_underflow_debug_data, }; static const struct hwseq_private_funcs dcn314_private_funcs = { @@ -141,6 +142,7 @@ static const struct hwseq_private_funcs dcn314_private_funcs = { .enable_power_gating_plane = dcn314_enable_power_gating_plane, .dpp_root_clock_control = dcn314_dpp_root_clock_control, .hubp_pg_control = dcn31_hubp_pg_control, + .dpp_pg_control = dcn314_dpp_pg_control, .program_all_writeback_pipes_in_tree = dcn30_program_all_writeback_pipes_in_tree, .update_odm = dcn314_update_odm, .dsc_pg_control = dcn314_dsc_pg_control, diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_init.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_init.c index b971356d30b1..c19ef075c882 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_init.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_init.c @@ -121,6 +121,7 @@ static const struct hw_sequencer_funcs dcn32_funcs = { .calculate_pix_rate_divider = dcn32_calculate_pix_rate_divider, .program_outstanding_updates = dcn32_program_outstanding_updates, .wait_for_all_pending_updates = dcn30_wait_for_all_pending_updates, + .get_underflow_debug_data = dcn30_get_underflow_debug_data, }; static const struct hwseq_private_funcs dcn32_private_funcs = { diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_init.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_init.c index a3ccf805bd16..52cc488416ac 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_init.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_init.c @@ -128,6 +128,7 @@ static const struct hw_sequencer_funcs dcn35_funcs = { .enable_plane = dcn20_enable_plane, .update_dchubp_dpp = dcn20_update_dchubp_dpp, .post_unlock_reset_opp = dcn20_post_unlock_reset_opp, + .get_underflow_debug_data = dcn30_get_underflow_debug_data, }; static const struct hwseq_private_funcs dcn35_private_funcs = { diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn351/dcn351_init.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn351/dcn351_init.c index 58f2be2a326b..e34efcb7bde5 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn351/dcn351_init.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn351/dcn351_init.c @@ -123,6 +123,7 @@ static const struct hw_sequencer_funcs dcn351_funcs = { .set_long_vtotal = dcn35_set_long_vblank, .calculate_pix_rate_divider = dcn32_calculate_pix_rate_divider, .setup_hpo_hw_control = dcn35_setup_hpo_hw_control, + .get_underflow_debug_data = dcn30_get_underflow_debug_data, }; static const struct hwseq_private_funcs dcn351_private_funcs = { diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c index cc9f40d97af2..d5b5e2ce6ff6 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c @@ -965,6 +965,8 @@ void dcn401_enable_stream(struct pipe_ctx *pipe_ctx) } } + link_hwss->setup_stream_attribute(pipe_ctx); + if (dc->res_pool->dccg->funcs->set_pixel_rate_div) { dc->res_pool->dccg->funcs->set_pixel_rate_div( dc->res_pool->dccg, @@ -1619,20 +1621,28 @@ void dcn401_unblank_stream(struct pipe_ctx *pipe_ctx, void dcn401_hardware_release(struct dc *dc) { - dc_dmub_srv_fams2_update_config(dc, dc->current_state, false); - - /* If pstate unsupported, or still supported - * by firmware, force it supported by dcn - */ - if (dc->current_state) { - if ((!dc->clk_mgr->clks.p_state_change_support || - dc->current_state->bw_ctx.bw.dcn.fams2_global_config.features.bits.enable) && - dc->res_pool->hubbub->funcs->force_pstate_change_control) - dc->res_pool->hubbub->funcs->force_pstate_change_control( - dc->res_pool->hubbub, true, true); - - dc->current_state->bw_ctx.bw.dcn.clk.p_state_change_support = true; - dc->clk_mgr->funcs->update_clocks(dc->clk_mgr, dc->current_state, true); + if (!dc->debug.disable_force_pstate_allow_on_hw_release) { + dc_dmub_srv_fams2_update_config(dc, dc->current_state, false); + + /* If pstate unsupported, or still supported + * by firmware, force it supported by dcn + */ + if (dc->current_state) { + if ((!dc->clk_mgr->clks.p_state_change_support || + dc->current_state->bw_ctx.bw.dcn.fams2_global_config.features.bits.enable) && + dc->res_pool->hubbub->funcs->force_pstate_change_control) + dc->res_pool->hubbub->funcs->force_pstate_change_control( + dc->res_pool->hubbub, true, true); + + dc->current_state->bw_ctx.bw.dcn.clk.p_state_change_support = true; + dc->clk_mgr->funcs->update_clocks(dc->clk_mgr, dc->current_state, true); + } + } else { + if (dc->current_state) { + dc->clk_mgr->clks.p_state_change_support = false; + dc->clk_mgr->funcs->update_clocks(dc->clk_mgr, dc->current_state, true); + } + dc_dmub_srv_fams2_update_config(dc, dc->current_state, false); } } diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_init.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_init.c index fe7aceb2f510..d6e11b7e4fce 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_init.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_init.c @@ -104,6 +104,7 @@ static const struct hw_sequencer_funcs dcn401_funcs = { .enable_plane = dcn20_enable_plane, .update_dchubp_dpp = dcn20_update_dchubp_dpp, .post_unlock_reset_opp = dcn20_post_unlock_reset_opp, + .get_underflow_debug_data = dcn30_get_underflow_debug_data, }; static const struct hwseq_private_funcs dcn401_private_funcs = { diff --git a/drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer.h b/drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer.h index 9df8030e37f7..1723bbcf2c46 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer.h @@ -47,6 +47,7 @@ struct link_resource; struct dc_dmub_cmd; struct pg_block_update; struct drr_params; +struct dc_underflow_debug_data; struct subvp_pipe_control_lock_fast_params { struct dc *dc; @@ -475,6 +476,9 @@ struct hw_sequencer_funcs { struct dc_state *context); void (*post_unlock_reset_opp)(struct dc *dc, struct pipe_ctx *opp_head); + void (*get_underflow_debug_data)(const struct dc *dc, + struct timing_generator *tg, + struct dc_underflow_debug_data *out_data); }; void color_space_to_black_color( diff --git a/drivers/gpu/drm/amd/display/dc/inc/core_types.h b/drivers/gpu/drm/amd/display/dc/inc/core_types.h index f0d7185153b2..d30f94c35f11 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/core_types.h +++ b/drivers/gpu/drm/amd/display/dc/inc/core_types.h @@ -228,8 +228,7 @@ struct resource_funcs { enum dc_status (*update_dc_state_for_encoder_switch)(struct dc_link *link, struct dc_link_settings *link_setting, uint8_t pipe_count, - struct pipe_ctx *pipes, - struct audio_output *audio_output); + struct pipe_ctx *pipes); }; struct audio_support{ @@ -361,6 +360,8 @@ struct stream_resource { uint8_t gsl_group; struct test_pattern_params test_pattern_params; + + struct audio_output audio_output; }; struct plane_resource { @@ -433,7 +434,7 @@ enum p_state_switch_method { P_STATE_V_ACTIVE, P_STATE_SUB_VP, P_STATE_DRR_SUB_VP, - P_STATE_V_BLANK_SUB_VP + P_STATE_V_BLANK_SUB_VP, }; struct pipe_ctx { @@ -683,6 +684,7 @@ struct replay_context { /* Controller Id used for Dig Fe source select */ enum controller_id controllerId; unsigned int line_time_in_ns; + bool os_request_force_ffu; }; enum dc_replay_enable { diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h b/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h index 52b745667ef7..9bee45b36629 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h @@ -229,6 +229,8 @@ struct hubbub_funcs { void (*program_compbuf_segments)(struct hubbub *hubbub, unsigned compbuf_size_seg, bool safe_to_increase); void (*wait_for_det_update)(struct hubbub *hubbub, int hubp_inst); bool (*program_arbiter)(struct hubbub *hubbub, struct dml2_display_arb_regs *arb_regs, bool safe_to_lower); + void (*get_det_sizes)(struct hubbub *hubbub, uint32_t *curr_det_sizes, uint32_t *target_det_sizes); + uint32_t (*compbuf_config_error)(struct hubbub *hubbub); }; struct hubbub { diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/dpp.h b/drivers/gpu/drm/amd/display/dc/inc/hw/dpp.h index 0c5675d1c593..1b7c085dc2cc 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/dpp.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dpp.h @@ -349,6 +349,9 @@ struct dpp_funcs { struct dpp *dpp_base, enum dc_color_space color_space, struct dc_csc_transform cursor_csc_color_matrix); + + void (*dpp_force_disable_cursor)(struct dpp *dpp_base); + }; diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h b/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h index cee29e89ec5c..2b874d2cc61c 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h @@ -89,7 +89,7 @@ enum hubp_3dlut_fl_addressing_mode { enum hubp_3dlut_fl_width { hubp_3dlut_fl_width_17 = 17, hubp_3dlut_fl_width_33 = 33, - hubp_3dlut_fl_width_transformed = 4916 + hubp_3dlut_fl_width_transformed = 4916, //mpc default }; enum hubp_3dlut_fl_crossbar_bit_slice { @@ -99,6 +99,22 @@ enum hubp_3dlut_fl_crossbar_bit_slice { hubp_3dlut_fl_crossbar_bit_slice_48_63 = 3 }; +struct hubp_fl_3dlut_config { + bool enabled; + enum hubp_3dlut_fl_width width; + enum hubp_3dlut_fl_mode mode; + enum hubp_3dlut_fl_format format; + uint16_t bias; + uint16_t scale; + struct dc_plane_address address; + enum hubp_3dlut_fl_addressing_mode addr_mode; + enum dc_cm2_gpu_mem_layout layout; + uint8_t protection_bits; + enum hubp_3dlut_fl_crossbar_bit_slice crossbar_bit_slice_y_g; + enum hubp_3dlut_fl_crossbar_bit_slice crossbar_bit_slice_cb_b; + enum hubp_3dlut_fl_crossbar_bit_slice crossbar_bit_slice_cr_r; +}; + struct hubp { const struct hubp_funcs *funcs; struct dc_context *ctx; @@ -288,7 +304,10 @@ struct hubp_funcs { enum hubp_3dlut_fl_crossbar_bit_slice bit_slice_cb_b, enum hubp_3dlut_fl_crossbar_bit_slice bit_slice_cr_r); int (*hubp_get_3dlut_fl_done)(struct hubp *hubp); + void (*hubp_program_3dlut_fl_config)(struct hubp *hubp, struct hubp_fl_3dlut_config *cfg); void (*hubp_clear_tiling)(struct hubp *hubp); + uint32_t (*hubp_get_current_read_line)(struct hubp *hubp); + uint32_t (*hubp_get_det_config_error)(struct hubp *hubp); }; #endif diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h b/drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h index 7641439f6ca0..22960ee03dee 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h @@ -115,6 +115,16 @@ enum MCM_LUT_ID { MCM_LUT_SHAPER }; +struct mpc_fl_3dlut_config { + bool enabled; + uint16_t width; + bool select_lut_bank_a; + uint16_t bit_depth; + int hubp_index; + uint16_t bias; + uint16_t scale; +}; + union mcm_lut_params { const struct pwl_params *pwl; const struct tetrahedral_params *lut3d; @@ -1059,21 +1069,6 @@ struct mpc_funcs { */ void (*program_lut_mode)(struct mpc *mpc, const enum MCM_LUT_ID id, const enum MCM_LUT_XABLE xable, bool lut_bank_a, int mpcc_id); - /** - * @program_3dlut_size: - * - * Program 3D LUT size. - * - * Parameters: - * - [in/out] mpc - MPC context. - * - [in] is_17x17x17 - is 3dlut 17x17x17 - * - [in] mpcc_id - * - * Return: - * - * void - */ - void (*program_3dlut_size)(struct mpc *mpc, bool is_17x17x17, int mpcc_id); /** * @mcm: @@ -1098,6 +1093,7 @@ struct mpc_funcs { * MPC RMCM new HW sequential programming functions */ struct { + void (*fl_3dlut_configure)(struct mpc *mpc, struct mpc_fl_3dlut_config *cfg, int mpcc_id); void (*enable_3dlut_fl)(struct mpc *mpc, bool enable, int mpcc_id); void (*update_3dlut_fast_load_select)(struct mpc *mpc, int mpcc_id, int hubp_idx); void (*program_lut_read_write_control)(struct mpc *mpc, const enum MCM_LUT_ID id, diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h b/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h index 267ace4eef8a..f2de2cf23859 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h @@ -374,6 +374,7 @@ struct timing_generator_funcs { void (*wait_drr_doublebuffer_pending_clear)(struct timing_generator *tg); void (*set_long_vtotal)(struct timing_generator *optc, const struct long_vtotal_params *params); void (*wait_odm_doublebuffer_pending_clear)(struct timing_generator *tg); + void (*wait_otg_disable)(struct timing_generator *optc); bool (*get_optc_double_buffer_pending)(struct timing_generator *tg); bool (*get_otg_double_buffer_pending)(struct timing_generator *tg); bool (*get_pipe_update_pending)(struct timing_generator *tg); diff --git a/drivers/gpu/drm/amd/display/dc/inc/link.h b/drivers/gpu/drm/amd/display/dc/inc/link.h index f2503402c10e..0cce49d95e26 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/link.h +++ b/drivers/gpu/drm/amd/display/dc/inc/link.h @@ -218,7 +218,10 @@ struct link_service { bool (*dp_overwrite_extended_receiver_cap)(struct dc_link *link); enum lttpr_mode (*dp_decide_lttpr_mode)(struct dc_link *link, struct dc_link_settings *link_setting); - + uint8_t (*dp_get_lttpr_count)(struct dc_link *link); + void (*edp_get_alpm_support)(struct dc_link *link, + bool *auxless_support, + bool *auxwake_support); /*************************** DP DPIA/PHY ******************************/ void (*dpia_handle_usb4_bandwidth_allocation_for_link)( diff --git a/drivers/gpu/drm/amd/display/dc/inc/soc_and_ip_translator.h b/drivers/gpu/drm/amd/display/dc/inc/soc_and_ip_translator.h new file mode 100644 index 000000000000..23daf98b8aa8 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/inc/soc_and_ip_translator.h @@ -0,0 +1,24 @@ +// SPDX-License-Identifier: MIT +// +// Copyright 2025 Advanced Micro Devices, Inc. + +#ifndef __SOC_AND_IP_TRANSLATOR_H__ +#define __SOC_AND_IP_TRANSLATOR_H__ + +#include "dc.h" +#include "dml_top_soc_parameter_types.h" + +struct soc_and_ip_translator_funcs { + void (*get_soc_bb)(struct dml2_soc_bb *soc_bb, const struct dc *dc, const struct dml2_configuration_options *config); + void (*get_ip_caps)(struct dml2_ip_capabilities *dml_ip_caps); +}; + +struct soc_and_ip_translator { + const struct soc_and_ip_translator_funcs *translator_funcs; +}; + +struct soc_and_ip_translator *dc_create_soc_and_ip_translator(enum dce_version dc_version); +void dc_destroy_soc_and_ip_translator(struct soc_and_ip_translator **soc_and_ip_translator); + + +#endif // __SOC_AND_IP_TRANSLATOR_H__ diff --git a/drivers/gpu/drm/amd/display/dc/link/accessories/link_dp_cts.c b/drivers/gpu/drm/amd/display/dc/link/accessories/link_dp_cts.c index 2956c2b3ad1a..23f41c99fa38 100644 --- a/drivers/gpu/drm/amd/display/dc/link/accessories/link_dp_cts.c +++ b/drivers/gpu/drm/amd/display/dc/link/accessories/link_dp_cts.c @@ -75,7 +75,9 @@ static void dp_retrain_link_dp_test(struct dc_link *link, bool is_hpo_acquired; uint8_t count; int i; - struct audio_output audio_output[MAX_PIPES]; + + struct dc_stream_state *streams_on_link[MAX_PIPES]; + int num_streams_on_link = 0; needs_divider_update = (link->dc->link_srv->dp_get_encoding_format(link_setting) != link->dc->link_srv->dp_get_encoding_format((const struct dc_link_settings *) &link->cur_link_settings)); @@ -99,7 +101,7 @@ static void dp_retrain_link_dp_test(struct dc_link *link, if (needs_divider_update && link->dc->res_pool->funcs->update_dc_state_for_encoder_switch) { link->dc->res_pool->funcs->update_dc_state_for_encoder_switch(link, link_setting, count, - *pipes, &audio_output[0]); + *pipes); for (i = 0; i < count; i++) { pipes[i]->clock_source->funcs->program_pix_clk( pipes[i]->clock_source, @@ -111,15 +113,16 @@ static void dp_retrain_link_dp_test(struct dc_link *link, const struct link_hwss *link_hwss = get_link_hwss( link, &pipes[i]->link_res); - link_hwss->setup_audio_output(pipes[i], &audio_output[i], - pipes[i]->stream_res.audio->inst); + link_hwss->setup_audio_output(pipes[i], + &pipes[i]->stream_res.audio_output, + pipes[i]->stream_res.audio->inst); pipes[i]->stream_res.audio->funcs->az_configure( pipes[i]->stream_res.audio, pipes[i]->stream->signal, - &audio_output[i].crtc_info, + &pipes[i]->stream_res.audio_output.crtc_info, &pipes[i]->stream->audio_info, - &audio_output[i].dp_link_info); + &pipes[i]->stream_res.audio_output.dp_link_info); if (link->dc->config.disable_hbr_audio_dp2 && pipes[i]->stream_res.audio->funcs->az_disable_hbr_audio && @@ -138,12 +141,19 @@ static void dp_retrain_link_dp_test(struct dc_link *link, pipes[i]->stream_res.tg->funcs->enable_crtc(pipes[i]->stream_res.tg); // Set DPMS on with stream update - for (i = 0; i < state->stream_count; i++) - if (state->streams[i] && state->streams[i]->link && state->streams[i]->link == link) { - stream_update.stream = state->streams[i]; + // Cache all streams on current link since dc_update_planes_and_stream might kill current_state + for (i = 0; i < MAX_PIPES; i++) { + if (state->streams[i] && state->streams[i]->link && state->streams[i]->link == link) + streams_on_link[num_streams_on_link++] = state->streams[i]; + } + + for (i = 0; i < num_streams_on_link; i++) { + if (streams_on_link[i] && streams_on_link[i]->link && streams_on_link[i]->link == link) { + stream_update.stream = streams_on_link[i]; stream_update.dpms_off = &dpms_off; - dc_update_planes_and_stream(state->clk_mgr->ctx->dc, NULL, 0, state->streams[i], &stream_update); + dc_update_planes_and_stream(state->clk_mgr->ctx->dc, NULL, 0, streams_on_link[i], &stream_update); } + } } static void dp_test_send_link_training(struct dc_link *link) diff --git a/drivers/gpu/drm/amd/display/dc/link/link_dpms.c b/drivers/gpu/drm/amd/display/dc/link/link_dpms.c index 8c8682f743d6..cb80b4599936 100644 --- a/drivers/gpu/drm/amd/display/dc/link/link_dpms.c +++ b/drivers/gpu/drm/amd/display/dc/link/link_dpms.c @@ -2458,7 +2458,6 @@ void link_set_dpms_on( struct link_encoder *link_enc = pipe_ctx->link_res.dio_link_enc; enum otg_out_mux_dest otg_out_dest = OUT_MUX_DIO; struct vpg *vpg = pipe_ctx->stream_res.stream_enc->vpg; - const struct link_hwss *link_hwss = get_link_hwss(link, &pipe_ctx->link_res); bool apply_edp_fast_boot_optimization = pipe_ctx->stream->apply_edp_fast_boot_optimization; @@ -2502,8 +2501,6 @@ void link_set_dpms_on( pipe_ctx->stream_res.tg->funcs->set_out_mux(pipe_ctx->stream_res.tg, otg_out_dest); } - link_hwss->setup_stream_attribute(pipe_ctx); - pipe_ctx->stream->apply_edp_fast_boot_optimization = false; // Enable VPG before building infoframe diff --git a/drivers/gpu/drm/amd/display/dc/link/link_factory.c b/drivers/gpu/drm/amd/display/dc/link/link_factory.c index de1143dbbd25..31a73867cd4c 100644 --- a/drivers/gpu/drm/amd/display/dc/link/link_factory.c +++ b/drivers/gpu/drm/amd/display/dc/link/link_factory.c @@ -165,6 +165,8 @@ static void construct_link_service_dp_capability(struct link_service *link_srv) link_srv->dp_overwrite_extended_receiver_cap = dp_overwrite_extended_receiver_cap; link_srv->dp_decide_lttpr_mode = dp_decide_lttpr_mode; + link_srv->dp_get_lttpr_count = dp_get_lttpr_count; + link_srv->edp_get_alpm_support = edp_get_alpm_support; } /* link dp phy/dpia implements basic dp phy/dpia functionality such as diff --git a/drivers/gpu/drm/amd/display/dc/link/link_validation.c b/drivers/gpu/drm/amd/display/dc/link/link_validation.c index aecaf37eee35..acdc162de535 100644 --- a/drivers/gpu/drm/amd/display/dc/link/link_validation.c +++ b/drivers/gpu/drm/amd/display/dc/link/link_validation.c @@ -408,8 +408,10 @@ enum dc_status link_validate_dp_tunnel_bandwidth(const struct dc *dc, const stru link = stream->link; if (!(link && (stream->signal == SIGNAL_TYPE_DISPLAY_PORT - || stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST) - && link->hpd_status)) + || stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST))) + continue; + + if ((link->ep_type == DISPLAY_ENDPOINT_USB4_DPIA) && (link->hpd_status == false)) continue; dp_tunnel_settings = get_dp_tunnel_settings(new_ctx, stream); diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c index 651926e547b9..caddb7dfb133 100644 --- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c +++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c @@ -1525,8 +1525,8 @@ bool read_is_mst_supported(struct dc_link *link) return false; } - rev.raw = 0; - cap.raw = 0; + rev.raw = 0; + cap.raw = 0; st = core_link_read_dpcd(link, DP_DPCD_REV, &rev.raw, sizeof(rev)); @@ -2125,13 +2125,13 @@ void detect_edp_sink_caps(struct dc_link *link) &backlight_adj_cap, sizeof(backlight_adj_cap)); link->dpcd_caps.dynamic_backlight_capable_edp = - (backlight_adj_cap & DP_EDP_DYNAMIC_BACKLIGHT_CAP) ? true:false; + (backlight_adj_cap & DP_EDP_DYNAMIC_BACKLIGHT_CAP) ? true : false; core_link_read_dpcd(link, DP_EDP_GENERAL_CAP_1, &general_edp_cap, sizeof(general_edp_cap)); link->dpcd_caps.set_power_state_capable_edp = - (general_edp_cap & DP_EDP_SET_POWER_CAP) ? true:false; + (general_edp_cap & DP_EDP_SET_POWER_CAP) ? true : false; set_default_brightness_aux(link); @@ -2506,3 +2506,40 @@ bool dp_is_sink_present(struct dc_link *link) return present; } + +uint8_t dp_get_lttpr_count(struct dc_link *link) +{ + if (dp_is_lttpr_present(link)) + return dp_parse_lttpr_repeater_count(link->dpcd_caps.lttpr_caps.phy_repeater_cnt); + + return 0; +} + +void edp_get_alpm_support(struct dc_link *link, + bool *auxless_support, + bool *auxwake_support) +{ + bool lttpr_present = dp_is_lttpr_present(link); + + if (auxless_support == NULL || auxwake_support == NULL) + return; + + *auxless_support = false; + *auxwake_support = false; + + if (!dc_is_embedded_signal(link->connector_signal)) + return; + + if (link->dpcd_caps.alpm_caps.bits.AUX_LESS_ALPM_CAP) { + if (lttpr_present) { + if (link->dpcd_caps.lttpr_caps.alpm.bits.AUX_LESS_ALPM_SUPPORTED) + *auxless_support = true; + } else + *auxless_support = true; + } + + if (link->dpcd_caps.alpm_caps.bits.AUX_WAKE_ALPM_CAP) { + if (!lttpr_present) + *auxwake_support = true; + } +} diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.h b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.h index 940b147cc5d4..7170db5a1c13 100644 --- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.h +++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.h @@ -108,4 +108,10 @@ uint32_t link_bw_kbps_from_raw_frl_link_rate_data(uint8_t bw); bool dp_overwrite_extended_receiver_cap(struct dc_link *link); +uint8_t dp_get_lttpr_count(struct dc_link *link); + +void edp_get_alpm_support(struct dc_link *link, + bool *auxless_support, + bool *auxwake_support); + #endif /* __DC_LINK_DP_CAPABILITY_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.c index 819bf2d8ba53..8a3c18ae97a7 100644 --- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.c +++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.c @@ -48,8 +48,7 @@ */ static bool link_dp_is_bw_alloc_available(struct dc_link *link) { - return (link && link->hpd_status - && link->dpcd_caps.usb4_dp_tun_info.dp_tun_cap.bits.dp_tunneling + return (link && link->dpcd_caps.usb4_dp_tun_info.dp_tun_cap.bits.dp_tunneling && link->dpcd_caps.usb4_dp_tun_info.dp_tun_cap.bits.dpia_bw_alloc && link->dpcd_caps.usb4_dp_tun_info.driver_bw_cap.bits.driver_bw_alloc_support); } @@ -226,35 +225,40 @@ bool link_dpia_enable_usb4_dp_bw_alloc_mode(struct dc_link *link) bool ret = false; uint8_t val; - if (link->hpd_status) { - val = DPTX_BW_ALLOC_MODE_ENABLE | DPTX_BW_ALLOC_UNMASK_IRQ; + if (link->dc->debug.dpia_debug.bits.enable_bw_allocation_mode == false) { + DC_LOG_DEBUG("%s: link[%d] DPTX BW allocation mode disabled", __func__, link->link_index); + return false; + } - if (core_link_write_dpcd(link, DPTX_BW_ALLOCATION_MODE_CONTROL, &val, sizeof(uint8_t)) == DC_OK) { - DC_LOG_DEBUG("%s: link[%d] DPTX BW allocation mode enabled", __func__, link->link_index); + val = DPTX_BW_ALLOC_MODE_ENABLE | DPTX_BW_ALLOC_UNMASK_IRQ; - retrieve_usb4_dp_bw_allocation_info(link); + if (core_link_write_dpcd(link, DPTX_BW_ALLOCATION_MODE_CONTROL, &val, sizeof(uint8_t)) == DC_OK) { + DC_LOG_DEBUG("%s: link[%d] DPTX BW allocation mode enabled", __func__, link->link_index); - if (link->dpia_bw_alloc_config.nrd_max_link_rate && link->dpia_bw_alloc_config.nrd_max_lane_count) { - link->reported_link_cap.link_rate = link->dpia_bw_alloc_config.nrd_max_link_rate; - link->reported_link_cap.lane_count = link->dpia_bw_alloc_config.nrd_max_lane_count; - } + retrieve_usb4_dp_bw_allocation_info(link); - link->dpia_bw_alloc_config.bw_alloc_enabled = true; - ret = true; - - if (link->dc->debug.dpia_debug.bits.enable_usb4_bw_zero_alloc_patch) { - /* - * During DP tunnel creation, the CM preallocates BW - * and reduces the estimated BW of other DPIAs. - * The CM releases the preallocation only when the allocation is complete. - * Perform a zero allocation to make the CM release the preallocation - * and correctly update the estimated BW for all DPIAs per host router. - */ - link_dp_dpia_allocate_usb4_bandwidth_for_stream(link, 0); - } - } else - DC_LOG_DEBUG("%s: link[%d] failed to enable DPTX BW allocation mode", __func__, link->link_index); - } + if ( + link->dpia_bw_alloc_config.nrd_max_link_rate + && link->dpia_bw_alloc_config.nrd_max_lane_count) { + link->reported_link_cap.link_rate = link->dpia_bw_alloc_config.nrd_max_link_rate; + link->reported_link_cap.lane_count = link->dpia_bw_alloc_config.nrd_max_lane_count; + } + + link->dpia_bw_alloc_config.bw_alloc_enabled = true; + ret = true; + + if (link->dc->debug.dpia_debug.bits.enable_usb4_bw_zero_alloc_patch) { + /* + * During DP tunnel creation, the CM preallocates BW + * and reduces the estimated BW of other DPIAs. + * The CM releases the preallocation only when the allocation is complete. + * Perform a zero allocation to make the CM release the preallocation + * and correctly update the estimated BW for all DPIAs per host router. + */ + link_dp_dpia_allocate_usb4_bandwidth_for_stream(link, 0); + } + } else + DC_LOG_DEBUG("%s: link[%d] failed to enable DPTX BW allocation mode", __func__, link->link_index); return ret; } @@ -297,15 +301,12 @@ void dpia_handle_usb4_bandwidth_allocation_for_link(struct dc_link *link, int pe { if (link && link->dpcd_caps.usb4_dp_tun_info.dp_tun_cap.bits.dp_tunneling && link->dpia_bw_alloc_config.bw_alloc_enabled) { - //1. Hot Plug - if (link->hpd_status && peak_bw > 0) { + if (peak_bw > 0) { // If DP over USB4 then we need to check BW allocation link->dpia_bw_alloc_config.link_max_bw = peak_bw; link_dpia_send_bw_alloc_request(link, peak_bw); - } - //2. Cold Unplug - else if (!link->hpd_status) + } else dpia_bw_alloc_unplug(link); } } diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.c index 2dc1a660e504..134093ce5a8e 100644 --- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.c +++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.c @@ -1018,7 +1018,12 @@ static enum link_training_result dpcd_exit_training_mode(struct dc_link *link, e { enum dc_status status; uint8_t sink_status = 0; - uint8_t i; + uint32_t i; + uint8_t lttpr_count = dp_parse_lttpr_repeater_count(link->dpcd_caps.lttpr_caps.phy_repeater_cnt); + uint32_t intra_hop_disable_time_ms = (lttpr_count > 0 ? lttpr_count * 300 : 10); + + // Each hop could theoretically take over 256ms (max 128b/132b AUX RD INTERVAL) + // To be safe, allow 300ms per LTTPR and 10ms for no LTTPR case /* clear training pattern set */ status = dpcd_set_training_pattern(link, DP_TRAINING_PATTERN_VIDEOIDLE); @@ -1028,7 +1033,7 @@ static enum link_training_result dpcd_exit_training_mode(struct dc_link *link, e if (encoding == DP_128b_132b_ENCODING) { /* poll for intra-hop disable */ - for (i = 0; i < 10; i++) { + for (i = 0; i < intra_hop_disable_time_ms; i++) { if ((core_link_read_dpcd(link, DP_SINK_STATUS, &sink_status, 1) == DC_OK) && (sink_status & DP_INTRA_HOP_AUX_REPLY_INDICATION) == 0) break; diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c index e7927b8f5ba3..8b7b87b21c2e 100644 --- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c +++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c @@ -870,6 +870,8 @@ bool edp_setup_psr(struct dc_link *link, psr_context->dsc_slice_height = psr_config->dsc_slice_height; + psr_context->os_request_force_ffu = psr_config->os_request_force_ffu; + if (psr) { link->psr_settings.psr_feature_enabled = psr->funcs->psr_copy_settings(psr, link, psr_context, panel_inst); @@ -944,7 +946,7 @@ bool edp_set_replay_allow_active(struct dc_link *link, const bool *allow_active, // TODO: Handle mux change case if force_static is set // If force_static is set, just change the replay_allow_active state directly if (replay != NULL && link->replay_settings.replay_feature_enabled) - replay->funcs->replay_enable(replay, *allow_active, wait, panel_inst, link); + replay->funcs->replay_enable(replay, *allow_active, wait, panel_inst); link->replay_settings.replay_allow_active = *allow_active; } @@ -1029,6 +1031,8 @@ bool edp_setup_replay(struct dc_link *link, const struct dc_stream_state *stream replay_context.line_time_in_ns = lineTimeInNs; + replay_context.os_request_force_ffu = link->replay_settings.config.os_request_force_ffu; + link->replay_settings.replay_feature_enabled = replay->funcs->replay_copy_settings(replay, link, &replay_context, panel_inst); if (link->replay_settings.replay_feature_enabled) { @@ -1042,7 +1046,13 @@ bool edp_setup_replay(struct dc_link *link, const struct dc_stream_state *stream (uint8_t *)&(replay_config.raw), sizeof(uint8_t)); memset(&alpm_config, 0, sizeof(alpm_config)); - alpm_config.bits.ENABLE = 1; + alpm_config.bits.ENABLE = link->replay_settings.config.alpm_mode != DC_ALPM_UNSUPPORTED ? 1 : 0; + + if (link->replay_settings.config.alpm_mode == DC_ALPM_AUXLESS) { + alpm_config.bits.ALPM_MODE_SEL = 1; + alpm_config.bits.ACDS_PERIOD_DURATION = 0; + } + dm_helpers_dp_write_dpcd( link->ctx, link, diff --git a/drivers/gpu/drm/amd/display/dc/mmhubbub/dcn20/dcn20_mmhubbub.c b/drivers/gpu/drm/amd/display/dc/mmhubbub/dcn20/dcn20_mmhubbub.c index 259a98e4ee2c..2a422e223bf2 100644 --- a/drivers/gpu/drm/amd/display/dc/mmhubbub/dcn20/dcn20_mmhubbub.c +++ b/drivers/gpu/drm/amd/display/dc/mmhubbub/dcn20/dcn20_mmhubbub.c @@ -284,7 +284,7 @@ void mcifwb2_dump_frame(struct mcif_wb *mcif_wb, REG_UPDATE(MCIF_WB_BUFMGR_SW_CONTROL, MCIF_WB_BUFMGR_SW_LOCK, 0xf); - memcpy(dest_luma_buffer, luma_buffer, mcif_params->luma_pitch * dest_height); + memcpy(dest_luma_buffer, luma_buffer, (size_t)mcif_params->luma_pitch * dest_height); memcpy(dest_chroma_buffer, chroma_buffer, mcif_params->chroma_pitch * dest_height / 2); REG_UPDATE(MCIF_WB_BUFMGR_SW_CONTROL, MCIF_WB_BUFMGR_SW_LOCK, 0x0); diff --git a/drivers/gpu/drm/amd/display/dc/mpc/dcn401/dcn401_mpc.c b/drivers/gpu/drm/amd/display/dc/mpc/dcn401/dcn401_mpc.c index f3fb3fe13757..e1a0308dee57 100644 --- a/drivers/gpu/drm/amd/display/dc/mpc/dcn401/dcn401_mpc.c +++ b/drivers/gpu/drm/amd/display/dc/mpc/dcn401/dcn401_mpc.c @@ -287,13 +287,6 @@ void mpc401_program_lut_read_write_control(struct mpc *mpc, const enum MCM_LUT_I } } -void mpc401_program_3dlut_size(struct mpc *mpc, bool is_17x17x17, int mpcc_id) -{ - struct dcn401_mpc *mpc401 = TO_DCN401_MPC(mpc); - - REG_UPDATE(MPCC_MCM_3DLUT_MODE[mpcc_id], MPCC_MCM_3DLUT_SIZE, is_17x17x17 ? 0 : 1); -} - void mpc_program_gamut_remap( struct mpc *mpc, unsigned int mpcc_id, @@ -611,7 +604,6 @@ static const struct mpc_funcs dcn401_mpc_funcs = { .populate_lut = mpc401_populate_lut, .program_lut_read_write_control = mpc401_program_lut_read_write_control, .program_lut_mode = mpc401_program_lut_mode, - .program_3dlut_size = mpc401_program_3dlut_size, }; diff --git a/drivers/gpu/drm/amd/display/dc/mpc/dcn401/dcn401_mpc.h b/drivers/gpu/drm/amd/display/dc/mpc/dcn401/dcn401_mpc.h index eb0c68d0b0c7..fdc42f8ab3ff 100644 --- a/drivers/gpu/drm/amd/display/dc/mpc/dcn401/dcn401_mpc.h +++ b/drivers/gpu/drm/amd/display/dc/mpc/dcn401/dcn401_mpc.h @@ -221,11 +221,6 @@ void mpc401_program_lut_read_write_control( bool lut_bank_a, int mpcc_id); -void mpc401_program_3dlut_size( - struct mpc *mpc, - bool is_17x17x17, - int mpcc_id); - void mpc401_set_gamut_remap( struct mpc *mpc, int mpcc_id, diff --git a/drivers/gpu/drm/amd/display/dc/optc/dcn32/dcn32_optc.h b/drivers/gpu/drm/amd/display/dc/optc/dcn32/dcn32_optc.h index d159e3ed3bb3..ead92ad78a23 100644 --- a/drivers/gpu/drm/amd/display/dc/optc/dcn32/dcn32_optc.h +++ b/drivers/gpu/drm/amd/display/dc/optc/dcn32/dcn32_optc.h @@ -62,6 +62,7 @@ SF(OTG0_OTG_CONTROL, OTG_DISABLE_POINT_CNTL, mask_sh),\ SF(OTG0_OTG_CONTROL, OTG_FIELD_NUMBER_CNTL, mask_sh),\ SF(OTG0_OTG_CONTROL, OTG_OUT_MUX, mask_sh),\ + SF(OTG0_OTG_CONTROL, OTG_CURRENT_MASTER_EN_STATE, mask_sh),\ SF(OTG0_OTG_STEREO_CONTROL, OTG_STEREO_EN, mask_sh),\ SF(OTG0_OTG_STEREO_CONTROL, OTG_STEREO_SYNC_OUTPUT_LINE_NUM, mask_sh),\ SF(OTG0_OTG_STEREO_CONTROL, OTG_STEREO_SYNC_OUTPUT_POLARITY, mask_sh),\ diff --git a/drivers/gpu/drm/amd/display/dc/optc/dcn35/dcn35_optc.c b/drivers/gpu/drm/amd/display/dc/optc/dcn35/dcn35_optc.c index 72bff94cb57d..52d5ea98c86b 100644 --- a/drivers/gpu/drm/amd/display/dc/optc/dcn35/dcn35_optc.c +++ b/drivers/gpu/drm/amd/display/dc/optc/dcn35/dcn35_optc.c @@ -162,6 +162,8 @@ static bool optc35_disable_crtc(struct timing_generator *optc) REG_WAIT(OTG_CLOCK_CONTROL, OTG_BUSY, 0, 1, 100000); + REG_WAIT(OTG_CONTROL, OTG_CURRENT_MASTER_EN_STATE, 0, 1, 100000); + optc1_clear_optc_underflow(optc); return true; @@ -428,6 +430,21 @@ static void optc35_set_long_vtotal( } } +static void optc35_wait_otg_disable(struct timing_generator *optc) +{ + struct optc *optc1; + uint32_t is_master_en; + + if (!optc || !optc->ctx) + return; + + optc1 = DCN10TG_FROM_TG(optc); + + REG_GET(OTG_CONTROL, OTG_MASTER_EN, &is_master_en); + if (!is_master_en) + REG_WAIT(OTG_CLOCK_CONTROL, OTG_CURRENT_MASTER_EN_STATE, 0, 1, 100000); +} + static const struct timing_generator_funcs dcn35_tg_funcs = { .validate_timing = optc1_validate_timing, .program_timing = optc1_program_timing, @@ -479,6 +496,7 @@ static const struct timing_generator_funcs dcn35_tg_funcs = { .set_odm_bypass = optc32_set_odm_bypass, .set_odm_combine = optc35_set_odm_combine, .get_optc_source = optc2_get_optc_source, + .wait_otg_disable = optc35_wait_otg_disable, .set_h_timing_div_manual_mode = optc32_set_h_timing_div_manual_mode, .set_out_mux = optc3_set_out_mux, .set_drr_trigger_window = optc3_set_drr_trigger_window, diff --git a/drivers/gpu/drm/amd/display/dc/optc/dcn401/dcn401_optc.c b/drivers/gpu/drm/amd/display/dc/optc/dcn401/dcn401_optc.c index ff79c38287df..5af13706e601 100644 --- a/drivers/gpu/drm/amd/display/dc/optc/dcn401/dcn401_optc.c +++ b/drivers/gpu/drm/amd/display/dc/optc/dcn401/dcn401_optc.c @@ -226,6 +226,11 @@ bool optc401_disable_crtc(struct timing_generator *optc) REG_UPDATE(CONTROL, VTG0_ENABLE, 0); + // wait until CRTC_CURRENT_MASTER_EN_STATE == 0 + REG_WAIT(OTG_CONTROL, + OTG_CURRENT_MASTER_EN_STATE, + 0, 10, 15000); + /* CRTC disabled, so disable clock. */ REG_WAIT(OTG_CLOCK_CONTROL, OTG_BUSY, 0, diff --git a/drivers/gpu/drm/amd/display/dc/resource/dce112/dce112_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dce112/dce112_resource.c index 164ba796f64c..869a8e515fc0 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dce112/dce112_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dce112/dce112_resource.c @@ -1111,12 +1111,12 @@ static void bw_calcs_data_update_from_pplib(struct dc *dc) &clks); dc->bw_vbios->low_yclk = bw_frc_to_fixed( - clks.clocks_in_khz[0] * memory_type_multiplier, 1000); + (int64_t)clks.clocks_in_khz[0] * memory_type_multiplier, 1000); dc->bw_vbios->mid_yclk = bw_frc_to_fixed( - clks.clocks_in_khz[clks.num_levels>>1] * memory_type_multiplier, + (int64_t)clks.clocks_in_khz[clks.num_levels>>1] * memory_type_multiplier, 1000); dc->bw_vbios->high_yclk = bw_frc_to_fixed( - clks.clocks_in_khz[clks.num_levels-1] * memory_type_multiplier, + (int64_t)clks.clocks_in_khz[clks.num_levels-1] * memory_type_multiplier, 1000); return; @@ -1152,12 +1152,12 @@ static void bw_calcs_data_update_from_pplib(struct dc *dc) * YCLK = UMACLK*m_memoryTypeMultiplier */ dc->bw_vbios->low_yclk = bw_frc_to_fixed( - mem_clks.data[0].clocks_in_khz * memory_type_multiplier, 1000); + (int64_t)mem_clks.data[0].clocks_in_khz * memory_type_multiplier, 1000); dc->bw_vbios->mid_yclk = bw_frc_to_fixed( - mem_clks.data[mem_clks.num_levels>>1].clocks_in_khz * memory_type_multiplier, + (int64_t)mem_clks.data[mem_clks.num_levels>>1].clocks_in_khz * memory_type_multiplier, 1000); dc->bw_vbios->high_yclk = bw_frc_to_fixed( - mem_clks.data[mem_clks.num_levels-1].clocks_in_khz * memory_type_multiplier, + (int64_t)mem_clks.data[mem_clks.num_levels-1].clocks_in_khz * memory_type_multiplier, 1000); /* Now notify PPLib/SMU about which Watermarks sets they should select diff --git a/drivers/gpu/drm/amd/display/dc/resource/dce120/dce120_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dce120/dce120_resource.c index eb1e158d3436..2f23cc6df571 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dce120/dce120_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dce120/dce120_resource.c @@ -990,12 +990,12 @@ static void bw_calcs_data_update_from_pplib(struct dc *dc) memory_type_multiplier = MEMORY_TYPE_HBM; dc->bw_vbios->low_yclk = bw_frc_to_fixed( - mem_clks.data[0].clocks_in_khz * memory_type_multiplier, 1000); + (int64_t)mem_clks.data[0].clocks_in_khz * memory_type_multiplier, 1000); dc->bw_vbios->mid_yclk = bw_frc_to_fixed( - mem_clks.data[mem_clks.num_levels>>1].clocks_in_khz * memory_type_multiplier, + (int64_t)mem_clks.data[mem_clks.num_levels>>1].clocks_in_khz * memory_type_multiplier, 1000); dc->bw_vbios->high_yclk = bw_frc_to_fixed( - mem_clks.data[mem_clks.num_levels-1].clocks_in_khz * memory_type_multiplier, + (int64_t)mem_clks.data[mem_clks.num_levels-1].clocks_in_khz * memory_type_multiplier, 1000); /* Now notify PPLib/SMU about which Watermarks sets they should select diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.c index 895349d9ca07..201ed863b69e 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.c @@ -2192,7 +2192,7 @@ void dcn30_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params j = 0; // create the final dcfclk and uclk table while (i < num_dcfclk_sta_targets && j < num_uclk_states && num_states < DC__VOLTAGE_STATES) { - if (dcfclk_sta_targets[i] < optimal_dcfclk_for_uclk[j] && i < num_dcfclk_sta_targets) { + if (dcfclk_sta_targets[i] < optimal_dcfclk_for_uclk[j]) { dcfclk_mhz[num_states] = dcfclk_sta_targets[i]; dram_speed_mts[num_states++] = optimal_uclk_for_dcfclk_sta_targets[i++]; } else { diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.c index 3ed7f50554e2..ca17e5d8fdc2 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.c @@ -2239,8 +2239,7 @@ struct resource_pool *dcn31_create_resource_pool( enum dc_status dcn31_update_dc_state_for_encoder_switch(struct dc_link *link, struct dc_link_settings *link_setting, uint8_t pipe_count, - struct pipe_ctx *pipes, - struct audio_output *audio_output) + struct pipe_ctx *pipes) { struct dc_state *state = link->dc->current_state; int i; @@ -2255,7 +2254,7 @@ enum dc_status dcn31_update_dc_state_for_encoder_switch(struct dc_link *link, // Setup audio if (pipes[i].stream_res.audio != NULL) - build_audio_output(state, &pipes[i], &audio_output[i]); + build_audio_output(state, &pipes[i], &pipes[i].stream_res.audio_output); } #else /* This DCN requires rate divider updates and audio reprogramming to allow DP1<-->DP2 link rate switching, diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.h b/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.h index c32c85ef0ba4..7e8fde65528f 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.h +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.h @@ -69,8 +69,7 @@ unsigned int dcn31_get_det_buffer_size( enum dc_status dcn31_update_dc_state_for_encoder_switch(struct dc_link *link, struct dc_link_settings *link_setting, uint8_t pipe_count, - struct pipe_ctx *pipes, - struct audio_output *audio_output); + struct pipe_ctx *pipes); /*temp: B0 specific before switch to dcn313 headers*/ #ifndef regPHYPLLF_PIXCLK_RESYNC_CNTL diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.h b/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.h index 82f966cf4ed2..20d714596021 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.h +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.h @@ -1141,7 +1141,8 @@ unsigned int dcn32_get_max_hw_cursor_size(const struct dc *dc, SRI_ARR(DCN_SURF1_TTU_CNTL1, HUBPREQ, id), \ SRI_ARR(DCN_CUR0_TTU_CNTL0, HUBPREQ, id), \ SRI_ARR(DCN_CUR0_TTU_CNTL1, HUBPREQ, id), \ - SRI_ARR(HUBP_CLK_CNTL, HUBP, id) + SRI_ARR(HUBP_CLK_CNTL, HUBP, id), \ + SRI_ARR(HUBPRET_READ_LINE_VALUE, HUBPRET, id) #define HUBP_REG_LIST_DCN2_COMMON_RI(id) \ HUBP_REG_LIST_DCN_RI(id), HUBP_REG_LIST_DCN_VM_RI(id), \ SRI_ARR(PREFETCH_SETTINGS, HUBPREQ, id), \ diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.c index b3988e38d0a6..068c123ea8a8 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.c @@ -708,6 +708,7 @@ static const struct dc_debug_options debug_defaults_drv = { }, .use_max_lb = true, .force_disable_subvp = false, + .disable_force_pstate_allow_on_hw_release = false, .exit_idle_opt_for_cursor_updates = true, .using_dml2 = true, .using_dml21 = true, diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.h b/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.h index 2ae6831c31ef..0fc66487d800 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.h +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.h @@ -140,7 +140,8 @@ void dcn401_prepare_mcache_programming(struct dc *dc, struct dc_state *context); SRI_ARR(UCLK_PSTATE_FORCE, HUBPREQ, id), \ HUBP_3DLUT_FL_REG_LIST_DCN401(id), \ SRI_ARR(DCSURF_VIEWPORT_MCACHE_SPLIT_COORDINATE, HUBP, id), \ - SRI_ARR(DCHUBP_MCACHEID_CONFIG, HUBP, id) + SRI_ARR(DCHUBP_MCACHEID_CONFIG, HUBP, id), \ + SRI_ARR(HUBPRET_READ_LINE_VALUE, HUBPRET, id) /* ABM */ #define ABM_DCN401_REG_LIST_RI(id) \ diff --git a/drivers/gpu/drm/amd/display/dc/soc_and_ip_translator/Makefile b/drivers/gpu/drm/amd/display/dc/soc_and_ip_translator/Makefile new file mode 100644 index 000000000000..bc93356a0b5b --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/soc_and_ip_translator/Makefile @@ -0,0 +1,19 @@ +# SPDX-License-Identifier: MIT +# +# Copyright 2025 Advanced Micro Devices, Inc. +# Makefile for bounding box component. +# Floating point required due to nature of bounding box values + +soc_and_ip_translator_ccflags := $(CC_FLAGS_FPU) +soc_and_ip_translator_rcflags := $(CC_FLAGS_NO_FPU) + +CFLAGS_$(AMDDALPATH)/dc/soc_and_ip_translator/dcn401/dcn401_soc_and_ip_translator.o := $(soc_and_ip_translator_ccflags) + +CFLAGS_REMOVE_$(AMDDALPATH)/dc/soc_and_ip_translator/dcn401/dcn401_soc_and_ip_translator.o := $(soc_and_ip_translator_rcflags) + +soc_and_ip_translator := soc_and_ip_translator.o +soc_and_ip_translator += dcn401/dcn401_soc_and_ip_translator.o + +AMD_DAL_soc_and_ip_translator := $(addprefix $(AMDDALPATH)/dc/soc_and_ip_translator/, $(soc_and_ip_translator)) + +AMD_DISPLAY_FILES += $(AMD_DAL_soc_and_ip_translator) diff --git a/drivers/gpu/drm/amd/display/dc/soc_and_ip_translator/dcn401/dcn401_soc_and_ip_translator.c b/drivers/gpu/drm/amd/display/dc/soc_and_ip_translator/dcn401/dcn401_soc_and_ip_translator.c new file mode 100644 index 000000000000..3190c76eb482 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/soc_and_ip_translator/dcn401/dcn401_soc_and_ip_translator.c @@ -0,0 +1,304 @@ +// SPDX-License-Identifier: MIT +// +// Copyright 2025 Advanced Micro Devices, Inc. + +#include "dcn401_soc_and_ip_translator.h" +#include "bounding_boxes/dcn4_soc_bb.h" + +/* soc_and_ip_translator component used to get up-to-date values for bounding box. + * Bounding box values are stored in several locations and locations can vary with DCN revision. + * This component provides an interface to get DCN-specific bounding box values. + */ + +static void get_default_soc_bb(struct dml2_soc_bb *soc_bb) +{ + memcpy(soc_bb, &dml2_socbb_dcn401, sizeof(struct dml2_soc_bb)); + memcpy(&soc_bb->qos_parameters, &dml_dcn4_variant_a_soc_qos_params, sizeof(struct dml2_soc_qos_parameters)); +} + +/* + * DC clock table is obtained from SMU during runtime. + * SMU stands for System Management Unit. It is a power management processor. + * It owns the initialization of dc's clock table and programming of clock values + * based on dc's requests. + * Our clock values in base soc bb is a dummy placeholder. The real clock values + * are retrieved from SMU firmware to dc clock table at runtime. + * This function overrides our dummy placeholder values with real values in dc + * clock table. + */ +static void dcn401_convert_dc_clock_table_to_soc_bb_clock_table( + struct dml2_soc_state_table *dml_clk_table, + const struct clk_bw_params *dc_bw_params, + bool use_clock_dc_limits) +{ + int i; + const struct clk_limit_table *dc_clk_table; + + if (dc_bw_params == NULL) + /* skip if bw params could not be obtained from smu */ + return; + + dc_clk_table = &dc_bw_params->clk_table; + + /* dcfclk */ + if (dc_clk_table->num_entries_per_clk.num_dcfclk_levels) { + dml_clk_table->dcfclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_dcfclk_levels; + for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) { + if (i < dml_clk_table->dcfclk.num_clk_values) { + if (use_clock_dc_limits && dc_bw_params->dc_mode_limit.dcfclk_mhz && + dc_clk_table->entries[i].dcfclk_mhz > dc_bw_params->dc_mode_limit.dcfclk_mhz) { + if (i == 0 || dc_clk_table->entries[i-1].dcfclk_mhz < dc_bw_params->dc_mode_limit.dcfclk_mhz) { + dml_clk_table->dcfclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.dcfclk_mhz * 1000; + dml_clk_table->dcfclk.num_clk_values = i + 1; + } else { + dml_clk_table->dcfclk.clk_values_khz[i] = 0; + dml_clk_table->dcfclk.num_clk_values = i; + } + } else { + dml_clk_table->dcfclk.clk_values_khz[i] = dc_clk_table->entries[i].dcfclk_mhz * 1000; + } + } else { + dml_clk_table->dcfclk.clk_values_khz[i] = 0; + } + } + } + + /* fclk */ + if (dc_clk_table->num_entries_per_clk.num_fclk_levels) { + dml_clk_table->fclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_fclk_levels; + for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) { + if (i < dml_clk_table->fclk.num_clk_values) { + if (use_clock_dc_limits && dc_bw_params->dc_mode_limit.fclk_mhz && + dc_clk_table->entries[i].fclk_mhz > dc_bw_params->dc_mode_limit.fclk_mhz) { + if (i == 0 || dc_clk_table->entries[i-1].fclk_mhz < dc_bw_params->dc_mode_limit.fclk_mhz) { + dml_clk_table->fclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.fclk_mhz * 1000; + dml_clk_table->fclk.num_clk_values = i + 1; + } else { + dml_clk_table->fclk.clk_values_khz[i] = 0; + dml_clk_table->fclk.num_clk_values = i; + } + } else { + dml_clk_table->fclk.clk_values_khz[i] = dc_clk_table->entries[i].fclk_mhz * 1000; + } + } else { + dml_clk_table->fclk.clk_values_khz[i] = 0; + } + } + } + + /* uclk */ + if (dc_clk_table->num_entries_per_clk.num_memclk_levels) { + dml_clk_table->uclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_memclk_levels; + for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) { + if (i < dml_clk_table->uclk.num_clk_values) { + if (use_clock_dc_limits && dc_bw_params->dc_mode_limit.memclk_mhz && + dc_clk_table->entries[i].memclk_mhz > dc_bw_params->dc_mode_limit.memclk_mhz) { + if (i == 0 || dc_clk_table->entries[i-1].memclk_mhz < dc_bw_params->dc_mode_limit.memclk_mhz) { + dml_clk_table->uclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.memclk_mhz * 1000; + dml_clk_table->uclk.num_clk_values = i + 1; + } else { + dml_clk_table->uclk.clk_values_khz[i] = 0; + dml_clk_table->uclk.num_clk_values = i; + } + } else { + dml_clk_table->uclk.clk_values_khz[i] = dc_clk_table->entries[i].memclk_mhz * 1000; + } + } else { + dml_clk_table->uclk.clk_values_khz[i] = 0; + } + } + } + + /* dispclk */ + if (dc_clk_table->num_entries_per_clk.num_dispclk_levels) { + dml_clk_table->dispclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_dispclk_levels; + for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) { + if (i < dml_clk_table->dispclk.num_clk_values) { + if (use_clock_dc_limits && dc_bw_params->dc_mode_limit.dispclk_mhz && + dc_clk_table->entries[i].dispclk_mhz > dc_bw_params->dc_mode_limit.dispclk_mhz) { + if (i == 0 || dc_clk_table->entries[i-1].dispclk_mhz < dc_bw_params->dc_mode_limit.dispclk_mhz) { + dml_clk_table->dispclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.dispclk_mhz * 1000; + dml_clk_table->dispclk.num_clk_values = i + 1; + } else { + dml_clk_table->dispclk.clk_values_khz[i] = 0; + dml_clk_table->dispclk.num_clk_values = i; + } + } else { + dml_clk_table->dispclk.clk_values_khz[i] = dc_clk_table->entries[i].dispclk_mhz * 1000; + } + } else { + dml_clk_table->dispclk.clk_values_khz[i] = 0; + } + } + } + + /* dppclk */ + if (dc_clk_table->num_entries_per_clk.num_dppclk_levels) { + dml_clk_table->dppclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_dppclk_levels; + for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) { + if (i < dml_clk_table->dppclk.num_clk_values) { + if (use_clock_dc_limits && dc_bw_params->dc_mode_limit.dppclk_mhz && + dc_clk_table->entries[i].dppclk_mhz > dc_bw_params->dc_mode_limit.dppclk_mhz) { + if (i == 0 || dc_clk_table->entries[i-1].dppclk_mhz < dc_bw_params->dc_mode_limit.dppclk_mhz) { + dml_clk_table->dppclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.dppclk_mhz * 1000; + dml_clk_table->dppclk.num_clk_values = i + 1; + } else { + dml_clk_table->dppclk.clk_values_khz[i] = 0; + dml_clk_table->dppclk.num_clk_values = i; + } + } else { + dml_clk_table->dppclk.clk_values_khz[i] = dc_clk_table->entries[i].dppclk_mhz * 1000; + } + } else { + dml_clk_table->dppclk.clk_values_khz[i] = 0; + } + } + } + + /* dtbclk */ + if (dc_clk_table->num_entries_per_clk.num_dtbclk_levels) { + dml_clk_table->dtbclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_dtbclk_levels; + for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) { + if (i < dml_clk_table->dtbclk.num_clk_values) { + if (use_clock_dc_limits && dc_bw_params->dc_mode_limit.dtbclk_mhz && + dc_clk_table->entries[i].dtbclk_mhz > dc_bw_params->dc_mode_limit.dtbclk_mhz) { + if (i == 0 || dc_clk_table->entries[i-1].dtbclk_mhz < dc_bw_params->dc_mode_limit.dtbclk_mhz) { + dml_clk_table->dtbclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.dtbclk_mhz * 1000; + dml_clk_table->dtbclk.num_clk_values = i + 1; + } else { + dml_clk_table->dtbclk.clk_values_khz[i] = 0; + dml_clk_table->dtbclk.num_clk_values = i; + } + } else { + dml_clk_table->dtbclk.clk_values_khz[i] = dc_clk_table->entries[i].dtbclk_mhz * 1000; + } + } else { + dml_clk_table->dtbclk.clk_values_khz[i] = 0; + } + } + } + + /* socclk */ + if (dc_clk_table->num_entries_per_clk.num_socclk_levels) { + dml_clk_table->socclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_socclk_levels; + for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) { + if (i < dml_clk_table->socclk.num_clk_values) { + if (use_clock_dc_limits && dc_bw_params->dc_mode_limit.socclk_mhz && + dc_clk_table->entries[i].socclk_mhz > dc_bw_params->dc_mode_limit.socclk_mhz) { + if (i == 0 || dc_clk_table->entries[i-1].socclk_mhz < dc_bw_params->dc_mode_limit.socclk_mhz) { + dml_clk_table->socclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.socclk_mhz * 1000; + dml_clk_table->socclk.num_clk_values = i + 1; + } else { + dml_clk_table->socclk.clk_values_khz[i] = 0; + dml_clk_table->socclk.num_clk_values = i; + } + } else { + dml_clk_table->socclk.clk_values_khz[i] = dc_clk_table->entries[i].socclk_mhz * 1000; + } + } else { + dml_clk_table->socclk.clk_values_khz[i] = 0; + } + } + } + + /* dram config */ + dml_clk_table->dram_config.channel_count = dc_bw_params->num_channels; + dml_clk_table->dram_config.channel_width_bytes = dc_bw_params->dram_channel_width_bytes; +} + +void dcn401_update_soc_bb_with_values_from_clk_mgr(struct dml2_soc_bb *soc_bb, const struct dc *dc, const struct dml2_configuration_options *config) +{ + soc_bb->dprefclk_mhz = dc->clk_mgr->dprefclk_khz / 1000; + soc_bb->dispclk_dppclk_vco_speed_mhz = dc->clk_mgr->dentist_vco_freq_khz / 1000.0; + soc_bb->mall_allocated_for_dcn_mbytes = dc->caps.mall_size_total / (1024 * 1024); + + if (dc->clk_mgr->funcs->is_smu_present && + dc->clk_mgr->funcs->is_smu_present(dc->clk_mgr)) { + dcn401_convert_dc_clock_table_to_soc_bb_clock_table(&soc_bb->clk_table, + dc->clk_mgr->bw_params, + config->use_clock_dc_limits); + } +} + +void dcn401_update_soc_bb_with_values_from_vbios(struct dml2_soc_bb *soc_bb, const struct dc *dc) +{ + soc_bb->dchub_refclk_mhz = dc->res_pool->ref_clocks.dchub_ref_clock_inKhz / 1000; + soc_bb->xtalclk_mhz = dc->ctx->dc_bios->fw_info.pll_info.crystal_frequency / 1000; + + /* latencies in vbios are platform specific and should be used if provided */ + if (dc->ctx->dc_bios->bb_info.dram_clock_change_latency_100ns) + soc_bb->power_management_parameters.dram_clk_change_blackout_us = + dc->ctx->dc_bios->bb_info.dram_clock_change_latency_100ns / 10.0; + + if (dc->ctx->dc_bios->bb_info.dram_sr_enter_exit_latency_100ns) + soc_bb->power_management_parameters.stutter_enter_plus_exit_latency_us = + dc->ctx->dc_bios->bb_info.dram_sr_enter_exit_latency_100ns / 10.0; + + if (dc->ctx->dc_bios->bb_info.dram_sr_exit_latency_100ns) + soc_bb->power_management_parameters.stutter_exit_latency_us = + dc->ctx->dc_bios->bb_info.dram_sr_exit_latency_100ns / 10.0; +} + +void dcn401_update_soc_bb_with_values_from_software_policy(struct dml2_soc_bb *soc_bb, const struct dc *dc) +{ + /* set if the value is provided */ + if (dc->bb_overrides.sr_exit_time_ns) + soc_bb->power_management_parameters.stutter_exit_latency_us = + dc->bb_overrides.sr_exit_time_ns / 1000.0; + + if (dc->bb_overrides.sr_enter_plus_exit_time_ns) + soc_bb->power_management_parameters.stutter_enter_plus_exit_latency_us = + dc->bb_overrides.sr_enter_plus_exit_time_ns / 1000.0; + + if (dc->bb_overrides.dram_clock_change_latency_ns) + soc_bb->power_management_parameters.dram_clk_change_blackout_us = + dc->bb_overrides.dram_clock_change_latency_ns / 1000.0; + + if (dc->bb_overrides.fclk_clock_change_latency_ns) + soc_bb->power_management_parameters.fclk_change_blackout_us = + dc->bb_overrides.fclk_clock_change_latency_ns / 1000.0; + + //Z8 values not expected nor used on DCN401 but still added for completeness + if (dc->bb_overrides.sr_exit_z8_time_ns) + soc_bb->power_management_parameters.z8_stutter_exit_latency_us = + dc->bb_overrides.sr_exit_z8_time_ns / 1000.0; + + if (dc->bb_overrides.sr_enter_plus_exit_z8_time_ns) + soc_bb->power_management_parameters.z8_stutter_enter_plus_exit_latency_us = + dc->bb_overrides.sr_enter_plus_exit_z8_time_ns / 1000.0; +} + +static void apply_soc_bb_updates(struct dml2_soc_bb *soc_bb, const struct dc *dc, const struct dml2_configuration_options *config) +{ + /* Individual modification can be overwritten even if it was obtained by a previous function. + * Modifications are acquired in order of priority (lowest to highest). + */ + dc_assert_fp_enabled(); + + dcn401_update_soc_bb_with_values_from_clk_mgr(soc_bb, dc, config); + dcn401_update_soc_bb_with_values_from_vbios(soc_bb, dc); + dcn401_update_soc_bb_with_values_from_software_policy(soc_bb, dc); +} + +void dcn401_get_soc_bb(struct dml2_soc_bb *soc_bb, const struct dc *dc, const struct dml2_configuration_options *config) +{ + //get default soc_bb with static values + get_default_soc_bb(soc_bb); + //update soc_bb values with more accurate values + apply_soc_bb_updates(soc_bb, dc, config); +} + +static void dcn401_get_ip_caps(struct dml2_ip_capabilities *ip_caps) +{ + *ip_caps = dml2_dcn401_max_ip_caps; +} + +static struct soc_and_ip_translator_funcs dcn401_translator_funcs = { + .get_soc_bb = dcn401_get_soc_bb, + .get_ip_caps = dcn401_get_ip_caps, +}; + +void dcn401_construct_soc_and_ip_translator(struct soc_and_ip_translator *soc_and_ip_translator) +{ + soc_and_ip_translator->translator_funcs = &dcn401_translator_funcs; +} diff --git a/drivers/gpu/drm/amd/display/dc/soc_and_ip_translator/dcn401/dcn401_soc_and_ip_translator.h b/drivers/gpu/drm/amd/display/dc/soc_and_ip_translator/dcn401/dcn401_soc_and_ip_translator.h new file mode 100644 index 000000000000..21d842857601 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/soc_and_ip_translator/dcn401/dcn401_soc_and_ip_translator.h @@ -0,0 +1,22 @@ +// SPDX-License-Identifier: MIT +// +// Copyright 2025 Advanced Micro Devices, Inc. + +#ifndef _DCN401_SOC_AND_IP_TRANSLATOR_H_ +#define _DCN401_SOC_AND_IP_TRANSLATOR_H_ + +#include "core_types.h" +#include "dc.h" +#include "clk_mgr.h" +#include "soc_and_ip_translator.h" +#include "dml2/dml21/inc/dml_top_soc_parameter_types.h" + +void dcn401_construct_soc_and_ip_translator(struct soc_and_ip_translator *soc_and_ip_translator); + +/* Functions that can be re-used by higher DCN revisions of this component */ +void dcn401_get_soc_bb(struct dml2_soc_bb *soc_bb, const struct dc *dc, const struct dml2_configuration_options *config); +void dcn401_update_soc_bb_with_values_from_clk_mgr(struct dml2_soc_bb *soc_bb, const struct dc *dc, const struct dml2_configuration_options *config); +void dcn401_update_soc_bb_with_values_from_vbios(struct dml2_soc_bb *soc_bb, const struct dc *dc); +void dcn401_update_soc_bb_with_values_from_software_policy(struct dml2_soc_bb *soc_bb, const struct dc *dc); + +#endif /* _DCN401_SOC_AND_IP_TRANSLATOR_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/soc_and_ip_translator/dcn42/dcn42_soc_and_ip_translator.c b/drivers/gpu/drm/amd/display/dc/soc_and_ip_translator/dcn42/dcn42_soc_and_ip_translator.c new file mode 100644 index 000000000000..c9e224d262c9 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/soc_and_ip_translator/dcn42/dcn42_soc_and_ip_translator.c @@ -0,0 +1,27 @@ +// SPDX-License-Identifier: MIT +// +// Copyright 2025 Advanced Micro Devices, Inc. + +#include "dcn42_soc_and_ip_translator.h" +#include "soc_and_ip_translator/dcn401/dcn401_soc_and_ip_translator.h" +#include "bounding_boxes/dcn42_soc_bb.h" + +/* soc_and_ip_translator component used to get up-to-date values for bounding box. + * Bounding box values are stored in several locations and locations can vary with DCN revision. + * This component provides an interface to get DCN-specific bounding box values. + */ + +static void dcn42_get_ip_caps(struct dml2_ip_capabilities *ip_caps) +{ + *ip_caps = dml2_dcn42_max_ip_caps; +} + +static struct soc_and_ip_translator_funcs dcn42_translator_funcs = { + .get_soc_bb = dcn401_get_soc_bb, + .get_ip_caps = dcn42_get_ip_caps, +}; + +void dcn42_construct_soc_and_ip_translator(struct soc_and_ip_translator *soc_and_ip_translator) +{ + soc_and_ip_translator->translator_funcs = &dcn42_translator_funcs; +} diff --git a/drivers/gpu/drm/amd/display/dc/soc_and_ip_translator/dcn42/dcn42_soc_and_ip_translator.h b/drivers/gpu/drm/amd/display/dc/soc_and_ip_translator/dcn42/dcn42_soc_and_ip_translator.h new file mode 100644 index 000000000000..914dcbb369a7 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/soc_and_ip_translator/dcn42/dcn42_soc_and_ip_translator.h @@ -0,0 +1,16 @@ +// SPDX-License-Identifier: MIT +// +// Copyright 2025 Advanced Micro Devices, Inc. + +#ifndef _DCN42_SOC_AND_IP_TRANSLATOR_H_ +#define _DCN42_SOC_AND_IP_TRANSLATOR_H_ + +#include "core_types.h" +#include "dc.h" +#include "clk_mgr.h" +#include "dml_top_soc_parameter_types.h" +#include "soc_and_ip_translator.h" + +void dcn42_construct_soc_and_ip_translator(struct soc_and_ip_translator *soc_and_ip_translator); + +#endif /* _DCN42_SOC_AND_IP_TRANSLATOR_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/soc_and_ip_translator/soc_and_ip_translator.c b/drivers/gpu/drm/amd/display/dc/soc_and_ip_translator/soc_and_ip_translator.c new file mode 100644 index 000000000000..0fc0e5a6c171 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/soc_and_ip_translator/soc_and_ip_translator.c @@ -0,0 +1,37 @@ +// SPDX-License-Identifier: MIT +// +// Copyright 2025 Advanced Micro Devices, Inc. + +#include "soc_and_ip_translator.h" +#include "soc_and_ip_translator/dcn401/dcn401_soc_and_ip_translator.h" + +static void dc_construct_soc_and_ip_translator(struct soc_and_ip_translator *soc_and_ip_translator, + enum dce_version dc_version) +{ + switch (dc_version) { + case DCN_VERSION_4_01: + dcn401_construct_soc_and_ip_translator(soc_and_ip_translator); + break; + default: + break; + } +} + +struct soc_and_ip_translator *dc_create_soc_and_ip_translator(enum dce_version dc_version) +{ + struct soc_and_ip_translator *soc_and_ip_translator; + + soc_and_ip_translator = kzalloc(sizeof(*soc_and_ip_translator), GFP_KERNEL); + if (!soc_and_ip_translator) + return NULL; + + dc_construct_soc_and_ip_translator(soc_and_ip_translator, dc_version); + + return soc_and_ip_translator; +} + +void dc_destroy_soc_and_ip_translator(struct soc_and_ip_translator **soc_and_ip_translator) +{ + kfree(*soc_and_ip_translator); + *soc_and_ip_translator = NULL; +} diff --git a/drivers/gpu/drm/amd/display/dc/virtual/virtual_stream_encoder.c b/drivers/gpu/drm/amd/display/dc/virtual/virtual_stream_encoder.c index ad088d70e189..6ffc74fc9dcd 100644 --- a/drivers/gpu/drm/amd/display/dc/virtual/virtual_stream_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/virtual/virtual_stream_encoder.c @@ -44,6 +44,11 @@ static void virtual_stream_encoder_dvi_set_stream_attribute( struct dc_crtc_timing *crtc_timing, bool is_dual_link) {} +static void virtual_stream_encoder_lvds_set_stream_attribute( + struct stream_encoder *enc, + struct dc_crtc_timing *crtc_timing) +{} + static void virtual_stream_encoder_set_throttled_vcp_size( struct stream_encoder *enc, struct fixed31_32 avg_time_slots_per_mtp) @@ -115,6 +120,8 @@ static const struct stream_encoder_funcs virtual_str_enc_funcs = { virtual_stream_encoder_hdmi_set_stream_attribute, .dvi_set_stream_attribute = virtual_stream_encoder_dvi_set_stream_attribute, + .lvds_set_stream_attribute = + virtual_stream_encoder_lvds_set_stream_attribute, .set_throttled_vcp_size = virtual_stream_encoder_set_throttled_vcp_size, .update_hdmi_info_packets = diff --git a/drivers/gpu/drm/amd/display/dmub/dmub_srv.h b/drivers/gpu/drm/amd/display/dmub/dmub_srv.h index 0bafb6710761..338fdc651f2c 100644 --- a/drivers/gpu/drm/amd/display/dmub/dmub_srv.h +++ b/drivers/gpu/drm/amd/display/dmub/dmub_srv.h @@ -316,6 +316,7 @@ struct dmub_srv_hw_params { bool disable_sldo_opt; bool enable_non_transparent_setconfig; bool lower_hbr3_phy_ssc; + bool override_hbr3_pll_vco; }; /** @@ -567,6 +568,7 @@ struct dmub_srv { bool sw_init; bool hw_init; + bool dpia_supported; uint64_t fb_base; uint64_t fb_offset; @@ -597,6 +599,8 @@ struct dmub_notification { enum dmub_notification_type type; uint8_t link_index; uint8_t result; + /* notify instance from DMUB */ + uint8_t instance; bool pending_notification; union { struct aux_reply_data aux_reply; diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h index c587b3441e07..02a4a20e3560 100644 --- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h +++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h @@ -843,7 +843,8 @@ union dmub_fw_boot_options { uint32_t ips_sequential_ono: 1; /**< 1 to enable sequential ONO IPS sequence */ uint32_t disable_sldo_opt: 1; /**< 1 to disable SLDO optimizations */ uint32_t lower_hbr3_phy_ssc: 1; /**< 1 to lower hbr3 phy ssc to 0.125 percent */ - uint32_t reserved : 6; /**< reserved */ + uint32_t override_hbr3_pll_vco: 1; /**< 1 to override the hbr3 pll vco to 0 */ + uint32_t reserved : 5; /**< reserved */ } bits; /**< boot bits */ uint32_t all; /**< 32-bit access to bits */ }; @@ -882,7 +883,7 @@ enum dmub_shared_state_feature_id { /** * struct dmub_shared_state_ips_fw - Firmware signals for IPS. */ - union dmub_shared_state_ips_fw_signals { +union dmub_shared_state_ips_fw_signals { struct { uint32_t ips1_commit : 1; /**< 1 if in IPS1 or IPS0 RCG */ uint32_t ips2_commit : 1; /**< 1 if in IPS2 */ @@ -897,7 +898,7 @@ enum dmub_shared_state_feature_id { /** * struct dmub_shared_state_ips_signals - Firmware signals for IPS. */ - union dmub_shared_state_ips_driver_signals { +union dmub_shared_state_ips_driver_signals { struct { uint32_t allow_pg : 1; /**< 1 if PG is allowed */ uint32_t allow_ips1 : 1; /**< 1 is IPS1 is allowed */ @@ -1990,18 +1991,19 @@ struct dmub_cmd_lsdma_data { struct lsdma_tiled_copy_data { uint32_t src_addr_lo; uint32_t src_addr_hi; + uint32_t dst_addr_lo; uint32_t dst_addr_hi; uint32_t src_x : 16; uint32_t src_y : 16; - uint32_t src_width : 16; - uint32_t src_height : 16; - uint32_t dst_x : 16; uint32_t dst_y : 16; + uint32_t src_width : 16; + uint32_t src_height : 16; + uint32_t dst_width : 16; uint32_t dst_height : 16; @@ -2034,23 +2036,58 @@ struct dmub_cmd_lsdma_data { uint32_t padding : 30; } tiled_copy_data; struct lsdma_linear_copy_data { + uint32_t src_lo; + uint32_t src_hi; + + uint32_t dst_lo; + uint32_t dst_hi; + uint32_t count : 30; uint32_t cache_policy_dst : 2; uint32_t tmz : 1; uint32_t cache_policy_src : 2; uint32_t padding : 29; - + } linear_copy_data; + struct lsdma_linear_sub_window_copy_data { uint32_t src_lo; uint32_t src_hi; + uint32_t dst_lo; uint32_t dst_hi; - } linear_copy_data; + + uint32_t src_x : 16; + uint32_t src_y : 16; + + uint32_t dst_x : 16; + uint32_t dst_y : 16; + + uint32_t rect_x : 16; + uint32_t rect_y : 16; + + uint32_t src_pitch : 16; + uint32_t dst_pitch : 16; + + uint32_t src_slice_pitch; + uint32_t dst_slice_pitch; + + uint32_t tmz : 1; + uint32_t element_size : 3; + uint32_t src_cache_policy : 3; + uint32_t dst_cache_policy : 3; + uint32_t reserved0 : 22; + } linear_sub_window_copy_data; struct lsdma_reg_write_data { uint32_t reg_addr; uint32_t reg_data; } reg_write_data; struct lsdma_pio_copy_data { + uint32_t src_lo; + uint32_t src_hi; + + uint32_t dst_lo; + uint32_t dst_hi; + union { struct { uint32_t byte_count : 26; @@ -2063,12 +2100,11 @@ struct dmub_cmd_lsdma_data { } fields; uint32_t raw; } packet; - uint32_t src_lo; - uint32_t src_hi; - uint32_t dst_lo; - uint32_t dst_hi; } pio_copy_data; struct lsdma_pio_constfill_data { + uint32_t dst_lo; + uint32_t dst_hi; + union { struct { uint32_t byte_count : 26; @@ -2081,14 +2117,12 @@ struct dmub_cmd_lsdma_data { } fields; uint32_t raw; } packet; - uint32_t dst_lo; - uint32_t dst_hi; + uint32_t data; } pio_constfill_data; uint32_t all[14]; } u; - }; struct dmub_rb_cmd_lsdma { @@ -2330,6 +2364,7 @@ struct dmub_cmd_fams2_global_config { union dmub_fams2_global_feature_config features; uint32_t recovery_timeout_us; uint32_t hwfq_flip_programming_delay_us; + uint32_t max_allow_to_target_delta_us; // how early DCN could assert P-State allow compared to the P-State target }; union dmub_cmd_fams2_config { @@ -3986,6 +4021,10 @@ enum dmub_cmd_replay_type { */ DMUB_CMD__REPLAY_DISABLED_ADAPTIVE_SYNC_SDP = 8, /** + * Set version + */ + DMUB_CMD__REPLAY_SET_VERSION = 9, + /** * Set Replay General command. */ DMUB_CMD__REPLAY_SET_GENERAL_CMD = 16, @@ -4015,6 +4054,10 @@ struct dmub_alpm_auxless_data { uint16_t lfps_t1_t2_override_us; short lfps_t1_t2_offset_us; uint8_t lttpr_count; + /* + * Padding to align structure to 4 byte boundary. + */ + uint8_t pad[1]; }; /** @@ -4048,14 +4091,6 @@ struct dmub_cmd_replay_copy_settings_data { */ uint8_t digbe_inst; /** - * @hpo_stream_enc_inst: HPO stream encoder instance - */ - uint8_t hpo_stream_enc_inst; - /** - * @hpo_link_enc_inst: HPO link encoder instance - */ - uint8_t hpo_link_enc_inst; - /** * AUX HW instance. */ uint8_t aux_inst; @@ -4099,13 +4134,77 @@ struct dmub_cmd_replay_copy_settings_data { * Use for AUX-less ALPM LFPS wake operation */ struct dmub_alpm_auxless_data auxless_alpm_data; - + /** + * @hpo_stream_enc_inst: HPO stream encoder instance + */ + uint8_t hpo_stream_enc_inst; + /** + * @hpo_link_enc_inst: HPO link encoder instance + */ + uint8_t hpo_link_enc_inst; /** * @pad: Align structure to 4 byte boundary. */ uint8_t pad[2]; }; + +/** + * Replay versions. + */ +enum replay_version { + /** + * FreeSync Replay + */ + REPLAY_VERSION_FREESYNC_REPLAY = 0, + /** + * Panel Replay + */ + REPLAY_VERSION_PANEL_REPLAY = 1, + /** + * Replay not supported. + */ + REPLAY_VERSION_UNSUPPORTED = 0xFF, +}; + +/** + * Data passed from driver to FW in a DMUB_CMD___SET_REPLAY_VERSION command. + */ +struct dmub_cmd_replay_set_version_data { + /** + * Panel Instance. + * Panel instance to identify which psr_state to use + * Currently the support is only for 0 or 1 + */ + uint8_t panel_inst; + /** + * PSR version that FW should implement. + */ + enum replay_version version; + /** + * PSR control version. + */ + uint8_t cmd_version; + /** + * Explicit padding to 4 byte boundary. + */ + uint8_t pad[2]; +}; + +/** + * Definition of a DMUB_CMD__REPLAY_SET_VERSION command. + */ +struct dmub_rb_cmd_replay_set_version { + /** + * Command header. + */ + struct dmub_cmd_header header; + /** + * Data passed from driver to FW in a DMUB_CMD__REPLAY_SET_VERSION command. + */ + struct dmub_cmd_replay_set_version_data replay_set_version_data; +}; + /** * Definition of a DMUB_CMD__REPLAY_COPY_SETTINGS command. */ @@ -4470,6 +4569,10 @@ union dmub_replay_cmd_set { */ struct dmub_cmd_replay_disabled_adaptive_sync_sdp_data disabled_adaptive_sync_sdp_data; /** + * Definition of DMUB_CMD__REPLAY_SET_VERSION command data. + */ + struct dmub_cmd_replay_set_version_data version_data; + /** * Definition of DMUB_CMD__REPLAY_SET_GENERAL_CMD command data. */ struct dmub_cmd_replay_set_general_cmd_data set_general_cmd_data; @@ -4685,21 +4788,25 @@ enum dmub_cmd_lsdma_type { */ DMUB_CMD__LSDMA_LINEAR_COPY = 1, /** + * LSDMA copies data from source to destination linearly in sub window + */ + DMUB_CMD__LSDMA_LINEAR_SUB_WINDOW_COPY = 2, + /** * Send the tiled-to-tiled copy command */ - DMUB_CMD__LSDMA_TILED_TO_TILED_COPY = 2, + DMUB_CMD__LSDMA_TILED_TO_TILED_COPY = 3, /** * Send the poll reg write command */ - DMUB_CMD__LSDMA_POLL_REG_WRITE = 3, + DMUB_CMD__LSDMA_POLL_REG_WRITE = 4, /** * Send the pio copy command */ - DMUB_CMD__LSDMA_PIO_COPY = 4, + DMUB_CMD__LSDMA_PIO_COPY = 5, /** * Send the pio constfill command */ - DMUB_CMD__LSDMA_PIO_CONSTFILL = 5, + DMUB_CMD__LSDMA_PIO_CONSTFILL = 6, }; struct abm_ace_curve { @@ -5914,6 +6021,9 @@ enum ips_residency_mode { IPS_RESIDENCY__IPS2, IPS_RESIDENCY__IPS1_RCG, IPS_RESIDENCY__IPS1_ONO2_ON, + IPS_RESIDENCY__IPS1_Z8_RETENTION, + IPS_RESIDENCY__PG_ONO_LAST_SEEN_IN_IPS, + IPS_RESIDENCY__PG_ONO_CURRENT_STATE }; #define NUM_IPS_HISTOGRAM_BUCKETS 16 @@ -5927,6 +6037,8 @@ struct dmub_ips_residency_info { uint32_t histogram[NUM_IPS_HISTOGRAM_BUCKETS]; uint64_t total_time_us; uint64_t total_inactive_time_us; + uint32_t ono_pg_state_at_collection; + uint32_t ono_pg_state_last_seen_in_ips; }; /** @@ -6223,6 +6335,10 @@ union dmub_rb_cmd { * Definition of a DMUB_CMD__IDLE_OPT_SET_DC_POWER_STATE command. */ struct dmub_rb_cmd_idle_opt_set_dc_power_state idle_opt_set_dc_power_state; + /** + * Definition of a DMUB_CMD__REPLAY_SET_VERSION command. + */ + struct dmub_rb_cmd_replay_set_version replay_set_version; /* * Definition of a DMUB_CMD__REPLAY_COPY_SETTINGS command. */ @@ -6431,15 +6547,18 @@ static inline bool dmub_rb_full(struct dmub_rb *rb) static inline bool dmub_rb_push_front(struct dmub_rb *rb, const union dmub_rb_cmd *cmd) { - uint64_t volatile *dst = (uint64_t volatile *)((uint8_t *)(rb->base_address) + rb->wrpt); - const uint64_t *src = (const uint64_t *)cmd; + uint8_t *dst = (uint8_t *)(rb->base_address) + rb->wrpt; + const uint8_t *src = (const uint8_t *)cmd; uint8_t i; + if (rb->capacity == 0) + return false; + if (dmub_rb_full(rb)) return false; // copying data - for (i = 0; i < DMUB_RB_CMD_SIZE / sizeof(uint64_t); i++) + for (i = 0; i < DMUB_RB_CMD_SIZE; i++) *dst++ = *src++; rb->wrpt += DMUB_RB_CMD_SIZE; @@ -6464,6 +6583,9 @@ static inline bool dmub_rb_out_push_front(struct dmub_rb *rb, uint8_t *dst = (uint8_t *)(rb->base_address) + rb->wrpt; const uint8_t *src = (const uint8_t *)cmd; + if (rb->capacity == 0) + return false; + if (dmub_rb_full(rb)) return false; @@ -6509,6 +6631,9 @@ static inline void dmub_rb_get_rptr_with_offset(struct dmub_rb *rb, uint32_t num_cmds, uint32_t *next_rptr) { + if (rb->capacity == 0) + return; + *next_rptr = rb->rptr + DMUB_RB_CMD_SIZE * num_cmds; if (*next_rptr >= rb->capacity) @@ -6572,6 +6697,9 @@ static inline bool dmub_rb_out_front(struct dmub_rb *rb, */ static inline bool dmub_rb_pop_front(struct dmub_rb *rb) { + if (rb->capacity == 0) + return false; + if (dmub_rb_empty(rb)) return false; @@ -6596,6 +6724,9 @@ static inline void dmub_rb_flush_pending(const struct dmub_rb *rb) uint32_t rptr = rb->rptr; uint32_t wptr = rb->wrpt; + if (rb->capacity == 0) + return; + while (rptr != wptr) { uint64_t *data = (uint64_t *)((uint8_t *)(rb->base_address) + rptr); uint8_t i; diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c index 3f38db752b84..4777c7203b2c 100644 --- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c @@ -377,6 +377,7 @@ void dmub_dcn31_enable_dmub_boot_options(struct dmub_srv *dmub, const struct dmu boot_options.bits.dpia_hpd_int_enable_supported = params->dpia_hpd_int_enable_supported; boot_options.bits.power_optimization = params->power_optimization; boot_options.bits.lower_hbr3_phy_ssc = params->lower_hbr3_phy_ssc; + boot_options.bits.override_hbr3_pll_vco = params->override_hbr3_pll_vco; boot_options.bits.sel_mux_phy_c_d_phy_f_g = (dmub->asic == DMUB_ASIC_DCN31B) ? 1 : 0; diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn35.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn35.c index 2228d62adc7e..834e5434ccb8 100644 --- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn35.c +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn35.c @@ -400,13 +400,14 @@ union dmub_fw_boot_options dmub_dcn35_get_fw_boot_option(struct dmub_srv *dmub) void dmub_dcn35_enable_dmub_boot_options(struct dmub_srv *dmub, const struct dmub_srv_hw_params *params) { union dmub_fw_boot_options boot_options = {0}; - union dmub_fw_boot_options cur_boot_options = {0}; - cur_boot_options = dmub_dcn35_get_fw_boot_option(dmub); + if (!dmub->dpia_supported) { + dmub->dpia_supported = dmub_dcn35_get_fw_boot_option(dmub).bits.enable_dpia; + } boot_options.bits.z10_disable = params->disable_z10; boot_options.bits.dpia_supported = params->dpia_supported; - boot_options.bits.enable_dpia = cur_boot_options.bits.enable_dpia && !params->disable_dpia; + boot_options.bits.enable_dpia = dmub->dpia_supported && !params->disable_dpia; boot_options.bits.usb4_cm_version = params->usb4_cm_version; boot_options.bits.dpia_hpd_int_enable_supported = params->dpia_hpd_int_enable_supported; boot_options.bits.power_optimization = params->power_optimization; diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_srv_stat.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_srv_stat.c index 567c5b1aeb7a..e7a58b140388 100644 --- a/drivers/gpu/drm/amd/display/dmub/src/dmub_srv_stat.c +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_srv_stat.c @@ -71,7 +71,7 @@ enum dmub_status dmub_srv_stat_get_notification(struct dmub_srv *dmub, switch (cmd.cmd_common.header.type) { case DMUB_OUT_CMD__DP_AUX_REPLY: notify->type = DMUB_NOTIFICATION_AUX_REPLY; - notify->link_index = cmd.dp_aux_reply.control.instance; + notify->instance = cmd.dp_aux_reply.control.instance; notify->result = cmd.dp_aux_reply.control.result; dmub_memcpy((void *)¬ify->aux_reply, (void *)&cmd.dp_aux_reply.reply_data, sizeof(struct aux_reply_data)); @@ -84,17 +84,17 @@ enum dmub_status dmub_srv_stat_get_notification(struct dmub_srv *dmub, notify->type = DMUB_NOTIFICATION_HPD_IRQ; } - notify->link_index = cmd.dp_hpd_notify.hpd_data.instance; + notify->instance = cmd.dp_hpd_notify.hpd_data.instance; notify->result = AUX_RET_SUCCESS; break; case DMUB_OUT_CMD__SET_CONFIG_REPLY: notify->type = DMUB_NOTIFICATION_SET_CONFIG_REPLY; - notify->link_index = cmd.set_config_reply.set_config_reply_control.instance; + notify->instance = cmd.set_config_reply.set_config_reply_control.instance; notify->sc_status = cmd.set_config_reply.set_config_reply_control.status; break; case DMUB_OUT_CMD__DPIA_NOTIFICATION: notify->type = DMUB_NOTIFICATION_DPIA_NOTIFICATION; - notify->link_index = cmd.dpia_notification.payload.header.instance; + notify->instance = cmd.dpia_notification.payload.header.instance; break; case DMUB_OUT_CMD__HPD_SENSE_NOTIFY: notify->type = DMUB_NOTIFICATION_HPD_SENSE_NOTIFY; diff --git a/drivers/gpu/drm/amd/display/include/dal_asic_id.h b/drivers/gpu/drm/amd/display/include/dal_asic_id.h index 5fc29164e4b4..8aea50aa9533 100644 --- a/drivers/gpu/drm/amd/display/include/dal_asic_id.h +++ b/drivers/gpu/drm/amd/display/include/dal_asic_id.h @@ -213,6 +213,11 @@ enum { #endif #define DEVICE_ID_NV_13FE 0x13FE // CYAN_SKILLFISH #define DEVICE_ID_NV_143F 0x143F +#define DEVICE_ID_NV_13F9 0x13F9 +#define DEVICE_ID_NV_13FA 0x13FA +#define DEVICE_ID_NV_13FB 0x13FB +#define DEVICE_ID_NV_13FC 0x13FC +#define DEVICE_ID_NV_13DB 0x13DB #define FAMILY_VGH 144 #define DEVICE_ID_VGH_163F 0x163F #define DEVICE_ID_VGH_1435 0x1435 diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_psp.c b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_psp.c index e58e7b93810b..6b7db8ec9a53 100644 --- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_psp.c +++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_psp.c @@ -260,6 +260,9 @@ enum mod_hdcp_status mod_hdcp_hdcp1_create_session(struct mod_hdcp *hdcp) return MOD_HDCP_STATUS_FAILURE; } + if (!display) + return MOD_HDCP_STATUS_DISPLAY_NOT_FOUND; + hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.context.mem_context.shared_buf; mutex_lock(&psp->hdcp_context.mutex); diff --git a/drivers/gpu/drm/amd/include/atomfirmware.h b/drivers/gpu/drm/amd/include/atomfirmware.h index 5c86423c2e92..3d083010e734 100644 --- a/drivers/gpu/drm/amd/include/atomfirmware.h +++ b/drivers/gpu/drm/amd/include/atomfirmware.h @@ -211,7 +211,7 @@ atom_bios_string = "ATOM" }; */ -#pragma pack(1) /* BIOS data must use byte aligment*/ +#pragma pack(1) /* BIOS data must use byte alignment*/ enum atombios_image_offset{ OFFSET_TO_ATOM_ROM_HEADER_POINTER = 0x00000048, @@ -255,8 +255,8 @@ struct atom_rom_header_v2_2 uint16_t subsystem_vendor_id; uint16_t subsystem_id; uint16_t pci_info_offset; - uint16_t masterhwfunction_offset; //Offest for SW to get all command function offsets, Don't change the position - uint16_t masterdatatable_offset; //Offest for SW to get all data table offsets, Don't change the position + uint16_t masterhwfunction_offset; //Offset for SW to get all command function offsets, Don't change the position + uint16_t masterdatatable_offset; //Offset for SW to get all data table offsets, Don't change the position uint16_t reserved; uint32_t pspdirtableoffset; }; @@ -453,7 +453,7 @@ struct atom_dtd_format uint8_t refreshrate; }; -/* atom_dtd_format.modemiscinfo defintion */ +/* atom_dtd_format.modemiscinfo definition */ enum atom_dtd_format_modemiscinfo{ ATOM_HSYNC_POLARITY = 0x0002, ATOM_VSYNC_POLARITY = 0x0004, @@ -678,7 +678,7 @@ struct lcd_info_v2_1 uint32_t reserved1[8]; }; -/* lcd_info_v2_1.panel_misc defintion */ +/* lcd_info_v2_1.panel_misc definition */ enum atom_lcd_info_panel_misc{ ATOM_PANEL_MISC_FPDI =0x0002, }; @@ -716,7 +716,7 @@ enum atom_gpio_pin_assignment_gpio_id { /* gpio_id pre-define id for multiple usage */ /* GPIO use to control PCIE_VDDC in certain SLT board */ PCIE_VDDC_CONTROL_GPIO_PINID = 56, - /* if PP_AC_DC_SWITCH_GPIO_PINID in Gpio_Pin_LutTable, AC/DC swithing feature is enable */ + /* if PP_AC_DC_SWITCH_GPIO_PINID in Gpio_Pin_LutTable, AC/DC switching feature is enable */ PP_AC_DC_SWITCH_GPIO_PINID = 60, /* VDDC_REGULATOR_VRHOT_GPIO_PINID in Gpio_Pin_LutTable, VRHot feature is enable */ VDDC_VRHOT_GPIO_PINID = 61, @@ -734,7 +734,7 @@ enum atom_gpio_pin_assignment_gpio_id { struct atom_gpio_pin_lut_v2_1 { struct atom_common_table_header table_header; - /*the real number of this included in the structure is calcualted by using the (whole structure size - the header size)/size of atom_gpio_pin_lut */ + /*the real number of this included in the structure is calculated by using the (whole structure size - the header size)/size of atom_gpio_pin_lut */ struct atom_gpio_pin_assignment gpio_pin[]; }; @@ -997,7 +997,7 @@ enum atom_connector_layout_info_mini_type_def { enum atom_display_device_tag_def{ ATOM_DISPLAY_LCD1_SUPPORT = 0x0002, //an embedded display is either an LVDS or eDP signal type of display - ATOM_DISPLAY_LCD2_SUPPORT = 0x0020, //second edp device tag 0x0020 for backward compability + ATOM_DISPLAY_LCD2_SUPPORT = 0x0020, //second edp device tag 0x0020 for backward compatibility ATOM_DISPLAY_DFP1_SUPPORT = 0x0008, ATOM_DISPLAY_DFP2_SUPPORT = 0x0080, ATOM_DISPLAY_DFP3_SUPPORT = 0x0200, @@ -1011,7 +1011,7 @@ struct atom_display_object_path_v2 { uint16_t display_objid; //Connector Object ID or Misc Object ID uint16_t disp_recordoffset; - uint16_t encoderobjid; //first encoder closer to the connector, could be either an external or intenal encoder + uint16_t encoderobjid; //first encoder closer to the connector, could be either an external or internal encoder uint16_t extencoderobjid; //2nd encoder after the first encoder, from the connector point of view; uint16_t encoder_recordoffset; uint16_t extencoder_recordoffset; @@ -1023,7 +1023,7 @@ struct atom_display_object_path_v2 struct atom_display_object_path_v3 { uint16_t display_objid; //Connector Object ID or Misc Object ID uint16_t disp_recordoffset; - uint16_t encoderobjid; //first encoder closer to the connector, could be either an external or intenal encoder + uint16_t encoderobjid; //first encoder closer to the connector, could be either an external or internal encoder uint16_t reserved1; //only on USBC case, otherwise always = 0 uint16_t reserved2; //reserved and always = 0 uint16_t reserved3; //reserved and always = 0 @@ -3547,7 +3547,7 @@ struct atom_voltage_object_header_v4{ enum atom_voltage_object_mode { VOLTAGE_OBJ_GPIO_LUT = 0, //VOLTAGE and GPIO Lookup table ->atom_gpio_voltage_object_v4 - VOLTAGE_OBJ_VR_I2C_INIT_SEQ = 3, //VOLTAGE REGULATOR INIT sequece through I2C -> atom_i2c_voltage_object_v4 + VOLTAGE_OBJ_VR_I2C_INIT_SEQ = 3, //VOLTAGE REGULATOR INIT sequence through I2C -> atom_i2c_voltage_object_v4 VOLTAGE_OBJ_PHASE_LUT = 4, //Set Vregulator Phase lookup table ->atom_gpio_voltage_object_v4 VOLTAGE_OBJ_SVID2 = 7, //Indicate voltage control by SVID2 ->atom_svid2_voltage_object_v4 VOLTAGE_OBJ_EVV = 8, @@ -3585,7 +3585,7 @@ struct atom_gpio_voltage_object_v4 { struct atom_voltage_object_header_v4 header; // voltage mode = VOLTAGE_OBJ_GPIO_LUT or VOLTAGE_OBJ_PHASE_LUT uint8_t gpio_control_id; // default is 0 which indicate control through CG VID mode - uint8_t gpio_entry_num; // indiate the entry numbers of Votlage/Gpio value Look up table + uint8_t gpio_entry_num; // indicate the entry numbers of Votlage/Gpio value Look up table uint8_t phase_delay_us; // phase delay in unit of micro second uint8_t reserved; uint32_t gpio_mask_val; // GPIO Mask value @@ -4507,8 +4507,8 @@ struct amd_acpi_description_header{ struct uefi_acpi_vfct{ struct amd_acpi_description_header sheader; uint8_t tableUUID[16]; //0x24 - uint32_t vbiosimageoffset; //0x34. Offset to the first GOP_VBIOS_CONTENT block from the beginning of the stucture. - uint32_t lib1Imageoffset; //0x38. Offset to the first GOP_LIB1_CONTENT block from the beginning of the stucture. + uint32_t vbiosimageoffset; //0x34. Offset to the first GOP_VBIOS_CONTENT block from the beginning of the structure. + uint32_t lib1Imageoffset; //0x38. Offset to the first GOP_LIB1_CONTENT block from the beginning of the structure. uint32_t reserved[4]; //0x3C }; @@ -4540,7 +4540,7 @@ struct gop_lib1_content { /* *************************************************************************** Scratch Register definitions - Each number below indicates which scratch regiser request, Active and + Each number below indicates which scratch register request, Active and Connect all share the same definitions as display_device_tag defines *************************************************************************** */ diff --git a/drivers/gpu/drm/amd/include/kgd_pp_interface.h b/drivers/gpu/drm/amd/include/kgd_pp_interface.h index e2b1ea7467b0..2f7e4b5bebf3 100644 --- a/drivers/gpu/drm/amd/include/kgd_pp_interface.h +++ b/drivers/gpu/drm/amd/include/kgd_pp_interface.h @@ -30,6 +30,12 @@ extern const struct amdgpu_ip_block_version smu_v12_0_ip_block; extern const struct amdgpu_ip_block_version smu_v13_0_ip_block; extern const struct amdgpu_ip_block_version smu_v14_0_ip_block; +enum smu_temp_metric_type { + SMU_TEMP_METRIC_BASEBOARD, + SMU_TEMP_METRIC_GPUBOARD, + SMU_TEMP_METRIC_MAX, +}; + enum smu_event_type { SMU_EVENT_RESET_COMPLETE = 0, }; @@ -496,6 +502,8 @@ struct amd_pm_funcs { int (*set_df_cstate)(void *handle, enum pp_df_cstate state); int (*set_xgmi_pstate)(void *handle, uint32_t pstate); ssize_t (*get_gpu_metrics)(void *handle, void **table); + ssize_t (*get_temp_metrics)(void *handle, enum smu_temp_metric_type type, void *table); + bool (*temp_metrics_is_supported)(void *handle, enum smu_temp_metric_type type); ssize_t (*get_xcp_metrics)(void *handle, int xcp_id, void *table); ssize_t (*get_pm_metrics)(void *handle, void *pmmetrics, size_t size); int (*set_watermarks_for_clock_ranges)(void *handle, @@ -1595,6 +1603,79 @@ struct amdgpu_pm_metrics { uint8_t data[]; }; +enum amdgpu_vr_temp { + AMDGPU_VDDCR_VDD0_TEMP, + AMDGPU_VDDCR_VDD1_TEMP, + AMDGPU_VDDCR_VDD2_TEMP, + AMDGPU_VDDCR_VDD3_TEMP, + AMDGPU_VDDCR_SOC_A_TEMP, + AMDGPU_VDDCR_SOC_C_TEMP, + AMDGPU_VDDCR_SOCIO_A_TEMP, + AMDGPU_VDDCR_SOCIO_C_TEMP, + AMDGPU_VDD_085_HBM_TEMP, + AMDGPU_VDDCR_11_HBM_B_TEMP, + AMDGPU_VDDCR_11_HBM_D_TEMP, + AMDGPU_VDD_USR_TEMP, + AMDGPU_VDDIO_11_E32_TEMP, + AMDGPU_VR_MAX_TEMP_ENTRIES, +}; + +enum amdgpu_system_temp { + AMDGPU_UBB_FPGA_TEMP, + AMDGPU_UBB_FRONT_TEMP, + AMDGPU_UBB_BACK_TEMP, + AMDGPU_UBB_OAM7_TEMP, + AMDGPU_UBB_IBC_TEMP, + AMDGPU_UBB_UFPGA_TEMP, + AMDGPU_UBB_OAM1_TEMP, + AMDGPU_OAM_0_1_HSC_TEMP, + AMDGPU_OAM_2_3_HSC_TEMP, + AMDGPU_OAM_4_5_HSC_TEMP, + AMDGPU_OAM_6_7_HSC_TEMP, + AMDGPU_UBB_FPGA_0V72_VR_TEMP, + AMDGPU_UBB_FPGA_3V3_VR_TEMP, + AMDGPU_RETIMER_0_1_2_3_1V2_VR_TEMP, + AMDGPU_RETIMER_4_5_6_7_1V2_VR_TEMP, + AMDGPU_RETIMER_0_1_0V9_VR_TEMP, + AMDGPU_RETIMER_4_5_0V9_VR_TEMP, + AMDGPU_RETIMER_2_3_0V9_VR_TEMP, + AMDGPU_RETIMER_6_7_0V9_VR_TEMP, + AMDGPU_OAM_0_1_2_3_3V3_VR_TEMP, + AMDGPU_OAM_4_5_6_7_3V3_VR_TEMP, + AMDGPU_IBC_HSC_TEMP, + AMDGPU_IBC_TEMP, + AMDGPU_SYSTEM_MAX_TEMP_ENTRIES = 32, +}; + +enum amdgpu_node_temp { + AMDGPU_RETIMER_X_TEMP, + AMDGPU_OAM_X_IBC_TEMP, + AMDGPU_OAM_X_IBC_2_TEMP, + AMDGPU_OAM_X_VDD18_VR_TEMP, + AMDGPU_OAM_X_04_HBM_B_VR_TEMP, + AMDGPU_OAM_X_04_HBM_D_VR_TEMP, + AMDGPU_NODE_MAX_TEMP_ENTRIES = 12, +}; + +struct amdgpu_gpuboard_temp_metrics_v1_0 { + struct metrics_table_header common_header; + uint16_t label_version; + uint16_t node_id; + uint64_t accumulation_counter; + /* Encoded temperature in Celcius, 24:31 is sensor id 0:23 is temp value */ + uint32_t node_temp[AMDGPU_NODE_MAX_TEMP_ENTRIES]; + uint32_t vr_temp[AMDGPU_VR_MAX_TEMP_ENTRIES]; +}; + +struct amdgpu_baseboard_temp_metrics_v1_0 { + struct metrics_table_header common_header; + uint16_t label_version; + uint16_t node_id; + uint64_t accumulation_counter; + /* Encoded temperature in Celcius, 24:31 is sensor id 0:23 is temp value */ + uint32_t system_temp[AMDGPU_SYSTEM_MAX_TEMP_ENTRIES]; +}; + struct amdgpu_partition_metrics_v1_0 { struct metrics_table_header common_header; /* Current clocks (Mhz) */ diff --git a/drivers/gpu/drm/amd/include/mes_v12_api_def.h b/drivers/gpu/drm/amd/include/mes_v12_api_def.h index d85ffab2aff9..c04bd351b250 100644 --- a/drivers/gpu/drm/amd/include/mes_v12_api_def.h +++ b/drivers/gpu/drm/amd/include/mes_v12_api_def.h @@ -66,6 +66,7 @@ enum MES_SCH_API_OPCODE { MES_SCH_API_SET_SE_MODE = 17, MES_SCH_API_SET_GANG_SUBMIT = 18, MES_SCH_API_SET_HW_RSRC_1 = 19, + MES_SCH_API_INV_TLBS = 20, MES_SCH_API_MAX = 0xFF }; @@ -870,6 +871,35 @@ union MESAPI__SET_GANG_SUBMIT { uint32_t max_dwords_in_api[API_FRAME_SIZE_IN_DWORDS]; }; +/* + * @inv_sel 0-select pasid as input to do the invalidation , 1-select vmid + * @flush_type 0-old style, 1-light weight, 2-heavyweight, 3-heavyweight2 + * @inv_sel_id specific pasid when inv_sel is 0 and specific vmid if inv_sel is 1 + * @hub_id 0-gc_hub, 1-mm_hub + */ +struct INV_TLBS { + uint8_t inv_sel; + uint8_t flush_type; + uint16_t inv_sel_id; + uint32_t hub_id; + /* If following two inv_range setting are all 0 , whole VM will be invalidated, + * otherwise only required range be invalidated + */ + uint64_t inv_range_va_start; + uint64_t inv_range_size; + uint64_t reserved; +}; + +union MESAPI__INV_TLBS { + struct { + union MES_API_HEADER header; + struct MES_API_STATUS api_status; + struct INV_TLBS invalidate_tlbs; + }; + + uint32_t max_dwords_in_api[API_FRAME_SIZE_IN_DWORDS]; +}; + #pragma pack(pop) #endif diff --git a/drivers/gpu/drm/amd/pm/amdgpu_dpm.c b/drivers/gpu/drm/amd/pm/amdgpu_dpm.c index 71d986dd7a6e..518d07afc7df 100644 --- a/drivers/gpu/drm/amd/pm/amdgpu_dpm.c +++ b/drivers/gpu/drm/amd/pm/amdgpu_dpm.c @@ -764,10 +764,6 @@ int amdgpu_dpm_send_rma_reason(struct amdgpu_device *adev) ret = smu_send_rma_reason(smu); mutex_unlock(&adev->pm.mutex); - if (adev->cper.enabled) - if (amdgpu_cper_generate_bp_threshold_record(adev)) - dev_warn(adev->dev, "fail to generate bad page threshold cper records\n"); - return ret; } @@ -824,6 +820,21 @@ int amdgpu_dpm_reset_vcn(struct amdgpu_device *adev, uint32_t inst_mask) return ret; } +bool amdgpu_dpm_reset_vcn_is_supported(struct amdgpu_device *adev) +{ + struct smu_context *smu = adev->powerplay.pp_handle; + bool ret; + + if (!is_support_sw_smu(adev)) + return false; + + mutex_lock(&adev->pm.mutex); + ret = smu_reset_vcn_is_supported(smu); + mutex_unlock(&adev->pm.mutex); + + return ret; +} + int amdgpu_dpm_get_dpm_freq_range(struct amdgpu_device *adev, enum pp_clock_type type, uint32_t *min, @@ -2038,6 +2049,66 @@ int amdgpu_dpm_get_dpm_clock_table(struct amdgpu_device *adev, } /** + * amdgpu_dpm_get_temp_metrics - Retrieve metrics for a specific compute + * partition + * @adev: Pointer to the device. + * @type: Identifier for the temperature type metrics to be fetched. + * @table: Pointer to a buffer where the metrics will be stored. If NULL, the + * function returns the size of the metrics structure. + * + * This function retrieves metrics for a specific temperature type, If the + * table parameter is NULL, the function returns the size of the metrics + * structure without populating it. + * + * Return: Size of the metrics structure on success, or a negative error code on failure. + */ +ssize_t amdgpu_dpm_get_temp_metrics(struct amdgpu_device *adev, + enum smu_temp_metric_type type, void *table) +{ + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + int ret; + + if (!pp_funcs->get_temp_metrics || + !amdgpu_dpm_is_temp_metrics_supported(adev, type)) + return -EOPNOTSUPP; + + mutex_lock(&adev->pm.mutex); + ret = pp_funcs->get_temp_metrics(adev->powerplay.pp_handle, type, table); + mutex_unlock(&adev->pm.mutex); + + return ret; +} + +/** + * amdgpu_dpm_is_temp_metrics_supported - Return if specific temperature metrics support + * is available + * @adev: Pointer to the device. + * @type: Identifier for the temperature type metrics to be fetched. + * + * This function returns metrics if specific temperature metrics type is supported or not. + * + * Return: True in case of metrics type supported else false. + */ +bool amdgpu_dpm_is_temp_metrics_supported(struct amdgpu_device *adev, + enum smu_temp_metric_type type) +{ + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + bool support_temp_metrics = false; + + if (!pp_funcs->temp_metrics_is_supported) + return support_temp_metrics; + + if (is_support_sw_smu(adev)) { + mutex_lock(&adev->pm.mutex); + support_temp_metrics = + pp_funcs->temp_metrics_is_supported(adev->powerplay.pp_handle, type); + mutex_unlock(&adev->pm.mutex); + } + + return support_temp_metrics; +} + +/** * amdgpu_dpm_get_xcp_metrics - Retrieve metrics for a specific compute * partition * @adev: Pointer to the device. diff --git a/drivers/gpu/drm/amd/pm/amdgpu_pm.c b/drivers/gpu/drm/amd/pm/amdgpu_pm.c index 4b64851fdb42..5230276628a3 100644 --- a/drivers/gpu/drm/amd/pm/amdgpu_pm.c +++ b/drivers/gpu/drm/amd/pm/amdgpu_pm.c @@ -2073,6 +2073,134 @@ static int pp_dpm_clk_default_attr_update(struct amdgpu_device *adev, struct amd return 0; } +/** + * DOC: board + * + * Certain SOCs can support various board attributes reporting. This is useful + * for user application to monitor various board reated attributes. + * + * The amdgpu driver provides a sysfs API for reporting board attributes. Presently, + * only two types of attributes are reported, baseboard temperature and + * gpu board temperature. Both of them are reported as binary files. + * + * * .. code-block:: console + * + * hexdump /sys/bus/pci/devices/.../board/baseboard_temp + * + * hexdump /sys/bus/pci/devices/.../board/gpuboard_temp + * + */ + +/** + * DOC: baseboard_temp + * + * The amdgpu driver provides a sysfs API for retrieving current baseboard + * temperature metrics data. The file baseboard_temp is used for this. + * Reading the file will dump all the current baseboard temperature metrics data. + */ +static ssize_t amdgpu_get_baseboard_temp_metrics(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct drm_device *ddev = dev_get_drvdata(dev); + struct amdgpu_device *adev = drm_to_adev(ddev); + ssize_t size; + int ret; + + ret = amdgpu_pm_get_access_if_active(adev); + if (ret) + return ret; + + size = amdgpu_dpm_get_temp_metrics(adev, SMU_TEMP_METRIC_BASEBOARD, NULL); + if (size <= 0) + goto out; + if (size >= PAGE_SIZE) { + ret = -ENOSPC; + goto out; + } + + amdgpu_dpm_get_temp_metrics(adev, SMU_TEMP_METRIC_BASEBOARD, buf); + +out: + amdgpu_pm_put_access(adev); + + if (ret) + return ret; + + return size; +} + +/** + * DOC: gpuboard_temp + * + * The amdgpu driver provides a sysfs API for retrieving current gpuboard + * temperature metrics data. The file gpuboard_temp is used for this. + * Reading the file will dump all the current gpuboard temperature metrics data. + */ +static ssize_t amdgpu_get_gpuboard_temp_metrics(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct drm_device *ddev = dev_get_drvdata(dev); + struct amdgpu_device *adev = drm_to_adev(ddev); + ssize_t size; + int ret; + + ret = amdgpu_pm_get_access_if_active(adev); + if (ret) + return ret; + + size = amdgpu_dpm_get_temp_metrics(adev, SMU_TEMP_METRIC_GPUBOARD, NULL); + if (size <= 0) + goto out; + if (size >= PAGE_SIZE) { + ret = -ENOSPC; + goto out; + } + + amdgpu_dpm_get_temp_metrics(adev, SMU_TEMP_METRIC_GPUBOARD, buf); + +out: + amdgpu_pm_put_access(adev); + + if (ret) + return ret; + + return size; +} + +static DEVICE_ATTR(baseboard_temp, 0444, amdgpu_get_baseboard_temp_metrics, NULL); +static DEVICE_ATTR(gpuboard_temp, 0444, amdgpu_get_gpuboard_temp_metrics, NULL); + +static struct attribute *board_attrs[] = { + &dev_attr_baseboard_temp.attr, + &dev_attr_gpuboard_temp.attr, + NULL +}; + +static umode_t amdgpu_board_attr_visible(struct kobject *kobj, struct attribute *attr, int n) +{ + struct device *dev = kobj_to_dev(kobj); + struct drm_device *ddev = dev_get_drvdata(dev); + struct amdgpu_device *adev = drm_to_adev(ddev); + + if (attr == &dev_attr_baseboard_temp.attr) { + if (!amdgpu_dpm_is_temp_metrics_supported(adev, SMU_TEMP_METRIC_BASEBOARD)) + return 0; + } + + if (attr == &dev_attr_gpuboard_temp.attr) { + if (!amdgpu_dpm_is_temp_metrics_supported(adev, SMU_TEMP_METRIC_GPUBOARD)) + return 0; + } + + return attr->mode; +} + +const struct attribute_group amdgpu_board_attr_group = { + .name = "board", + .attrs = board_attrs, + .is_visible = amdgpu_board_attr_visible, +}; + /* pm policy attributes */ struct amdgpu_pm_policy_attr { struct device_attribute dev_attr; @@ -3458,14 +3586,16 @@ static umode_t hwmon_attributes_visible(struct kobject *kobj, effective_mode &= ~S_IWUSR; /* not implemented yet for APUs other than GC 10.3.1 (vangogh) and 9.4.3 */ - if (((adev->family == AMDGPU_FAMILY_SI) || - ((adev->flags & AMD_IS_APU) && (gc_ver != IP_VERSION(10, 3, 1)) && - (gc_ver != IP_VERSION(9, 4, 3) && gc_ver != IP_VERSION(9, 4, 4)))) && - (attr == &sensor_dev_attr_power1_cap_max.dev_attr.attr || - attr == &sensor_dev_attr_power1_cap_min.dev_attr.attr || - attr == &sensor_dev_attr_power1_cap.dev_attr.attr || - attr == &sensor_dev_attr_power1_cap_default.dev_attr.attr)) - return 0; + if (attr == &sensor_dev_attr_power1_cap_max.dev_attr.attr || + attr == &sensor_dev_attr_power1_cap_min.dev_attr.attr || + attr == &sensor_dev_attr_power1_cap.dev_attr.attr || + attr == &sensor_dev_attr_power1_cap_default.dev_attr.attr) { + if (adev->family == AMDGPU_FAMILY_SI || + ((adev->flags & AMD_IS_APU) && gc_ver != IP_VERSION(10, 3, 1) && + (gc_ver != IP_VERSION(9, 4, 3) && gc_ver != IP_VERSION(9, 4, 4))) || + (amdgpu_sriov_vf(adev) && gc_ver == IP_VERSION(11, 0, 3))) + return 0; + } /* not implemented yet for APUs having < GC 9.3.0 (Renoir) */ if (((adev->family == AMDGPU_FAMILY_SI) || @@ -4461,6 +4591,13 @@ int amdgpu_pm_sysfs_init(struct amdgpu_device *adev) goto err_out0; } + if (amdgpu_dpm_is_temp_metrics_supported(adev, SMU_TEMP_METRIC_GPUBOARD)) { + ret = devm_device_add_group(adev->dev, + &amdgpu_board_attr_group); + if (ret) + goto err_out0; + } + adev->pm.sysfs_initialized = true; return 0; diff --git a/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm.h b/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm.h index 768317ee1486..9748744133d9 100644 --- a/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm.h +++ b/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm.h @@ -526,6 +526,8 @@ int amdgpu_dpm_set_power_profile_mode(struct amdgpu_device *adev, int amdgpu_dpm_get_gpu_metrics(struct amdgpu_device *adev, void **table); ssize_t amdgpu_dpm_get_xcp_metrics(struct amdgpu_device *adev, int xcp_id, void *table); +ssize_t amdgpu_dpm_get_temp_metrics(struct amdgpu_device *adev, + enum smu_temp_metric_type type, void *table); /** * @get_pm_metrics: Get one snapshot of power management metrics from PMFW. The @@ -613,5 +615,8 @@ ssize_t amdgpu_dpm_get_pm_policy_info(struct amdgpu_device *adev, int amdgpu_dpm_reset_sdma(struct amdgpu_device *adev, uint32_t inst_mask); bool amdgpu_dpm_reset_sdma_is_supported(struct amdgpu_device *adev); int amdgpu_dpm_reset_vcn(struct amdgpu_device *adev, uint32_t inst_mask); +bool amdgpu_dpm_reset_vcn_is_supported(struct amdgpu_device *adev); +bool amdgpu_dpm_is_temp_metrics_supported(struct amdgpu_device *adev, + enum smu_temp_metric_type type); #endif diff --git a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c index b47cb4a5f488..c5965924e7c6 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c +++ b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c @@ -766,6 +766,7 @@ static int smu_set_funcs(struct amdgpu_device *adev) case IP_VERSION(13, 0, 14): case IP_VERSION(13, 0, 12): smu_v13_0_6_set_ppt_funcs(smu); + smu_v13_0_6_set_temp_funcs(smu); /* Enable pp_od_clk_voltage node */ smu->od_enabled = true; break; @@ -3831,6 +3832,51 @@ int smu_set_pm_policy(struct smu_context *smu, enum pp_pm_policy p_type, return ret; } +static ssize_t smu_sys_get_temp_metrics(void *handle, enum smu_temp_metric_type type, void *table) +{ + struct smu_context *smu = handle; + struct smu_table_context *smu_table = &smu->smu_table; + struct smu_table *tables = smu_table->tables; + enum smu_table_id table_id; + + if (!smu->pm_enabled || !smu->adev->pm.dpm_enabled) + return -EOPNOTSUPP; + + if (!smu->smu_temp.temp_funcs || !smu->smu_temp.temp_funcs->get_temp_metrics) + return -EOPNOTSUPP; + + table_id = smu_metrics_get_temp_table_id(type); + + if (table_id == SMU_TABLE_COUNT) + return -EINVAL; + + /* If the request is to get size alone, return the cached table size */ + if (!table && tables[table_id].cache.size) + return tables[table_id].cache.size; + + if (smu_table_cache_is_valid(&tables[table_id])) { + memcpy(table, tables[table_id].cache.buffer, + tables[table_id].cache.size); + return tables[table_id].cache.size; + } + + return smu->smu_temp.temp_funcs->get_temp_metrics(smu, type, table); +} + +static bool smu_temp_metrics_is_supported(void *handle, enum smu_temp_metric_type type) +{ + struct smu_context *smu = handle; + bool ret = false; + + if (!smu->pm_enabled) + return false; + + if (smu->smu_temp.temp_funcs && smu->smu_temp.temp_funcs->temp_metrics_is_supported) + ret = smu->smu_temp.temp_funcs->temp_metrics_is_supported(smu, type); + + return ret; +} + static ssize_t smu_sys_get_xcp_metrics(void *handle, int xcp_id, void *table) { struct smu_context *smu = handle; @@ -3903,6 +3949,8 @@ static const struct amd_pm_funcs swsmu_pm_funcs = { .get_dpm_clock_table = smu_get_dpm_clock_table, .get_smu_prv_buf_details = smu_get_prv_buffer_details, .get_xcp_metrics = smu_sys_get_xcp_metrics, + .get_temp_metrics = smu_sys_get_temp_metrics, + .temp_metrics_is_supported = smu_temp_metrics_is_supported, }; int smu_wait_for_event(struct smu_context *smu, enum smu_event_type event, @@ -4076,6 +4124,16 @@ int smu_reset_sdma(struct smu_context *smu, uint32_t inst_mask) return ret; } +bool smu_reset_vcn_is_supported(struct smu_context *smu) +{ + bool ret = false; + + if (smu->ppt_funcs && smu->ppt_funcs->reset_vcn_is_supported) + ret = smu->ppt_funcs->reset_vcn_is_supported(smu); + + return ret; +} + int smu_reset_vcn(struct smu_context *smu, uint32_t inst_mask) { if (smu->ppt_funcs && smu->ppt_funcs->dpm_reset_vcn) diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h b/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h index b52e194397e2..5dd49eca598d 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h @@ -249,6 +249,14 @@ struct smu_user_dpm_profile { tables[table_id].domain = d; \ } while (0) +struct smu_table_cache { + void *buffer; + size_t size; + /* interval in ms*/ + uint32_t interval; + unsigned long last_cache_time; +}; + struct smu_table { uint64_t size; uint32_t align; @@ -257,6 +265,7 @@ struct smu_table { void *cpu_addr; struct amdgpu_bo *bo; uint32_t version; + struct smu_table_cache cache; }; enum smu_perf_level_designation { @@ -322,6 +331,8 @@ enum smu_table_id { SMU_TABLE_ECCINFO, SMU_TABLE_COMBO_PPTABLE, SMU_TABLE_WIFIBAND, + SMU_TABLE_GPUBOARD_TEMP_METRICS, + SMU_TABLE_BASEBOARD_TEMP_METRICS, SMU_TABLE_COUNT, }; @@ -396,6 +407,10 @@ struct smu_dpm_context { struct smu_dpm_policy_ctxt *dpm_policies; }; +struct smu_temp_context { + const struct smu_temp_funcs *temp_funcs; +}; + struct smu_power_gate { bool uvd_gated; bool vce_gated; @@ -529,6 +544,7 @@ struct smu_context { struct smu_table_context smu_table; struct smu_dpm_context smu_dpm; struct smu_power_context smu_power; + struct smu_temp_context smu_temp; struct smu_feature smu_feature; struct amd_pp_display_configuration *display_config; struct smu_baco_context smu_baco; @@ -624,6 +640,28 @@ struct smu_context { struct i2c_adapter; /** + * struct smu_temp_funcs - Callbacks used to get temperature data. + */ +struct smu_temp_funcs { + /** + * @get_temp_metrics: Calibrate voltage/frequency curve to fit the system's + * power delivery and voltage margins. Required for adaptive + * @type Temperature metrics type(baseboard/gpuboard) + * Return: Size of &table + */ + ssize_t (*get_temp_metrics)(struct smu_context *smu, + enum smu_temp_metric_type type, void *table); + + /** + * @temp_metrics_is_support: Get if specific temperature metrics is supported + * @type Temperature metrics type(baseboard/gpuboard) + * Return: true if supported else false + */ + bool (*temp_metrics_is_supported)(struct smu_context *smu, enum smu_temp_metric_type type); + +}; + +/** * struct pptable_funcs - Callbacks used to interact with the SMU. */ struct pptable_funcs { @@ -1397,6 +1435,10 @@ struct pptable_funcs { * @reset_vcn: message SMU to soft reset vcn instance. */ int (*dpm_reset_vcn)(struct smu_context *smu, uint32_t inst_mask); + /** + * @reset_vcn_is_supported: Check if support resets vcn. + */ + bool (*reset_vcn_is_supported)(struct smu_context *smu); /** * @get_ecc_table: message SMU to get ECC INFO table. @@ -1622,6 +1664,71 @@ typedef struct { struct smu_dpm_policy *smu_get_pm_policy(struct smu_context *smu, enum pp_pm_policy p_type); +static inline enum smu_table_id +smu_metrics_get_temp_table_id(enum smu_temp_metric_type type) +{ + switch (type) { + case SMU_TEMP_METRIC_BASEBOARD: + return SMU_TABLE_BASEBOARD_TEMP_METRICS; + case SMU_TEMP_METRIC_GPUBOARD: + return SMU_TABLE_GPUBOARD_TEMP_METRICS; + default: + return SMU_TABLE_COUNT; + } + + return SMU_TABLE_COUNT; +} + +static inline void smu_table_cache_update_time(struct smu_table *table, + unsigned long time) +{ + table->cache.last_cache_time = time; +} + +static inline bool smu_table_cache_is_valid(struct smu_table *table) +{ + if (!table->cache.buffer || !table->cache.last_cache_time || + !table->cache.interval || !table->cache.size || + time_after(jiffies, + table->cache.last_cache_time + + msecs_to_jiffies(table->cache.interval))) + return false; + + return true; +} + +static inline int smu_table_cache_init(struct smu_context *smu, + enum smu_table_id table_id, size_t size, + uint32_t cache_interval) +{ + struct smu_table_context *smu_table = &smu->smu_table; + struct smu_table *tables = smu_table->tables; + + tables[table_id].cache.buffer = kzalloc(size, GFP_KERNEL); + if (!tables[table_id].cache.buffer) + return -ENOMEM; + + tables[table_id].cache.last_cache_time = 0; + tables[table_id].cache.interval = cache_interval; + tables[table_id].cache.size = size; + + return 0; +} + +static inline void smu_table_cache_fini(struct smu_context *smu, + enum smu_table_id table_id) +{ + struct smu_table_context *smu_table = &smu->smu_table; + struct smu_table *tables = smu_table->tables; + + if (tables[table_id].cache.buffer) { + kfree(tables[table_id].cache.buffer); + tables[table_id].cache.buffer = NULL; + tables[table_id].cache.last_cache_time = 0; + tables[table_id].cache.interval = 0; + } +} + #if !defined(SWSMU_CODE_LAYER_L2) && !defined(SWSMU_CODE_LAYER_L3) && !defined(SWSMU_CODE_LAYER_L4) int smu_get_power_limit(void *handle, uint32_t *limit, @@ -1673,6 +1780,7 @@ int smu_send_rma_reason(struct smu_context *smu); int smu_reset_sdma(struct smu_context *smu, uint32_t inst_mask); bool smu_reset_sdma_is_supported(struct smu_context *smu); int smu_reset_vcn(struct smu_context *smu, uint32_t inst_mask); +bool smu_reset_vcn_is_supported(struct smu_context *smu); int smu_set_pm_policy(struct smu_context *smu, enum pp_pm_policy p_type, int level); ssize_t smu_get_pm_policy_info(struct smu_context *smu, diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_12_pmfw.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_12_pmfw.h index 0a2ca544f4e3..1c407a8e96ee 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_12_pmfw.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_12_pmfw.h @@ -135,7 +135,63 @@ typedef enum { GFX_DVM_MARGIN_COUNT } GFX_DVM_MARGIN_e; -#define SMU_METRICS_TABLE_VERSION 0x13 +typedef enum{ + SYSTEM_TEMP_UBB_FPGA, + SYSTEM_TEMP_UBB_FRONT, + SYSTEM_TEMP_UBB_BACK, + SYSTEM_TEMP_UBB_OAM7, + SYSTEM_TEMP_UBB_IBC, + SYSTEM_TEMP_UBB_UFPGA, + SYSTEM_TEMP_UBB_OAM1, + SYSTEM_TEMP_OAM_0_1_HSC, + SYSTEM_TEMP_OAM_2_3_HSC, + SYSTEM_TEMP_OAM_4_5_HSC, + SYSTEM_TEMP_OAM_6_7_HSC, + SYSTEM_TEMP_UBB_FPGA_0V72_VR, + SYSTEM_TEMP_UBB_FPGA_3V3_VR, + SYSTEM_TEMP_RETIMER_0_1_2_3_1V2_VR, + SYSTEM_TEMP_RETIMER_4_5_6_7_1V2_VR, + SYSTEM_TEMP_RETIMER_0_1_0V9_VR, + SYSTEM_TEMP_RETIMER_4_5_0V9_VR, + SYSTEM_TEMP_RETIMER_2_3_0V9_VR, + SYSTEM_TEMP_RETIMER_6_7_0V9_VR, + SYSTEM_TEMP_OAM_0_1_2_3_3V3_VR, + SYSTEM_TEMP_OAM_4_5_6_7_3V3_VR, + SYSTEM_TEMP_IBC_HSC, + SYSTEM_TEMP_IBC, + SYSTEM_TEMP_MAX_ENTRIES = 32 +} SYSTEM_TEMP_e; + +typedef enum{ + NODE_TEMP_RETIMER, + NODE_TEMP_IBC_TEMP, + NODE_TEMP_IBC_2_TEMP, + NODE_TEMP_VDD18_VR_TEMP, + NODE_TEMP_04_HBM_B_VR_TEMP, + NODE_TEMP_04_HBM_D_VR_TEMP, + NODE_TEMP_MAX_TEMP_ENTRIES = 12 +} NODE_TEMP_e; + +typedef enum { + SVI_VDDCR_VDD0_TEMP, + SVI_VDDCR_VDD1_TEMP, + SVI_VDDCR_VDD2_TEMP, + SVI_VDDCR_VDD3_TEMP, + SVI_VDDCR_SOC_A_TEMP, + SVI_VDDCR_SOC_C_TEMP, + SVI_VDDCR_SOCIO_A_TEMP, + SVI_VDDCR_SOCIO_C_TEMP, + SVI_VDD_085_HBM_TEMP, + SVI_VDDCR_11_HBM_B_TEMP, + SVI_VDDCR_11_HBM_D_TEMP, + SVI_VDD_USR_TEMP, + SVI_VDDIO_11_E32_TEMP, + SVI_MAX_TEMP_ENTRIES, // 13 +} SVI_TEMP_e; + +#define SMU_METRICS_TABLE_VERSION 0x14 + +#define SMU_SYSTEM_METRICS_TABLE_VERSION 0x0 typedef struct __attribute__((packed, aligned(4))) { uint64_t AccumulationCounter; @@ -231,11 +287,27 @@ typedef struct __attribute__((packed, aligned(4))) { uint64_t GfxclkBelowHostLimitThmAcc[8]; uint64_t GfxclkBelowHostLimitTotalAcc[8]; uint64_t GfxclkLowUtilizationAcc[8]; + + uint32_t AidTemperature[4]; + uint32_t XcdTemperature[8]; + uint32_t HbmTemperature[8]; } MetricsTable_t; #define SMU_VF_METRICS_TABLE_MASK (1 << 31) #define SMU_VF_METRICS_TABLE_VERSION (0x6 | SMU_VF_METRICS_TABLE_MASK) +#pragma pack(push, 4) +typedef struct { + uint64_t AccumulationCounter; // Last update timestamp + uint16_t LabelVersion; // Defaults to 0. + uint16_t NodeIdentifier; // Unique identifier to each node on system. + int16_t SystemTemperatures[SYSTEM_TEMP_MAX_ENTRIES]; // Signed integer temperature value in Celsius, unused fields are set to 0xFFFF + int16_t NodeTemperatures[NODE_TEMP_MAX_TEMP_ENTRIES]; // Signed integer temperature value in Celsius, unused fields are set to 0xFFFF + int16_t VrTemperatures[SVI_MAX_TEMP_ENTRIES]; // Signed integer temperature value in Celsius + int16_t spare[3]; +} SystemMetricsTable_t; +#pragma pack(pop) + typedef struct __attribute__((packed, aligned(4))) { uint32_t AccumulationCounter; uint32_t InstGfxclk_TargFreq; diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_12_ppsmc.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_12_ppsmc.h index e1f490b6ce64..aff2776a8b6f 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_12_ppsmc.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_12_ppsmc.h @@ -116,7 +116,11 @@ #define PPSMC_MSG_DumpErrorRecord 0x57 #define PPSMC_MSG_EraseRasTable 0x58 #define PPSMC_MSG_GetStaticMetricsTable 0x59 -#define PPSMC_Message_Count 0x5A +#define PPSMC_MSG_ResetVfArbitersByIndex 0x5A +#define PPSMC_MSG_GetBadPageSeverity 0x5B +#define PPSMC_MSG_GetSystemMetricsTable 0x5C +#define PPSMC_MSG_GetSystemMetricsVersion 0x5D +#define PPSMC_Message_Count 0x5E //PPSMC Reset Types for driver msg argument #define PPSMC_RESET_TYPE_DRIVER_MODE_1_RESET 0x1 diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_ppsmc.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_ppsmc.h index 41f268313613..63a088ef7169 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_ppsmc.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_ppsmc.h @@ -94,9 +94,9 @@ #define PPSMC_MSG_RmaDueToBadPageThreshold 0x43 #define PPSMC_MSG_SetThrottlingPolicy 0x44 #define PPSMC_MSG_ResetSDMA 0x4D -#define PPSMC_MSG_ResetVCN 0x4E #define PPSMC_MSG_GetStaticMetricsTable 0x59 -#define PPSMC_Message_Count 0x5A +#define PPSMC_MSG_ResetVCN 0x5B +#define PPSMC_Message_Count 0x5C //PPSMC Reset Types for driver msg argument #define PPSMC_RESET_TYPE_DRIVER_MODE_1_RESET 0x1 diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h index d7a9e41820fa..2256c77da636 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h @@ -278,7 +278,8 @@ __SMU_DUMMY_MAP(MALLPowerState), \ __SMU_DUMMY_MAP(ResetSDMA), \ __SMU_DUMMY_MAP(ResetVCN), \ - __SMU_DUMMY_MAP(GetStaticMetricsTable), + __SMU_DUMMY_MAP(GetStaticMetricsTable), \ + __SMU_DUMMY_MAP(GetSystemMetricsTable), #undef __SMU_DUMMY_MAP #define __SMU_DUMMY_MAP(type) SMU_MSG_##type @@ -469,6 +470,7 @@ enum smu_feature_mask { /* Message category flags */ #define SMU_MSG_VF_FLAG (1U << 0) #define SMU_MSG_RAS_PRI (1U << 1) +#define SMU_MSG_NO_PRECHECK (1U << 2) /* Firmware capability flags */ #define SMU_FW_CAP_RAS_PRI (1U << 0) diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c index 9ad46f545d15..599eddb5a67d 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c @@ -1897,7 +1897,7 @@ static ssize_t arcturus_get_gpu_metrics(struct smu_context *smu, ret = smu_cmn_get_metrics_table(smu, &metrics, - true); + false); if (ret) return ret; diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c index c63d2e28954d..b067147b7c41 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c @@ -1781,7 +1781,7 @@ static ssize_t aldebaran_get_gpu_metrics(struct smu_context *smu, ret = smu_cmn_get_metrics_table(smu, &metrics, - true); + false); if (ret) return ret; diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c index 02a455a31c25..32fd0be05cff 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c @@ -83,7 +83,6 @@ const struct cmn2asic_mapping smu_v13_0_12_feature_mask_map[SMU_FEATURE_COUNT] = SMU_13_0_12_FEA_MAP(SMU_FEATURE_PIT_BIT, FEATURE_PIT), }; -// clang-format off const struct cmn2asic_msg_mapping smu_v13_0_12_message_map[SMU_MSG_MAX_COUNT] = { MSG_MAP(TestMessage, PPSMC_MSG_TestMessage, 0), MSG_MAP(GetSmuVersion, PPSMC_MSG_GetSmuVersion, 1), @@ -106,7 +105,7 @@ const struct cmn2asic_msg_mapping smu_v13_0_12_message_map[SMU_MSG_MAX_COUNT] = MSG_MAP(GetDpmFreqByIndex, PPSMC_MSG_GetDpmFreqByIndex, 1), MSG_MAP(SetPptLimit, PPSMC_MSG_SetPptLimit, 0), MSG_MAP(GetPptLimit, PPSMC_MSG_GetPptLimit, 1), - MSG_MAP(GfxDeviceDriverReset, PPSMC_MSG_GfxDriverReset, SMU_MSG_RAS_PRI), + MSG_MAP(GfxDeviceDriverReset, PPSMC_MSG_GfxDriverReset, SMU_MSG_RAS_PRI | SMU_MSG_NO_PRECHECK), MSG_MAP(DramLogSetDramAddrHigh, PPSMC_MSG_DramLogSetDramAddrHigh, 0), MSG_MAP(DramLogSetDramAddrLow, PPSMC_MSG_DramLogSetDramAddrLow, 0), MSG_MAP(DramLogSetDramSize, PPSMC_MSG_DramLogSetDramSize, 0), @@ -138,8 +137,47 @@ const struct cmn2asic_msg_mapping smu_v13_0_12_message_map[SMU_MSG_MAX_COUNT] = MSG_MAP(SetThrottlingPolicy, PPSMC_MSG_SetThrottlingPolicy, 0), MSG_MAP(ResetSDMA, PPSMC_MSG_ResetSDMA, 0), MSG_MAP(GetStaticMetricsTable, PPSMC_MSG_GetStaticMetricsTable, 1), + MSG_MAP(GetSystemMetricsTable, PPSMC_MSG_GetSystemMetricsTable, 0), }; +int smu_v13_0_12_tables_init(struct smu_context *smu) +{ + struct amdgpu_baseboard_temp_metrics_v1_0 *baseboard_temp_metrics; + struct amdgpu_gpuboard_temp_metrics_v1_0 *gpuboard_temp_metrics; + struct smu_table_context *smu_table = &smu->smu_table; + struct smu_table *tables = smu_table->tables; + struct smu_table_cache *cache; + int ret; + + ret = smu_table_cache_init(smu, SMU_TABLE_BASEBOARD_TEMP_METRICS, + sizeof(*baseboard_temp_metrics), 50); + if (ret) + return ret; + /* Initialize base board temperature metrics */ + cache = &(tables[SMU_TABLE_BASEBOARD_TEMP_METRICS].cache); + baseboard_temp_metrics = + (struct amdgpu_baseboard_temp_metrics_v1_0 *) cache->buffer; + smu_cmn_init_baseboard_temp_metrics(baseboard_temp_metrics, 1, 0); + /* Initialize GPU board temperature metrics */ + ret = smu_table_cache_init(smu, SMU_TABLE_GPUBOARD_TEMP_METRICS, + sizeof(*gpuboard_temp_metrics), 50); + if (ret) { + smu_table_cache_fini(smu, SMU_TABLE_BASEBOARD_TEMP_METRICS); + return ret; + } + cache = &(tables[SMU_TABLE_GPUBOARD_TEMP_METRICS].cache); + gpuboard_temp_metrics = (struct amdgpu_gpuboard_temp_metrics_v1_0 *)cache->buffer; + smu_cmn_init_gpuboard_temp_metrics(gpuboard_temp_metrics, 1, 0); + + return 0; +} + +void smu_v13_0_12_tables_fini(struct smu_context *smu) +{ + smu_table_cache_fini(smu, SMU_TABLE_BASEBOARD_TEMP_METRICS); + smu_table_cache_fini(smu, SMU_TABLE_GPUBOARD_TEMP_METRICS); +} + static int smu_v13_0_12_get_enabled_mask(struct smu_context *smu, uint64_t *feature_mask) { @@ -184,7 +222,8 @@ static int smu_v13_0_12_fru_get_product_info(struct smu_context *smu, int smu_v13_0_12_get_max_metrics_size(void) { - return max(sizeof(StaticMetricsTable_t), sizeof(MetricsTable_t)); + return max3(sizeof(StaticMetricsTable_t), sizeof(MetricsTable_t), + sizeof(SystemMetricsTable_t)); } static void smu_v13_0_12_init_xgmi_data(struct smu_context *smu, @@ -220,7 +259,7 @@ int smu_v13_0_12_setup_driver_pptable(struct smu_context *smu) struct PPTable_t *pptable = (struct PPTable_t *)smu_table->driver_pptable; uint32_t table_version; - int ret, i; + int ret, i, n; if (!pptable->Init) { ret = smu_v13_0_6_get_static_metrics_table(smu); @@ -259,6 +298,22 @@ int smu_v13_0_12_setup_driver_pptable(struct smu_context *smu) /* use AID0 serial number by default */ pptable->PublicSerialNumber_AID = static_metrics->PublicSerialNumber_AID[0]; + + amdgpu_device_set_uid(smu->adev->uid_info, AMDGPU_UID_TYPE_SOC, + 0, pptable->PublicSerialNumber_AID); + n = ARRAY_SIZE(static_metrics->PublicSerialNumber_AID); + for (i = 0; i < n; i++) { + amdgpu_device_set_uid( + smu->adev->uid_info, AMDGPU_UID_TYPE_AID, i, + static_metrics->PublicSerialNumber_AID[i]); + } + n = ARRAY_SIZE(static_metrics->PublicSerialNumber_XCD); + for (i = 0; i < n; i++) { + amdgpu_device_set_uid( + smu->adev->uid_info, AMDGPU_UID_TYPE_XCD, i, + static_metrics->PublicSerialNumber_XCD[i]); + } + ret = smu_v13_0_12_fru_get_product_info(smu, static_metrics); if (ret) return ret; @@ -359,6 +414,243 @@ int smu_v13_0_12_get_smu_metrics_data(struct smu_context *smu, return 0; } +static int smu_v13_0_12_get_system_metrics_table(struct smu_context *smu, + void *metrics_table) +{ + struct smu_table_context *smu_table = &smu->smu_table; + uint32_t table_size = smu_table->tables[SMU_TABLE_SMU_METRICS].size; + struct smu_table *table = &smu_table->driver_table; + int ret; + + ret = smu_cmn_send_smc_msg(smu, SMU_MSG_GetSystemMetricsTable, NULL); + if (ret) { + dev_info(smu->adev->dev, + "Failed to export system metrics table!\n"); + return ret; + } + + amdgpu_asic_invalidate_hdp(smu->adev, NULL); + memcpy(smu_table->metrics_table, table->cpu_addr, table_size); + + if (metrics_table) + memcpy(metrics_table, smu_table->metrics_table, sizeof(SystemMetricsTable_t)); + + return 0; +} + +static enum amdgpu_node_temp smu_v13_0_12_get_node_sensor_type(NODE_TEMP_e type) +{ + switch (type) { + case NODE_TEMP_RETIMER: + return AMDGPU_RETIMER_X_TEMP; + case NODE_TEMP_IBC_TEMP: + return AMDGPU_OAM_X_IBC_TEMP; + case NODE_TEMP_IBC_2_TEMP: + return AMDGPU_OAM_X_IBC_2_TEMP; + case NODE_TEMP_VDD18_VR_TEMP: + return AMDGPU_OAM_X_VDD18_VR_TEMP; + case NODE_TEMP_04_HBM_B_VR_TEMP: + return AMDGPU_OAM_X_04_HBM_B_VR_TEMP; + case NODE_TEMP_04_HBM_D_VR_TEMP: + return AMDGPU_OAM_X_04_HBM_D_VR_TEMP; + default: + return -EINVAL; + } +} + +static enum amdgpu_vr_temp smu_v13_0_12_get_vr_sensor_type(SVI_TEMP_e type) +{ + switch (type) { + case SVI_VDDCR_VDD0_TEMP: + return AMDGPU_VDDCR_VDD0_TEMP; + case SVI_VDDCR_VDD1_TEMP: + return AMDGPU_VDDCR_VDD1_TEMP; + case SVI_VDDCR_VDD2_TEMP: + return AMDGPU_VDDCR_VDD2_TEMP; + case SVI_VDDCR_VDD3_TEMP: + return AMDGPU_VDDCR_VDD3_TEMP; + case SVI_VDDCR_SOC_A_TEMP: + return AMDGPU_VDDCR_SOC_A_TEMP; + case SVI_VDDCR_SOC_C_TEMP: + return AMDGPU_VDDCR_SOC_C_TEMP; + case SVI_VDDCR_SOCIO_A_TEMP: + return AMDGPU_VDDCR_SOCIO_A_TEMP; + case SVI_VDDCR_SOCIO_C_TEMP: + return AMDGPU_VDDCR_SOCIO_C_TEMP; + case SVI_VDD_085_HBM_TEMP: + return AMDGPU_VDD_085_HBM_TEMP; + case SVI_VDDCR_11_HBM_B_TEMP: + return AMDGPU_VDDCR_11_HBM_B_TEMP; + case SVI_VDDCR_11_HBM_D_TEMP: + return AMDGPU_VDDCR_11_HBM_D_TEMP; + case SVI_VDD_USR_TEMP: + return AMDGPU_VDD_USR_TEMP; + case SVI_VDDIO_11_E32_TEMP: + return AMDGPU_VDDIO_11_E32_TEMP; + default: + return -EINVAL; + } +} + +static enum amdgpu_system_temp smu_v13_0_12_get_system_sensor_type(SYSTEM_TEMP_e type) +{ + switch (type) { + case SYSTEM_TEMP_UBB_FPGA: + return AMDGPU_UBB_FPGA_TEMP; + case SYSTEM_TEMP_UBB_FRONT: + return AMDGPU_UBB_FRONT_TEMP; + case SYSTEM_TEMP_UBB_BACK: + return AMDGPU_UBB_BACK_TEMP; + case SYSTEM_TEMP_UBB_OAM7: + return AMDGPU_UBB_OAM7_TEMP; + case SYSTEM_TEMP_UBB_IBC: + return AMDGPU_UBB_IBC_TEMP; + case SYSTEM_TEMP_UBB_UFPGA: + return AMDGPU_UBB_UFPGA_TEMP; + case SYSTEM_TEMP_UBB_OAM1: + return AMDGPU_UBB_OAM1_TEMP; + case SYSTEM_TEMP_OAM_0_1_HSC: + return AMDGPU_OAM_0_1_HSC_TEMP; + case SYSTEM_TEMP_OAM_2_3_HSC: + return AMDGPU_OAM_2_3_HSC_TEMP; + case SYSTEM_TEMP_OAM_4_5_HSC: + return AMDGPU_OAM_4_5_HSC_TEMP; + case SYSTEM_TEMP_OAM_6_7_HSC: + return AMDGPU_OAM_6_7_HSC_TEMP; + case SYSTEM_TEMP_UBB_FPGA_0V72_VR: + return AMDGPU_UBB_FPGA_0V72_VR_TEMP; + case SYSTEM_TEMP_UBB_FPGA_3V3_VR: + return AMDGPU_UBB_FPGA_3V3_VR_TEMP; + case SYSTEM_TEMP_RETIMER_0_1_2_3_1V2_VR: + return AMDGPU_RETIMER_0_1_2_3_1V2_VR_TEMP; + case SYSTEM_TEMP_RETIMER_4_5_6_7_1V2_VR: + return AMDGPU_RETIMER_4_5_6_7_1V2_VR_TEMP; + case SYSTEM_TEMP_RETIMER_0_1_0V9_VR: + return AMDGPU_RETIMER_0_1_0V9_VR_TEMP; + case SYSTEM_TEMP_RETIMER_4_5_0V9_VR: + return AMDGPU_RETIMER_4_5_0V9_VR_TEMP; + case SYSTEM_TEMP_RETIMER_2_3_0V9_VR: + return AMDGPU_RETIMER_2_3_0V9_VR_TEMP; + case SYSTEM_TEMP_RETIMER_6_7_0V9_VR: + return AMDGPU_RETIMER_6_7_0V9_VR_TEMP; + case SYSTEM_TEMP_OAM_0_1_2_3_3V3_VR: + return AMDGPU_OAM_0_1_2_3_3V3_VR_TEMP; + case SYSTEM_TEMP_OAM_4_5_6_7_3V3_VR: + return AMDGPU_OAM_4_5_6_7_3V3_VR_TEMP; + case SYSTEM_TEMP_IBC_HSC: + return AMDGPU_IBC_HSC_TEMP; + case SYSTEM_TEMP_IBC: + return AMDGPU_IBC_TEMP; + default: + return -EINVAL; + } +} + +static bool smu_v13_0_12_is_temp_metrics_supported(struct smu_context *smu, + enum smu_temp_metric_type type) +{ + switch (type) { + case SMU_TEMP_METRIC_BASEBOARD: + if (smu->adev->gmc.xgmi.physical_node_id == 0 && + smu->adev->gmc.xgmi.num_physical_nodes > 1 && + smu_v13_0_6_cap_supported(smu, SMU_CAP(TEMP_METRICS))) + return true; + break; + case SMU_TEMP_METRIC_GPUBOARD: + return smu_v13_0_6_cap_supported(smu, SMU_CAP(TEMP_METRICS)); + default: + break; + } + + return false; +} + +static ssize_t smu_v13_0_12_get_temp_metrics(struct smu_context *smu, + enum smu_temp_metric_type type, void *table) +{ + struct amdgpu_baseboard_temp_metrics_v1_0 *baseboard_temp_metrics; + struct amdgpu_gpuboard_temp_metrics_v1_0 *gpuboard_temp_metrics; + struct smu_table_context *smu_table = &smu->smu_table; + SystemMetricsTable_t *metrics = + (SystemMetricsTable_t *)smu_table->metrics_table; + + struct smu_table *data_table; + int ret, sensor_type; + u32 idx, sensors; + ssize_t size; + + if (type == SMU_TEMP_METRIC_BASEBOARD) { + /* Initialize base board temperature metrics */ + data_table = + &smu->smu_table.tables[SMU_TABLE_BASEBOARD_TEMP_METRICS]; + baseboard_temp_metrics = + (struct amdgpu_baseboard_temp_metrics_v1_0 *) + data_table->cache.buffer; + size = sizeof(*baseboard_temp_metrics); + } else { + data_table = + &smu->smu_table.tables[SMU_TABLE_GPUBOARD_TEMP_METRICS]; + gpuboard_temp_metrics = + (struct amdgpu_gpuboard_temp_metrics_v1_0 *) + data_table->cache.buffer; + size = sizeof(*baseboard_temp_metrics); + } + + ret = smu_v13_0_12_get_system_metrics_table(smu, NULL); + if (ret) + return ret; + + smu_table_cache_update_time(data_table, jiffies); + + if (type == SMU_TEMP_METRIC_GPUBOARD) { + gpuboard_temp_metrics->accumulation_counter = metrics->AccumulationCounter; + gpuboard_temp_metrics->label_version = metrics->LabelVersion; + gpuboard_temp_metrics->node_id = metrics->NodeIdentifier; + + idx = 0; + for (sensors = 0; sensors < NODE_TEMP_MAX_TEMP_ENTRIES; sensors++) { + if (metrics->NodeTemperatures[sensors] != -1) { + sensor_type = smu_v13_0_12_get_node_sensor_type(sensors); + gpuboard_temp_metrics->node_temp[idx] = + ((int)metrics->NodeTemperatures[sensors]) & 0xFFFFFF; + gpuboard_temp_metrics->node_temp[idx] |= (sensor_type << 24); + idx++; + } + } + + idx = 0; + + for (sensors = 0; sensors < SVI_MAX_TEMP_ENTRIES; sensors++) { + if (metrics->VrTemperatures[sensors] != -1) { + sensor_type = smu_v13_0_12_get_vr_sensor_type(sensors); + gpuboard_temp_metrics->vr_temp[idx] = + ((int)metrics->VrTemperatures[sensors]) & 0xFFFFFF; + gpuboard_temp_metrics->vr_temp[idx] |= (sensor_type << 24); + idx++; + } + } + } else if (type == SMU_TEMP_METRIC_BASEBOARD) { + baseboard_temp_metrics->accumulation_counter = metrics->AccumulationCounter; + baseboard_temp_metrics->label_version = metrics->LabelVersion; + baseboard_temp_metrics->node_id = metrics->NodeIdentifier; + + idx = 0; + for (sensors = 0; sensors < SYSTEM_TEMP_MAX_ENTRIES; sensors++) { + if (metrics->SystemTemperatures[sensors] != -1) { + sensor_type = smu_v13_0_12_get_system_sensor_type(sensors); + baseboard_temp_metrics->system_temp[idx] = + ((int)metrics->SystemTemperatures[sensors]) & 0xFFFFFF; + baseboard_temp_metrics->system_temp[idx] |= (sensor_type << 24); + idx++; + } + } + } + + memcpy(table, data_table->cache.buffer, size); + + return size; +} + ssize_t smu_v13_0_12_get_xcp_metrics(struct smu_context *smu, struct amdgpu_xcp *xcp, void *table, void *smu_metrics) { const u8 num_jpeg_rings = NUM_JPEG_RINGS_FW; @@ -572,3 +864,8 @@ ssize_t smu_v13_0_12_get_gpu_metrics(struct smu_context *smu, void **table, void return sizeof(*gpu_metrics); } + +const struct smu_temp_funcs smu_v13_0_12_temp_funcs = { + .temp_metrics_is_supported = smu_v13_0_12_is_temp_metrics_supported, + .get_temp_metrics = smu_v13_0_12_get_temp_metrics, +}; diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c index 9cc294f4708b..e37b7b5358ea 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c @@ -145,7 +145,7 @@ static const struct cmn2asic_msg_mapping smu_v13_0_6_message_map[SMU_MSG_MAX_COU MSG_MAP(GetDpmFreqByIndex, PPSMC_MSG_GetDpmFreqByIndex, 1), MSG_MAP(SetPptLimit, PPSMC_MSG_SetPptLimit, 0), MSG_MAP(GetPptLimit, PPSMC_MSG_GetPptLimit, 1), - MSG_MAP(GfxDeviceDriverReset, PPSMC_MSG_GfxDriverReset, SMU_MSG_RAS_PRI), + MSG_MAP(GfxDeviceDriverReset, PPSMC_MSG_GfxDriverReset, SMU_MSG_RAS_PRI | SMU_MSG_NO_PRECHECK), MSG_MAP(DramLogSetDramAddrHigh, PPSMC_MSG_DramLogSetDramAddrHigh, 0), MSG_MAP(DramLogSetDramAddrLow, PPSMC_MSG_DramLogSetDramAddrLow, 0), MSG_MAP(DramLogSetDramSize, PPSMC_MSG_DramLogSetDramSize, 0), @@ -177,7 +177,7 @@ static const struct cmn2asic_msg_mapping smu_v13_0_6_message_map[SMU_MSG_MAX_COU MSG_MAP(SetThrottlingPolicy, PPSMC_MSG_SetThrottlingPolicy, 0), MSG_MAP(ResetSDMA, PPSMC_MSG_ResetSDMA, 0), MSG_MAP(ResetVCN, PPSMC_MSG_ResetVCN, 0), - MSG_MAP(GetStaticMetricsTable, PPSMC_MSG_GetStaticMetricsTable, 0), + MSG_MAP(GetStaticMetricsTable, PPSMC_MSG_GetStaticMetricsTable, 1), }; // clang-format on @@ -312,6 +312,8 @@ static void smu_v13_0_14_init_caps(struct smu_context *smu) smu_v13_0_6_cap_set(smu, SMU_CAP(PER_INST_METRICS)); if (fw_ver >= 0x5551200) smu_v13_0_6_cap_set(smu, SMU_CAP(SDMA_RESET)); + if (fw_ver >= 0x5551800) + smu_v13_0_6_cap_set(smu, SMU_CAP(VCN_RESET)); if (fw_ver >= 0x5551600) { smu_v13_0_6_cap_set(smu, SMU_CAP(STATIC_METRICS)); smu_v13_0_6_cap_set(smu, SMU_CAP(BOARD_VOLTAGE)); @@ -350,6 +352,13 @@ static void smu_v13_0_12_init_caps(struct smu_context *smu) smu_v13_0_6_cap_set(smu, SMU_CAP(BOARD_VOLTAGE)); smu_v13_0_6_cap_set(smu, SMU_CAP(PLDM_VERSION)); } + + if (fw_ver >= 0x04560700) { + if (!amdgpu_sriov_vf(smu->adev)) + smu_v13_0_6_cap_set(smu, SMU_CAP(TEMP_METRICS)); + } else { + smu_v13_0_12_tables_fini(smu); + } } static void smu_v13_0_6_init_caps(struct smu_context *smu) @@ -402,19 +411,37 @@ static void smu_v13_0_6_init_caps(struct smu_context *smu) if ((pgm == 7 && fw_ver >= 0x7550E00) || (pgm == 0 && fw_ver >= 0x00557E00)) smu_v13_0_6_cap_set(smu, SMU_CAP(HST_LIMIT_METRICS)); - if ((pgm == 0 && fw_ver >= 0x00557F01) || - (pgm == 7 && fw_ver >= 0x7551000)) { - smu_v13_0_6_cap_set(smu, SMU_CAP(STATIC_METRICS)); - smu_v13_0_6_cap_set(smu, SMU_CAP(BOARD_VOLTAGE)); + + if (amdgpu_sriov_vf(adev)) { + if ((pgm == 0 && fw_ver >= 0x00558000) || + (pgm == 7 && fw_ver >= 0x7551000)) { + smu_v13_0_6_cap_set(smu, + SMU_CAP(STATIC_METRICS)); + smu_v13_0_6_cap_set(smu, + SMU_CAP(BOARD_VOLTAGE)); + smu_v13_0_6_cap_set(smu, SMU_CAP(PLDM_VERSION)); + } + } else { + if ((pgm == 0 && fw_ver >= 0x00557F01) || + (pgm == 7 && fw_ver >= 0x7551000)) { + smu_v13_0_6_cap_set(smu, + SMU_CAP(STATIC_METRICS)); + smu_v13_0_6_cap_set(smu, + SMU_CAP(BOARD_VOLTAGE)); + } + if ((pgm == 0 && fw_ver >= 0x00558000) || + (pgm == 7 && fw_ver >= 0x7551000)) + smu_v13_0_6_cap_set(smu, SMU_CAP(PLDM_VERSION)); } - if ((pgm == 0 && fw_ver >= 0x00558000) || - (pgm == 7 && fw_ver >= 0x7551000)) - smu_v13_0_6_cap_set(smu, SMU_CAP(PLDM_VERSION)); } if (((pgm == 7) && (fw_ver >= 0x7550700)) || ((pgm == 0) && (fw_ver >= 0x00557900)) || ((pgm == 4) && (fw_ver >= 0x4557000))) smu_v13_0_6_cap_set(smu, SMU_CAP(SDMA_RESET)); + + if (((pgm == 0) && (fw_ver >= 0x00558200)) || + ((pgm == 4) && (fw_ver >= 0x04557100))) + smu_v13_0_6_cap_set(smu, SMU_CAP(VCN_RESET)); } static void smu_v13_0_x_init_caps(struct smu_context *smu) @@ -511,8 +538,12 @@ static int smu_v13_0_6_tables_init(struct smu_context *smu) { struct smu_table_context *smu_table = &smu->smu_table; struct smu_table *tables = smu_table->tables; + void *gpu_metrics_table __free(kfree) = NULL; + void *driver_pptable __free(kfree) = NULL; + void *metrics_table __free(kfree) = NULL; struct amdgpu_device *adev = smu->adev; int gpu_metrcs_size = METRICS_TABLE_SIZE; + int ret; if (!(adev->flags & AMD_IS_APU)) SMU_TABLE_INIT(tables, SMU_TABLE_PMSTATUSLOG, SMU13_TOOL_SIZE, @@ -528,27 +559,32 @@ static int smu_v13_0_6_tables_init(struct smu_context *smu) PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM | AMDGPU_GEM_DOMAIN_GTT); - smu_table->metrics_table = kzalloc(METRICS_TABLE_SIZE, GFP_KERNEL); - if (!smu_table->metrics_table) + metrics_table = kzalloc(METRICS_TABLE_SIZE, GFP_KERNEL); + if (!metrics_table) return -ENOMEM; smu_table->metrics_time = 0; smu_table->gpu_metrics_table_size = sizeof(struct gpu_metrics_v1_8); - smu_table->gpu_metrics_table = + gpu_metrics_table = kzalloc(smu_table->gpu_metrics_table_size, GFP_KERNEL); - if (!smu_table->gpu_metrics_table) { - kfree(smu_table->metrics_table); + if (!gpu_metrics_table) return -ENOMEM; - } - smu_table->driver_pptable = - kzalloc(sizeof(struct PPTable_t), GFP_KERNEL); - if (!smu_table->driver_pptable) { - kfree(smu_table->metrics_table); - kfree(smu_table->gpu_metrics_table); + driver_pptable = kzalloc(sizeof(struct PPTable_t), GFP_KERNEL); + if (!driver_pptable) return -ENOMEM; + + if (amdgpu_ip_version(smu->adev, MP1_HWIP, 0) == + IP_VERSION(13, 0, 12)) { + ret = smu_v13_0_12_tables_init(smu); + if (ret) + return ret; } + smu_table->gpu_metrics_table = no_free_ptr(gpu_metrics_table); + smu_table->metrics_table = no_free_ptr(metrics_table); + smu_table->driver_pptable = no_free_ptr(driver_pptable); + return 0; } @@ -677,6 +713,13 @@ static int smu_v13_0_6_init_smc_tables(struct smu_context *smu) return ret; } +static int smu_v13_0_6_fini_smc_tables(struct smu_context *smu) +{ + if (amdgpu_ip_version(smu->adev, MP1_HWIP, 0) == IP_VERSION(13, 0, 12)) + smu_v13_0_12_tables_fini(smu); + return smu_v13_0_fini_smc_tables(smu); +} + static int smu_v13_0_6_get_allowed_feature_mask(struct smu_context *smu, uint32_t *feature_mask, uint32_t num) @@ -803,7 +846,7 @@ static int smu_v13_0_6_setup_driver_pptable(struct smu_context *smu) struct PPTable_t *pptable = (struct PPTable_t *)smu_table->driver_pptable; int version = smu_v13_0_6_get_metrics_version(smu); - int ret, i, retry = 100; + int ret, i, retry = 100, n; uint32_t table_version; uint16_t max_speed; uint8_t max_width; @@ -865,6 +908,23 @@ static int smu_v13_0_6_setup_driver_pptable(struct smu_context *smu) pptable->PublicSerialNumber_AID = GET_METRIC_FIELD(PublicSerialNumber_AID, version)[0]; + amdgpu_device_set_uid(smu->adev->uid_info, AMDGPU_UID_TYPE_SOC, + 0, pptable->PublicSerialNumber_AID); + n = ARRAY_SIZE(metrics_v0->PublicSerialNumber_AID); + for (i = 0; i < n; i++) { + amdgpu_device_set_uid( + smu->adev->uid_info, AMDGPU_UID_TYPE_AID, i, + GET_METRIC_FIELD(PublicSerialNumber_AID, + version)[i]); + } + n = ARRAY_SIZE(metrics_v0->PublicSerialNumber_XCD); + for (i = 0; i < n; i++) { + amdgpu_device_set_uid( + smu->adev->uid_info, AMDGPU_UID_TYPE_XCD, i, + GET_METRIC_FIELD(PublicSerialNumber_XCD, + version)[i]); + } + pptable->Init = true; if (smu_v13_0_6_cap_supported(smu, SMU_CAP(STATIC_METRICS))) { ret = smu_v13_0_6_get_static_metrics_table(smu); @@ -2560,9 +2620,9 @@ static ssize_t smu_v13_0_6_get_xcp_metrics(struct smu_context *smu, int xcp_id, const u8 num_jpeg_rings = AMDGPU_MAX_JPEG_RINGS_4_0_3; int version = smu_v13_0_6_get_metrics_version(smu); struct amdgpu_partition_metrics_v1_0 *xcp_metrics; + MetricsTableV0_t *metrics_v0 __free(kfree) = NULL; struct amdgpu_device *adev = smu->adev; int ret, inst, i, j, k, idx; - MetricsTableV0_t *metrics_v0; MetricsTableV1_t *metrics_v1; MetricsTableV2_t *metrics_v2; struct amdgpu_xcp *xcp; @@ -2587,17 +2647,14 @@ static ssize_t smu_v13_0_6_get_xcp_metrics(struct smu_context *smu, int xcp_id, return -ENOMEM; ret = smu_v13_0_6_get_metrics_table(smu, metrics_v0, false); - if (ret) { - kfree(metrics_v0); + if (ret) return ret; - } if (amdgpu_ip_version(smu->adev, MP1_HWIP, 0) == IP_VERSION(13, 0, 12) && - smu_v13_0_6_cap_supported(smu, SMU_CAP(STATIC_METRICS))) { - ret = smu_v13_0_12_get_xcp_metrics(smu, xcp, table, metrics_v0); - goto out; - } + smu_v13_0_6_cap_supported(smu, SMU_CAP(STATIC_METRICS))) + return smu_v13_0_12_get_xcp_metrics(smu, xcp, table, + metrics_v0); metrics_v1 = (MetricsTableV1_t *)metrics_v0; metrics_v2 = (MetricsTableV2_t *)metrics_v0; @@ -2668,8 +2725,6 @@ static ssize_t smu_v13_0_6_get_xcp_metrics(struct smu_context *smu, int xcp_id, idx++; } } -out: - kfree(metrics_v0); return sizeof(*xcp_metrics); } @@ -2680,31 +2735,26 @@ static ssize_t smu_v13_0_6_get_gpu_metrics(struct smu_context *smu, void **table struct gpu_metrics_v1_8 *gpu_metrics = (struct gpu_metrics_v1_8 *)smu_table->gpu_metrics_table; int version = smu_v13_0_6_get_metrics_version(smu); + MetricsTableV0_t *metrics_v0 __free(kfree) = NULL; int ret = 0, xcc_id, inst, i, j, k, idx; struct amdgpu_device *adev = smu->adev; - MetricsTableV0_t *metrics_v0; MetricsTableV1_t *metrics_v1; MetricsTableV2_t *metrics_v2; struct amdgpu_xcp *xcp; u16 link_width_level; - ssize_t num_bytes; u8 num_jpeg_rings; u32 inst_mask; bool per_inst; metrics_v0 = kzalloc(METRICS_TABLE_SIZE, GFP_KERNEL); ret = smu_v13_0_6_get_metrics_table(smu, metrics_v0, false); - if (ret) { - kfree(metrics_v0); + if (ret) return ret; - } - if (amdgpu_ip_version(smu->adev, MP1_HWIP, 0) == IP_VERSION(13, 0, 12) && - smu_v13_0_6_cap_supported(smu, SMU_CAP(STATIC_METRICS))) { - num_bytes = smu_v13_0_12_get_gpu_metrics(smu, table, metrics_v0); - kfree(metrics_v0); - return num_bytes; - } + if (amdgpu_ip_version(smu->adev, MP1_HWIP, 0) == + IP_VERSION(13, 0, 12) && + smu_v13_0_6_cap_supported(smu, SMU_CAP(STATIC_METRICS))) + return smu_v13_0_12_get_gpu_metrics(smu, table, metrics_v0); metrics_v1 = (MetricsTableV1_t *)metrics_v0; metrics_v2 = (MetricsTableV2_t *)metrics_v0; @@ -2890,7 +2940,6 @@ static ssize_t smu_v13_0_6_get_gpu_metrics(struct smu_context *smu, void **table gpu_metrics->firmware_timestamp = GET_METRIC_FIELD(Timestamp, version); *table = (void *)gpu_metrics; - kfree(metrics_v0); return sizeof(*gpu_metrics); } @@ -3076,7 +3125,7 @@ static inline bool smu_v13_0_6_is_link_reset_supported(struct smu_context *smu) struct amdgpu_device *adev = smu->adev; int var = (adev->pdev->device & 0xF); - if (var == 0x1) + if (var == 0x0 || var == 0x1 || var == 0x3) return true; return false; @@ -3152,6 +3201,11 @@ static int smu_v13_0_6_reset_sdma(struct smu_context *smu, uint32_t inst_mask) return ret; } +static bool smu_v13_0_6_reset_vcn_is_supported(struct smu_context *smu) +{ + return smu_v13_0_6_cap_supported(smu, SMU_CAP(VCN_RESET)); +} + static int smu_v13_0_6_reset_vcn(struct smu_context *smu, uint32_t inst_mask) { int ret = 0; @@ -3797,7 +3851,7 @@ static const struct pptable_funcs smu_v13_0_6_ppt_funcs = { .init_microcode = smu_v13_0_6_init_microcode, .fini_microcode = smu_v13_0_fini_microcode, .init_smc_tables = smu_v13_0_6_init_smc_tables, - .fini_smc_tables = smu_v13_0_fini_smc_tables, + .fini_smc_tables = smu_v13_0_6_fini_smc_tables, .init_power = smu_v13_0_init_power, .fini_power = smu_v13_0_fini_power, .check_fw_status = smu_v13_0_6_check_fw_status, @@ -3840,6 +3894,7 @@ static const struct pptable_funcs smu_v13_0_6_ppt_funcs = { .reset_sdma = smu_v13_0_6_reset_sdma, .reset_sdma_is_supported = smu_v13_0_6_reset_sdma_is_supported, .dpm_reset_vcn = smu_v13_0_6_reset_vcn, + .reset_vcn_is_supported = smu_v13_0_6_reset_vcn_is_supported, }; void smu_v13_0_6_set_ppt_funcs(struct smu_context *smu) @@ -3857,3 +3912,9 @@ void smu_v13_0_6_set_ppt_funcs(struct smu_context *smu) amdgpu_mca_smu_init_funcs(smu->adev, &smu_v13_0_6_mca_smu_funcs); amdgpu_aca_set_smu_funcs(smu->adev, &smu_v13_0_6_aca_smu_funcs); } + +void smu_v13_0_6_set_temp_funcs(struct smu_context *smu) +{ + smu->smu_temp.temp_funcs = (amdgpu_ip_version(smu->adev, MP1_HWIP, 0) + == IP_VERSION(13, 0, 12)) ? &smu_v13_0_12_temp_funcs : NULL; +} diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.h b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.h index 67b30674fd31..bcb8246c0804 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.h +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.h @@ -64,14 +64,17 @@ enum smu_v13_0_6_caps { SMU_CAP(RMA_MSG), SMU_CAP(ACA_SYND), SMU_CAP(SDMA_RESET), + SMU_CAP(VCN_RESET), SMU_CAP(STATIC_METRICS), SMU_CAP(HST_LIMIT_METRICS), SMU_CAP(BOARD_VOLTAGE), SMU_CAP(PLDM_VERSION), + SMU_CAP(TEMP_METRICS), SMU_CAP(ALL), }; extern void smu_v13_0_6_set_ppt_funcs(struct smu_context *smu); +extern void smu_v13_0_6_set_temp_funcs(struct smu_context *smu); bool smu_v13_0_6_cap_supported(struct smu_context *smu, enum smu_v13_0_6_caps cap); int smu_v13_0_6_get_static_metrics_table(struct smu_context *smu); int smu_v13_0_6_get_metrics_table(struct smu_context *smu, void *metrics_table, @@ -86,6 +89,9 @@ ssize_t smu_v13_0_12_get_gpu_metrics(struct smu_context *smu, void **table, void ssize_t smu_v13_0_12_get_xcp_metrics(struct smu_context *smu, struct amdgpu_xcp *xcp, void *table, void *smu_metrics); +int smu_v13_0_12_tables_init(struct smu_context *smu); +void smu_v13_0_12_tables_fini(struct smu_context *smu); extern const struct cmn2asic_mapping smu_v13_0_12_feature_mask_map[]; extern const struct cmn2asic_msg_mapping smu_v13_0_12_message_map[]; +extern const struct smu_temp_funcs smu_v13_0_12_temp_funcs; #endif diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c index 3aea32baea3d..f32474af90b3 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c @@ -1697,9 +1697,11 @@ static int smu_v14_0_2_get_power_limit(struct smu_context *smu, uint32_t *min_power_limit) { struct smu_table_context *table_context = &smu->smu_table; + struct smu_14_0_2_powerplay_table *powerplay_table = + table_context->power_play_table; PPTable_t *pptable = table_context->driver_pptable; CustomSkuTable_t *skutable = &pptable->CustomSkuTable; - uint32_t power_limit; + uint32_t power_limit, od_percent_upper = 0, od_percent_lower = 0; uint32_t msg_limit = pptable->SkuTable.MsgLimits.Power[PPT_THROTTLER_PPT0][POWER_SOURCE_AC]; if (smu_v14_0_get_current_power_limit(smu, &power_limit)) @@ -1712,11 +1714,29 @@ static int smu_v14_0_2_get_power_limit(struct smu_context *smu, if (default_power_limit) *default_power_limit = power_limit; - if (max_power_limit) - *max_power_limit = msg_limit; + if (powerplay_table) { + if (smu->od_enabled && + smu_v14_0_2_is_od_feature_supported(smu, PP_OD_FEATURE_PPT_BIT)) { + od_percent_upper = pptable->SkuTable.OverDriveLimitsBasicMax.Ppt; + od_percent_lower = pptable->SkuTable.OverDriveLimitsBasicMin.Ppt; + } else if (smu_v14_0_2_is_od_feature_supported(smu, PP_OD_FEATURE_PPT_BIT)) { + od_percent_upper = 0; + od_percent_lower = pptable->SkuTable.OverDriveLimitsBasicMin.Ppt; + } + } + + dev_dbg(smu->adev->dev, "od percent upper:%d, od percent lower:%d (default power: %d)\n", + od_percent_upper, od_percent_lower, power_limit); + + if (max_power_limit) { + *max_power_limit = msg_limit * (100 + od_percent_upper); + *max_power_limit /= 100; + } - if (min_power_limit) - *min_power_limit = 0; + if (min_power_limit) { + *min_power_limit = power_limit * (100 + od_percent_lower); + *min_power_limit /= 100; + } return 0; } diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c index 59f9abd0f7b8..f532f7c69259 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c @@ -256,11 +256,12 @@ static int __smu_cmn_ras_filter_msg(struct smu_context *smu, { struct amdgpu_device *adev = smu->adev; uint32_t flags, resp; - bool fed_status; + bool fed_status, pri; flags = __smu_cmn_get_msg_flags(smu, msg); *poll = true; + pri = !!(flags & SMU_MSG_NO_PRECHECK); /* When there is RAS fatal error, FW won't process non-RAS priority * messages. Don't allow any messages other than RAS priority messages. */ @@ -272,15 +273,18 @@ static int __smu_cmn_ras_filter_msg(struct smu_context *smu, smu_get_message_name(smu, msg)); return -EACCES; } + } + if (pri || fed_status) { /* FW will ignore non-priority messages when a RAS fatal error - * is detected. Hence it is possible that a previous message - * wouldn't have got response. Allow to continue without polling - * for response status for priority messages. + * or reset condition is detected. Hence it is possible that a + * previous message wouldn't have got response. Allow to + * continue without polling for response status for priority + * messages. */ resp = RREG32(smu->resp_reg); dev_dbg(adev->dev, - "Sending RAS priority message %s response status: %x", + "Sending priority message %s response status: %x", smu_get_message_name(smu, msg), resp); if (resp == 0) *poll = false; diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.h b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.h index a608cdbdada4..d588f74b98de 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.h +++ b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.h @@ -65,6 +65,32 @@ header->structure_size = sizeof(*tmp); \ } while (0) +#define smu_cmn_init_baseboard_temp_metrics(ptr, fr, cr) \ + do { \ + typecheck(struct amdgpu_baseboard_temp_metrics_v##fr##_##cr *, \ + (ptr)); \ + struct amdgpu_baseboard_temp_metrics_v##fr##_##cr *tmp = (ptr); \ + struct metrics_table_header *header = \ + (struct metrics_table_header *)tmp; \ + memset(header, 0xFF, sizeof(*tmp)); \ + header->format_revision = fr; \ + header->content_revision = cr; \ + header->structure_size = sizeof(*tmp); \ + } while (0) + +#define smu_cmn_init_gpuboard_temp_metrics(ptr, fr, cr) \ + do { \ + typecheck(struct amdgpu_gpuboard_temp_metrics_v##fr##_##cr *, \ + (ptr)); \ + struct amdgpu_gpuboard_temp_metrics_v##fr##_##cr *tmp = (ptr); \ + struct metrics_table_header *header = \ + (struct metrics_table_header *)tmp; \ + memset(header, 0xFF, sizeof(*tmp)); \ + header->format_revision = fr; \ + header->content_revision = cr; \ + header->structure_size = sizeof(*tmp); \ + } while (0) + extern const int link_speed[]; /* Helper to Convert from PCIE Gen 1/2/3/4/5/6 to 0.1 GT/s speed units */ diff --git a/drivers/gpu/drm/drm_drv.c b/drivers/gpu/drm/drm_drv.c index cdd591b11488..0ac723a46a91 100644 --- a/drivers/gpu/drm/drm_drv.c +++ b/drivers/gpu/drm/drm_drv.c @@ -532,6 +532,8 @@ static const char *drm_get_wedge_recovery(unsigned int opt) return "rebind"; case DRM_WEDGE_RECOVERY_BUS_RESET: return "bus-reset"; + case DRM_WEDGE_RECOVERY_VENDOR: + return "vendor-specific"; default: return NULL; } diff --git a/drivers/gpu/drm/drm_gpusvm.c b/drivers/gpu/drm/drm_gpusvm.c index 647b49ff2da5..e2a9a6ae1d54 100644 --- a/drivers/gpu/drm/drm_gpusvm.c +++ b/drivers/gpu/drm/drm_gpusvm.c @@ -975,7 +975,7 @@ static void __drm_gpusvm_range_unmap_pages(struct drm_gpusvm *gpusvm, }; for (i = 0, j = 0; i < npages; j++) { - struct drm_pagemap_device_addr *addr = &range->dma_addr[j]; + struct drm_pagemap_addr *addr = &range->dma_addr[j]; if (addr->proto == DRM_INTERCONNECT_SYSTEM) dma_unmap_page(dev, @@ -1328,7 +1328,7 @@ map_pages: goto err_unmap; } - range->dma_addr[j] = drm_pagemap_device_addr_encode + range->dma_addr[j] = drm_pagemap_addr_encode (addr, DRM_INTERCONNECT_SYSTEM, order, DMA_BIDIRECTIONAL); } diff --git a/drivers/gpu/drm/drm_pagemap.c b/drivers/gpu/drm/drm_pagemap.c index 1da55322af12..22c44807e3fe 100644 --- a/drivers/gpu/drm/drm_pagemap.c +++ b/drivers/gpu/drm/drm_pagemap.c @@ -202,7 +202,7 @@ static void drm_pagemap_get_devmem_page(struct page *page, /** * drm_pagemap_migrate_map_pages() - Map migration pages for GPU SVM migration * @dev: The device for which the pages are being mapped - * @dma_addr: Array to store DMA addresses corresponding to mapped pages + * @pagemap_addr: Array to store DMA information corresponding to mapped pages * @migrate_pfn: Array of migrate page frame numbers to map * @npages: Number of pages to map * @dir: Direction of data transfer (e.g., DMA_BIDIRECTIONAL) @@ -215,25 +215,39 @@ static void drm_pagemap_get_devmem_page(struct page *page, * Returns: 0 on success, -EFAULT if an error occurs during mapping. */ static int drm_pagemap_migrate_map_pages(struct device *dev, - dma_addr_t *dma_addr, + struct drm_pagemap_addr *pagemap_addr, unsigned long *migrate_pfn, unsigned long npages, enum dma_data_direction dir) { unsigned long i; - for (i = 0; i < npages; ++i) { + for (i = 0; i < npages;) { struct page *page = migrate_pfn_to_page(migrate_pfn[i]); + dma_addr_t dma_addr; + struct folio *folio; + unsigned int order = 0; if (!page) - continue; + goto next; if (WARN_ON_ONCE(is_zone_device_page(page))) return -EFAULT; - dma_addr[i] = dma_map_page(dev, page, 0, PAGE_SIZE, dir); - if (dma_mapping_error(dev, dma_addr[i])) + folio = page_folio(page); + order = folio_order(folio); + + dma_addr = dma_map_page(dev, page, 0, page_size(page), dir); + if (dma_mapping_error(dev, dma_addr)) return -EFAULT; + + pagemap_addr[i] = + drm_pagemap_addr_encode(dma_addr, + DRM_INTERCONNECT_SYSTEM, + order, dir); + +next: + i += NR_PAGES(order); } return 0; @@ -242,7 +256,7 @@ static int drm_pagemap_migrate_map_pages(struct device *dev, /** * drm_pagemap_migrate_unmap_pages() - Unmap pages previously mapped for GPU SVM migration * @dev: The device for which the pages were mapped - * @dma_addr: Array of DMA addresses corresponding to mapped pages + * @pagemap_addr: Array of DMA information corresponding to mapped pages * @npages: Number of pages to unmap * @dir: Direction of data transfer (e.g., DMA_BIDIRECTIONAL) * @@ -251,17 +265,20 @@ static int drm_pagemap_migrate_map_pages(struct device *dev, * if it's valid and not already unmapped, and unmaps the corresponding page. */ static void drm_pagemap_migrate_unmap_pages(struct device *dev, - dma_addr_t *dma_addr, + struct drm_pagemap_addr *pagemap_addr, unsigned long npages, enum dma_data_direction dir) { unsigned long i; - for (i = 0; i < npages; ++i) { - if (!dma_addr[i] || dma_mapping_error(dev, dma_addr[i])) - continue; + for (i = 0; i < npages;) { + if (!pagemap_addr[i].addr || dma_mapping_error(dev, pagemap_addr[i].addr)) + goto next; - dma_unmap_page(dev, dma_addr[i], PAGE_SIZE, dir); + dma_unmap_page(dev, pagemap_addr[i].addr, PAGE_SIZE << pagemap_addr[i].order, dir); + +next: + i += NR_PAGES(pagemap_addr[i].order); } } @@ -314,7 +331,7 @@ int drm_pagemap_migrate_to_devmem(struct drm_pagemap_devmem *devmem_allocation, struct vm_area_struct *vas; struct drm_pagemap_zdd *zdd = NULL; struct page **pages; - dma_addr_t *dma_addr; + struct drm_pagemap_addr *pagemap_addr; void *buf; int err; @@ -340,14 +357,14 @@ int drm_pagemap_migrate_to_devmem(struct drm_pagemap_devmem *devmem_allocation, goto err_out; } - buf = kvcalloc(npages, 2 * sizeof(*migrate.src) + sizeof(*dma_addr) + + buf = kvcalloc(npages, 2 * sizeof(*migrate.src) + sizeof(*pagemap_addr) + sizeof(*pages), GFP_KERNEL); if (!buf) { err = -ENOMEM; goto err_out; } - dma_addr = buf + (2 * sizeof(*migrate.src) * npages); - pages = buf + (2 * sizeof(*migrate.src) + sizeof(*dma_addr)) * npages; + pagemap_addr = buf + (2 * sizeof(*migrate.src) * npages); + pages = buf + (2 * sizeof(*migrate.src) + sizeof(*pagemap_addr)) * npages; zdd = drm_pagemap_zdd_alloc(pgmap_owner); if (!zdd) { @@ -377,8 +394,9 @@ int drm_pagemap_migrate_to_devmem(struct drm_pagemap_devmem *devmem_allocation, if (err) goto err_finalize; - err = drm_pagemap_migrate_map_pages(devmem_allocation->dev, dma_addr, + err = drm_pagemap_migrate_map_pages(devmem_allocation->dev, pagemap_addr, migrate.src, npages, DMA_TO_DEVICE); + if (err) goto err_finalize; @@ -390,7 +408,7 @@ int drm_pagemap_migrate_to_devmem(struct drm_pagemap_devmem *devmem_allocation, drm_pagemap_get_devmem_page(page, zdd); } - err = ops->copy_to_devmem(pages, dma_addr, npages); + err = ops->copy_to_devmem(pages, pagemap_addr, npages); if (err) goto err_finalize; @@ -404,7 +422,7 @@ err_finalize: drm_pagemap_migration_unlock_put_pages(npages, migrate.dst); migrate_vma_pages(&migrate); migrate_vma_finalize(&migrate); - drm_pagemap_migrate_unmap_pages(devmem_allocation->dev, dma_addr, npages, + drm_pagemap_migrate_unmap_pages(devmem_allocation->dev, pagemap_addr, npages, DMA_TO_DEVICE); err_free: if (zdd) @@ -442,54 +460,80 @@ static int drm_pagemap_migrate_populate_ram_pfn(struct vm_area_struct *vas, { unsigned long i; - for (i = 0; i < npages; ++i, addr += PAGE_SIZE) { - struct page *page, *src_page; + for (i = 0; i < npages;) { + struct page *page = NULL, *src_page; + struct folio *folio; + unsigned int order = 0; if (!(src_mpfn[i] & MIGRATE_PFN_MIGRATE)) - continue; + goto next; src_page = migrate_pfn_to_page(src_mpfn[i]); if (!src_page) - continue; + goto next; if (fault_page) { if (src_page->zone_device_data != fault_page->zone_device_data) - continue; + goto next; } + order = folio_order(page_folio(src_page)); + + /* TODO: Support fallback to single pages if THP allocation fails */ if (vas) - page = alloc_page_vma(GFP_HIGHUSER, vas, addr); + folio = vma_alloc_folio(GFP_HIGHUSER, order, vas, addr); else - page = alloc_page(GFP_HIGHUSER); + folio = folio_alloc(GFP_HIGHUSER, order); - if (!page) + if (!folio) goto free_pages; + page = folio_page(folio, 0); mpfn[i] = migrate_pfn(page_to_pfn(page)); + +next: + if (page) + addr += page_size(page); + else + addr += PAGE_SIZE; + + i += NR_PAGES(order); } - for (i = 0; i < npages; ++i) { + for (i = 0; i < npages;) { struct page *page = migrate_pfn_to_page(mpfn[i]); + unsigned int order = 0; if (!page) - continue; + goto next_lock; - WARN_ON_ONCE(!trylock_page(page)); - ++*mpages; + WARN_ON_ONCE(!folio_trylock(page_folio(page))); + + order = folio_order(page_folio(page)); + *mpages += NR_PAGES(order); + +next_lock: + i += NR_PAGES(order); } return 0; free_pages: - for (i = 0; i < npages; ++i) { + for (i = 0; i < npages;) { struct page *page = migrate_pfn_to_page(mpfn[i]); + unsigned int order = 0; if (!page) - continue; + goto next_put; put_page(page); mpfn[i] = 0; + + order = folio_order(page_folio(page)); + +next_put: + i += NR_PAGES(order); } return -ENOMEM; } @@ -509,7 +553,7 @@ int drm_pagemap_evict_to_ram(struct drm_pagemap_devmem *devmem_allocation) unsigned long npages, mpages = 0; struct page **pages; unsigned long *src, *dst; - dma_addr_t *dma_addr; + struct drm_pagemap_addr *pagemap_addr; void *buf; int i, err = 0; unsigned int retry_count = 2; @@ -520,7 +564,7 @@ retry: if (!mmget_not_zero(devmem_allocation->mm)) return -EFAULT; - buf = kvcalloc(npages, 2 * sizeof(*src) + sizeof(*dma_addr) + + buf = kvcalloc(npages, 2 * sizeof(*src) + sizeof(*pagemap_addr) + sizeof(*pages), GFP_KERNEL); if (!buf) { err = -ENOMEM; @@ -528,8 +572,8 @@ retry: } src = buf; dst = buf + (sizeof(*src) * npages); - dma_addr = buf + (2 * sizeof(*src) * npages); - pages = buf + (2 * sizeof(*src) + sizeof(*dma_addr)) * npages; + pagemap_addr = buf + (2 * sizeof(*src) * npages); + pages = buf + (2 * sizeof(*src) + sizeof(*pagemap_addr)) * npages; err = ops->populate_devmem_pfn(devmem_allocation, npages, src); if (err) @@ -544,7 +588,7 @@ retry: if (err || !mpages) goto err_finalize; - err = drm_pagemap_migrate_map_pages(devmem_allocation->dev, dma_addr, + err = drm_pagemap_migrate_map_pages(devmem_allocation->dev, pagemap_addr, dst, npages, DMA_FROM_DEVICE); if (err) goto err_finalize; @@ -552,7 +596,7 @@ retry: for (i = 0; i < npages; ++i) pages[i] = migrate_pfn_to_page(src[i]); - err = ops->copy_to_ram(pages, dma_addr, npages); + err = ops->copy_to_ram(pages, pagemap_addr, npages); if (err) goto err_finalize; @@ -561,7 +605,7 @@ err_finalize: drm_pagemap_migration_unlock_put_pages(npages, dst); migrate_device_pages(src, dst, npages); migrate_device_finalize(src, dst, npages); - drm_pagemap_migrate_unmap_pages(devmem_allocation->dev, dma_addr, npages, + drm_pagemap_migrate_unmap_pages(devmem_allocation->dev, pagemap_addr, npages, DMA_FROM_DEVICE); err_free: kvfree(buf); @@ -612,7 +656,7 @@ static int __drm_pagemap_migrate_to_ram(struct vm_area_struct *vas, struct device *dev = NULL; unsigned long npages, mpages = 0; struct page **pages; - dma_addr_t *dma_addr; + struct drm_pagemap_addr *pagemap_addr; unsigned long start, end; void *buf; int i, err = 0; @@ -637,14 +681,14 @@ static int __drm_pagemap_migrate_to_ram(struct vm_area_struct *vas, migrate.end = end; npages = npages_in_range(start, end); - buf = kvcalloc(npages, 2 * sizeof(*migrate.src) + sizeof(*dma_addr) + + buf = kvcalloc(npages, 2 * sizeof(*migrate.src) + sizeof(*pagemap_addr) + sizeof(*pages), GFP_KERNEL); if (!buf) { err = -ENOMEM; goto err_out; } - dma_addr = buf + (2 * sizeof(*migrate.src) * npages); - pages = buf + (2 * sizeof(*migrate.src) + sizeof(*dma_addr)) * npages; + pagemap_addr = buf + (2 * sizeof(*migrate.src) * npages); + pages = buf + (2 * sizeof(*migrate.src) + sizeof(*pagemap_addr)) * npages; migrate.vma = vas; migrate.src = buf; @@ -680,7 +724,7 @@ static int __drm_pagemap_migrate_to_ram(struct vm_area_struct *vas, if (err) goto err_finalize; - err = drm_pagemap_migrate_map_pages(dev, dma_addr, migrate.dst, npages, + err = drm_pagemap_migrate_map_pages(dev, pagemap_addr, migrate.dst, npages, DMA_FROM_DEVICE); if (err) goto err_finalize; @@ -688,7 +732,7 @@ static int __drm_pagemap_migrate_to_ram(struct vm_area_struct *vas, for (i = 0; i < npages; ++i) pages[i] = migrate_pfn_to_page(migrate.src[i]); - err = ops->copy_to_ram(pages, dma_addr, npages); + err = ops->copy_to_ram(pages, pagemap_addr, npages); if (err) goto err_finalize; @@ -698,7 +742,7 @@ err_finalize: migrate_vma_pages(&migrate); migrate_vma_finalize(&migrate); if (dev) - drm_pagemap_migrate_unmap_pages(dev, dma_addr, npages, + drm_pagemap_migrate_unmap_pages(dev, pagemap_addr, npages, DMA_FROM_DEVICE); err_free: kvfree(buf); diff --git a/drivers/gpu/drm/i915/Kconfig.debug b/drivers/gpu/drm/i915/Kconfig.debug index 1852e0804942..3562a02ef7ad 100644 --- a/drivers/gpu/drm/i915/Kconfig.debug +++ b/drivers/gpu/drm/i915/Kconfig.debug @@ -50,7 +50,7 @@ config DRM_I915_DEBUG select DRM_VGEM # used by igt/prime_vgem (dmabuf interop checks) select DRM_DEBUG_MM if DRM=y select DRM_EXPORT_FOR_TESTS if m - select DRM_DEBUG_SELFTEST + select DRM_KUNIT_TEST if KUNIT select DMABUF_SELFTESTS select SW_SYNC # signaling validation framework (igt/syncobj*) select DRM_I915_WERROR diff --git a/drivers/gpu/drm/i915/display/g4x_dp.c b/drivers/gpu/drm/i915/display/g4x_dp.c index 87f6b9602b16..aa159f9ce12f 100644 --- a/drivers/gpu/drm/i915/display/g4x_dp.c +++ b/drivers/gpu/drm/i915/display/g4x_dp.c @@ -424,17 +424,6 @@ intel_dp_link_down(struct intel_encoder *encoder, drm_dbg_kms(display->drm, "\n"); - if ((display->platform.ivybridge && port == PORT_A) || - (HAS_PCH_CPT(display) && port != PORT_A)) { - intel_dp->DP &= ~DP_LINK_TRAIN_MASK_CPT; - intel_dp->DP |= DP_LINK_TRAIN_PAT_IDLE_CPT; - } else { - intel_dp->DP &= ~DP_LINK_TRAIN_MASK; - intel_dp->DP |= DP_LINK_TRAIN_PAT_IDLE; - } - intel_de_write(display, intel_dp->output_reg, intel_dp->DP); - intel_de_posting_read(display, intel_dp->output_reg); - intel_dp->DP &= ~DP_PORT_EN; intel_de_write(display, intel_dp->output_reg, intel_dp->DP); intel_de_posting_read(display, intel_dp->output_reg); @@ -612,6 +601,19 @@ cpt_set_link_train(struct intel_dp *intel_dp, } static void +cpt_set_idle_link_train(struct intel_dp *intel_dp, + const struct intel_crtc_state *crtc_state) +{ + struct intel_display *display = to_intel_display(intel_dp); + + intel_dp->DP &= ~DP_LINK_TRAIN_MASK_CPT; + intel_dp->DP |= DP_LINK_TRAIN_PAT_IDLE_CPT; + + intel_de_write(display, intel_dp->output_reg, intel_dp->DP); + intel_de_posting_read(display, intel_dp->output_reg); +} + +static void g4x_set_link_train(struct intel_dp *intel_dp, const struct intel_crtc_state *crtc_state, u8 dp_train_pat) @@ -639,6 +641,19 @@ g4x_set_link_train(struct intel_dp *intel_dp, intel_de_posting_read(display, intel_dp->output_reg); } +static void +g4x_set_idle_link_train(struct intel_dp *intel_dp, + const struct intel_crtc_state *crtc_state) +{ + struct intel_display *display = to_intel_display(intel_dp); + + intel_dp->DP &= ~DP_LINK_TRAIN_MASK; + intel_dp->DP |= DP_LINK_TRAIN_PAT_IDLE; + + intel_de_write(display, intel_dp->output_reg, intel_dp->DP); + intel_de_posting_read(display, intel_dp->output_reg); +} + static void intel_dp_enable_port(struct intel_dp *intel_dp, const struct intel_crtc_state *crtc_state) { @@ -1285,12 +1300,10 @@ bool g4x_dp_init(struct intel_display *display, drm_dbg_kms(display->drm, "No VBT child device for DP-%c\n", port_name(port)); - dig_port = kzalloc(sizeof(*dig_port), GFP_KERNEL); + dig_port = intel_dig_port_alloc(); if (!dig_port) return false; - dig_port->aux_ch = AUX_CH_NONE; - intel_connector = intel_connector_alloc(); if (!intel_connector) goto err_connector_alloc; @@ -1300,8 +1313,6 @@ bool g4x_dp_init(struct intel_display *display, intel_encoder->devdata = devdata; - mutex_init(&dig_port->hdcp.mutex); - if (drm_encoder_init(display->drm, &intel_encoder->base, &intel_dp_enc_funcs, DRM_MODE_ENCODER_TMDS, "DP %c", port_name(port))) @@ -1342,10 +1353,13 @@ bool g4x_dp_init(struct intel_display *display, intel_encoder->audio_disable = g4x_dp_audio_disable; if ((display->platform.ivybridge && port == PORT_A) || - (HAS_PCH_CPT(display) && port != PORT_A)) + (HAS_PCH_CPT(display) && port != PORT_A)) { dig_port->dp.set_link_train = cpt_set_link_train; - else + dig_port->dp.set_idle_link_train = cpt_set_idle_link_train; + } else { dig_port->dp.set_link_train = g4x_set_link_train; + dig_port->dp.set_idle_link_train = g4x_set_idle_link_train; + } if (display->platform.cherryview) intel_encoder->set_signal_levels = chv_set_signal_levels; @@ -1368,7 +1382,6 @@ bool g4x_dp_init(struct intel_display *display, } dig_port->dp.output_reg = output_reg; - dig_port->max_lanes = 4; intel_encoder->type = INTEL_OUTPUT_DP; intel_encoder->power_domain = intel_display_power_ddi_lanes_domain(display, port); diff --git a/drivers/gpu/drm/i915/display/g4x_hdmi.c b/drivers/gpu/drm/i915/display/g4x_hdmi.c index 2610f5702fb9..f6e2d1ed5639 100644 --- a/drivers/gpu/drm/i915/display/g4x_hdmi.c +++ b/drivers/gpu/drm/i915/display/g4x_hdmi.c @@ -19,7 +19,7 @@ #include "intel_display_types.h" #include "intel_dp_aux.h" #include "intel_dpio_phy.h" -#include "intel_fdi.h" +#include "intel_encoder.h" #include "intel_fifo_underrun.h" #include "intel_hdmi.h" #include "intel_hotplug.h" @@ -135,11 +135,8 @@ static int g4x_hdmi_compute_config(struct intel_encoder *encoder, struct intel_atomic_state *state = to_intel_atomic_state(crtc_state->uapi.state); struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); - if (HAS_PCH_SPLIT(display)) { + if (HAS_PCH_SPLIT(display)) crtc_state->has_pch_encoder = true; - if (!intel_fdi_compute_pipe_bpp(crtc_state)) - return -EINVAL; - } if (display->platform.g4x) crtc_state->has_hdmi_sink = g4x_compute_has_hdmi_sink(state, crtc); @@ -690,12 +687,10 @@ bool g4x_hdmi_init(struct intel_display *display, drm_dbg_kms(display->drm, "No VBT child device for HDMI-%c\n", port_name(port)); - dig_port = kzalloc(sizeof(*dig_port), GFP_KERNEL); + dig_port = intel_dig_port_alloc(); if (!dig_port) return false; - dig_port->aux_ch = AUX_CH_NONE; - intel_connector = intel_connector_alloc(); if (!intel_connector) goto err_connector_alloc; @@ -704,8 +699,6 @@ bool g4x_hdmi_init(struct intel_display *display, intel_encoder->devdata = devdata; - mutex_init(&dig_port->hdcp.mutex); - if (drm_encoder_init(display->drm, &intel_encoder->base, &intel_hdmi_enc_funcs, DRM_MODE_ENCODER_TMDS, "HDMI %c", port_name(port))) @@ -767,8 +760,6 @@ bool g4x_hdmi_init(struct intel_display *display, intel_encoder->cloneable |= BIT(INTEL_OUTPUT_HDMI); dig_port->hdmi.hdmi_reg = hdmi_reg; - dig_port->dp.output_reg = INVALID_MMIO_REG; - dig_port->max_lanes = 4; intel_infoframe_init(dig_port); diff --git a/drivers/gpu/drm/i915/display/i9xx_plane.c b/drivers/gpu/drm/i915/display/i9xx_plane.c index f291ced989dc..3eb96d8abba8 100644 --- a/drivers/gpu/drm/i915/display/i9xx_plane.c +++ b/drivers/gpu/drm/i915/display/i9xx_plane.c @@ -155,8 +155,7 @@ static bool i9xx_plane_has_windowing(struct intel_plane *plane) i9xx_plane == PLANE_C; } -static u32 i9xx_plane_ctl(const struct intel_crtc_state *crtc_state, - const struct intel_plane_state *plane_state) +static u32 i9xx_plane_ctl(const struct intel_plane_state *plane_state) { struct intel_display *display = to_intel_display(plane_state); const struct drm_framebuffer *fb = plane_state->hw.fb; @@ -355,11 +354,24 @@ i9xx_plane_check(struct intel_crtc_state *crtc_state, if (ret) return ret; - plane_state->ctl = i9xx_plane_ctl(crtc_state, plane_state); + plane_state->ctl = i9xx_plane_ctl(plane_state); return 0; } +static u32 i8xx_plane_surf_offset(const struct intel_plane_state *plane_state) +{ + int x = plane_state->view.color_plane[0].x; + int y = plane_state->view.color_plane[0].y; + + return intel_fb_xy_to_linear(x, y, plane_state, 0); +} + +u32 i965_plane_surf_offset(const struct intel_plane_state *plane_state) +{ + return plane_state->view.color_plane[0].offset; +} + static u32 i9xx_plane_ctl_crtc(const struct intel_crtc_state *crtc_state) { struct intel_display *display = to_intel_display(crtc_state); @@ -463,7 +475,7 @@ static void i9xx_plane_update_arm(struct intel_dsb *dsb, enum i9xx_plane_id i9xx_plane = plane->i9xx_plane; int x = plane_state->view.color_plane[0].x; int y = plane_state->view.color_plane[0].y; - u32 dspcntr, dspaddr_offset, linear_offset; + u32 dspcntr; dspcntr = plane_state->ctl | i9xx_plane_ctl_crtc(crtc_state); @@ -472,13 +484,6 @@ static void i9xx_plane_update_arm(struct intel_dsb *dsb, crtc_state->async_flip_planes & BIT(plane->id)) dspcntr |= DISP_ASYNC_FLIP; - linear_offset = intel_fb_xy_to_linear(x, y, plane_state, 0); - - if (DISPLAY_VER(display) >= 4) - dspaddr_offset = plane_state->view.color_plane[0].offset; - else - dspaddr_offset = linear_offset; - if (display->platform.cherryview && i9xx_plane == PLANE_B) { int crtc_x = plane_state->uapi.dst.x1; int crtc_y = plane_state->uapi.dst.y1; @@ -498,7 +503,7 @@ static void i9xx_plane_update_arm(struct intel_dsb *dsb, DISP_OFFSET_Y(y) | DISP_OFFSET_X(x)); } else if (DISPLAY_VER(display) >= 4) { intel_de_write_fw(display, DSPLINOFF(display, i9xx_plane), - linear_offset); + intel_fb_xy_to_linear(x, y, plane_state, 0)); intel_de_write_fw(display, DSPTILEOFF(display, i9xx_plane), DISP_OFFSET_Y(y) | DISP_OFFSET_X(x)); } @@ -511,11 +516,9 @@ static void i9xx_plane_update_arm(struct intel_dsb *dsb, intel_de_write_fw(display, DSPCNTR(display, i9xx_plane), dspcntr); if (DISPLAY_VER(display) >= 4) - intel_de_write_fw(display, DSPSURF(display, i9xx_plane), - intel_plane_ggtt_offset(plane_state) + dspaddr_offset); + intel_de_write_fw(display, DSPSURF(display, i9xx_plane), plane_state->surf); else - intel_de_write_fw(display, DSPADDR(display, i9xx_plane), - intel_plane_ggtt_offset(plane_state) + dspaddr_offset); + intel_de_write_fw(display, DSPADDR(display, i9xx_plane), plane_state->surf); } static void i830_plane_update_arm(struct intel_dsb *dsb, @@ -604,16 +607,13 @@ g4x_primary_async_flip(struct intel_dsb *dsb, { struct intel_display *display = to_intel_display(plane); u32 dspcntr = plane_state->ctl | i9xx_plane_ctl_crtc(crtc_state); - u32 dspaddr_offset = plane_state->view.color_plane[0].offset; enum i9xx_plane_id i9xx_plane = plane->i9xx_plane; if (async_flip) dspcntr |= DISP_ASYNC_FLIP; intel_de_write_fw(display, DSPCNTR(display, i9xx_plane), dspcntr); - - intel_de_write_fw(display, DSPSURF(display, i9xx_plane), - intel_plane_ggtt_offset(plane_state) + dspaddr_offset); + intel_de_write_fw(display, DSPSURF(display, i9xx_plane), plane_state->surf); } static void @@ -624,11 +624,9 @@ vlv_primary_async_flip(struct intel_dsb *dsb, bool async_flip) { struct intel_display *display = to_intel_display(plane); - u32 dspaddr_offset = plane_state->view.color_plane[0].offset; enum i9xx_plane_id i9xx_plane = plane->i9xx_plane; - intel_de_write_fw(display, DSPADDR_VLV(display, i9xx_plane), - intel_plane_ggtt_offset(plane_state) + dspaddr_offset); + intel_de_write_fw(display, DSPADDR_VLV(display, i9xx_plane), plane_state->surf); } static void @@ -1037,6 +1035,11 @@ intel_primary_plane_create(struct intel_display *display, enum pipe pipe) plane->get_hw_state = i9xx_plane_get_hw_state; plane->check_plane = i9xx_plane_check; + if (DISPLAY_VER(display) >= 4) + plane->surf_offset = i965_plane_surf_offset; + else + plane->surf_offset = i8xx_plane_surf_offset; + if (DISPLAY_VER(display) >= 5 || display->platform.g4x) plane->capture_error = g4x_primary_capture_error; else if (DISPLAY_VER(display) >= 4) @@ -1254,24 +1257,21 @@ bool i9xx_fixup_initial_plane_config(struct intel_crtc *crtc, const struct intel_plane_state *plane_state = to_intel_plane_state(plane->base.state); enum i9xx_plane_id i9xx_plane = plane->i9xx_plane; - u32 base; if (!plane_state->uapi.visible) return false; - base = intel_plane_ggtt_offset(plane_state); - /* * We may have moved the surface to a different * part of ggtt, make the plane aware of that. */ - if (plane_config->base == base) + if (plane_config->base == plane_state->surf) return false; if (DISPLAY_VER(display) >= 4) - intel_de_write(display, DSPSURF(display, i9xx_plane), base); + intel_de_write(display, DSPSURF(display, i9xx_plane), plane_state->surf); else - intel_de_write(display, DSPADDR(display, i9xx_plane), base); + intel_de_write(display, DSPADDR(display, i9xx_plane), plane_state->surf); return true; } diff --git a/drivers/gpu/drm/i915/display/i9xx_plane.h b/drivers/gpu/drm/i915/display/i9xx_plane.h index d90546d60855..565dab751301 100644 --- a/drivers/gpu/drm/i915/display/i9xx_plane.h +++ b/drivers/gpu/drm/i915/display/i9xx_plane.h @@ -24,6 +24,7 @@ unsigned int vlv_plane_min_alignment(struct intel_plane *plane, const struct drm_framebuffer *fb, int colot_plane); int i9xx_check_plane_surface(struct intel_plane_state *plane_state); +u32 i965_plane_surf_offset(const struct intel_plane_state *plane_state); struct intel_plane * intel_primary_plane_create(struct intel_display *display, enum pipe pipe); diff --git a/drivers/gpu/drm/i915/display/i9xx_wm.c b/drivers/gpu/drm/i915/display/i9xx_wm.c index 1f9db5118777..fd3b7b35f351 100644 --- a/drivers/gpu/drm/i915/display/i9xx_wm.c +++ b/drivers/gpu/drm/i915/display/i9xx_wm.c @@ -3,6 +3,10 @@ * Copyright © 2023 Intel Corporation */ +#include <linux/iopoll.h> + +#include "soc/intel_dram.h" + #include "i915_drv.h" #include "i915_reg.h" #include "i9xx_wm.h" @@ -85,7 +89,8 @@ static const struct cxsr_latency cxsr_latency_table[] = { static const struct cxsr_latency *pnv_get_cxsr_latency(struct intel_display *display) { - struct drm_i915_private *i915 = to_i915(display->drm); + const struct dram_info *dram_info = intel_dram_info(display->drm); + bool is_ddr3 = dram_info->type == INTEL_DRAM_DDR3; int i; for (i = 0; i < ARRAY_SIZE(cxsr_latency_table); i++) { @@ -93,15 +98,16 @@ static const struct cxsr_latency *pnv_get_cxsr_latency(struct intel_display *dis bool is_desktop = !display->platform.mobile; if (is_desktop == latency->is_desktop && - i915->is_ddr3 == latency->is_ddr3 && - DIV_ROUND_CLOSEST(i915->fsb_freq, 1000) == latency->fsb_freq && - DIV_ROUND_CLOSEST(i915->mem_freq, 1000) == latency->mem_freq) + is_ddr3 == latency->is_ddr3 && + DIV_ROUND_CLOSEST(dram_info->fsb_freq, 1000) == latency->fsb_freq && + DIV_ROUND_CLOSEST(dram_info->mem_freq, 1000) == latency->mem_freq) return latency; } drm_dbg_kms(display->drm, - "Could not find CxSR latency for DDR%s, FSB %u kHz, MEM %u kHz\n", - i915->is_ddr3 ? "3" : "2", i915->fsb_freq, i915->mem_freq); + "Could not find CxSR latency for %s, FSB %u kHz, MEM %u kHz\n", + intel_dram_type_str(dram_info->type), + dram_info->fsb_freq, dram_info->mem_freq); return NULL; } @@ -109,6 +115,7 @@ static const struct cxsr_latency *pnv_get_cxsr_latency(struct intel_display *dis static void chv_set_memory_dvfs(struct intel_display *display, bool enable) { u32 val; + int ret; vlv_punit_get(display->drm); @@ -121,8 +128,10 @@ static void chv_set_memory_dvfs(struct intel_display *display, bool enable) val |= FORCE_DDR_FREQ_REQ_ACK; vlv_punit_write(display->drm, PUNIT_REG_DDR_SETUP2, val); - if (wait_for((vlv_punit_read(display->drm, PUNIT_REG_DDR_SETUP2) & - FORCE_DDR_FREQ_REQ_ACK) == 0, 3)) + ret = poll_timeout_us(val = vlv_punit_read(display->drm, PUNIT_REG_DDR_SETUP2), + (val & FORCE_DDR_FREQ_REQ_ACK) == 0, + 500, 3000, false); + if (ret) drm_err(display->drm, "timed out waiting for Punit DDR DVFS request\n"); @@ -3902,6 +3911,7 @@ static void vlv_wm_get_hw_state(struct intel_display *display) struct vlv_wm_values *wm = &display->wm.vlv; struct intel_crtc *crtc; u32 val; + int ret; vlv_read_wm_values(display, wm); @@ -3928,8 +3938,10 @@ static void vlv_wm_get_hw_state(struct intel_display *display) val |= FORCE_DDR_FREQ_REQ_ACK; vlv_punit_write(display->drm, PUNIT_REG_DDR_SETUP2, val); - if (wait_for((vlv_punit_read(display->drm, PUNIT_REG_DDR_SETUP2) & - FORCE_DDR_FREQ_REQ_ACK) == 0, 3)) { + ret = poll_timeout_us(val = vlv_punit_read(display->drm, PUNIT_REG_DDR_SETUP2), + (val & FORCE_DDR_FREQ_REQ_ACK) == 0, + 500, 3000, false); + if (ret) { drm_dbg_kms(display->drm, "Punit not acking DDR DVFS request, " "assuming DDR DVFS is disabled\n"); diff --git a/drivers/gpu/drm/i915/display/icl_dsi.c b/drivers/gpu/drm/i915/display/icl_dsi.c index 8d9cb73a93a7..37faa8f19f6e 100644 --- a/drivers/gpu/drm/i915/display/icl_dsi.c +++ b/drivers/gpu/drm/i915/display/icl_dsi.c @@ -25,6 +25,8 @@ * Jani Nikula <jani.nikula@intel.com> */ +#include <linux/iopoll.h> + #include <drm/display/drm_dsc_helper.h> #include <drm/drm_atomic_helper.h> #include <drm/drm_fixed.h> @@ -72,8 +74,12 @@ static int payload_credits_available(struct intel_display *display, static bool wait_for_header_credits(struct intel_display *display, enum transcoder dsi_trans, int hdr_credit) { - if (wait_for_us(header_credits_available(display, dsi_trans) >= - hdr_credit, 100)) { + int ret, available; + + ret = poll_timeout_us(available = header_credits_available(display, dsi_trans), + available >= hdr_credit, + 10, 100, false); + if (ret) { drm_err(display->drm, "DSI header credits not released\n"); return false; } @@ -84,8 +90,12 @@ static bool wait_for_header_credits(struct intel_display *display, static bool wait_for_payload_credits(struct intel_display *display, enum transcoder dsi_trans, int payld_credit) { - if (wait_for_us(payload_credits_available(display, dsi_trans) >= - payld_credit, 100)) { + int ret, available; + + ret = poll_timeout_us(available = payload_credits_available(display, dsi_trans), + available >= payld_credit, + 10, 100, false); + if (ret) { drm_err(display->drm, "DSI payload credits not released\n"); return false; } @@ -137,8 +147,11 @@ static void wait_for_cmds_dispatched_to_panel(struct intel_encoder *encoder) /* wait for LP TX in progress bit to be cleared */ for_each_dsi_port(port, intel_dsi->ports) { dsi_trans = dsi_port_to_transcoder(port); - if (wait_for_us(!(intel_de_read(display, DSI_LP_MSG(dsi_trans)) & - LPTX_IN_PROGRESS), 20)) + + ret = intel_de_wait_custom(display, DSI_LP_MSG(dsi_trans), + LPTX_IN_PROGRESS, 0, + 20, 0, NULL); + if (ret) drm_err(display->drm, "LPTX bit not cleared\n"); } } @@ -516,13 +529,15 @@ static void gen11_dsi_enable_ddi_buffer(struct intel_encoder *encoder) struct intel_display *display = to_intel_display(encoder); struct intel_dsi *intel_dsi = enc_to_intel_dsi(encoder); enum port port; + int ret; for_each_dsi_port(port, intel_dsi->ports) { intel_de_rmw(display, DDI_BUF_CTL(port), 0, DDI_BUF_CTL_ENABLE); - if (wait_for_us(!(intel_de_read(display, DDI_BUF_CTL(port)) & - DDI_BUF_IS_IDLE), - 500)) + ret = intel_de_wait_custom(display, DDI_BUF_CTL(port), + DDI_BUF_IS_IDLE, 0, + 500, 0, NULL); + if (ret) drm_err(display->drm, "DDI port:%c buffer idle\n", port_name(port)); } @@ -838,9 +853,14 @@ gen11_dsi_configure_transcoder(struct intel_encoder *encoder, /* wait for link ready */ for_each_dsi_port(port, intel_dsi->ports) { + int ret; + dsi_trans = dsi_port_to_transcoder(port); - if (wait_for_us((intel_de_read(display, DSI_TRANS_FUNC_CONF(dsi_trans)) & - LINK_READY), 2500)) + + ret = intel_de_wait_custom(display, DSI_TRANS_FUNC_CONF(dsi_trans), + LINK_READY, LINK_READY, + 2500, 0, NULL); + if (ret) drm_err(display->drm, "DSI link not ready\n"); } } @@ -1321,6 +1341,7 @@ static void gen11_dsi_deconfigure_trancoder(struct intel_encoder *encoder) enum port port; enum transcoder dsi_trans; u32 tmp; + int ret; /* disable periodic update mode */ if (is_cmd_mode(intel_dsi)) { @@ -1337,9 +1358,10 @@ static void gen11_dsi_deconfigure_trancoder(struct intel_encoder *encoder) tmp &= ~LINK_ULPS_TYPE_LP11; intel_de_write(display, DSI_LP_MSG(dsi_trans), tmp); - if (wait_for_us((intel_de_read(display, DSI_LP_MSG(dsi_trans)) & - LINK_IN_ULPS), - 10)) + ret = intel_de_wait_custom(display, DSI_LP_MSG(dsi_trans), + LINK_IN_ULPS, LINK_IN_ULPS, + 10, 0, NULL); + if (ret) drm_err(display->drm, "DSI link not in ULPS\n"); } @@ -1367,14 +1389,17 @@ static void gen11_dsi_disable_port(struct intel_encoder *encoder) struct intel_display *display = to_intel_display(encoder); struct intel_dsi *intel_dsi = enc_to_intel_dsi(encoder); enum port port; + int ret; gen11_dsi_ungate_clocks(encoder); for_each_dsi_port(port, intel_dsi->ports) { intel_de_rmw(display, DDI_BUF_CTL(port), DDI_BUF_CTL_ENABLE, 0); - if (wait_for_us((intel_de_read(display, DDI_BUF_CTL(port)) & - DDI_BUF_IS_IDLE), - 8)) + ret = intel_de_wait_custom(display, DDI_BUF_CTL(port), + DDI_BUF_IS_IDLE, DDI_BUF_IS_IDLE, + 8, 0, NULL); + + if (ret) drm_err(display->drm, "DDI port:%c buffer not idle\n", port_name(port)); diff --git a/drivers/gpu/drm/i915/display/intel_alpm.c b/drivers/gpu/drm/i915/display/intel_alpm.c index dfdde8e4eabe..ed7a7ed486b5 100644 --- a/drivers/gpu/drm/i915/display/intel_alpm.c +++ b/drivers/gpu/drm/i915/display/intel_alpm.c @@ -16,6 +16,14 @@ #include "intel_psr.h" #include "intel_psr_regs.h" +#define SILENCE_PERIOD_MIN_TIME 80 +#define SILENCE_PERIOD_MAX_TIME 180 +#define SILENCE_PERIOD_TIME (SILENCE_PERIOD_MIN_TIME + \ + (SILENCE_PERIOD_MAX_TIME - \ + SILENCE_PERIOD_MIN_TIME) / 2) + +#define LFPS_CYCLE_COUNT 10 + bool intel_alpm_aux_wake_supported(struct intel_dp *intel_dp) { return intel_dp->alpm_dpcd & DP_ALPM_CAP; @@ -44,72 +52,49 @@ void intel_alpm_init(struct intel_dp *intel_dp) mutex_init(&intel_dp->alpm_parameters.lock); } -/* - * See Bspec: 71632 for the table - * - * Silence_period = tSilence,Min + ((tSilence,Max - tSilence,Min) / 2) - * - * Half cycle duration: - * - * Link rates 1.62 - 4.32 and tLFPS_Cycle = 70 ns - * FLOOR( (Link Rate * tLFPS_Cycle) / (2 * 10) ) - * - * Link rates 5.4 - 8.1 - * PORT_ALPM_LFPS_CTL[ LFPS Cycle Count ] = 10 - * LFPS Period chosen is the mid-point of the min:max values from the table - * FLOOR( LFPS Period in Symbol clocks / - * (2 * PORT_ALPM_LFPS_CTL[ LFPS Cycle Count ]) ) - */ -static bool _lnl_get_silence_period_and_lfps_half_cycle(int link_rate, - int *silence_period, - int *lfps_half_cycle) +static int get_silence_period_symbols(const struct intel_crtc_state *crtc_state) { - switch (link_rate) { - case 162000: - *silence_period = 20; - *lfps_half_cycle = 5; - break; - case 216000: - *silence_period = 27; - *lfps_half_cycle = 7; - break; - case 243000: - *silence_period = 31; - *lfps_half_cycle = 8; - break; - case 270000: - *silence_period = 34; - *lfps_half_cycle = 9; - break; - case 324000: - *silence_period = 41; - *lfps_half_cycle = 11; - break; - case 432000: - *silence_period = 56; - *lfps_half_cycle = 15; - break; - case 540000: - *silence_period = 69; - *lfps_half_cycle = 12; - break; - case 648000: - *silence_period = 84; - *lfps_half_cycle = 15; - break; - case 675000: - *silence_period = 87; - *lfps_half_cycle = 15; - break; - case 810000: - *silence_period = 104; - *lfps_half_cycle = 19; - break; - default: - *silence_period = *lfps_half_cycle = -1; - return false; + return SILENCE_PERIOD_TIME * intel_dp_link_symbol_clock(crtc_state->port_clock) / + 1000 / 1000; +} + +static int get_lfps_cycle_min_max_time(const struct intel_crtc_state *crtc_state, + int *min, int *max) +{ + if (crtc_state->port_clock < 540000) { + *min = 65 * LFPS_CYCLE_COUNT; + *max = 75 * LFPS_CYCLE_COUNT; + } else if (crtc_state->port_clock <= 810000) { + *min = 140; + *max = 800; + } else { + *min = *max = -1; + return -1; } - return true; + + return 0; +} + +static int get_lfps_cycle_time(const struct intel_crtc_state *crtc_state) +{ + int tlfps_cycle_min, tlfps_cycle_max, ret; + + ret = get_lfps_cycle_min_max_time(crtc_state, &tlfps_cycle_min, + &tlfps_cycle_max); + if (ret) + return ret; + + return tlfps_cycle_min + (tlfps_cycle_max - tlfps_cycle_min) / 2; +} + +static int get_lfps_half_cycle_clocks(const struct intel_crtc_state *crtc_state) +{ + int lfps_cycle_time = get_lfps_cycle_time(crtc_state); + + if (lfps_cycle_time < 0) + return -1; + + return lfps_cycle_time * crtc_state->port_clock / 1000 / 1000 / (2 * LFPS_CYCLE_COUNT); } /* @@ -131,21 +116,19 @@ static bool _lnl_get_silence_period_and_lfps_half_cycle(int link_rate, * tML_PHY_LOCK = TPS4 Length * ( 10 / (Link Rate in MHz) ) * TPS4 Length = 252 Symbols */ -static int _lnl_compute_aux_less_wake_time(int port_clock) +static int _lnl_compute_aux_less_wake_time(const struct intel_crtc_state *crtc_state) { int tphy2_p2_to_p0 = 12 * 1000; - int tlfps_period_max = 800; - int tsilence_max = 180; int t1 = 50 * 1000; int tps4 = 252; /* port_clock is link rate in 10kbit/s units */ - int tml_phy_lock = 1000 * 1000 * tps4 / port_clock; + int tml_phy_lock = 1000 * 1000 * tps4 / crtc_state->port_clock; int num_ml_phy_lock = 7 + DIV_ROUND_UP(6500, tml_phy_lock) + 1; int t2 = num_ml_phy_lock * tml_phy_lock; int tcds = 1 * t2; - return DIV_ROUND_UP(tphy2_p2_to_p0 + tlfps_period_max + tsilence_max + - t1 + tcds, 1000); + return DIV_ROUND_UP(tphy2_p2_to_p0 + get_lfps_cycle_time(crtc_state) + + SILENCE_PERIOD_TIME + t1 + tcds, 1000); } static int @@ -157,13 +140,13 @@ _lnl_compute_aux_less_alpm_params(struct intel_dp *intel_dp, lfps_half_cycle; aux_less_wake_time = - _lnl_compute_aux_less_wake_time(crtc_state->port_clock); + _lnl_compute_aux_less_wake_time(crtc_state); aux_less_wake_lines = intel_usecs_to_scanlines(&crtc_state->hw.adjusted_mode, aux_less_wake_time); + silence_period = get_silence_period_symbols(crtc_state); - if (!_lnl_get_silence_period_and_lfps_half_cycle(crtc_state->port_clock, - &silence_period, - &lfps_half_cycle)) + lfps_half_cycle = get_lfps_half_cycle_clocks(crtc_state); + if (lfps_half_cycle < 0) return false; if (aux_less_wake_lines > ALPM_CTL_AUX_LESS_WAKE_TIME_MASK || @@ -406,7 +389,7 @@ void intel_alpm_port_configure(struct intel_dp *intel_dp, PORT_ALPM_CTL_MAX_PHY_SWING_HOLD(0) | PORT_ALPM_CTL_SILENCE_PERIOD( intel_dp->alpm_parameters.silence_period_sym_clocks); - lfps_ctl_val = PORT_ALPM_LFPS_CTL_LFPS_CYCLE_COUNT(10) | + lfps_ctl_val = PORT_ALPM_LFPS_CTL_LFPS_CYCLE_COUNT(LFPS_CYCLE_COUNT) | PORT_ALPM_LFPS_CTL_LFPS_HALF_CYCLE_DURATION( intel_dp->alpm_parameters.lfps_half_cycle_num_of_syms) | PORT_ALPM_LFPS_CTL_FIRST_LFPS_HALF_CYCLE_DURATION( diff --git a/drivers/gpu/drm/i915/display/intel_backlight.c b/drivers/gpu/drm/i915/display/intel_backlight.c index e007380e9a63..3b14f929825a 100644 --- a/drivers/gpu/drm/i915/display/intel_backlight.c +++ b/drivers/gpu/drm/i915/display/intel_backlight.c @@ -236,7 +236,8 @@ static void i9xx_set_backlight(const struct drm_connector_state *conn_state, u32 struct intel_panel *panel = &connector->panel; u32 tmp, mask; - drm_WARN_ON(display->drm, panel->backlight.pwm_level_max == 0); + if (drm_WARN_ON(display->drm, panel->backlight.pwm_level_max == 0)) + return; if (panel->backlight.combination_mode) { struct pci_dev *pdev = to_pci_dev(display->drm->dev); diff --git a/drivers/gpu/drm/i915/display/intel_bios.c b/drivers/gpu/drm/i915/display/intel_bios.c index 9c268bed091d..3596dce84c28 100644 --- a/drivers/gpu/drm/i915/display/intel_bios.c +++ b/drivers/gpu/drm/i915/display/intel_bios.c @@ -36,6 +36,7 @@ #include "soc/intel_rom.h" #include "i915_drv.h" +#include "i915_utils.h" #include "intel_display.h" #include "intel_display_core.h" #include "intel_display_rpm.h" @@ -1566,10 +1567,7 @@ parse_psr(struct intel_display *display, panel->vbt.psr.full_link = psr_table->full_link; panel->vbt.psr.require_aux_wakeup = psr_table->require_aux_to_wakeup; - - /* Allowed VBT values goes from 0 to 15 */ - panel->vbt.psr.idle_frames = psr_table->idle_frames < 0 ? 0 : - psr_table->idle_frames > 15 ? 15 : psr_table->idle_frames; + panel->vbt.psr.idle_frames = psr_table->idle_frames; /* * New psr options 0=500us, 1=100us, 2=2500us, 3=0us @@ -2480,6 +2478,25 @@ static int parse_bdb_216_dp_max_link_rate(const int vbt_max_link_rate) } } +static u32 edp_rate_override_mask(int rate) +{ + switch (rate) { + case 2000000: return BDB_263_VBT_EDP_LINK_RATE_20; + case 1350000: return BDB_263_VBT_EDP_LINK_RATE_13_5; + case 1000000: return BDB_263_VBT_EDP_LINK_RATE_10; + case 810000: return BDB_263_VBT_EDP_LINK_RATE_8_1; + case 675000: return BDB_263_VBT_EDP_LINK_RATE_6_75; + case 540000: return BDB_263_VBT_EDP_LINK_RATE_5_4; + case 432000: return BDB_263_VBT_EDP_LINK_RATE_4_32; + case 324000: return BDB_263_VBT_EDP_LINK_RATE_3_24; + case 270000: return BDB_263_VBT_EDP_LINK_RATE_2_7; + case 243000: return BDB_263_VBT_EDP_LINK_RATE_2_43; + case 216000: return BDB_263_VBT_EDP_LINK_RATE_2_16; + case 162000: return BDB_263_VBT_EDP_LINK_RATE_1_62; + default: return 0; + } +} + int intel_bios_dp_max_link_rate(const struct intel_bios_encoder_data *devdata) { if (!devdata || devdata->display->vbt.version < 216) @@ -2499,6 +2516,19 @@ int intel_bios_dp_max_lane_count(const struct intel_bios_encoder_data *devdata) return devdata->child.dp_max_lane_count + 1; } +bool +intel_bios_encoder_reject_edp_rate(const struct intel_bios_encoder_data *devdata, + int rate) +{ + if (!devdata || devdata->display->vbt.version < 263) + return false; + + if (devdata->child.edp_data_rate_override == BDB_263_VBT_EDP_RATES_MASK) + return false; + + return devdata->child.edp_data_rate_override & edp_rate_override_mask(rate); +} + static void sanitize_device_type(struct intel_bios_encoder_data *devdata, enum port port) { @@ -2747,8 +2777,10 @@ static int child_device_expected_size(u16 version) { BUILD_BUG_ON(sizeof(struct child_device_config) < 40); - if (version > 256) + if (version > 263) return -ENOENT; + else if (version >= 263) + return 44; else if (version >= 256) return 40; else if (version >= 216) @@ -3743,8 +3775,6 @@ DEFINE_SHOW_ATTRIBUTE(intel_bios_vbt); void intel_bios_debugfs_register(struct intel_display *display) { - struct drm_minor *minor = display->drm->primary; - - debugfs_create_file("i915_vbt", 0444, minor->debugfs_root, + debugfs_create_file("i915_vbt", 0444, display->drm->debugfs_root, display, &intel_bios_vbt_fops); } diff --git a/drivers/gpu/drm/i915/display/intel_bios.h b/drivers/gpu/drm/i915/display/intel_bios.h index 6cd7a011b8c4..f9e438b2787b 100644 --- a/drivers/gpu/drm/i915/display/intel_bios.h +++ b/drivers/gpu/drm/i915/display/intel_bios.h @@ -50,180 +50,6 @@ enum intel_backlight_type { INTEL_BACKLIGHT_VESA_EDP_AUX_INTERFACE, }; -/* - * MIPI Sequence Block definitions - * - * Note the VBT spec has AssertReset / DeassertReset swapped from their - * usual naming, we use the proper names here to avoid confusion when - * reading the code. - */ -enum mipi_seq { - MIPI_SEQ_END = 0, - MIPI_SEQ_DEASSERT_RESET, /* Spec says MipiAssertResetPin */ - MIPI_SEQ_INIT_OTP, - MIPI_SEQ_DISPLAY_ON, - MIPI_SEQ_DISPLAY_OFF, - MIPI_SEQ_ASSERT_RESET, /* Spec says MipiDeassertResetPin */ - MIPI_SEQ_BACKLIGHT_ON, /* sequence block v2+ */ - MIPI_SEQ_BACKLIGHT_OFF, /* sequence block v2+ */ - MIPI_SEQ_TEAR_ON, /* sequence block v2+ */ - MIPI_SEQ_TEAR_OFF, /* sequence block v3+ */ - MIPI_SEQ_POWER_ON, /* sequence block v3+ */ - MIPI_SEQ_POWER_OFF, /* sequence block v3+ */ - MIPI_SEQ_MAX -}; - -enum mipi_seq_element { - MIPI_SEQ_ELEM_END = 0, - MIPI_SEQ_ELEM_SEND_PKT, - MIPI_SEQ_ELEM_DELAY, - MIPI_SEQ_ELEM_GPIO, - MIPI_SEQ_ELEM_I2C, /* sequence block v2+ */ - MIPI_SEQ_ELEM_SPI, /* sequence block v3+ */ - MIPI_SEQ_ELEM_PMIC, /* sequence block v3+ */ - MIPI_SEQ_ELEM_MAX -}; - -#define MIPI_DSI_UNDEFINED_PANEL_ID 0 -#define MIPI_DSI_GENERIC_PANEL_ID 1 - -struct mipi_config { - u16 panel_id; - - /* General Params */ - u32 enable_dithering:1; - u32 rsvd1:1; - u32 is_bridge:1; - - u32 panel_arch_type:2; - u32 is_cmd_mode:1; - -#define NON_BURST_SYNC_PULSE 0x1 -#define NON_BURST_SYNC_EVENTS 0x2 -#define BURST_MODE 0x3 - u32 video_transfer_mode:2; - - u32 cabc_supported:1; -#define PPS_BLC_PMIC 0 -#define PPS_BLC_SOC 1 - u32 pwm_blc:1; - - /* Bit 13:10 */ -#define PIXEL_FORMAT_RGB565 0x1 -#define PIXEL_FORMAT_RGB666 0x2 -#define PIXEL_FORMAT_RGB666_LOOSELY_PACKED 0x3 -#define PIXEL_FORMAT_RGB888 0x4 - u32 videomode_color_format:4; - - /* Bit 15:14 */ -#define ENABLE_ROTATION_0 0x0 -#define ENABLE_ROTATION_90 0x1 -#define ENABLE_ROTATION_180 0x2 -#define ENABLE_ROTATION_270 0x3 - u32 rotation:2; - u32 bta_enabled:1; - u32 rsvd2:15; - - /* 2 byte Port Description */ -#define DUAL_LINK_NOT_SUPPORTED 0 -#define DUAL_LINK_FRONT_BACK 1 -#define DUAL_LINK_PIXEL_ALT 2 - u16 dual_link:2; - u16 lane_cnt:2; - u16 pixel_overlap:3; - u16 rgb_flip:1; -#define DL_DCS_PORT_A 0x00 -#define DL_DCS_PORT_C 0x01 -#define DL_DCS_PORT_A_AND_C 0x02 - u16 dl_dcs_cabc_ports:2; - u16 dl_dcs_backlight_ports:2; - u16 rsvd3:4; - - u16 rsvd4; - - u8 rsvd5; - u32 target_burst_mode_freq; - u32 dsi_ddr_clk; - u32 bridge_ref_clk; - -#define BYTE_CLK_SEL_20MHZ 0 -#define BYTE_CLK_SEL_10MHZ 1 -#define BYTE_CLK_SEL_5MHZ 2 - u8 byte_clk_sel:2; - - u8 rsvd6:6; - - /* DPHY Flags */ - u16 dphy_param_valid:1; - u16 eot_pkt_disabled:1; - u16 enable_clk_stop:1; - u16 rsvd7:13; - - u32 hs_tx_timeout; - u32 lp_rx_timeout; - u32 turn_around_timeout; - u32 device_reset_timer; - u32 master_init_timer; - u32 dbi_bw_timer; - u32 lp_byte_clk_val; - - /* 4 byte Dphy Params */ - u32 prepare_cnt:6; - u32 rsvd8:2; - u32 clk_zero_cnt:8; - u32 trail_cnt:5; - u32 rsvd9:3; - u32 exit_zero_cnt:6; - u32 rsvd10:2; - - u32 clk_lane_switch_cnt; - u32 hl_switch_cnt; - - u32 rsvd11[6]; - - /* timings based on dphy spec */ - u8 tclk_miss; - u8 tclk_post; - u8 rsvd12; - u8 tclk_pre; - u8 tclk_prepare; - u8 tclk_settle; - u8 tclk_term_enable; - u8 tclk_trail; - u16 tclk_prepare_clkzero; - u8 rsvd13; - u8 td_term_enable; - u8 teot; - u8 ths_exit; - u8 ths_prepare; - u16 ths_prepare_hszero; - u8 rsvd14; - u8 ths_settle; - u8 ths_skip; - u8 ths_trail; - u8 tinit; - u8 tlpx; - u8 rsvd15[3]; - - /* GPIOs */ - u8 panel_enable; - u8 bl_enable; - u8 pwm_enable; - u8 reset_r_n; - u8 pwr_down_r; - u8 stdby_r_n; - -} __packed; - -/* all delays have a unit of 100us */ -struct mipi_pps_data { - u16 panel_on_delay; - u16 bl_enable_delay; - u16 bl_disable_delay; - u16 panel_off_delay; - u16 panel_power_cycle_delay; -} __packed; - void intel_bios_init(struct intel_display *display); void intel_bios_init_panel_early(struct intel_display *display, struct intel_panel *panel, @@ -259,6 +85,8 @@ bool intel_bios_encoder_is_lspcon(const struct intel_bios_encoder_data *devdata) bool intel_bios_encoder_lane_reversal(const struct intel_bios_encoder_data *devdata); bool intel_bios_encoder_hpd_invert(const struct intel_bios_encoder_data *devdata); enum port intel_bios_encoder_port(const struct intel_bios_encoder_data *devdata); +bool intel_bios_encoder_reject_edp_rate(const struct intel_bios_encoder_data *devdata, + int rate); enum aux_ch intel_bios_dp_aux_ch(const struct intel_bios_encoder_data *devdata); int intel_bios_dp_boost_level(const struct intel_bios_encoder_data *devdata); int intel_bios_dp_max_lane_count(const struct intel_bios_encoder_data *devdata); diff --git a/drivers/gpu/drm/i915/display/intel_bo.c b/drivers/gpu/drm/i915/display/intel_bo.c index 65d64f79a4bd..d29c1508ccb9 100644 --- a/drivers/gpu/drm/i915/display/intel_bo.c +++ b/drivers/gpu/drm/i915/display/intel_bo.c @@ -2,7 +2,7 @@ /* Copyright © 2024 Intel Corporation */ #include <drm/drm_panic.h> -#include "display/intel_display_types.h" + #include "gem/i915_gem_mman.h" #include "gem/i915_gem_object.h" #include "gem/i915_gem_object_frontbuffer.h" diff --git a/drivers/gpu/drm/i915/display/intel_bw.c b/drivers/gpu/drm/i915/display/intel_bw.c index d29a755612de..ac6da20d9529 100644 --- a/drivers/gpu/drm/i915/display/intel_bw.c +++ b/drivers/gpu/drm/i915/display/intel_bw.c @@ -359,7 +359,7 @@ static int icl_get_qgv_points(struct intel_display *display, for (i = 0; i < qi->num_psf_points; i++) drm_dbg_kms(display->drm, - "PSF GV %d: CLK=%d \n", + "PSF GV %d: CLK=%d\n", i, qi->psf_points[i].clk); } diff --git a/drivers/gpu/drm/i915/display/intel_cdclk.c b/drivers/gpu/drm/i915/display/intel_cdclk.c index 228aa64c1349..9725eebe5706 100644 --- a/drivers/gpu/drm/i915/display/intel_cdclk.c +++ b/drivers/gpu/drm/i915/display/intel_cdclk.c @@ -22,6 +22,7 @@ */ #include <linux/debugfs.h> +#include <linux/iopoll.h> #include <linux/time.h> #include <drm/drm_fixed.h> @@ -31,6 +32,7 @@ #include "hsw_ips.h" #include "i915_drv.h" #include "i915_reg.h" +#include "i915_utils.h" #include "intel_atomic.h" #include "intel_audio.h" #include "intel_bw.h" @@ -672,6 +674,7 @@ static void vlv_set_cdclk(struct intel_display *display, int cdclk = cdclk_config->cdclk; u32 val, cmd = cdclk_config->voltage_level; intel_wakeref_t wakeref; + int ret; switch (cdclk) { case 400000: @@ -702,12 +705,12 @@ static void vlv_set_cdclk(struct intel_display *display, val &= ~DSPFREQGUAR_MASK; val |= (cmd << DSPFREQGUAR_SHIFT); vlv_punit_write(display->drm, PUNIT_REG_DSPSSPM, val); - if (wait_for((vlv_punit_read(display->drm, PUNIT_REG_DSPSSPM) & - DSPFREQSTAT_MASK) == (cmd << DSPFREQSTAT_SHIFT), - 50)) { - drm_err(display->drm, - "timed out waiting for CDclk change\n"); - } + + ret = poll_timeout_us(val = vlv_punit_read(display->drm, PUNIT_REG_DSPSSPM), + (val & DSPFREQSTAT_MASK) == (cmd << DSPFREQSTAT_SHIFT), + 500, 50 * 1000, false); + if (ret) + drm_err(display->drm, "timed out waiting for CDCLK change\n"); if (cdclk == 400000) { u32 divider; @@ -721,11 +724,11 @@ static void vlv_set_cdclk(struct intel_display *display, val |= divider; vlv_cck_write(display->drm, CCK_DISPLAY_CLOCK_CONTROL, val); - if (wait_for((vlv_cck_read(display->drm, CCK_DISPLAY_CLOCK_CONTROL) & - CCK_FREQUENCY_STATUS) == (divider << CCK_FREQUENCY_STATUS_SHIFT), - 50)) - drm_err(display->drm, - "timed out waiting for CDclk change\n"); + ret = poll_timeout_us(val = vlv_cck_read(display->drm, CCK_DISPLAY_CLOCK_CONTROL), + (val & CCK_FREQUENCY_STATUS) == (divider << CCK_FREQUENCY_STATUS_SHIFT), + 500, 50 * 1000, false); + if (ret) + drm_err(display->drm, "timed out waiting for CDCLK change\n"); } /* adjust self-refresh exit latency value */ @@ -761,6 +764,7 @@ static void chv_set_cdclk(struct intel_display *display, int cdclk = cdclk_config->cdclk; u32 val, cmd = cdclk_config->voltage_level; intel_wakeref_t wakeref; + int ret; switch (cdclk) { case 333333: @@ -786,12 +790,12 @@ static void chv_set_cdclk(struct intel_display *display, val &= ~DSPFREQGUAR_MASK_CHV; val |= (cmd << DSPFREQGUAR_SHIFT_CHV); vlv_punit_write(display->drm, PUNIT_REG_DSPSSPM, val); - if (wait_for((vlv_punit_read(display->drm, PUNIT_REG_DSPSSPM) & - DSPFREQSTAT_MASK_CHV) == (cmd << DSPFREQSTAT_SHIFT_CHV), - 50)) { - drm_err(display->drm, - "timed out waiting for CDclk change\n"); - } + + ret = poll_timeout_us(val = vlv_punit_read(display->drm, PUNIT_REG_DSPSSPM), + (val & DSPFREQSTAT_MASK_CHV) == (cmd << DSPFREQSTAT_SHIFT_CHV), + 500, 50 * 1000, false); + if (ret) + drm_err(display->drm, "timed out waiting for CDCLK change\n"); vlv_punit_put(display->drm); @@ -903,8 +907,10 @@ static void bdw_set_cdclk(struct intel_display *display, * According to the spec, it should be enough to poll for this 1 us. * However, extensive testing shows that this can take longer. */ - if (wait_for_us(intel_de_read(display, LCPLL_CTL) & - LCPLL_CD_SOURCE_FCLK_DONE, 100)) + ret = intel_de_wait_custom(display, LCPLL_CTL, + LCPLL_CD_SOURCE_FCLK_DONE, LCPLL_CD_SOURCE_FCLK_DONE, + 100, 0, NULL); + if (ret) drm_err(display->drm, "Switching to FCLK failed\n"); intel_de_rmw(display, LCPLL_CTL, @@ -913,8 +919,10 @@ static void bdw_set_cdclk(struct intel_display *display, intel_de_rmw(display, LCPLL_CTL, LCPLL_CD_SOURCE_FCLK, 0); - if (wait_for_us((intel_de_read(display, LCPLL_CTL) & - LCPLL_CD_SOURCE_FCLK_DONE) == 0, 1)) + ret = intel_de_wait_custom(display, LCPLL_CTL, + LCPLL_CD_SOURCE_FCLK_DONE, 0, + 1, 0, NULL); + if (ret) drm_err(display->drm, "Switching back to LCPLL failed\n"); intel_pcode_write(display->drm, HSW_PCODE_DE_WRITE_FREQ_REQ, @@ -3569,7 +3577,7 @@ static int i9xx_hrawclk(struct intel_display *display) struct drm_i915_private *i915 = to_i915(display->drm); /* hrawclock is 1/4 the FSB frequency */ - return DIV_ROUND_CLOSEST(i9xx_fsb_freq(i915), 4); + return DIV_ROUND_CLOSEST(intel_fsb_freq(i915), 4); } /** @@ -3622,9 +3630,7 @@ DEFINE_SHOW_ATTRIBUTE(i915_cdclk_info); void intel_cdclk_debugfs_register(struct intel_display *display) { - struct drm_minor *minor = display->drm->primary; - - debugfs_create_file("i915_cdclk_info", 0444, minor->debugfs_root, + debugfs_create_file("i915_cdclk_info", 0444, display->drm->debugfs_root, display, &i915_cdclk_info_fops); } diff --git a/drivers/gpu/drm/i915/display/intel_connector.c b/drivers/gpu/drm/i915/display/intel_connector.c index 42c923f416b3..6a55854db5b6 100644 --- a/drivers/gpu/drm/i915/display/intel_connector.c +++ b/drivers/gpu/drm/i915/display/intel_connector.c @@ -77,7 +77,7 @@ void intel_connector_cancel_modeset_retry_work(struct intel_connector *connector drm_connector_put(&connector->base); } -int intel_connector_init(struct intel_connector *connector) +static int intel_connector_init(struct intel_connector *connector) { struct intel_digital_connector_state *conn_state; diff --git a/drivers/gpu/drm/i915/display/intel_connector.h b/drivers/gpu/drm/i915/display/intel_connector.h index aafb25a814fa..0aa86626e646 100644 --- a/drivers/gpu/drm/i915/display/intel_connector.h +++ b/drivers/gpu/drm/i915/display/intel_connector.h @@ -14,7 +14,6 @@ struct i2c_adapter; struct intel_connector; struct intel_encoder; -int intel_connector_init(struct intel_connector *connector); struct intel_connector *intel_connector_alloc(void); void intel_connector_free(struct intel_connector *connector); void intel_connector_destroy(struct drm_connector *connector); diff --git a/drivers/gpu/drm/i915/display/intel_crt.c b/drivers/gpu/drm/i915/display/intel_crt.c index 898c5d9e8f7a..31e68047f217 100644 --- a/drivers/gpu/drm/i915/display/intel_crt.c +++ b/drivers/gpu/drm/i915/display/intel_crt.c @@ -50,6 +50,7 @@ #include "intel_gmbus.h" #include "intel_hotplug.h" #include "intel_hotplug_irq.h" +#include "intel_link_bw.h" #include "intel_load_detect.h" #include "intel_pch_display.h" #include "intel_pch_refclk.h" @@ -421,7 +422,7 @@ static int pch_crt_compute_config(struct intel_encoder *encoder, return -EINVAL; crtc_state->has_pch_encoder = true; - if (!intel_fdi_compute_pipe_bpp(crtc_state)) + if (!intel_link_bw_compute_pipe_bpp(crtc_state)) return -EINVAL; crtc_state->output_format = INTEL_OUTPUT_FORMAT_RGB; @@ -446,7 +447,7 @@ static int hsw_crt_compute_config(struct intel_encoder *encoder, return -EINVAL; crtc_state->has_pch_encoder = true; - if (!intel_fdi_compute_pipe_bpp(crtc_state)) + if (!intel_link_bw_compute_pipe_bpp(crtc_state)) return -EINVAL; crtc_state->output_format = INTEL_OUTPUT_FORMAT_RGB; diff --git a/drivers/gpu/drm/i915/display/intel_cursor.c b/drivers/gpu/drm/i915/display/intel_cursor.c index 198e69efe9ac..d4d181f9dca5 100644 --- a/drivers/gpu/drm/i915/display/intel_cursor.c +++ b/drivers/gpu/drm/i915/display/intel_cursor.c @@ -33,17 +33,9 @@ static const u32 intel_cursor_formats[] = { DRM_FORMAT_ARGB8888, }; -static u32 intel_cursor_base(const struct intel_plane_state *plane_state) +static u32 intel_cursor_surf_offset(const struct intel_plane_state *plane_state) { - struct intel_display *display = to_intel_display(plane_state); - u32 base; - - if (DISPLAY_INFO(display)->cursor_needs_physical) - base = plane_state->phys_dma_addr; - else - base = intel_plane_ggtt_offset(plane_state); - - return base + plane_state->view.color_plane[0].offset; + return plane_state->view.color_plane[0].offset; } static u32 intel_cursor_position(const struct intel_crtc_state *crtc_state, @@ -213,8 +205,7 @@ static u32 i845_cursor_ctl_crtc(const struct intel_crtc_state *crtc_state) return cntl; } -static u32 i845_cursor_ctl(const struct intel_crtc_state *crtc_state, - const struct intel_plane_state *plane_state) +static u32 i845_cursor_ctl(const struct intel_plane_state *plane_state) { return CURSOR_ENABLE | CURSOR_FORMAT_ARGB | @@ -274,7 +265,7 @@ static int i845_check_cursor(struct intel_crtc_state *crtc_state, return -EINVAL; } - plane_state->ctl = i845_cursor_ctl(crtc_state, plane_state); + plane_state->ctl = i845_cursor_ctl(plane_state); return 0; } @@ -297,7 +288,7 @@ static void i845_cursor_update_arm(struct intel_dsb *dsb, size = CURSOR_HEIGHT(height) | CURSOR_WIDTH(width); - base = intel_cursor_base(plane_state); + base = plane_state->surf; pos = intel_cursor_position(crtc_state, plane_state, false); } @@ -406,8 +397,7 @@ static u32 i9xx_cursor_ctl_crtc(const struct intel_crtc_state *crtc_state) return cntl; } -static u32 i9xx_cursor_ctl(const struct intel_crtc_state *crtc_state, - const struct intel_plane_state *plane_state) +static u32 i9xx_cursor_ctl(const struct intel_plane_state *plane_state) { struct intel_display *display = to_intel_display(plane_state); u32 cntl = 0; @@ -534,7 +524,7 @@ static int i9xx_check_cursor(struct intel_crtc_state *crtc_state, return -EINVAL; } - plane_state->ctl = i9xx_cursor_ctl(crtc_state, plane_state); + plane_state->ctl = i9xx_cursor_ctl(plane_state); return 0; } @@ -675,7 +665,7 @@ static void i9xx_cursor_update_arm(struct intel_dsb *dsb, if (width != height) fbc_ctl = CUR_FBC_EN | CUR_FBC_HEIGHT(height - 1); - base = intel_cursor_base(plane_state); + base = plane_state->surf; pos = intel_cursor_position(crtc_state, plane_state, false); } @@ -1051,6 +1041,8 @@ intel_cursor_plane_create(struct intel_display *display, cursor->check_plane = i9xx_check_cursor; } + cursor->surf_offset = intel_cursor_surf_offset; + if (DISPLAY_VER(display) >= 5 || display->platform.g4x) cursor->capture_error = g4x_cursor_capture_error; else diff --git a/drivers/gpu/drm/i915/display/intel_ddi.c b/drivers/gpu/drm/i915/display/intel_ddi.c index 0405396c7750..46017091bb0b 100644 --- a/drivers/gpu/drm/i915/display/intel_ddi.c +++ b/drivers/gpu/drm/i915/display/intel_ddi.c @@ -2166,7 +2166,8 @@ icl_program_mg_dp_mode(struct intel_digital_port *dig_port, { struct intel_display *display = to_intel_display(crtc_state); enum tc_port tc_port = intel_encoder_to_tc(&dig_port->base); - u32 ln0, ln1, pin_assignment; + enum intel_tc_pin_assignment pin_assignment; + u32 ln0, ln1; u8 width; if (DISPLAY_VER(display) >= 14) @@ -2188,11 +2189,11 @@ icl_program_mg_dp_mode(struct intel_digital_port *dig_port, ln1 &= ~(MG_DP_MODE_CFG_DP_X1_MODE | MG_DP_MODE_CFG_DP_X2_MODE); /* DPPATC */ - pin_assignment = intel_tc_port_get_pin_assignment_mask(dig_port); + pin_assignment = intel_tc_port_get_pin_assignment(dig_port); width = crtc_state->lane_count; switch (pin_assignment) { - case 0x0: + case INTEL_TC_PIN_ASSIGNMENT_NONE: drm_WARN_ON(display->drm, !intel_tc_port_in_legacy_mode(dig_port)); if (width == 1) { @@ -2202,20 +2203,20 @@ icl_program_mg_dp_mode(struct intel_digital_port *dig_port, ln1 |= MG_DP_MODE_CFG_DP_X2_MODE; } break; - case 0x1: + case INTEL_TC_PIN_ASSIGNMENT_A: if (width == 4) { ln0 |= MG_DP_MODE_CFG_DP_X2_MODE; ln1 |= MG_DP_MODE_CFG_DP_X2_MODE; } break; - case 0x2: + case INTEL_TC_PIN_ASSIGNMENT_B: if (width == 2) { ln0 |= MG_DP_MODE_CFG_DP_X2_MODE; ln1 |= MG_DP_MODE_CFG_DP_X2_MODE; } break; - case 0x3: - case 0x5: + case INTEL_TC_PIN_ASSIGNMENT_C: + case INTEL_TC_PIN_ASSIGNMENT_E: if (width == 1) { ln0 |= MG_DP_MODE_CFG_DP_X1_MODE; ln1 |= MG_DP_MODE_CFG_DP_X1_MODE; @@ -2224,8 +2225,8 @@ icl_program_mg_dp_mode(struct intel_digital_port *dig_port, ln1 |= MG_DP_MODE_CFG_DP_X2_MODE; } break; - case 0x4: - case 0x6: + case INTEL_TC_PIN_ASSIGNMENT_D: + case INTEL_TC_PIN_ASSIGNMENT_F: if (width == 1) { ln0 |= MG_DP_MODE_CFG_DP_X1_MODE; ln1 |= MG_DP_MODE_CFG_DP_X1_MODE; @@ -2339,34 +2340,24 @@ static void intel_dp_sink_set_fec_ready(struct intel_dp *intel_dp, drm_dbg_kms(display->drm, "Failed to clear FEC detected flags\n"); } -static int read_fec_detected_status(struct drm_dp_aux *aux) -{ - int ret; - u8 status; - - ret = drm_dp_dpcd_readb(aux, DP_FEC_STATUS, &status); - if (ret < 0) - return ret; - - return status; -} - static int wait_for_fec_detected(struct drm_dp_aux *aux, bool enabled) { struct intel_display *display = to_intel_display(aux->drm_dev); int mask = enabled ? DP_FEC_DECODE_EN_DETECTED : DP_FEC_DECODE_DIS_DETECTED; - int status; - int err; + u8 status = 0; + int ret, err; - err = readx_poll_timeout(read_fec_detected_status, aux, status, - status & mask || status < 0, - 10000, 200000); + ret = poll_timeout_us(err = drm_dp_dpcd_read_byte(aux, DP_FEC_STATUS, &status), + err || (status & mask), + 10 * 1000, 200 * 1000, false); - if (err || status < 0) { + /* Either can be non-zero, but not both */ + ret = ret ?: err; + if (ret) { drm_dbg_kms(display->drm, - "Failed waiting for FEC %s to get detected: %d (status %d)\n", - str_enabled_disabled(enabled), err, status); - return err ? err : status; + "Failed waiting for FEC %s to get detected: %d (status 0x%02x)\n", + str_enabled_disabled(enabled), ret, status); + return ret; } return 0; @@ -2561,6 +2552,7 @@ mtl_ddi_enable_d2d(struct intel_encoder *encoder) enum port port = encoder->port; i915_reg_t reg; u32 set_bits, wait_bits; + int ret; if (DISPLAY_VER(display) < 14) return; @@ -2576,7 +2568,11 @@ mtl_ddi_enable_d2d(struct intel_encoder *encoder) } intel_de_rmw(display, reg, 0, set_bits); - if (wait_for_us(intel_de_read(display, reg) & wait_bits, 100)) { + + ret = intel_de_wait_custom(display, reg, + wait_bits, wait_bits, + 100, 0, NULL); + if (ret) { drm_err(display->drm, "Timeout waiting for D2D Link enable for DDI/PORT_BUF_CTL %c\n", port_name(port)); } @@ -3058,6 +3054,7 @@ mtl_ddi_disable_d2d(struct intel_encoder *encoder) enum port port = encoder->port; i915_reg_t reg; u32 clr_bits, wait_bits; + int ret; if (DISPLAY_VER(display) < 14) return; @@ -3073,7 +3070,11 @@ mtl_ddi_disable_d2d(struct intel_encoder *encoder) } intel_de_rmw(display, reg, clr_bits, 0); - if (wait_for_us(!(intel_de_read(display, reg) & wait_bits), 100)) + + ret = intel_de_wait_custom(display, reg, + wait_bits, 0, + 100, 0, NULL); + if (ret) drm_err(display->drm, "Timeout waiting for D2D Link disable for DDI/PORT_BUF_CTL %c\n", port_name(port)); } @@ -5148,12 +5149,10 @@ void intel_ddi_init(struct intel_display *display, phy_name(phy)); } - dig_port = kzalloc(sizeof(*dig_port), GFP_KERNEL); + dig_port = intel_dig_port_alloc(); if (!dig_port) return; - dig_port->aux_ch = AUX_CH_NONE; - encoder = &dig_port->base; encoder->devdata = devdata; @@ -5191,9 +5190,6 @@ void intel_ddi_init(struct intel_display *display, intel_encoder_link_check_init(encoder, intel_ddi_link_check); - mutex_init(&dig_port->hdcp.mutex); - dig_port->hdcp.num_streams = 0; - encoder->hotplug = intel_ddi_hotplug; encoder->compute_output_type = intel_ddi_compute_output_type; encoder->compute_config = intel_ddi_compute_config; @@ -5331,7 +5327,6 @@ void intel_ddi_init(struct intel_display *display, dig_port->ddi_a_4_lanes = DISPLAY_VER(display) < 11 && ddi_buf_ctl & DDI_A_4_LANES; - dig_port->dp.output_reg = INVALID_MMIO_REG; dig_port->max_lanes = intel_ddi_max_lanes(dig_port); if (need_aux_ch(encoder, init_dp)) { diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c index 7035c1fc9033..c1a3a95c65f0 100644 --- a/drivers/gpu/drm/i915/display/intel_display.c +++ b/drivers/gpu/drm/i915/display/intel_display.c @@ -76,6 +76,7 @@ #include "intel_display_regs.h" #include "intel_display_rpm.h" #include "intel_display_types.h" +#include "intel_display_wa.h" #include "intel_dmc.h" #include "intel_dp.h" #include "intel_dp_link_training.h" @@ -1081,6 +1082,11 @@ static void intel_post_plane_update(struct intel_atomic_state *state, if (audio_enabling(old_crtc_state, new_crtc_state)) intel_encoders_audio_enable(state, crtc); + if (intel_display_wa(display, 14011503117)) { + if (old_crtc_state->pch_pfit.enabled != new_crtc_state->pch_pfit.enabled) + adl_scaler_ecc_unmask(new_crtc_state); + } + intel_alpm_post_plane_update(state, crtc); intel_psr_post_plane_update(state, crtc); diff --git a/drivers/gpu/drm/i915/display/intel_display_conversion.c b/drivers/gpu/drm/i915/display/intel_display_conversion.c index 4d565935e2cc..d56065f22655 100644 --- a/drivers/gpu/drm/i915/display/intel_display_conversion.c +++ b/drivers/gpu/drm/i915/display/intel_display_conversion.c @@ -4,7 +4,7 @@ #include "i915_drv.h" #include "intel_display_conversion.h" -struct intel_display *__i915_to_display(struct drm_i915_private *i915) +static struct intel_display *__i915_to_display(struct drm_i915_private *i915) { return i915->display; } diff --git a/drivers/gpu/drm/i915/display/intel_display_conversion.h b/drivers/gpu/drm/i915/display/intel_display_conversion.h index 46c7208d42ba..d497bc58a73f 100644 --- a/drivers/gpu/drm/i915/display/intel_display_conversion.h +++ b/drivers/gpu/drm/i915/display/intel_display_conversion.h @@ -9,20 +9,8 @@ #define __INTEL_DISPLAY_CONVERSION__ struct drm_device; -struct drm_i915_private; struct intel_display; -struct intel_display *__i915_to_display(struct drm_i915_private *i915); struct intel_display *__drm_to_display(struct drm_device *drm); -/* - * Transitional macro to optionally convert struct drm_i915_private * to struct - * intel_display *, also accepting the latter. - */ -#define __to_intel_display(p) \ - _Generic(p, \ - const struct drm_i915_private *: __i915_to_display((struct drm_i915_private *)(p)), \ - struct drm_i915_private *: __i915_to_display((struct drm_i915_private *)(p)), \ - const struct intel_display *: (p), \ - struct intel_display *: (p)) #endif /* __INTEL_DISPLAY_CONVERSION__ */ diff --git a/drivers/gpu/drm/i915/display/intel_display_debugfs.c b/drivers/gpu/drm/i915/display/intel_display_debugfs.c index ce3f9810c42d..10dddec3796f 100644 --- a/drivers/gpu/drm/i915/display/intel_display_debugfs.c +++ b/drivers/gpu/drm/i915/display/intel_display_debugfs.c @@ -820,14 +820,14 @@ static const struct drm_info_list intel_display_debugfs_list[] = { void intel_display_debugfs_register(struct intel_display *display) { - struct drm_minor *minor = display->drm->primary; + struct dentry *debugfs_root = display->drm->debugfs_root; - debugfs_create_file("i915_fifo_underrun_reset", 0644, minor->debugfs_root, + debugfs_create_file("i915_fifo_underrun_reset", 0644, debugfs_root, display, &i915_fifo_underrun_reset_ops); drm_debugfs_create_files(intel_display_debugfs_list, ARRAY_SIZE(intel_display_debugfs_list), - minor->debugfs_root, minor); + debugfs_root, display->drm->primary); intel_bios_debugfs_register(display); intel_cdclk_debugfs_register(display); diff --git a/drivers/gpu/drm/i915/display/intel_display_debugfs_params.c b/drivers/gpu/drm/i915/display/intel_display_debugfs_params.c index 88914a1f3f62..de62b774272d 100644 --- a/drivers/gpu/drm/i915/display/intel_display_debugfs_params.c +++ b/drivers/gpu/drm/i915/display/intel_display_debugfs_params.c @@ -7,7 +7,6 @@ #include <linux/kernel.h> #include <drm/drm_drv.h> -#include <drm/drm_file.h> #include "intel_display_core.h" #include "intel_display_debugfs_params.h" @@ -154,14 +153,14 @@ intel_display_debugfs_create_uint(const char *name, umode_t mode, /* add a subdirectory with files for each intel display param */ void intel_display_debugfs_params(struct intel_display *display) { - struct drm_minor *minor = display->drm->primary; + struct dentry *debugfs_root = display->drm->debugfs_root; struct dentry *dir; char dirname[16]; snprintf(dirname, sizeof(dirname), "%s_params", display->drm->driver->name); - dir = debugfs_lookup(dirname, minor->debugfs_root); + dir = debugfs_lookup(dirname, debugfs_root); if (!dir) - dir = debugfs_create_dir(dirname, minor->debugfs_root); + dir = debugfs_create_dir(dirname, debugfs_root); if (IS_ERR(dir)) return; diff --git a/drivers/gpu/drm/i915/display/intel_display_device.c b/drivers/gpu/drm/i915/display/intel_display_device.c index 089cffabbad5..65f0efc35bb7 100644 --- a/drivers/gpu/drm/i915/display/intel_display_device.c +++ b/drivers/gpu/drm/i915/display/intel_display_device.c @@ -1354,6 +1354,19 @@ static const struct intel_display_device_info xe2_lpd_display = { .__runtime_defaults.has_dbuf_overlap_detection = true, }; +static const struct intel_display_device_info wcl_display = { + XE_LPDP_FEATURES, + + .__runtime_defaults.cpu_transcoder_mask = + BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | BIT(TRANSCODER_C), + .__runtime_defaults.pipe_mask = + BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C), + .__runtime_defaults.fbc_mask = + BIT(INTEL_FBC_A) | BIT(INTEL_FBC_B) | BIT(INTEL_FBC_C), + .__runtime_defaults.port_mask = + BIT(PORT_A) | BIT(PORT_B) | BIT(PORT_TC1) | BIT(PORT_TC2), +}; + static const struct intel_display_device_info xe2_hpd_display = { XE_LPDP_FEATURES, .__runtime_defaults.port_mask = BIT(PORT_A) | @@ -1480,7 +1493,7 @@ static const struct { { 14, 1, &xe2_hpd_display }, { 20, 0, &xe2_lpd_display }, { 30, 0, &xe2_lpd_display }, - { 30, 2, &xe2_lpd_display }, + { 30, 2, &wcl_display }, }; static const struct intel_display_device_info * diff --git a/drivers/gpu/drm/i915/display/intel_display_device.h b/drivers/gpu/drm/i915/display/intel_display_device.h index 4308822f0415..6e87b763fe7c 100644 --- a/drivers/gpu/drm/i915/display/intel_display_device.h +++ b/drivers/gpu/drm/i915/display/intel_display_device.h @@ -9,7 +9,6 @@ #include <linux/bitops.h> #include <linux/types.h> -#include "intel_display_conversion.h" #include "intel_display_limits.h" struct drm_printer; @@ -224,8 +223,8 @@ struct intel_display_platforms { (IS_DISPLAY_VERx100((__display), (ipver), (ipver)) && \ IS_DISPLAY_STEP((__display), (from), (until))) -#define DISPLAY_INFO(__display) (__to_intel_display(__display)->info.__device_info) -#define DISPLAY_RUNTIME_INFO(__display) (&__to_intel_display(__display)->info.__runtime_info) +#define DISPLAY_INFO(__display) ((__display)->info.__device_info) +#define DISPLAY_RUNTIME_INFO(__display) (&(__display)->info.__runtime_info) #define DISPLAY_VER(__display) (DISPLAY_RUNTIME_INFO(__display)->ip.ver) #define DISPLAY_VERx100(__display) (DISPLAY_RUNTIME_INFO(__display)->ip.ver * 100 + \ @@ -236,7 +235,7 @@ struct intel_display_platforms { #define INTEL_DISPLAY_STEP(__display) (DISPLAY_RUNTIME_INFO(__display)->step) #define IS_DISPLAY_STEP(__display, since, until) \ - (drm_WARN_ON(__to_intel_display(__display)->drm, INTEL_DISPLAY_STEP(__display) == STEP_NONE), \ + (drm_WARN_ON((__display)->drm, INTEL_DISPLAY_STEP(__display) == STEP_NONE), \ INTEL_DISPLAY_STEP(__display) >= (since) && INTEL_DISPLAY_STEP(__display) < (until)) #define ARLS_HOST_BRIDGE_PCI_ID1 0x7D1C diff --git a/drivers/gpu/drm/i915/display/intel_display_driver.c b/drivers/gpu/drm/i915/display/intel_display_driver.c index 8586ba102605..cf1c14412abe 100644 --- a/drivers/gpu/drm/i915/display/intel_display_driver.c +++ b/drivers/gpu/drm/i915/display/intel_display_driver.c @@ -18,6 +18,7 @@ #include <drm/drm_vblank.h> #include "i915_drv.h" +#include "i915_utils.h" #include "i9xx_wm.h" #include "intel_acpi.h" #include "intel_atomic.h" diff --git a/drivers/gpu/drm/i915/display/intel_display_irq.c b/drivers/gpu/drm/i915/display/intel_display_irq.c index fb25ec8adae3..123e054affbe 100644 --- a/drivers/gpu/drm/i915/display/intel_display_irq.c +++ b/drivers/gpu/drm/i915/display/intel_display_irq.c @@ -1506,10 +1506,14 @@ u32 gen11_gu_misc_irq_ack(struct intel_display *display, const u32 master_ctl) if (!(master_ctl & GEN11_GU_MISC_IRQ)) return 0; + intel_display_rpm_assert_block(display); + iir = intel_de_read(display, GEN11_GU_MISC_IIR); if (likely(iir)) intel_de_write(display, GEN11_GU_MISC_IIR, iir); + intel_display_rpm_assert_unblock(display); + return iir; } @@ -1986,20 +1990,17 @@ void vlv_display_irq_postinstall(struct intel_display *display) void ibx_display_irq_reset(struct intel_display *display) { - struct drm_i915_private *i915 = to_i915(display->drm); - - if (HAS_PCH_NOP(i915)) + if (HAS_PCH_NOP(display)) return; gen2_irq_reset(to_intel_uncore(display->drm), SDE_IRQ_REGS); - if (HAS_PCH_CPT(i915) || HAS_PCH_LPT(i915)) + if (HAS_PCH_CPT(display) || HAS_PCH_LPT(display)) intel_de_write(display, SERR_INT, 0xffffffff); } void gen8_display_irq_reset(struct intel_display *display) { - struct drm_i915_private *i915 = to_i915(display->drm); enum pipe pipe; if (!HAS_DISPLAY(display)) @@ -2016,7 +2017,7 @@ void gen8_display_irq_reset(struct intel_display *display) intel_display_irq_regs_reset(display, GEN8_DE_PORT_IRQ_REGS); intel_display_irq_regs_reset(display, GEN8_DE_MISC_IRQ_REGS); - if (HAS_PCH_SPLIT(i915)) + if (HAS_PCH_SPLIT(display)) ibx_display_irq_reset(display); } diff --git a/drivers/gpu/drm/i915/display/intel_display_params.c b/drivers/gpu/drm/i915/display/intel_display_params.c index 75316247ee8a..2aed110c5b09 100644 --- a/drivers/gpu/drm/i915/display/intel_display_params.c +++ b/drivers/gpu/drm/i915/display/intel_display_params.c @@ -120,6 +120,9 @@ intel_display_param_named_unsafe(enable_psr, int, 0400, "(0=disabled, 1=enable up to PSR1, 2=enable up to PSR2) " "Default: -1 (use per-chip default)"); +intel_display_param_named_unsafe(enable_panel_replay, int, 0400, + "Enable Panel Replay (0=disabled, 1=enabled). Default: -1 (use per-chip default)"); + intel_display_param_named(psr_safest_params, bool, 0400, "Replace PSR VBT parameters by the safest and not optimal ones. This " "is helpful to detect if PSR issues are related to bad values set in " diff --git a/drivers/gpu/drm/i915/display/intel_display_params.h b/drivers/gpu/drm/i915/display/intel_display_params.h index 784e6bae8615..b01bc5700c52 100644 --- a/drivers/gpu/drm/i915/display/intel_display_params.h +++ b/drivers/gpu/drm/i915/display/intel_display_params.h @@ -46,6 +46,7 @@ struct drm_printer; param(bool, enable_dp_mst, true, 0600) \ param(int, enable_fbc, -1, 0600) \ param(int, enable_psr, -1, 0600) \ + param(int, enable_panel_replay, -1, 0600) \ param(bool, psr_safest_params, false, 0400) \ param(bool, enable_psr2_sel_fetch, true, 0400) \ param(int, enable_dmc_wl, -1, 0400) \ diff --git a/drivers/gpu/drm/i915/display/intel_display_power.c b/drivers/gpu/drm/i915/display/intel_display_power.c index 273054c22325..7f9a2d743bac 100644 --- a/drivers/gpu/drm/i915/display/intel_display_power.c +++ b/drivers/gpu/drm/i915/display/intel_display_power.c @@ -3,6 +3,7 @@ * Copyright © 2019 Intel Corporation */ +#include <linux/iopoll.h> #include <linux/string_helpers.h> #include "soc/intel_dram.h" @@ -10,6 +11,7 @@ #include "i915_drv.h" #include "i915_irq.h" #include "i915_reg.h" +#include "i915_utils.h" #include "intel_backlight_regs.h" #include "intel_cdclk.h" #include "intel_clock_gating.h" @@ -1278,6 +1280,7 @@ static void hsw_disable_lcpll(struct intel_display *display, bool switch_to_fclk, bool allow_power_down) { u32 val; + int ret; assert_can_disable_lcpll(display); @@ -1287,8 +1290,10 @@ static void hsw_disable_lcpll(struct intel_display *display, val |= LCPLL_CD_SOURCE_FCLK; intel_de_write(display, LCPLL_CTL, val); - if (wait_for_us(intel_de_read(display, LCPLL_CTL) & - LCPLL_CD_SOURCE_FCLK_DONE, 1)) + ret = intel_de_wait_custom(display, LCPLL_CTL, + LCPLL_CD_SOURCE_FCLK_DONE, LCPLL_CD_SOURCE_FCLK_DONE, + 1, 0, NULL); + if (ret) drm_err(display->drm, "Switching to FCLK failed\n"); val = intel_de_read(display, LCPLL_CTL); @@ -1306,8 +1311,10 @@ static void hsw_disable_lcpll(struct intel_display *display, hsw_write_dcomp(display, val); ndelay(100); - if (wait_for((hsw_read_dcomp(display) & - D_COMP_RCOMP_IN_PROGRESS) == 0, 1)) + ret = poll_timeout_us(val = hsw_read_dcomp(display), + (val & D_COMP_RCOMP_IN_PROGRESS) == 0, + 100, 1000, false); + if (ret) drm_err(display->drm, "D_COMP RCOMP still in progress\n"); if (allow_power_down) { @@ -1324,6 +1331,7 @@ static void hsw_restore_lcpll(struct intel_display *display) { struct drm_i915_private __maybe_unused *dev_priv = to_i915(display->drm); u32 val; + int ret; val = intel_de_read(display, LCPLL_CTL); @@ -1358,8 +1366,10 @@ static void hsw_restore_lcpll(struct intel_display *display) if (val & LCPLL_CD_SOURCE_FCLK) { intel_de_rmw(display, LCPLL_CTL, LCPLL_CD_SOURCE_FCLK, 0); - if (wait_for_us((intel_de_read(display, LCPLL_CTL) & - LCPLL_CD_SOURCE_FCLK_DONE) == 0, 1)) + ret = intel_de_wait_custom(display, LCPLL_CTL, + LCPLL_CD_SOURCE_FCLK_DONE, 0, + 1, 0, NULL); + if (ret) drm_err(display->drm, "Switching back to LCPLL failed\n"); } @@ -2155,8 +2165,6 @@ void intel_power_domains_resume(struct intel_display *display) power_domains->init_wakeref = intel_display_power_get(display, POWER_DOMAIN_INIT); } - - intel_power_domains_verify_state(display); } #if IS_ENABLED(CONFIG_DRM_I915_DEBUG_RUNTIME_PM) diff --git a/drivers/gpu/drm/i915/display/intel_display_power_map.c b/drivers/gpu/drm/i915/display/intel_display_power_map.c index 77268802b55e..39b71fffa2cd 100644 --- a/drivers/gpu/drm/i915/display/intel_display_power_map.c +++ b/drivers/gpu/drm/i915/display/intel_display_power_map.c @@ -1717,6 +1717,59 @@ static const struct i915_power_well_desc_list xe3lpd_power_wells[] = { I915_PW_DESCRIPTORS(xe2lpd_power_wells_pica), }; +static const struct i915_power_well_desc wcl_power_wells_main[] = { + { + .instances = &I915_PW_INSTANCES( + I915_PW("PW_2", &xe3lpd_pwdoms_pw_2, + .hsw.idx = ICL_PW_CTL_IDX_PW_2, + .id = SKL_DISP_PW_2), + ), + .ops = &hsw_power_well_ops, + .has_vga = true, + .has_fuses = true, + }, { + .instances = &I915_PW_INSTANCES( + I915_PW("PW_A", &xelpd_pwdoms_pw_a, + .hsw.idx = XELPD_PW_CTL_IDX_PW_A), + ), + .ops = &hsw_power_well_ops, + .irq_pipe_mask = BIT(PIPE_A), + .has_fuses = true, + }, { + .instances = &I915_PW_INSTANCES( + I915_PW("PW_B", &xe3lpd_pwdoms_pw_b, + .hsw.idx = XELPD_PW_CTL_IDX_PW_B), + ), + .ops = &hsw_power_well_ops, + .irq_pipe_mask = BIT(PIPE_B), + .has_fuses = true, + }, { + .instances = &I915_PW_INSTANCES( + I915_PW("PW_C", &xe3lpd_pwdoms_pw_c, + .hsw.idx = XELPD_PW_CTL_IDX_PW_C), + ), + .ops = &hsw_power_well_ops, + .irq_pipe_mask = BIT(PIPE_C), + .has_fuses = true, + }, { + .instances = &I915_PW_INSTANCES( + I915_PW("AUX_A", &icl_pwdoms_aux_a, .xelpdp.aux_ch = AUX_CH_A), + I915_PW("AUX_B", &icl_pwdoms_aux_b, .xelpdp.aux_ch = AUX_CH_B), + I915_PW("AUX_TC1", &xelpdp_pwdoms_aux_tc1, .xelpdp.aux_ch = AUX_CH_USBC1), + I915_PW("AUX_TC2", &xelpdp_pwdoms_aux_tc2, .xelpdp.aux_ch = AUX_CH_USBC2), + ), + .ops = &xelpdp_aux_power_well_ops, + }, +}; + +static const struct i915_power_well_desc_list wcl_power_wells[] = { + I915_PW_DESCRIPTORS(i9xx_power_wells_always_on), + I915_PW_DESCRIPTORS(icl_power_wells_pw_1), + I915_PW_DESCRIPTORS(xe3lpd_power_wells_dcoff), + I915_PW_DESCRIPTORS(wcl_power_wells_main), + I915_PW_DESCRIPTORS(xe2lpd_power_wells_pica), +}; + static void init_power_well_domains(const struct i915_power_well_instance *inst, struct i915_power_well *power_well) { @@ -1824,7 +1877,9 @@ int intel_display_power_map_init(struct i915_power_domains *power_domains) return 0; } - if (DISPLAY_VER(display) >= 30) + if (DISPLAY_VERx100(display) == 3002) + return set_power_wells(power_domains, wcl_power_wells); + else if (DISPLAY_VER(display) >= 30) return set_power_wells(power_domains, xe3lpd_power_wells); else if (DISPLAY_VER(display) >= 20) return set_power_wells(power_domains, xe2lpd_power_wells); diff --git a/drivers/gpu/drm/i915/display/intel_display_power_well.c b/drivers/gpu/drm/i915/display/intel_display_power_well.c index 48cac225a809..5e88b930f5aa 100644 --- a/drivers/gpu/drm/i915/display/intel_display_power_well.c +++ b/drivers/gpu/drm/i915/display/intel_display_power_well.c @@ -3,6 +3,8 @@ * Copyright © 2022 Intel Corporation */ +#include <linux/iopoll.h> + #include "i915_drv.h" #include "i915_irq.h" #include "i915_reg.h" @@ -499,7 +501,6 @@ static void icl_tc_port_assert_ref_held(struct intel_display *display, static void icl_tc_cold_exit(struct intel_display *display) { - struct drm_i915_private *i915 = to_i915(display->drm); int ret, tries = 0; while (1) { @@ -514,7 +515,7 @@ static void icl_tc_cold_exit(struct intel_display *display) msleep(1); /* TODO: turn failure into a error as soon i915 CI updates ICL IFWI */ - drm_dbg_kms(&i915->drm, "TC cold block %s\n", ret ? "failed" : + drm_dbg_kms(display->drm, "TC cold block %s\n", ret ? "failed" : "succeeded"); } @@ -527,6 +528,8 @@ icl_tc_phy_aux_power_well_enable(struct intel_display *display, const struct i915_power_well_regs *regs = power_well->desc->ops->regs; bool is_tbt = power_well->desc->is_tc_tbt; bool timeout_expected; + u32 val; + int ret; icl_tc_port_assert_ref_held(display, power_well, dig_port); @@ -553,10 +556,11 @@ icl_tc_phy_aux_power_well_enable(struct intel_display *display, tc_port = TGL_AUX_PW_TO_TC_PORT(i915_power_well_instance(power_well)->hsw.idx); - if (wait_for(intel_dkl_phy_read(display, DKL_CMN_UC_DW_27(tc_port)) & - DKL_CMN_UC_DW27_UC_HEALTH, 1)) - drm_warn(display->drm, - "Timeout waiting TC uC health\n"); + ret = poll_timeout_us(val = intel_dkl_phy_read(display, DKL_CMN_UC_DW_27(tc_port)), + val & DKL_CMN_UC_DW27_UC_HEALTH, + 100, 1000, false); + if (ret) + drm_warn(display->drm, "Timeout waiting TC uC health\n"); } } @@ -1122,6 +1126,8 @@ static void vlv_set_power_well(struct intel_display *display, u32 mask; u32 state; u32 ctrl; + u32 val; + int ret; mask = PUNIT_PWRGT_MASK(pw_idx); state = enable ? PUNIT_PWRGT_PWR_ON(pw_idx) : @@ -1129,10 +1135,8 @@ static void vlv_set_power_well(struct intel_display *display, vlv_punit_get(display->drm); -#define COND \ - ((vlv_punit_read(display->drm, PUNIT_REG_PWRGT_STATUS) & mask) == state) - - if (COND) + val = vlv_punit_read(display->drm, PUNIT_REG_PWRGT_STATUS); + if ((val & mask) == state) goto out; ctrl = vlv_punit_read(display->drm, PUNIT_REG_PWRGT_CTRL); @@ -1140,14 +1144,15 @@ static void vlv_set_power_well(struct intel_display *display, ctrl |= state; vlv_punit_write(display->drm, PUNIT_REG_PWRGT_CTRL, ctrl); - if (wait_for(COND, 100)) + ret = poll_timeout_us(val = vlv_punit_read(display->drm, PUNIT_REG_PWRGT_STATUS), + (val & mask) == state, + 500, 100 * 1000, false); + if (ret) drm_err(display->drm, "timeout setting power well state %08x (%08x)\n", state, vlv_punit_read(display->drm, PUNIT_REG_PWRGT_CTRL)); -#undef COND - out: vlv_punit_put(display->drm); } @@ -1208,7 +1213,7 @@ static void vlv_init_display_clock_gating(struct intel_display *display) * (and never recovering) in this case. intel_dsi_post_disable() will * clear it when we turn off the display. */ - intel_de_rmw(display, DSPCLK_GATE_D(display), + intel_de_rmw(display, VLV_DSPCLK_GATE_D, ~DPOUNIT_CLOCK_GATE_DISABLE, VRHUNIT_CLOCK_GATE_DISABLE); /* @@ -1711,23 +1716,24 @@ static void chv_set_pipe_power_well(struct intel_display *display, enum pipe pipe = PIPE_A; u32 state; u32 ctrl; + int ret; state = enable ? DP_SSS_PWR_ON(pipe) : DP_SSS_PWR_GATE(pipe); vlv_punit_get(display->drm); -#define COND \ - ((vlv_punit_read(display->drm, PUNIT_REG_DSPSSPM) & DP_SSS_MASK(pipe)) == state) - - if (COND) + ctrl = vlv_punit_read(display->drm, PUNIT_REG_DSPSSPM); + if ((ctrl & DP_SSS_MASK(pipe)) == state) goto out; - ctrl = vlv_punit_read(display->drm, PUNIT_REG_DSPSSPM); ctrl &= ~DP_SSC_MASK(pipe); ctrl |= enable ? DP_SSC_PWR_ON(pipe) : DP_SSC_PWR_GATE(pipe); vlv_punit_write(display->drm, PUNIT_REG_DSPSSPM, ctrl); - if (wait_for(COND, 100)) + ret = poll_timeout_us(ctrl = vlv_punit_read(display->drm, PUNIT_REG_DSPSSPM), + (ctrl & DP_SSS_MASK(pipe)) == state, + 500, 100 * 1000, false); + if (ret) drm_err(display->drm, "timeout setting power well state %08x (%08x)\n", state, @@ -1765,7 +1771,6 @@ static void chv_pipe_power_well_disable(struct intel_display *display, static void tgl_tc_cold_request(struct intel_display *display, bool block) { - struct drm_i915_private *i915 = to_i915(display->drm); u8 tries = 0; int ret; @@ -1798,10 +1803,9 @@ tgl_tc_cold_request(struct intel_display *display, bool block) } if (ret) - drm_err(&i915->drm, "TC cold %sblock failed\n", - block ? "" : "un"); + drm_err(display->drm, "TC cold %sblock failed\n", block ? "" : "un"); else - drm_dbg_kms(&i915->drm, "TC cold %sblock succeeded\n", + drm_dbg_kms(display->drm, "TC cold %sblock succeeded\n", block ? "" : "un"); } diff --git a/drivers/gpu/drm/i915/display/intel_display_regs.h b/drivers/gpu/drm/i915/display/intel_display_regs.h index 7bd09d981cd2..9d71e26a4fa2 100644 --- a/drivers/gpu/drm/i915/display/intel_display_regs.h +++ b/drivers/gpu/drm/i915/display/intel_display_regs.h @@ -2890,6 +2890,7 @@ enum skl_power_gate { #define DP_PIN_ASSIGNMENT_SHIFT(idx) ((idx) * 4) #define DP_PIN_ASSIGNMENT_MASK(idx) (0xf << ((idx) * 4)) #define DP_PIN_ASSIGNMENT(idx, x) ((x) << ((idx) * 4)) +/* See enum intel_tc_pin_assignment for the pin assignment field values. */ #define _TCSS_DDI_STATUS_1 0x161500 #define _TCSS_DDI_STATUS_2 0x161504 @@ -2897,6 +2898,7 @@ enum skl_power_gate { _TCSS_DDI_STATUS_1, \ _TCSS_DDI_STATUS_2)) #define TCSS_DDI_STATUS_PIN_ASSIGNMENT_MASK REG_GENMASK(28, 25) +/* See enum intel_tc_pin_assignment for the pin assignment field values. */ #define TCSS_DDI_STATUS_READY REG_BIT(2) #define TCSS_DDI_STATUS_HPD_LIVE_STATUS_TBT REG_BIT(1) #define TCSS_DDI_STATUS_HPD_LIVE_STATUS_ALT REG_BIT(0) diff --git a/drivers/gpu/drm/i915/display/intel_display_types.h b/drivers/gpu/drm/i915/display/intel_display_types.h index ce45261c4a8f..fd9d2527889b 100644 --- a/drivers/gpu/drm/i915/display/intel_display_types.h +++ b/drivers/gpu/drm/i915/display/intel_display_types.h @@ -50,14 +50,15 @@ #include "intel_display_limits.h" #include "intel_display_power.h" #include "intel_dpll_mgr.h" +#include "intel_dsi_vbt_defs.h" #include "intel_wm_types.h" struct cec_notifier; struct drm_printer; -struct __intel_global_objs_state; struct intel_connector; struct intel_ddi_buf_trans; struct intel_fbc; +struct intel_global_objs_state; struct intel_hdcp_shim; struct intel_tc_port; @@ -593,7 +594,7 @@ struct intel_atomic_state { struct ref_tracker *wakeref; - struct __intel_global_objs_state *global_objs; + struct intel_global_objs_state *global_objs; int num_global_objs; /* Internal commit, as opposed to userspace/client initiated one */ @@ -642,7 +643,6 @@ struct intel_plane_state { #define PLANE_HAS_FENCE BIT(0) struct intel_fb_view view; - u32 phys_dma_addr; /* for cursor_needs_physical */ /* for legacy cursor fb unpin */ struct drm_vblank_work unpin_work; @@ -665,6 +665,9 @@ struct intel_plane_state { /* chroma upsampler control register */ u32 cus_ctl; + /* surface address register */ + u32 surf; + /* * scaler_id * = -1 : not using a scaler @@ -941,10 +944,6 @@ struct intel_csc_matrix { u16 postoff[3]; }; -void intel_io_mmio_fw_write(void *ctx, i915_reg_t reg, u32 val); - -typedef void (*intel_io_reg_write)(void *ctx, i915_reg_t reg, u32 val); - struct intel_crtc_state { /* * uapi (drm) state. This is the software state shown to userspace. @@ -1122,6 +1121,7 @@ struct intel_crtc_state { bool req_psr2_sdp_prior_scanline; bool has_panel_replay; bool wm_level_disabled; + bool pkg_c_latency_used; u32 dc3co_exitline; u16 su_y_granularity; u8 active_non_psr_pipes; @@ -1534,6 +1534,7 @@ struct intel_plane { bool (*get_hw_state)(struct intel_plane *plane, enum pipe *pipe); int (*check_plane)(struct intel_crtc_state *crtc_state, struct intel_plane_state *plane_state); + u32 (*surf_offset)(const struct intel_plane_state *plane_state); int (*min_cdclk)(const struct intel_crtc_state *crtc_state, const struct intel_plane_state *plane_state); void (*async_flip)(struct intel_dsb *dsb, @@ -1683,6 +1684,7 @@ struct intel_psr { u8 entry_setup_frames; bool link_ok; + bool pkg_c_latency_used; u8 active_non_psr_pipes; }; diff --git a/drivers/gpu/drm/i915/display/intel_display_wa.c b/drivers/gpu/drm/i915/display/intel_display_wa.c index f57280e9d041..31cd2c9cd488 100644 --- a/drivers/gpu/drm/i915/display/intel_display_wa.c +++ b/drivers/gpu/drm/i915/display/intel_display_wa.c @@ -3,6 +3,8 @@ * Copyright © 2023 Intel Corporation */ +#include <drm/drm_print.h> + #include "i915_reg.h" #include "intel_de.h" #include "intel_display_core.h" @@ -39,3 +41,36 @@ void intel_display_wa_apply(struct intel_display *display) else if (DISPLAY_VER(display) == 11) gen11_display_wa_apply(display); } + +/* + * Wa_16025573575: + * Fixes: Issue with bitbashing on Xe3 based platforms. + * Workaround: Set masks bits in GPIO CTL and preserve it during bitbashing sequence. + */ +static bool intel_display_needs_wa_16025573575(struct intel_display *display) +{ + return DISPLAY_VERx100(display) == 3000 || DISPLAY_VERx100(display) == 3002; +} + +/* + * Wa_14011503117: + * Fixes: Before enabling the scaler DE fatal error is masked + * Workaround: Unmask the DE fatal error register after enabling the scaler + * and after waiting of at least 1 frame. + */ +bool __intel_display_wa(struct intel_display *display, enum intel_display_wa wa, const char *name) +{ + switch (wa) { + case INTEL_DISPLAY_WA_16023588340: + return intel_display_needs_wa_16023588340(display); + case INTEL_DISPLAY_WA_16025573575: + return intel_display_needs_wa_16025573575(display); + case INTEL_DISPLAY_WA_14011503117: + return DISPLAY_VER(display) == 13; + default: + drm_WARN(display->drm, 1, "Missing Wa number: %s\n", name); + break; + } + + return false; +} diff --git a/drivers/gpu/drm/i915/display/intel_display_wa.h b/drivers/gpu/drm/i915/display/intel_display_wa.h index babd9d16603d..abc1df83f066 100644 --- a/drivers/gpu/drm/i915/display/intel_display_wa.h +++ b/drivers/gpu/drm/i915/display/intel_display_wa.h @@ -21,4 +21,15 @@ static inline bool intel_display_needs_wa_16023588340(struct intel_display *disp bool intel_display_needs_wa_16023588340(struct intel_display *display); #endif +enum intel_display_wa { + INTEL_DISPLAY_WA_16023588340, + INTEL_DISPLAY_WA_16025573575, + INTEL_DISPLAY_WA_14011503117, +}; + +bool __intel_display_wa(struct intel_display *display, enum intel_display_wa wa, const char *name); + +#define intel_display_wa(__display, __wa) \ + __intel_display_wa((__display), INTEL_DISPLAY_WA_##__wa, __stringify(__wa)) + #endif diff --git a/drivers/gpu/drm/i915/display/intel_dmc.c b/drivers/gpu/drm/i915/display/intel_dmc.c index 744f51c0eab8..77a0199f9ea5 100644 --- a/drivers/gpu/drm/i915/display/intel_dmc.c +++ b/drivers/gpu/drm/i915/display/intel_dmc.c @@ -1603,9 +1603,7 @@ DEFINE_SHOW_ATTRIBUTE(intel_dmc_debugfs_status); void intel_dmc_debugfs_register(struct intel_display *display) { - struct drm_minor *minor = display->drm->primary; - - debugfs_create_file("i915_dmc_info", 0444, minor->debugfs_root, + debugfs_create_file("i915_dmc_info", 0444, display->drm->debugfs_root, display, &intel_dmc_debugfs_status_fops); } diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c index 7976fec88606..2eab591a8ef5 100644 --- a/drivers/gpu/drm/i915/display/intel_dp.c +++ b/drivers/gpu/drm/i915/display/intel_dp.c @@ -27,6 +27,7 @@ #include <linux/export.h> #include <linux/i2c.h> +#include <linux/iopoll.h> #include <linux/log2.h> #include <linux/math.h> #include <linux/notifier.h> @@ -174,7 +175,6 @@ int intel_dp_link_symbol_clock(int rate) static int max_dprx_rate(struct intel_dp *intel_dp) { struct intel_display *display = to_intel_display(intel_dp); - struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base; int max_rate; if (intel_dp_tunnel_bw_alloc_is_enabled(intel_dp)) @@ -183,16 +183,13 @@ static int max_dprx_rate(struct intel_dp *intel_dp) max_rate = drm_dp_bw_code_to_link_rate(intel_dp->dpcd[DP_MAX_LINK_RATE]); /* - * Some broken eDP sinks illegally declare support for - * HBR3 without TPS4, and are unable to produce a stable - * output. Reject HBR3 when TPS4 is not available. + * Some platforms + eDP panels may not reliably support HBR3 + * due to signal integrity limitations, despite advertising it. + * Cap the link rate to HBR2 to avoid unstable configurations for the + * known machines. */ - if (max_rate >= 810000 && !drm_dp_tps4_supported(intel_dp->dpcd)) { - drm_dbg_kms(display->drm, - "[ENCODER:%d:%s] Rejecting HBR3 due to missing TPS4 support\n", - encoder->base.base.id, encoder->base.name); - max_rate = 540000; - } + if (intel_dp_is_edp(intel_dp) && intel_has_quirk(display, QUIRK_EDP_LIMIT_RATE_HBR2)) + max_rate = min(max_rate, 540000); return max_rate; } @@ -1418,6 +1415,7 @@ intel_dp_mode_valid(struct drm_connector *_connector, struct intel_display *display = to_intel_display(_connector->dev); struct intel_connector *connector = to_intel_connector(_connector); struct intel_dp *intel_dp = intel_attached_dp(connector); + enum intel_output_format sink_format, output_format; const struct drm_display_mode *fixed_mode; int target_clock = mode->clock; int max_rate, mode_rate, max_lanes, max_link_clock; @@ -1451,6 +1449,13 @@ intel_dp_mode_valid(struct drm_connector *_connector, mode->hdisplay, target_clock); max_dotclk *= num_joined_pipes; + sink_format = intel_dp_sink_format(connector, mode); + output_format = intel_dp_output_format(connector, sink_format); + + status = intel_pfit_mode_valid(display, mode, output_format, num_joined_pipes); + if (status != MODE_OK) + return status; + if (target_clock > max_dotclk) return MODE_CLOCK_HIGH; @@ -1466,11 +1471,8 @@ intel_dp_mode_valid(struct drm_connector *_connector, intel_dp_mode_min_output_bpp(connector, mode)); if (intel_dp_has_dsc(connector)) { - enum intel_output_format sink_format, output_format; int pipe_bpp; - sink_format = intel_dp_sink_format(connector, mode); - output_format = intel_dp_output_format(connector, sink_format); /* * TBD pass the connector BPC, * for now U8_MAX so that max BPC on that platform would be picked @@ -2535,13 +2537,15 @@ intel_dp_dsc_compute_pipe_bpp_limits(struct intel_dp *intel_dp, bool intel_dp_compute_config_limits(struct intel_dp *intel_dp, - struct intel_connector *connector, + struct drm_connector_state *conn_state, struct intel_crtc_state *crtc_state, bool respect_downstream_limits, bool dsc, struct link_config_limits *limits) { bool is_mst = intel_crtc_has_type(crtc_state, INTEL_OUTPUT_DP_MST); + struct intel_connector *connector = + to_intel_connector(conn_state->connector); limits->min_rate = intel_dp_min_link_rate(intel_dp); limits->max_rate = intel_dp_max_link_rate(intel_dp); @@ -2551,7 +2555,8 @@ intel_dp_compute_config_limits(struct intel_dp *intel_dp, limits->min_lane_count = intel_dp_min_lane_count(intel_dp); limits->max_lane_count = intel_dp_max_lane_count(intel_dp); - limits->pipe.min_bpp = intel_dp_min_bpp(crtc_state->output_format); + limits->pipe.min_bpp = intel_dp_in_hdr_mode(conn_state) ? 30 : + intel_dp_min_bpp(crtc_state->output_format); if (is_mst) { /* * FIXME: If all the streams can't fit into the link with their @@ -2650,7 +2655,7 @@ intel_dp_compute_link_config(struct intel_encoder *encoder, joiner_needs_dsc = intel_dp_joiner_needs_dsc(display, num_joined_pipes); dsc_needed = joiner_needs_dsc || intel_dp->force_dsc_en || - !intel_dp_compute_config_limits(intel_dp, connector, pipe_config, + !intel_dp_compute_config_limits(intel_dp, conn_state, pipe_config, respect_downstream_limits, false, &limits); @@ -2684,7 +2689,7 @@ intel_dp_compute_link_config(struct intel_encoder *encoder, str_yes_no(ret), str_yes_no(joiner_needs_dsc), str_yes_no(intel_dp->force_dsc_en)); - if (!intel_dp_compute_config_limits(intel_dp, connector, pipe_config, + if (!intel_dp_compute_config_limits(intel_dp, conn_state, pipe_config, respect_downstream_limits, true, &limits)) @@ -2916,6 +2921,19 @@ static void intel_dp_compute_vsc_sdp(struct intel_dp *intel_dp, } } +bool +intel_dp_in_hdr_mode(const struct drm_connector_state *conn_state) +{ + struct hdr_output_metadata *hdr_metadata; + + if (!conn_state->hdr_output_metadata) + return false; + + hdr_metadata = conn_state->hdr_output_metadata->data; + + return hdr_metadata->hdmi_metadata_type1.eotf == HDMI_EOTF_SMPTE_ST2084; +} + static void intel_dp_compute_hdr_metadata_infoframe_sdp(struct intel_dp *intel_dp, struct intel_crtc_state *crtc_state, @@ -3181,7 +3199,26 @@ int intel_dp_compute_min_hblank(struct intel_crtc_state *crtc_state, */ min_hblank = min_hblank - 2; - min_hblank = min(10, min_hblank); + /* + * min_hblank formula is undergoing a change, to avoid underrun use the + * recomended value in spec to compare with the calculated one and use the + * minimum value + */ + if (intel_dp_is_uhbr(crtc_state)) { + /* + * Note: Bspec requires a min_hblank of 2 for YCBCR420 + * with compressed bpp 6, but the minimum compressed bpp + * supported by the driver is 8. + */ + drm_WARN_ON(display->drm, + (crtc_state->dsc.compression_enable && + crtc_state->output_format == INTEL_OUTPUT_FORMAT_YCBCR420 && + crtc_state->dsc.compressed_bpp_x16 < fxp_q4_from_int(8))); + min_hblank = min(3, min_hblank); + } else { + min_hblank = min(10, min_hblank); + } + crtc_state->min_hblank = min_hblank; return 0; @@ -3842,10 +3879,11 @@ static int intel_dp_pcon_start_frl_training(struct intel_dp *intel_dp) if (ret < 0) return ret; /* Wait for PCON to be FRL Ready */ - wait_for(is_active = drm_dp_pcon_is_frl_ready(&intel_dp->aux) == true, TIMEOUT_FRL_READY_MS); - - if (!is_active) - return -ETIMEDOUT; + ret = poll_timeout_us(is_active = drm_dp_pcon_is_frl_ready(&intel_dp->aux), + is_active, + 1000, TIMEOUT_FRL_READY_MS * 1000, false); + if (ret) + return ret; ret = drm_dp_pcon_frl_configure_1(&intel_dp->aux, max_frl_bw, DP_PCON_ENABLE_SEQUENTIAL_LINK); @@ -3862,12 +3900,11 @@ static int intel_dp_pcon_start_frl_training(struct intel_dp *intel_dp) * Wait for FRL to be completed * Check if the HDMI Link is up and active. */ - wait_for(is_active = - intel_dp_pcon_is_frl_trained(intel_dp, max_frl_bw_mask, &frl_trained_mask), - TIMEOUT_HDMI_LINK_ACTIVE_MS); - - if (!is_active) - return -ETIMEDOUT; + ret = poll_timeout_us(is_active = intel_dp_pcon_is_frl_trained(intel_dp, max_frl_bw_mask, &frl_trained_mask), + is_active, + 1000, TIMEOUT_HDMI_LINK_ACTIVE_MS * 1000, false); + if (ret) + return ret; frl_trained: drm_dbg(display->drm, "FRL_TRAINED_MASK = %u\n", frl_trained_mask); @@ -4277,10 +4314,26 @@ static void intel_edp_mso_init(struct intel_dp *intel_dp) } static void +intel_edp_set_data_override_rates(struct intel_dp *intel_dp) +{ + struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base; + int *sink_rates = intel_dp->sink_rates; + int i, count = 0; + + for (i = 0; i < intel_dp->num_sink_rates; i++) { + if (intel_bios_encoder_reject_edp_rate(encoder->devdata, + intel_dp->sink_rates[i])) + continue; + + sink_rates[count++] = intel_dp->sink_rates[i]; + } + intel_dp->num_sink_rates = count; +} + +static void intel_edp_set_sink_rates(struct intel_dp *intel_dp) { struct intel_display *display = to_intel_display(intel_dp); - struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base; intel_dp->num_sink_rates = 0; @@ -4306,16 +4359,13 @@ intel_edp_set_sink_rates(struct intel_dp *intel_dp) break; /* - * Some broken eDP sinks illegally declare support for - * HBR3 without TPS4, and are unable to produce a stable - * output. Reject HBR3 when TPS4 is not available. + * Some platforms cannot reliably drive HBR3 rates due to PHY limitations, + * even if the sink advertises support. Reject any sink rates above HBR2 on + * the known machines for stable output. */ - if (rate >= 810000 && !drm_dp_tps4_supported(intel_dp->dpcd)) { - drm_dbg_kms(display->drm, - "[ENCODER:%d:%s] Rejecting HBR3 due to missing TPS4 support\n", - encoder->base.base.id, encoder->base.name); + if (rate > 540000 && + intel_has_quirk(display, QUIRK_EDP_LIMIT_RATE_HBR2)) break; - } intel_dp->sink_rates[i] = rate; } @@ -4330,6 +4380,8 @@ intel_edp_set_sink_rates(struct intel_dp *intel_dp) intel_dp->use_rate_select = true; else intel_dp_set_sink_rates(intel_dp); + + intel_edp_set_data_override_rates(intel_dp); } static bool @@ -5611,14 +5663,9 @@ bool intel_digital_port_connected_locked(struct intel_encoder *encoder) intel_wakeref_t wakeref; with_intel_display_power(display, POWER_DOMAIN_DISPLAY_CORE, wakeref) { - unsigned long wait_expires = jiffies + msecs_to_jiffies_timeout(4); - - do { - is_connected = dig_port->connected(encoder); - if (is_connected || is_glitch_free) - break; - usleep_range(10, 30); - } while (time_before(jiffies, wait_expires)); + poll_timeout_us(is_connected = dig_port->connected(encoder), + is_connected || is_glitch_free, + 30, 4000, false); } return is_connected; diff --git a/drivers/gpu/drm/i915/display/intel_dp.h b/drivers/gpu/drm/i915/display/intel_dp.h index 0657f5681196..f90cfd1dbbd0 100644 --- a/drivers/gpu/drm/i915/display/intel_dp.h +++ b/drivers/gpu/drm/i915/display/intel_dp.h @@ -193,7 +193,7 @@ void intel_dp_wait_source_oui(struct intel_dp *intel_dp); int intel_dp_output_bpp(enum intel_output_format output_format, int bpp); bool intel_dp_compute_config_limits(struct intel_dp *intel_dp, - struct intel_connector *connector, + struct drm_connector_state *conn_state, struct intel_crtc_state *crtc_state, bool respect_downstream_limits, bool dsc, @@ -214,5 +214,6 @@ int intel_dp_compute_min_hblank(struct intel_crtc_state *crtc_state, int intel_dp_dsc_bpp_step_x16(const struct intel_connector *connector); void intel_dp_dpcd_set_probe(struct intel_dp *intel_dp, bool force_on_external); +bool intel_dp_in_hdr_mode(const struct drm_connector_state *conn_state); #endif /* __INTEL_DP_H__ */ diff --git a/drivers/gpu/drm/i915/display/intel_dp_aux_backlight.c b/drivers/gpu/drm/i915/display/intel_dp_aux_backlight.c index 41228478b21c..12084a542fc5 100644 --- a/drivers/gpu/drm/i915/display/intel_dp_aux_backlight.c +++ b/drivers/gpu/drm/i915/display/intel_dp_aux_backlight.c @@ -225,19 +225,6 @@ intel_dp_aux_hdr_set_aux_backlight(const struct drm_connector_state *conn_state, connector->base.base.id, connector->base.name); } -static bool -intel_dp_in_hdr_mode(const struct drm_connector_state *conn_state) -{ - struct hdr_output_metadata *hdr_metadata; - - if (!conn_state->hdr_output_metadata) - return false; - - hdr_metadata = conn_state->hdr_output_metadata->data; - - return hdr_metadata->hdmi_metadata_type1.eotf == HDMI_EOTF_SMPTE_ST2084; -} - static void intel_dp_aux_hdr_set_backlight(const struct drm_connector_state *conn_state, u32 level) { diff --git a/drivers/gpu/drm/i915/display/intel_dp_link_training.c b/drivers/gpu/drm/i915/display/intel_dp_link_training.c index a479b63112ea..27f3716bdc1f 100644 --- a/drivers/gpu/drm/i915/display/intel_dp_link_training.c +++ b/drivers/gpu/drm/i915/display/intel_dp_link_training.c @@ -22,6 +22,7 @@ */ #include <linux/debugfs.h> +#include <linux/iopoll.h> #include <drm/display/drm_dp_helper.h> #include <drm/drm_print.h> @@ -478,12 +479,13 @@ static u8 intel_dp_get_lane_adjust_train(struct intel_dp *intel_dp, _TRAIN_REQ_TX_FFE_ARGS(link_status, 2), \ _TRAIN_REQ_TX_FFE_ARGS(link_status, 3) -void +bool intel_dp_get_adjust_train(struct intel_dp *intel_dp, const struct intel_crtc_state *crtc_state, enum drm_dp_phy dp_phy, const u8 link_status[DP_LINK_STATUS_SIZE]) { + bool changed = false; int lane; if (intel_dp_is_uhbr(crtc_state)) { @@ -502,10 +504,17 @@ intel_dp_get_adjust_train(struct intel_dp *intel_dp, TRAIN_REQ_PREEMPH_ARGS(link_status)); } - for (lane = 0; lane < 4; lane++) - intel_dp->train_set[lane] = - intel_dp_get_lane_adjust_train(intel_dp, crtc_state, - dp_phy, link_status, lane); + for (lane = 0; lane < 4; lane++) { + u8 new = intel_dp_get_lane_adjust_train(intel_dp, crtc_state, + dp_phy, link_status, lane); + if (intel_dp->train_set[lane] == new) + continue; + + intel_dp->train_set[lane] = new; + changed = true; + } + + return changed; } static int intel_dp_training_pattern_set_reg(struct intel_dp *intel_dp, @@ -758,6 +767,63 @@ void intel_dp_link_training_set_bw(struct intel_dp *intel_dp, } } +/* + * Pick Training Pattern Sequence (TPS) for channel equalization. 128b/132b TPS2 + * for UHBR+, TPS4 for HBR3 or for 1.4 devices that support it, TPS3 for HBR2 or + * 1.2 devices that support it, TPS2 otherwise. + */ +static u32 intel_dp_training_pattern(struct intel_dp *intel_dp, + const struct intel_crtc_state *crtc_state, + enum drm_dp_phy dp_phy) +{ + struct intel_display *display = to_intel_display(intel_dp); + bool source_tps3, sink_tps3, source_tps4, sink_tps4; + + /* UHBR+ use separate 128b/132b TPS2 */ + if (intel_dp_is_uhbr(crtc_state)) + return DP_TRAINING_PATTERN_2; + + /* + * TPS4 support is mandatory for all downstream devices that + * support HBR3. There are no known eDP panels that support + * TPS4 as of Feb 2018 as per VESA eDP_v1.4b_E1 specification. + * LTTPRs must support TPS4. + */ + source_tps4 = intel_dp_source_supports_tps4(display); + sink_tps4 = dp_phy != DP_PHY_DPRX || + drm_dp_tps4_supported(intel_dp->dpcd); + if (source_tps4 && sink_tps4) { + return DP_TRAINING_PATTERN_4; + } else if (crtc_state->port_clock == 810000) { + if (!source_tps4) + lt_dbg(intel_dp, dp_phy, + "8.1 Gbps link rate without source TPS4 support\n"); + if (!sink_tps4) + lt_dbg(intel_dp, dp_phy, + "8.1 Gbps link rate without sink TPS4 support\n"); + } + + /* + * TPS3 support is mandatory for downstream devices that + * support HBR2. However, not all sinks follow the spec. + */ + source_tps3 = intel_dp_source_supports_tps3(display); + sink_tps3 = dp_phy != DP_PHY_DPRX || + drm_dp_tps3_supported(intel_dp->dpcd); + if (source_tps3 && sink_tps3) { + return DP_TRAINING_PATTERN_3; + } else if (crtc_state->port_clock >= 540000) { + if (!source_tps3) + lt_dbg(intel_dp, dp_phy, + ">=5.4/6.48 Gbps link rate without source TPS3 support\n"); + if (!sink_tps3) + lt_dbg(intel_dp, dp_phy, + ">=5.4/6.48 Gbps link rate without sink TPS3 support\n"); + } + + return DP_TRAINING_PATTERN_2; +} + static void intel_dp_update_link_bw_set(struct intel_dp *intel_dp, const struct intel_crtc_state *crtc_state, u8 link_bw, u8 rate_select) @@ -950,63 +1016,6 @@ intel_dp_link_training_clock_recovery(struct intel_dp *intel_dp, } /* - * Pick Training Pattern Sequence (TPS) for channel equalization. 128b/132b TPS2 - * for UHBR+, TPS4 for HBR3 or for 1.4 devices that support it, TPS3 for HBR2 or - * 1.2 devices that support it, TPS2 otherwise. - */ -static u32 intel_dp_training_pattern(struct intel_dp *intel_dp, - const struct intel_crtc_state *crtc_state, - enum drm_dp_phy dp_phy) -{ - struct intel_display *display = to_intel_display(intel_dp); - bool source_tps3, sink_tps3, source_tps4, sink_tps4; - - /* UHBR+ use separate 128b/132b TPS2 */ - if (intel_dp_is_uhbr(crtc_state)) - return DP_TRAINING_PATTERN_2; - - /* - * TPS4 support is mandatory for all downstream devices that - * support HBR3. There are no known eDP panels that support - * TPS4 as of Feb 2018 as per VESA eDP_v1.4b_E1 specification. - * LTTPRs must support TPS4. - */ - source_tps4 = intel_dp_source_supports_tps4(display); - sink_tps4 = dp_phy != DP_PHY_DPRX || - drm_dp_tps4_supported(intel_dp->dpcd); - if (source_tps4 && sink_tps4) { - return DP_TRAINING_PATTERN_4; - } else if (crtc_state->port_clock == 810000) { - if (!source_tps4) - lt_dbg(intel_dp, dp_phy, - "8.1 Gbps link rate without source TPS4 support\n"); - if (!sink_tps4) - lt_dbg(intel_dp, dp_phy, - "8.1 Gbps link rate without sink TPS4 support\n"); - } - - /* - * TPS3 support is mandatory for downstream devices that - * support HBR2. However, not all sinks follow the spec. - */ - source_tps3 = intel_dp_source_supports_tps3(display); - sink_tps3 = dp_phy != DP_PHY_DPRX || - drm_dp_tps3_supported(intel_dp->dpcd); - if (source_tps3 && sink_tps3) { - return DP_TRAINING_PATTERN_3; - } else if (crtc_state->port_clock >= 540000) { - if (!source_tps3) - lt_dbg(intel_dp, dp_phy, - ">=5.4/6.48 Gbps link rate without source TPS3 support\n"); - if (!sink_tps3) - lt_dbg(intel_dp, dp_phy, - ">=5.4/6.48 Gbps link rate without sink TPS3 support\n"); - } - - return DP_TRAINING_PATTERN_2; -} - -/* * Perform the link training channel equalization phase on the given DP PHY * using one of training pattern 2, 3 or 4 depending on the source and * sink capabilities. @@ -1127,16 +1136,19 @@ void intel_dp_stop_link_train(struct intel_dp *intel_dp, { struct intel_display *display = to_intel_display(intel_dp); struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base; + int ret; intel_dp->link.active = true; - intel_dp_disable_dpcd_training_pattern(intel_dp, DP_PHY_DPRX); intel_dp_program_link_training_pattern(intel_dp, crtc_state, DP_PHY_DPRX, DP_TRAINING_PATTERN_DISABLE); - if (intel_dp_is_uhbr(crtc_state) && - wait_for(intel_dp_128b132b_intra_hop(intel_dp, crtc_state) == 0, 500)) { - lt_dbg(intel_dp, DP_PHY_DPRX, "128b/132b intra-hop not clearing\n"); + if (intel_dp_is_uhbr(crtc_state)) { + ret = poll_timeout_us(ret = intel_dp_128b132b_intra_hop(intel_dp, crtc_state), + ret == 0, + 500, 500 * 1000, false); + if (ret) + lt_dbg(intel_dp, DP_PHY_DPRX, "128b/132b intra-hop not clearing\n"); } intel_hpd_unblock(encoder); @@ -1371,8 +1383,8 @@ intel_dp_link_train_all_phys(struct intel_dp *intel_dp, if (ret) ret = intel_dp_link_train_phy(intel_dp, crtc_state, DP_PHY_DPRX); - if (intel_dp->set_idle_link_train) - intel_dp->set_idle_link_train(intel_dp, crtc_state); + intel_dp_disable_dpcd_training_pattern(intel_dp, DP_PHY_DPRX); + intel_dp->set_idle_link_train(intel_dp, crtc_state); return ret; } @@ -1574,8 +1586,12 @@ intel_dp_128b132b_link_train(struct intel_dp *intel_dp, int lttpr_count) { bool passed = false; + int ret; - if (wait_for(intel_dp_128b132b_intra_hop(intel_dp, crtc_state) == 0, 500)) { + ret = poll_timeout_us(ret = intel_dp_128b132b_intra_hop(intel_dp, crtc_state), + ret == 0, + 500, 500 * 1000, false); + if (ret) { lt_err(intel_dp, DP_PHY_DPRX, "128b/132b intra-hop not clear\n"); goto out; } @@ -1602,6 +1618,8 @@ out: intel_dp_program_link_training_pattern(intel_dp, crtc_state, DP_PHY_DPRX, DP_TRAINING_PATTERN_2); + intel_dp_disable_dpcd_training_pattern(intel_dp, DP_PHY_DPRX); + return passed; } diff --git a/drivers/gpu/drm/i915/display/intel_dp_link_training.h b/drivers/gpu/drm/i915/display/intel_dp_link_training.h index 46614124569f..1ba22ed6db08 100644 --- a/drivers/gpu/drm/i915/display/intel_dp_link_training.h +++ b/drivers/gpu/drm/i915/display/intel_dp_link_training.h @@ -23,7 +23,7 @@ void intel_dp_link_training_set_bw(struct intel_dp *intel_dp, int link_bw, int rate_select, int lane_count, bool enhanced_framing); -void intel_dp_get_adjust_train(struct intel_dp *intel_dp, +bool intel_dp_get_adjust_train(struct intel_dp *intel_dp, const struct intel_crtc_state *crtc_state, enum drm_dp_phy dp_phy, const u8 link_status[DP_LINK_STATUS_SIZE]); diff --git a/drivers/gpu/drm/i915/display/intel_dp_mst.c b/drivers/gpu/drm/i915/display/intel_dp_mst.c index 74497c9a0554..352f7ef29c28 100644 --- a/drivers/gpu/drm/i915/display/intel_dp_mst.c +++ b/drivers/gpu/drm/i915/display/intel_dp_mst.c @@ -611,12 +611,15 @@ adjust_limits_for_dsc_hblank_expansion_quirk(struct intel_dp *intel_dp, static bool mst_stream_compute_config_limits(struct intel_dp *intel_dp, - struct intel_connector *connector, + struct drm_connector_state *conn_state, struct intel_crtc_state *crtc_state, bool dsc, struct link_config_limits *limits) { - if (!intel_dp_compute_config_limits(intel_dp, connector, + struct intel_connector *connector = + to_intel_connector(conn_state->connector); + + if (!intel_dp_compute_config_limits(intel_dp, conn_state, crtc_state, false, dsc, limits)) return false; @@ -665,7 +668,7 @@ static int mst_stream_compute_config(struct intel_encoder *encoder, joiner_needs_dsc = intel_dp_joiner_needs_dsc(display, num_joined_pipes); dsc_needed = joiner_needs_dsc || intel_dp->force_dsc_en || - !mst_stream_compute_config_limits(intel_dp, connector, + !mst_stream_compute_config_limits(intel_dp, conn_state, pipe_config, false, &limits); if (!dsc_needed) { @@ -691,7 +694,7 @@ static int mst_stream_compute_config(struct intel_encoder *encoder, str_yes_no(intel_dp->force_dsc_en)); - if (!mst_stream_compute_config_limits(intel_dp, connector, + if (!mst_stream_compute_config_limits(intel_dp, conn_state, pipe_config, true, &limits)) return -EINVAL; diff --git a/drivers/gpu/drm/i915/display/intel_dp_test.c b/drivers/gpu/drm/i915/display/intel_dp_test.c index 6ed5012c5fac..5cfa1dd411da 100644 --- a/drivers/gpu/drm/i915/display/intel_dp_test.c +++ b/drivers/gpu/drm/i915/display/intel_dp_test.c @@ -6,7 +6,6 @@ #include <drm/display/drm_dp.h> #include <drm/display/drm_dp_helper.h> #include <drm/drm_edid.h> -#include <drm/drm_file.h> #include <drm/drm_print.h> #include <drm/drm_probe_helper.h> @@ -753,13 +752,12 @@ static const struct { void intel_dp_test_debugfs_register(struct intel_display *display) { - struct drm_minor *minor = display->drm->primary; int i; for (i = 0; i < ARRAY_SIZE(intel_display_debugfs_files); i++) { debugfs_create_file(intel_display_debugfs_files[i].name, 0644, - minor->debugfs_root, + display->drm->debugfs_root, display, intel_display_debugfs_files[i].fops); } diff --git a/drivers/gpu/drm/i915/display/intel_dpll_mgr.c b/drivers/gpu/drm/i915/display/intel_dpll_mgr.c index 33e0398120c8..8ea96cc524a1 100644 --- a/drivers/gpu/drm/i915/display/intel_dpll_mgr.c +++ b/drivers/gpu/drm/i915/display/intel_dpll_mgr.c @@ -2046,6 +2046,7 @@ static void bxt_ddi_pll_enable(struct intel_display *display, enum dpio_phy phy = DPIO_PHY0; enum dpio_channel ch = DPIO_CH0; u32 temp; + int ret; bxt_port_to_phy_channel(display, port, &phy, &ch); @@ -2056,8 +2057,10 @@ static void bxt_ddi_pll_enable(struct intel_display *display, intel_de_rmw(display, BXT_PORT_PLL_ENABLE(port), 0, PORT_PLL_POWER_ENABLE); - if (wait_for_us((intel_de_read(display, BXT_PORT_PLL_ENABLE(port)) & - PORT_PLL_POWER_STATE), 200)) + ret = intel_de_wait_custom(display, BXT_PORT_PLL_ENABLE(port), + PORT_PLL_POWER_STATE, PORT_PLL_POWER_STATE, + 200, 0, NULL); + if (ret) drm_err(display->drm, "Power state not set for PLL:%d\n", port); } @@ -2119,8 +2122,10 @@ static void bxt_ddi_pll_enable(struct intel_display *display, intel_de_rmw(display, BXT_PORT_PLL_ENABLE(port), 0, PORT_PLL_ENABLE); intel_de_posting_read(display, BXT_PORT_PLL_ENABLE(port)); - if (wait_for_us((intel_de_read(display, BXT_PORT_PLL_ENABLE(port)) & PORT_PLL_LOCK), - 200)) + ret = intel_de_wait_custom(display, BXT_PORT_PLL_ENABLE(port), + PORT_PLL_LOCK, PORT_PLL_LOCK, + 200, 0, NULL); + if (ret) drm_err(display->drm, "PLL %d not locked\n", port); if (display->platform.geminilake) { @@ -2144,6 +2149,7 @@ static void bxt_ddi_pll_disable(struct intel_display *display, struct intel_dpll *pll) { enum port port = (enum port)pll->info->id; /* 1:1 port->PLL mapping */ + int ret; intel_de_rmw(display, BXT_PORT_PLL_ENABLE(port), PORT_PLL_ENABLE, 0); intel_de_posting_read(display, BXT_PORT_PLL_ENABLE(port)); @@ -2152,8 +2158,10 @@ static void bxt_ddi_pll_disable(struct intel_display *display, intel_de_rmw(display, BXT_PORT_PLL_ENABLE(port), PORT_PLL_POWER_ENABLE, 0); - if (wait_for_us(!(intel_de_read(display, BXT_PORT_PLL_ENABLE(port)) & - PORT_PLL_POWER_STATE), 200)) + ret = intel_de_wait_custom(display, BXT_PORT_PLL_ENABLE(port), + PORT_PLL_POWER_STATE, 0, + 200, 0, NULL); + if (ret) drm_err(display->drm, "Power state not reset for PLL:%d\n", port); } diff --git a/drivers/gpu/drm/i915/display/intel_dpt.c b/drivers/gpu/drm/i915/display/intel_dpt.c index aea249e2699f..c0a817018d08 100644 --- a/drivers/gpu/drm/i915/display/intel_dpt.c +++ b/drivers/gpu/drm/i915/display/intel_dpt.c @@ -33,8 +33,6 @@ i915_vm_to_dpt(struct i915_address_space *vm) return container_of(vm, struct i915_dpt, vm); } -#define dpt_total_entries(dpt) ((dpt)->vm.total >> PAGE_SHIFT) - static void gen8_set_pte(void __iomem *addr, gen8_pte_t pte) { writeq(pte, addr); @@ -322,5 +320,5 @@ void intel_dpt_destroy(struct i915_address_space *vm) u64 intel_dpt_offset(struct i915_vma *dpt_vma) { - return dpt_vma->node.start; + return i915_vma_offset(dpt_vma); } diff --git a/drivers/gpu/drm/i915/display/intel_dsb.c b/drivers/gpu/drm/i915/display/intel_dsb.c index 53d8ae3a70e9..dee44d45b668 100644 --- a/drivers/gpu/drm/i915/display/intel_dsb.c +++ b/drivers/gpu/drm/i915/display/intel_dsb.c @@ -4,10 +4,11 @@ * */ +#include <linux/iopoll.h> + #include <drm/drm_print.h> #include <drm/drm_vblank.h> -#include "i915_utils.h" #include "intel_crtc.h" #include "intel_de.h" #include "intel_display_regs.h" @@ -871,8 +872,13 @@ void intel_dsb_wait(struct intel_dsb *dsb) struct intel_crtc *crtc = dsb->crtc; struct intel_display *display = to_intel_display(crtc->base.dev); enum pipe pipe = crtc->pipe; + bool is_busy; + int ret; - if (wait_for(!is_dsb_busy(display, pipe, dsb->id), 1)) { + ret = poll_timeout_us(is_busy = is_dsb_busy(display, pipe, dsb->id), + !is_busy, + 100, 1000, false); + if (ret) { u32 offset = intel_dsb_buffer_ggtt_offset(&dsb->dsb_buf); intel_de_write_fw(display, DSB_CTRL(pipe, dsb->id), diff --git a/drivers/gpu/drm/i915/display/intel_dsi_vbt.c b/drivers/gpu/drm/i915/display/intel_dsi_vbt.c index e6a851d276f8..23402408e172 100644 --- a/drivers/gpu/drm/i915/display/intel_dsi_vbt.c +++ b/drivers/gpu/drm/i915/display/intel_dsi_vbt.c @@ -777,7 +777,7 @@ bool intel_dsi_vbt_init(struct intel_dsi *intel_dsi, u16 panel_id) intel_dsi->init_count = mipi_config->master_init_timer; intel_dsi->bw_timer = mipi_config->dbi_bw_timer; intel_dsi->video_frmt_cfg_bits = - mipi_config->bta_enabled ? DISABLE_VIDEO_BTA : 0; + mipi_config->bta_disable ? DISABLE_VIDEO_BTA : 0; intel_dsi->bgr_enabled = mipi_config->rgb_flip; /* Starting point, adjusted depending on dual link and burst mode */ diff --git a/drivers/gpu/drm/i915/display/intel_dsi_vbt_defs.h b/drivers/gpu/drm/i915/display/intel_dsi_vbt_defs.h new file mode 100644 index 000000000000..edc7331dcca2 --- /dev/null +++ b/drivers/gpu/drm/i915/display/intel_dsi_vbt_defs.h @@ -0,0 +1,197 @@ +/* SPDX-License-Identifier: MIT */ +/* Copyright © 2025 Intel Corporation */ + +#ifndef __INTEL_DSI_VBT_DEFS_H__ +#define __INTEL_DSI_VBT_DEFS_H__ + +#include <linux/types.h> + +/* + * MIPI Sequence Block definitions + * + * Note the VBT spec has AssertReset / DeassertReset swapped from their + * usual naming, we use the proper names here to avoid confusion when + * reading the code. + */ +enum mipi_seq { + MIPI_SEQ_END = 0, + MIPI_SEQ_DEASSERT_RESET, /* Spec says MipiAssertResetPin */ + MIPI_SEQ_INIT_OTP, + MIPI_SEQ_DISPLAY_ON, + MIPI_SEQ_DISPLAY_OFF, + MIPI_SEQ_ASSERT_RESET, /* Spec says MipiDeassertResetPin */ + MIPI_SEQ_BACKLIGHT_ON, /* sequence block v2+ */ + MIPI_SEQ_BACKLIGHT_OFF, /* sequence block v2+ */ + MIPI_SEQ_TEAR_ON, /* sequence block v2+ */ + MIPI_SEQ_TEAR_OFF, /* sequence block v3+ */ + MIPI_SEQ_POWER_ON, /* sequence block v3+ */ + MIPI_SEQ_POWER_OFF, /* sequence block v3+ */ + MIPI_SEQ_MAX +}; + +enum mipi_seq_element { + MIPI_SEQ_ELEM_END = 0, + MIPI_SEQ_ELEM_SEND_PKT, + MIPI_SEQ_ELEM_DELAY, + MIPI_SEQ_ELEM_GPIO, + MIPI_SEQ_ELEM_I2C, /* sequence block v2+ */ + MIPI_SEQ_ELEM_SPI, /* sequence block v3+ */ + MIPI_SEQ_ELEM_PMIC, /* sequence block v3+ */ + MIPI_SEQ_ELEM_MAX +}; + +#define MIPI_DSI_UNDEFINED_PANEL_ID 0 +#define MIPI_DSI_GENERIC_PANEL_ID 1 + +struct mipi_config { + u16 panel_id; + + /* General Params */ + struct { + u32 enable_dithering:1; + u32 rsvd1:1; + u32 is_bridge:1; + + u32 panel_arch_type:2; + u32 is_cmd_mode:1; + +#define NON_BURST_SYNC_PULSE 0x1 +#define NON_BURST_SYNC_EVENTS 0x2 +#define BURST_MODE 0x3 + u32 video_transfer_mode:2; + + u32 cabc_supported:1; +#define PPS_BLC_PMIC 0 +#define PPS_BLC_SOC 1 + u32 pwm_blc:1; + +#define PIXEL_FORMAT_RGB565 0x1 +#define PIXEL_FORMAT_RGB666 0x2 +#define PIXEL_FORMAT_RGB666_LOOSELY_PACKED 0x3 +#define PIXEL_FORMAT_RGB888 0x4 + u32 videomode_color_format:4; + +#define ENABLE_ROTATION_0 0x0 +#define ENABLE_ROTATION_90 0x1 +#define ENABLE_ROTATION_180 0x2 +#define ENABLE_ROTATION_270 0x3 + u32 rotation:2; + u32 bta_disable:1; + u32 rsvd2:15; + } __packed; + + /* Port Desc */ + struct { +#define DUAL_LINK_NOT_SUPPORTED 0 +#define DUAL_LINK_FRONT_BACK 1 +#define DUAL_LINK_PIXEL_ALT 2 + u16 dual_link:2; + u16 lane_cnt:2; + u16 pixel_overlap:3; + u16 rgb_flip:1; +#define DL_DCS_PORT_A 0x00 +#define DL_DCS_PORT_C 0x01 +#define DL_DCS_PORT_A_AND_C 0x02 + u16 dl_dcs_cabc_ports:2; + u16 dl_dcs_backlight_ports:2; + u16 port_sync:1; /* 219-230 */ + u16 rsvd3:3; + } __packed; + + /* DSI Controller Parameters */ + struct { + u16 dsi_usage:1; + u16 rsvd4:15; + } __packed; + + u8 rsvd5; + u32 target_burst_mode_freq; + u32 dsi_ddr_clk; + u32 bridge_ref_clk; + + /* LP Byte Clock */ + struct { +#define BYTE_CLK_SEL_20MHZ 0 +#define BYTE_CLK_SEL_10MHZ 1 +#define BYTE_CLK_SEL_5MHZ 2 + u8 byte_clk_sel:2; + u8 rsvd6:6; + } __packed; + + /* DPhy Flags */ + struct { + u16 dphy_param_valid:1; + u16 eot_pkt_disabled:1; + u16 enable_clk_stop:1; + u16 blanking_packets_during_bllp:1; /* 219+ */ + u16 lp_clock_during_lpm:1; /* 219+ */ + u16 rsvd7:11; + } __packed; + + u32 hs_tx_timeout; + u32 lp_rx_timeout; + u32 turn_around_timeout; + u32 device_reset_timer; + u32 master_init_timer; + u32 dbi_bw_timer; + u32 lp_byte_clk_val; + + /* DPhy Params */ + struct { + u32 prepare_cnt:6; + u32 rsvd8:2; + u32 clk_zero_cnt:8; + u32 trail_cnt:5; + u32 rsvd9:3; + u32 exit_zero_cnt:6; + u32 rsvd10:2; + } __packed; + + u32 clk_lane_switch_cnt; + u32 hl_switch_cnt; + + u32 rsvd11[6]; + + /* timings based on dphy spec */ + u8 tclk_miss; + u8 tclk_post; + u8 rsvd12; + u8 tclk_pre; + u8 tclk_prepare; + u8 tclk_settle; + u8 tclk_term_enable; + u8 tclk_trail; + u16 tclk_prepare_clkzero; + u8 rsvd13; + u8 td_term_enable; + u8 teot; + u8 ths_exit; + u8 ths_prepare; + u16 ths_prepare_hszero; + u8 rsvd14; + u8 ths_settle; + u8 ths_skip; + u8 ths_trail; + u8 tinit; + u8 tlpx; + u8 rsvd15[3]; + + /* GPIOs */ + u8 panel_enable; + u8 bl_enable; + u8 pwm_enable; + u8 reset_r_n; + u8 pwr_down_r; + u8 stdby_r_n; +} __packed; + +/* all delays have a unit of 100us */ +struct mipi_pps_data { + u16 panel_on_delay; + u16 bl_enable_delay; + u16 bl_disable_delay; + u16 panel_off_delay; + u16 panel_power_cycle_delay; +} __packed; + +#endif /* __INTEL_DSI_VBT_DEFS_H__ */ diff --git a/drivers/gpu/drm/i915/display/intel_encoder.c b/drivers/gpu/drm/i915/display/intel_encoder.c index 0b7bd26f4339..2ffe1f251ef8 100644 --- a/drivers/gpu/drm/i915/display/intel_encoder.c +++ b/drivers/gpu/drm/i915/display/intel_encoder.c @@ -8,6 +8,7 @@ #include "intel_display_core.h" #include "intel_display_types.h" #include "intel_encoder.h" +#include "intel_hotplug.h" static void intel_encoder_link_check_work_fn(struct work_struct *work) { @@ -37,6 +38,28 @@ void intel_encoder_link_check_queue_work(struct intel_encoder *encoder, int dela &encoder->link_check_work, msecs_to_jiffies(delay_ms)); } +void intel_encoder_unblock_all_hpds(struct intel_display *display) +{ + struct intel_encoder *encoder; + + if (!HAS_DISPLAY(display)) + return; + + for_each_intel_encoder(display->drm, encoder) + intel_hpd_unblock(encoder); +} + +void intel_encoder_block_all_hpds(struct intel_display *display) +{ + struct intel_encoder *encoder; + + if (!HAS_DISPLAY(display)) + return; + + for_each_intel_encoder(display->drm, encoder) + intel_hpd_block(encoder); +} + void intel_encoder_suspend_all(struct intel_display *display) { struct intel_encoder *encoder; @@ -80,3 +103,21 @@ void intel_encoder_shutdown_all(struct intel_display *display) if (encoder->shutdown_complete) encoder->shutdown_complete(encoder); } + +struct intel_digital_port *intel_dig_port_alloc(void) +{ + struct intel_digital_port *dig_port; + + dig_port = kzalloc(sizeof(*dig_port), GFP_KERNEL); + if (!dig_port) + return NULL; + + dig_port->hdmi.hdmi_reg = INVALID_MMIO_REG; + dig_port->dp.output_reg = INVALID_MMIO_REG; + dig_port->aux_ch = AUX_CH_NONE; + dig_port->max_lanes = 4; + + mutex_init(&dig_port->hdcp.mutex); + + return dig_port; +} diff --git a/drivers/gpu/drm/i915/display/intel_encoder.h b/drivers/gpu/drm/i915/display/intel_encoder.h index 3fa5589f0b1c..ace0fe1a8f27 100644 --- a/drivers/gpu/drm/i915/display/intel_encoder.h +++ b/drivers/gpu/drm/i915/display/intel_encoder.h @@ -6,6 +6,7 @@ #ifndef __INTEL_ENCODER_H__ #define __INTEL_ENCODER_H__ +struct intel_digital_port; struct intel_display; struct intel_encoder; @@ -17,4 +18,9 @@ void intel_encoder_link_check_flush_work(struct intel_encoder *encoder); void intel_encoder_suspend_all(struct intel_display *display); void intel_encoder_shutdown_all(struct intel_display *display); +void intel_encoder_block_all_hpds(struct intel_display *display); +void intel_encoder_unblock_all_hpds(struct intel_display *display); + +struct intel_digital_port *intel_dig_port_alloc(void); + #endif /* __INTEL_ENCODER_H__ */ diff --git a/drivers/gpu/drm/i915/display/intel_fb.c b/drivers/gpu/drm/i915/display/intel_fb.c index 0da842bd2f2f..b210c3250501 100644 --- a/drivers/gpu/drm/i915/display/intel_fb.c +++ b/drivers/gpu/drm/i915/display/intel_fb.c @@ -11,6 +11,7 @@ #include <drm/drm_modeset_helper.h> #include "i915_drv.h" +#include "i915_utils.h" #include "intel_bo.h" #include "intel_display.h" #include "intel_display_core.h" diff --git a/drivers/gpu/drm/i915/display/intel_fb_pin.c b/drivers/gpu/drm/i915/display/intel_fb_pin.c index 5a0151775a3a..45af04cb0fb2 100644 --- a/drivers/gpu/drm/i915/display/intel_fb_pin.c +++ b/drivers/gpu/drm/i915/display/intel_fb_pin.c @@ -11,6 +11,7 @@ #include "gem/i915_gem_object.h" #include "i915_drv.h" +#include "i915_vma.h" #include "intel_display_core.h" #include "intel_display_rpm.h" #include "intel_display_types.h" @@ -151,7 +152,7 @@ intel_fb_pin_to_ggtt(const struct drm_framebuffer *fb, * happy to scanout from anywhere within its global aperture. */ pinctl = 0; - if (HAS_GMCH(dev_priv)) + if (HAS_GMCH(display)) pinctl |= PIN_MAPPABLE; i915_gem_ww_ctx_init(&ww, true); @@ -192,7 +193,7 @@ retry: * mode that matches the user configuration. */ ret = i915_vma_pin_fence(vma); - if (ret != 0 && DISPLAY_VER(dev_priv) < 4) { + if (ret != 0 && DISPLAY_VER(display) < 4) { i915_vma_unpin(vma); goto err_unpin; } @@ -260,6 +261,7 @@ intel_plane_fb_vtd_guard(const struct intel_plane_state *plane_state) int intel_plane_pin_fb(struct intel_plane_state *plane_state, const struct intel_plane_state *old_plane_state) { + struct intel_display *display = to_intel_display(plane_state); struct intel_plane *plane = to_intel_plane(plane_state->uapi.plane); const struct intel_framebuffer *fb = to_intel_framebuffer(plane_state->hw.fb); @@ -277,17 +279,6 @@ int intel_plane_pin_fb(struct intel_plane_state *plane_state, plane_state->ggtt_vma = vma; - /* - * Pre-populate the dma address before we enter the vblank - * evade critical section as i915_gem_object_get_dma_address() - * will trigger might_sleep() even if it won't actually sleep, - * which is the case when the fb has already been pinned. - */ - if (intel_plane_needs_physical(plane)) { - struct drm_i915_gem_object *obj = to_intel_bo(intel_fb_bo(&fb->base)); - - plane_state->phys_dma_addr = i915_gem_object_get_dma_address(obj, 0); - } } else { unsigned int alignment = intel_plane_fb_min_alignment(plane_state); @@ -309,6 +300,28 @@ int intel_plane_pin_fb(struct intel_plane_state *plane_state, plane_state->dpt_vma = vma; WARN_ON(plane_state->ggtt_vma == plane_state->dpt_vma); + + /* + * The DPT object contains only one vma, and there is no VT-d + * guard, so the VMA's offset within the DPT is always 0. + */ + drm_WARN_ON(display->drm, intel_dpt_offset(plane_state->dpt_vma)); + } + + /* + * Pre-populate the dma address before we enter the vblank + * evade critical section as i915_gem_object_get_dma_address() + * will trigger might_sleep() even if it won't actually sleep, + * which is the case when the fb has already been pinned. + */ + if (intel_plane_needs_physical(plane)) { + struct drm_i915_gem_object *obj = to_intel_bo(intel_fb_bo(&fb->base)); + + plane_state->surf = i915_gem_object_get_dma_address(obj, 0) + + plane->surf_offset(plane_state); + } else { + plane_state->surf = i915_ggtt_offset(plane_state->ggtt_vma) + + plane->surf_offset(plane_state); } return 0; diff --git a/drivers/gpu/drm/i915/display/intel_fbc.c b/drivers/gpu/drm/i915/display/intel_fbc.c index 685ac98bd001..d4c5deff9cbe 100644 --- a/drivers/gpu/drm/i915/display/intel_fbc.c +++ b/drivers/gpu/drm/i915/display/intel_fbc.c @@ -1460,7 +1460,7 @@ static int intel_fbc_check_plane(struct intel_atomic_state *state, return 0; } - if (intel_display_needs_wa_16023588340(display)) { + if (intel_display_wa(display, 16023588340)) { plane_state->no_fbc_reason = "Wa_16023588340"; return 0; } @@ -2240,10 +2240,9 @@ void intel_fbc_crtc_debugfs_add(struct intel_crtc *crtc) /* FIXME: remove this once igt is on board with per-crtc stuff */ void intel_fbc_debugfs_register(struct intel_display *display) { - struct drm_minor *minor = display->drm->primary; struct intel_fbc *fbc; fbc = display->fbc[INTEL_FBC_A]; if (fbc) - intel_fbc_debugfs_add(fbc, minor->debugfs_root); + intel_fbc_debugfs_add(fbc, display->drm->debugfs_root); } diff --git a/drivers/gpu/drm/i915/display/intel_fdi.c b/drivers/gpu/drm/i915/display/intel_fdi.c index 8039a84671cc..59a36b3a22c1 100644 --- a/drivers/gpu/drm/i915/display/intel_fdi.c +++ b/drivers/gpu/drm/i915/display/intel_fdi.c @@ -292,34 +292,6 @@ int intel_fdi_link_freq(struct intel_display *display, return display->fdi.pll_freq; } -/** - * intel_fdi_compute_pipe_bpp - compute pipe bpp limited by max link bpp - * @crtc_state: the crtc state - * - * Compute the pipe bpp limited by the CRTC's maximum link bpp. Encoders can - * call this function during state computation in the simple case where the - * link bpp will always match the pipe bpp. This is the case for all non-DP - * encoders, while DP encoders will use a link bpp lower than pipe bpp in case - * of DSC compression. - * - * Returns %true in case of success, %false if pipe bpp would need to be - * reduced below its valid range. - */ -bool intel_fdi_compute_pipe_bpp(struct intel_crtc_state *crtc_state) -{ - int pipe_bpp = min(crtc_state->pipe_bpp, - fxp_q4_to_int(crtc_state->max_link_bpp_x16)); - - pipe_bpp = rounddown(pipe_bpp, 2 * 3); - - if (pipe_bpp < 6 * 3) - return false; - - crtc_state->pipe_bpp = pipe_bpp; - - return true; -} - int ilk_fdi_compute_config(struct intel_crtc *crtc, struct intel_crtc_state *pipe_config) { diff --git a/drivers/gpu/drm/i915/display/intel_fdi.h b/drivers/gpu/drm/i915/display/intel_fdi.h index ad5e103c38a8..1cd08df9b0c2 100644 --- a/drivers/gpu/drm/i915/display/intel_fdi.h +++ b/drivers/gpu/drm/i915/display/intel_fdi.h @@ -20,7 +20,6 @@ struct intel_link_bw_limits; int intel_fdi_add_affected_crtcs(struct intel_atomic_state *state); int intel_fdi_link_freq(struct intel_display *display, const struct intel_crtc_state *pipe_config); -bool intel_fdi_compute_pipe_bpp(struct intel_crtc_state *crtc_state); int ilk_fdi_compute_config(struct intel_crtc *intel_crtc, struct intel_crtc_state *pipe_config); int intel_fdi_atomic_check_link(struct intel_atomic_state *state, diff --git a/drivers/gpu/drm/i915/display/intel_global_state.c b/drivers/gpu/drm/i915/display/intel_global_state.c index 000a898c9480..30eff6009e87 100644 --- a/drivers/gpu/drm/i915/display/intel_global_state.c +++ b/drivers/gpu/drm/i915/display/intel_global_state.c @@ -13,6 +13,36 @@ #include "intel_display_types.h" #include "intel_global_state.h" +#define for_each_new_global_obj_in_state(__state, obj, new_obj_state, __i) \ + for ((__i) = 0; \ + (__i) < (__state)->num_global_objs && \ + ((obj) = (__state)->global_objs[__i].ptr, \ + (new_obj_state) = (__state)->global_objs[__i].new_state, 1); \ + (__i)++) \ + for_each_if(obj) + +#define for_each_old_global_obj_in_state(__state, obj, old_obj_state, __i) \ + for ((__i) = 0; \ + (__i) < (__state)->num_global_objs && \ + ((obj) = (__state)->global_objs[__i].ptr, \ + (old_obj_state) = (__state)->global_objs[__i].old_state, 1); \ + (__i)++) \ + for_each_if(obj) + +#define for_each_oldnew_global_obj_in_state(__state, obj, old_obj_state, new_obj_state, __i) \ + for ((__i) = 0; \ + (__i) < (__state)->num_global_objs && \ + ((obj) = (__state)->global_objs[__i].ptr, \ + (old_obj_state) = (__state)->global_objs[__i].old_state, \ + (new_obj_state) = (__state)->global_objs[__i].new_state, 1); \ + (__i)++) \ + for_each_if(obj) + +struct intel_global_objs_state { + struct intel_global_obj *ptr; + struct intel_global_state *state, *old_state, *new_state; +}; + struct intel_global_commit { struct kref ref; struct completion done; @@ -148,7 +178,7 @@ intel_atomic_get_global_obj_state(struct intel_atomic_state *state, struct intel_display *display = to_intel_display(state); int index, num_objs, i; size_t size; - struct __intel_global_objs_state *arr; + struct intel_global_objs_state *arr; struct intel_global_state *obj_state; for (i = 0; i < state->num_global_objs; i++) diff --git a/drivers/gpu/drm/i915/display/intel_global_state.h b/drivers/gpu/drm/i915/display/intel_global_state.h index d42fb2547ee9..e1efa530cc86 100644 --- a/drivers/gpu/drm/i915/display/intel_global_state.h +++ b/drivers/gpu/drm/i915/display/intel_global_state.h @@ -11,6 +11,7 @@ struct intel_atomic_state; struct intel_display; +struct intel_global_commit; struct intel_global_obj; struct intel_global_state; @@ -26,36 +27,6 @@ struct intel_global_obj { const struct intel_global_state_funcs *funcs; }; -#define intel_for_each_global_obj(obj, dev_priv) \ - list_for_each_entry(obj, &(dev_priv)->display.global.obj_list, head) - -#define for_each_new_global_obj_in_state(__state, obj, new_obj_state, __i) \ - for ((__i) = 0; \ - (__i) < (__state)->num_global_objs && \ - ((obj) = (__state)->global_objs[__i].ptr, \ - (new_obj_state) = (__state)->global_objs[__i].new_state, 1); \ - (__i)++) \ - for_each_if(obj) - -#define for_each_old_global_obj_in_state(__state, obj, old_obj_state, __i) \ - for ((__i) = 0; \ - (__i) < (__state)->num_global_objs && \ - ((obj) = (__state)->global_objs[__i].ptr, \ - (old_obj_state) = (__state)->global_objs[__i].old_state, 1); \ - (__i)++) \ - for_each_if(obj) - -#define for_each_oldnew_global_obj_in_state(__state, obj, old_obj_state, new_obj_state, __i) \ - for ((__i) = 0; \ - (__i) < (__state)->num_global_objs && \ - ((obj) = (__state)->global_objs[__i].ptr, \ - (old_obj_state) = (__state)->global_objs[__i].old_state, \ - (new_obj_state) = (__state)->global_objs[__i].new_state, 1); \ - (__i)++) \ - for_each_if(obj) - -struct intel_global_commit; - struct intel_global_state { struct intel_global_obj *obj; struct intel_atomic_state *state; @@ -64,11 +35,6 @@ struct intel_global_state { bool changed, serialized; }; -struct __intel_global_objs_state { - struct intel_global_obj *ptr; - struct intel_global_state *state, *old_state, *new_state; -}; - void intel_atomic_global_obj_init(struct intel_display *display, struct intel_global_obj *obj, struct intel_global_state *state, diff --git a/drivers/gpu/drm/i915/display/intel_gmbus.c b/drivers/gpu/drm/i915/display/intel_gmbus.c index 0d73f32fe7f1..358210adb8f8 100644 --- a/drivers/gpu/drm/i915/display/intel_gmbus.c +++ b/drivers/gpu/drm/i915/display/intel_gmbus.c @@ -30,6 +30,7 @@ #include <linux/export.h> #include <linux/i2c-algo-bit.h> #include <linux/i2c.h> +#include <linux/iopoll.h> #include <drm/display/drm_hdcp_helper.h> @@ -39,6 +40,7 @@ #include "intel_de.h" #include "intel_display_regs.h" #include "intel_display_types.h" +#include "intel_display_wa.h" #include "intel_gmbus.h" #include "intel_gmbus_regs.h" @@ -217,7 +219,7 @@ static void pnv_gmbus_clock_gating(struct intel_display *display, bool enable) { /* When using bit bashing for I2C, this bit needs to be set to 1 */ - intel_de_rmw(display, DSPCLK_GATE_D(display), + intel_de_rmw(display, DSPCLK_GATE_D, PNV_GMBUSUNIT_CLOCK_GATE_DISABLE, !enable ? PNV_GMBUSUNIT_CLOCK_GATE_DISABLE : 0); } @@ -240,14 +242,20 @@ static void bxt_gmbus_clock_gating(struct intel_display *display, static u32 get_reserved(struct intel_gmbus *bus) { struct intel_display *display = bus->display; - u32 reserved = 0; + u32 preserve_bits = 0; + + if (display->platform.i830 || display->platform.i845g) + return 0; /* On most chips, these bits must be preserved in software. */ - if (!display->platform.i830 && !display->platform.i845g) - reserved = intel_de_read_notrace(display, bus->gpio_reg) & - (GPIO_DATA_PULLUP_DISABLE | GPIO_CLOCK_PULLUP_DISABLE); + preserve_bits |= GPIO_DATA_PULLUP_DISABLE | GPIO_CLOCK_PULLUP_DISABLE; + + /* Wa_16025573575: the masks bits need to be preserved through out */ + if (intel_display_wa(display, 16025573575)) + preserve_bits |= GPIO_CLOCK_DIR_MASK | GPIO_CLOCK_VAL_MASK | + GPIO_DATA_DIR_MASK | GPIO_DATA_VAL_MASK; - return reserved; + return intel_de_read_notrace(display, bus->gpio_reg) & preserve_bits; } static int get_clock(void *data) @@ -308,6 +316,22 @@ static void set_data(void *data, int state_high) intel_de_posting_read(display, bus->gpio_reg); } +static void +ptl_handle_mask_bits(struct intel_gmbus *bus, bool set) +{ + struct intel_display *display = bus->display; + u32 reg_val = intel_de_read_notrace(display, bus->gpio_reg); + u32 mask_bits = GPIO_CLOCK_DIR_MASK | GPIO_CLOCK_VAL_MASK | + GPIO_DATA_DIR_MASK | GPIO_DATA_VAL_MASK; + if (set) + reg_val |= mask_bits; + else + reg_val &= ~mask_bits; + + intel_de_write_notrace(display, bus->gpio_reg, reg_val); + intel_de_posting_read(display, bus->gpio_reg); +} + static int intel_gpio_pre_xfer(struct i2c_adapter *adapter) { @@ -319,6 +343,9 @@ intel_gpio_pre_xfer(struct i2c_adapter *adapter) if (display->platform.pineview) pnv_gmbus_clock_gating(display, false); + if (intel_display_wa(display, 16025573575)) + ptl_handle_mask_bits(bus, true); + set_data(bus, 1); set_clock(bus, 1); udelay(I2C_RISEFALL_TIME); @@ -336,6 +363,9 @@ intel_gpio_post_xfer(struct i2c_adapter *adapter) if (display->platform.pineview) pnv_gmbus_clock_gating(display, true); + + if (intel_display_wa(display, 16025573575)) + ptl_handle_mask_bits(bus, false); } static void @@ -385,11 +415,14 @@ static int gmbus_wait(struct intel_display *display, u32 status, u32 irq_en) intel_de_write_fw(display, GMBUS4(display), irq_en); status |= GMBUS_SATOER; - ret = wait_for_us((gmbus2 = intel_de_read_fw(display, GMBUS2(display))) & status, - 2); + + ret = poll_timeout_us_atomic(gmbus2 = intel_de_read_fw(display, GMBUS2(display)), + gmbus2 & status, + 0, 2, false); if (ret) - ret = wait_for((gmbus2 = intel_de_read_fw(display, GMBUS2(display))) & status, - 50); + ret = poll_timeout_us(gmbus2 = intel_de_read_fw(display, GMBUS2(display)), + gmbus2 & status, + 500, 50 * 1000, false); intel_de_write_fw(display, GMBUS4(display), 0); remove_wait_queue(&display->gmbus.wait_queue, &wait); diff --git a/drivers/gpu/drm/i915/display/intel_hdcp.c b/drivers/gpu/drm/i915/display/intel_hdcp.c index 42202c8bb066..531ee122bf82 100644 --- a/drivers/gpu/drm/i915/display/intel_hdcp.c +++ b/drivers/gpu/drm/i915/display/intel_hdcp.c @@ -11,6 +11,7 @@ #include <linux/component.h> #include <linux/debugfs.h> #include <linux/i2c.h> +#include <linux/iopoll.h> #include <linux/random.h> #include <drm/display/drm_hdcp_helper.h> @@ -326,16 +327,13 @@ static int intel_hdcp_poll_ksv_fifo(struct intel_digital_port *dig_port, bool ksv_ready; /* Poll for ksv list ready (spec says max time allowed is 5s) */ - ret = __wait_for(read_ret = shim->read_ksv_ready(dig_port, - &ksv_ready), - read_ret || ksv_ready, 5 * 1000 * 1000, 1000, - 100 * 1000); + ret = poll_timeout_us(read_ret = shim->read_ksv_ready(dig_port, &ksv_ready), + read_ret || ksv_ready, + 100 * 1000, 5 * 1000 * 1000, false); if (ret) return ret; if (read_ret) return read_ret; - if (!ksv_ready) - return -ETIMEDOUT; return 0; } @@ -817,6 +815,7 @@ static int intel_hdcp_auth(struct intel_connector *connector) enum port port = dig_port->base.port; unsigned long r0_prime_gen_start; int ret, i, tries = 2; + u32 val; union { u32 reg[2]; u8 shim[DRM_HDCP_AN_LEN]; @@ -905,8 +904,10 @@ static int intel_hdcp_auth(struct intel_connector *connector) HDCP_CONF_AUTH_AND_ENC); /* Wait for R0 ready */ - if (wait_for(intel_de_read(display, HDCP_STATUS(display, cpu_transcoder, port)) & - (HDCP_STATUS_R0_READY | HDCP_STATUS_ENC), 1)) { + ret = poll_timeout_us(val = intel_de_read(display, HDCP_STATUS(display, cpu_transcoder, port)), + val & (HDCP_STATUS_R0_READY | HDCP_STATUS_ENC), + 100, 1000, false); + if (ret) { drm_err(display->drm, "Timed out waiting for R0 ready\n"); return -ETIMEDOUT; } @@ -938,16 +939,16 @@ static int intel_hdcp_auth(struct intel_connector *connector) ri.reg); /* Wait for Ri prime match */ - if (!wait_for(intel_de_read(display, HDCP_STATUS(display, cpu_transcoder, port)) & - (HDCP_STATUS_RI_MATCH | HDCP_STATUS_ENC), 1)) + ret = poll_timeout_us(val = intel_de_read(display, HDCP_STATUS(display, cpu_transcoder, port)), + val & (HDCP_STATUS_RI_MATCH | HDCP_STATUS_ENC), + 100, 1000, false); + if (!ret) break; } if (i == tries) { drm_dbg_kms(display->drm, - "Timed out waiting for Ri prime match (%x)\n", - intel_de_read(display, - HDCP_STATUS(display, cpu_transcoder, port))); + "Timed out waiting for Ri prime match (%x)\n", val); return -ETIMEDOUT; } @@ -2446,12 +2447,6 @@ static int _intel_hdcp_enable(struct intel_atomic_state *state, if (!hdcp->shim) return -ENOENT; - if (!connector->encoder) { - drm_err(display->drm, "[CONNECTOR:%d:%s] encoder is not initialized\n", - connector->base.base.id, connector->base.name); - return -ENODEV; - } - mutex_lock(&hdcp->mutex); mutex_lock(&dig_port->hdcp.mutex); drm_WARN_ON(display->drm, diff --git a/drivers/gpu/drm/i915/display/intel_hdmi.c b/drivers/gpu/drm/i915/display/intel_hdmi.c index 9961ff259298..4ab7e2e3bfd4 100644 --- a/drivers/gpu/drm/i915/display/intel_hdmi.c +++ b/drivers/gpu/drm/i915/display/intel_hdmi.c @@ -29,6 +29,7 @@ #include <linux/delay.h> #include <linux/hdmi.h> #include <linux/i2c.h> +#include <linux/iopoll.h> #include <linux/slab.h> #include <linux/string_helpers.h> @@ -60,6 +61,7 @@ #include "intel_hdcp_regs.h" #include "intel_hdcp_shim.h" #include "intel_hdmi.h" +#include "intel_link_bw.h" #include "intel_lspcon.h" #include "intel_panel.h" #include "intel_pfit.h" @@ -1582,9 +1584,9 @@ bool intel_hdmi_hdcp_check_link_once(struct intel_digital_port *dig_port, intel_de_write(display, HDCP_RPRIME(display, cpu_transcoder, port), ri.reg); /* Wait for Ri prime match */ - if (wait_for((intel_de_read(display, HDCP_STATUS(display, cpu_transcoder, port)) & - (HDCP_STATUS_RI_MATCH | HDCP_STATUS_ENC)) == - (HDCP_STATUS_RI_MATCH | HDCP_STATUS_ENC), 1)) { + ret = intel_de_wait_for_set(display, HDCP_STATUS(display, cpu_transcoder, port), + HDCP_STATUS_RI_MATCH | HDCP_STATUS_ENC, 1); + if (ret) { drm_dbg_kms(display->drm, "Ri' mismatch detected (%x)\n", intel_de_read(display, HDCP_STATUS(display, cpu_transcoder, port))); @@ -1689,11 +1691,10 @@ intel_hdmi_hdcp2_wait_for_msg(struct intel_digital_port *dig_port, if (timeout < 0) return timeout; - ret = __wait_for(ret = hdcp2_detect_msg_availability(dig_port, - msg_id, &msg_ready, - &msg_sz), - !ret && msg_ready && msg_sz, timeout * 1000, - 1000, 5 * 1000); + ret = poll_timeout_us(ret = hdcp2_detect_msg_availability(dig_port, msg_id, + &msg_ready, &msg_sz), + !ret && msg_ready && msg_sz, + 4000, timeout * 1000, false); if (ret) drm_dbg_kms(display->drm, "msg_id: %d, ret: %d, timeout: %d\n", @@ -2053,6 +2054,10 @@ intel_hdmi_mode_valid(struct drm_connector *_connector, else sink_format = INTEL_OUTPUT_FORMAT_RGB; + status = intel_pfit_mode_valid(display, mode, sink_format, 0); + if (status != MODE_OK) + return status; + status = intel_hdmi_mode_clock_valid(&connector->base, clock, has_hdmi_sink, sink_format); if (status != MODE_OK) { if (ycbcr_420_only || @@ -2341,6 +2346,9 @@ int intel_hdmi_compute_config(struct intel_encoder *encoder, if (adjusted_mode->flags & DRM_MODE_FLAG_DBLCLK) pipe_config->pixel_multiplier = 2; + if (!intel_link_bw_compute_pipe_bpp(pipe_config)) + return -EINVAL; + pipe_config->has_audio = intel_hdmi_has_audio(encoder, pipe_config, conn_state) && intel_audio_compute_config(encoder, pipe_config, conn_state); diff --git a/drivers/gpu/drm/i915/display/intel_hotplug.c b/drivers/gpu/drm/i915/display/intel_hotplug.c index 265aa97fcc75..4451a792600a 100644 --- a/drivers/gpu/drm/i915/display/intel_hotplug.c +++ b/drivers/gpu/drm/i915/display/intel_hotplug.c @@ -28,6 +28,7 @@ #include "i915_drv.h" #include "i915_irq.h" +#include "i915_utils.h" #include "intel_connector.h" #include "intel_display_power.h" #include "intel_display_core.h" @@ -971,8 +972,6 @@ void intel_hpd_cancel_work(struct intel_display *display) spin_lock_irq(&display->irq.lock); - drm_WARN_ON(display->drm, get_blocked_hpd_pin_mask(display)); - display->hotplug.long_hpd_pin_mask = 0; display->hotplug.short_hpd_pin_mask = 0; display->hotplug.event_bits = 0; @@ -1333,12 +1332,12 @@ static const struct file_operations i915_hpd_short_storm_ctl_fops = { void intel_hpd_debugfs_register(struct intel_display *display) { - struct drm_minor *minor = display->drm->primary; + struct dentry *debugfs_root = display->drm->debugfs_root; - debugfs_create_file("i915_hpd_storm_ctl", 0644, minor->debugfs_root, + debugfs_create_file("i915_hpd_storm_ctl", 0644, debugfs_root, display, &i915_hpd_storm_ctl_fops); - debugfs_create_file("i915_hpd_short_storm_ctl", 0644, minor->debugfs_root, + debugfs_create_file("i915_hpd_short_storm_ctl", 0644, debugfs_root, display, &i915_hpd_short_storm_ctl_fops); - debugfs_create_bool("i915_ignore_long_hpd", 0644, minor->debugfs_root, + debugfs_create_bool("i915_ignore_long_hpd", 0644, debugfs_root, &display->hotplug.ignore_long_hpd); } diff --git a/drivers/gpu/drm/i915/display/intel_hotplug_irq.c b/drivers/gpu/drm/i915/display/intel_hotplug_irq.c index 43aee70597bf..4f72f3fb9af5 100644 --- a/drivers/gpu/drm/i915/display/intel_hotplug_irq.c +++ b/drivers/gpu/drm/i915/display/intel_hotplug_irq.c @@ -1025,7 +1025,7 @@ static void mtp_tc_hpd_enable_detection(struct intel_encoder *encoder) { struct intel_display *display = to_intel_display(encoder); - intel_de_rmw(display, SHOTPLUG_CTL_DDI, + intel_de_rmw(display, SHOTPLUG_CTL_TC, mtp_tc_hotplug_mask(encoder->hpd_pin), mtp_tc_hotplug_enables(encoder)); } diff --git a/drivers/gpu/drm/i915/display/intel_link_bw.c b/drivers/gpu/drm/i915/display/intel_link_bw.c index 3caef7f9c7c4..f52dee0ea412 100644 --- a/drivers/gpu/drm/i915/display/intel_link_bw.c +++ b/drivers/gpu/drm/i915/display/intel_link_bw.c @@ -165,6 +165,34 @@ int intel_link_bw_reduce_bpp(struct intel_atomic_state *state, } /** + * intel_link_bw_compute_pipe_bpp - compute pipe bpp limited by max link bpp + * @crtc_state: the crtc state + * + * Compute the pipe bpp limited by the CRTC's maximum link bpp. Encoders can + * call this function during state computation in the simple case where the + * link bpp will always match the pipe bpp. This is the case for all non-DP + * encoders, while DP encoders will use a link bpp lower than pipe bpp in case + * of DSC compression. + * + * Returns %true in case of success, %false if pipe bpp would need to be + * reduced below its valid range. + */ +bool intel_link_bw_compute_pipe_bpp(struct intel_crtc_state *crtc_state) +{ + int pipe_bpp = min(crtc_state->pipe_bpp, + fxp_q4_to_int(crtc_state->max_link_bpp_x16)); + + pipe_bpp = rounddown(pipe_bpp, 2 * 3); + + if (pipe_bpp < 6 * 3) + return false; + + crtc_state->pipe_bpp = pipe_bpp; + + return true; +} + +/** * intel_link_bw_set_bpp_limit_for_pipe - set link bpp limit for a pipe to its minimum * @state: atomic state * @old_limits: link BW limits @@ -449,6 +477,7 @@ void intel_link_bw_connector_debugfs_add(struct intel_connector *connector) switch (connector->base.connector_type) { case DRM_MODE_CONNECTOR_DisplayPort: case DRM_MODE_CONNECTOR_eDP: + case DRM_MODE_CONNECTOR_HDMIA: break; case DRM_MODE_CONNECTOR_VGA: case DRM_MODE_CONNECTOR_SVIDEO: @@ -458,11 +487,6 @@ void intel_link_bw_connector_debugfs_add(struct intel_connector *connector) break; return; - case DRM_MODE_CONNECTOR_HDMIA: - if (HAS_FDI(display) && !HAS_DDI(display)) - break; - - return; default: return; } diff --git a/drivers/gpu/drm/i915/display/intel_link_bw.h b/drivers/gpu/drm/i915/display/intel_link_bw.h index b499042e62b1..95ab7c50c61d 100644 --- a/drivers/gpu/drm/i915/display/intel_link_bw.h +++ b/drivers/gpu/drm/i915/display/intel_link_bw.h @@ -27,6 +27,7 @@ int intel_link_bw_reduce_bpp(struct intel_atomic_state *state, struct intel_link_bw_limits *limits, u8 pipe_mask, const char *reason); +bool intel_link_bw_compute_pipe_bpp(struct intel_crtc_state *crtc_state); bool intel_link_bw_set_bpp_limit_for_pipe(struct intel_atomic_state *state, const struct intel_link_bw_limits *old_limits, struct intel_link_bw_limits *new_limits, diff --git a/drivers/gpu/drm/i915/display/intel_lpe_audio.c b/drivers/gpu/drm/i915/display/intel_lpe_audio.c index 666148a14522..42284e9928f2 100644 --- a/drivers/gpu/drm/i915/display/intel_lpe_audio.c +++ b/drivers/gpu/drm/i915/display/intel_lpe_audio.c @@ -68,9 +68,9 @@ #include <linux/platform_device.h> #include <linux/pm_runtime.h> +#include <drm/drm_print.h> #include <drm/intel/intel_lpe_audio.h> -#include "i915_drv.h" #include "i915_irq.h" #include "intel_audio_regs.h" #include "intel_de.h" @@ -170,14 +170,11 @@ static struct irq_chip lpe_audio_irqchip = { static int lpe_audio_irq_init(struct intel_display *display) { - struct drm_i915_private *dev_priv = to_i915(display->drm); int irq = display->audio.lpe.irq; - drm_WARN_ON(display->drm, !intel_irqs_enabled(dev_priv)); - irq_set_chip_and_handler_name(irq, - &lpe_audio_irqchip, - handle_simple_irq, - "hdmi_lpe_audio_irq_handler"); + irq_set_chip_and_handler_name(irq, &lpe_audio_irqchip, + handle_simple_irq, + "hdmi_lpe_audio_irq_handler"); return 0; } diff --git a/drivers/gpu/drm/i915/display/intel_lspcon.c b/drivers/gpu/drm/i915/display/intel_lspcon.c index abc4b562083d..d56026c4efdd 100644 --- a/drivers/gpu/drm/i915/display/intel_lspcon.c +++ b/drivers/gpu/drm/i915/display/intel_lspcon.c @@ -23,6 +23,8 @@ * */ +#include <linux/iopoll.h> + #include <drm/display/drm_dp_dual_mode_helper.h> #include <drm/display/drm_hdmi_helper.h> #include <drm/drm_atomic_helper.h> @@ -181,6 +183,8 @@ static enum drm_lspcon_mode lspcon_wait_mode(struct intel_lspcon *lspcon, struct intel_dp *intel_dp = lspcon_to_intel_dp(lspcon); struct intel_display *display = to_intel_display(intel_dp); enum drm_lspcon_mode current_mode; + int timeout_us; + int ret; current_mode = lspcon_get_current_mode(lspcon); if (current_mode == mode) @@ -189,9 +193,12 @@ static enum drm_lspcon_mode lspcon_wait_mode(struct intel_lspcon *lspcon, drm_dbg_kms(display->drm, "Waiting for LSPCON mode %s to settle\n", lspcon_mode_name(mode)); - wait_for((current_mode = lspcon_get_current_mode(lspcon)) == mode, - lspcon_get_mode_settle_timeout(lspcon)); - if (current_mode != mode) + timeout_us = lspcon_get_mode_settle_timeout(lspcon) * 1000; + + ret = poll_timeout_us(current_mode = lspcon_get_current_mode(lspcon), + current_mode == mode, + 5000, timeout_us, false); + if (ret) drm_err(display->drm, "LSPCON mode hasn't settled\n"); out: diff --git a/drivers/gpu/drm/i915/display/intel_lvds.c b/drivers/gpu/drm/i915/display/intel_lvds.c index 7e48a235c99f..48f4d8ed4f15 100644 --- a/drivers/gpu/drm/i915/display/intel_lvds.c +++ b/drivers/gpu/drm/i915/display/intel_lvds.c @@ -48,6 +48,7 @@ #include "intel_dpll.h" #include "intel_fdi.h" #include "intel_gmbus.h" +#include "intel_link_bw.h" #include "intel_lvds.h" #include "intel_lvds_regs.h" #include "intel_panel.h" @@ -433,7 +434,7 @@ static int intel_lvds_compute_config(struct intel_encoder *encoder, if (HAS_PCH_SPLIT(display)) { crtc_state->has_pch_encoder = true; - if (!intel_fdi_compute_pipe_bpp(crtc_state)) + if (!intel_link_bw_compute_pipe_bpp(crtc_state)) return -EINVAL; } diff --git a/drivers/gpu/drm/i915/display/intel_opregion.c b/drivers/gpu/drm/i915/display/intel_opregion.c index 81efdb17fc0c..cbc220310813 100644 --- a/drivers/gpu/drm/i915/display/intel_opregion.c +++ b/drivers/gpu/drm/i915/display/intel_opregion.c @@ -28,13 +28,13 @@ #include <linux/acpi.h> #include <linux/debugfs.h> #include <linux/dmi.h> +#include <linux/iopoll.h> #include <acpi/video.h> #include <drm/drm_edid.h> #include <drm/drm_file.h> #include <drm/drm_print.h> -#include "i915_utils.h" #include "intel_acpi.h" #include "intel_backlight.h" #include "intel_display_core.h" @@ -357,10 +357,12 @@ static int swsci(struct intel_display *display, pci_write_config_word(pdev, SWSCI, swsci_val); /* Poll for the result. */ -#define C (((scic = swsci->scic) & SWSCI_SCIC_INDICATOR) == 0) - if (wait_for(C, dslp)) { + ret = poll_timeout_us(scic = swsci->scic, + (scic & SWSCI_SCIC_INDICATOR) == 0, + 1000, dslp * 1000, false); + if (ret) { drm_dbg(display->drm, "SWSCI request timed out\n"); - return -ETIMEDOUT; + return ret; } scic = (scic & SWSCI_SCIC_EXIT_STATUS_MASK) >> @@ -1299,8 +1301,6 @@ DEFINE_SHOW_ATTRIBUTE(intel_opregion); void intel_opregion_debugfs_register(struct intel_display *display) { - struct drm_minor *minor = display->drm->primary; - - debugfs_create_file("i915_opregion", 0444, minor->debugfs_root, + debugfs_create_file("i915_opregion", 0444, display->drm->debugfs_root, display, &intel_opregion_fops); } diff --git a/drivers/gpu/drm/i915/display/intel_overlay.c b/drivers/gpu/drm/i915/display/intel_overlay.c index 159a5f998ea0..272f9e7af4d4 100644 --- a/drivers/gpu/drm/i915/display/intel_overlay.c +++ b/drivers/gpu/drm/i915/display/intel_overlay.c @@ -217,10 +217,9 @@ static void i830_overlay_clock_gating(struct intel_display *display, /* WA_OVERLAY_CLKGATE:alm */ if (enable) - intel_de_write(display, DSPCLK_GATE_D(display), 0); + intel_de_write(display, DSPCLK_GATE_D, 0); else - intel_de_write(display, DSPCLK_GATE_D(display), - OVRUNIT_CLOCK_GATE_DISABLE); + intel_de_write(display, DSPCLK_GATE_D, OVRUNIT_CLOCK_GATE_DISABLE); /* WA_DISABLE_L2CACHE_CLOCK_GATING:alm */ pci_bus_read_config_byte(pdev->bus, diff --git a/drivers/gpu/drm/i915/display/intel_pch.h b/drivers/gpu/drm/i915/display/intel_pch.h index cf4dab1b98bf..19cac7412d0a 100644 --- a/drivers/gpu/drm/i915/display/intel_pch.h +++ b/drivers/gpu/drm/i915/display/intel_pch.h @@ -6,8 +6,6 @@ #ifndef __INTEL_PCH__ #define __INTEL_PCH__ -#include "intel_display_conversion.h" - struct intel_display; /* @@ -36,7 +34,7 @@ enum intel_pch { PCH_LNL, }; -#define INTEL_PCH_TYPE(_display) (__to_intel_display(_display)->pch_type) +#define INTEL_PCH_TYPE(_display) ((_display)->pch_type) #define HAS_PCH_DG2(display) (INTEL_PCH_TYPE(display) == PCH_DG2) #define HAS_PCH_ADP(display) (INTEL_PCH_TYPE(display) == PCH_ADP) #define HAS_PCH_DG1(display) (INTEL_PCH_TYPE(display) == PCH_DG1) diff --git a/drivers/gpu/drm/i915/display/intel_pch_refclk.c b/drivers/gpu/drm/i915/display/intel_pch_refclk.c index d3c5255bf1a8..9ae53679a041 100644 --- a/drivers/gpu/drm/i915/display/intel_pch_refclk.c +++ b/drivers/gpu/drm/i915/display/intel_pch_refclk.c @@ -17,16 +17,22 @@ static void lpt_fdi_reset_mphy(struct intel_display *display) { + int ret; + intel_de_rmw(display, SOUTH_CHICKEN2, 0, FDI_MPHY_IOSFSB_RESET_CTL); - if (wait_for_us(intel_de_read(display, SOUTH_CHICKEN2) & - FDI_MPHY_IOSFSB_RESET_STATUS, 100)) + ret = intel_de_wait_custom(display, SOUTH_CHICKEN2, + FDI_MPHY_IOSFSB_RESET_STATUS, FDI_MPHY_IOSFSB_RESET_STATUS, + 100, 0, NULL); + if (ret) drm_err(display->drm, "FDI mPHY reset assert timeout\n"); intel_de_rmw(display, SOUTH_CHICKEN2, FDI_MPHY_IOSFSB_RESET_CTL, 0); - if (wait_for_us((intel_de_read(display, SOUTH_CHICKEN2) & - FDI_MPHY_IOSFSB_RESET_STATUS) == 0, 100)) + ret = intel_de_wait_custom(display, SOUTH_CHICKEN2, + FDI_MPHY_IOSFSB_RESET_STATUS, 0, + 100, 0, NULL); + if (ret) drm_err(display->drm, "FDI mPHY reset de-assert timeout\n"); } diff --git a/drivers/gpu/drm/i915/display/intel_pfit.c b/drivers/gpu/drm/i915/display/intel_pfit.c index 13541be4d6df..68539e7c2a24 100644 --- a/drivers/gpu/drm/i915/display/intel_pfit.c +++ b/drivers/gpu/drm/i915/display/intel_pfit.c @@ -14,6 +14,7 @@ #include "intel_lvds_regs.h" #include "intel_pfit.h" #include "intel_pfit_regs.h" +#include "skl_scaler.h" static int intel_pch_pfit_check_dst_window(const struct intel_crtc_state *crtc_state) { @@ -546,6 +547,16 @@ out: return intel_gmch_pfit_check_timings(crtc_state); } +enum drm_mode_status +intel_pfit_mode_valid(struct intel_display *display, + const struct drm_display_mode *mode, + enum intel_output_format output_format, + int num_joined_pipes) +{ + return skl_scaler_mode_valid(display, mode, output_format, + num_joined_pipes); +} + int intel_pfit_compute_config(struct intel_crtc_state *crtc_state, const struct drm_connector_state *conn_state) { diff --git a/drivers/gpu/drm/i915/display/intel_pfit.h b/drivers/gpu/drm/i915/display/intel_pfit.h index ef34f9b49d09..c1bb0d1f344e 100644 --- a/drivers/gpu/drm/i915/display/intel_pfit.h +++ b/drivers/gpu/drm/i915/display/intel_pfit.h @@ -6,8 +6,12 @@ #ifndef __INTEL_PFIT_H__ #define __INTEL_PFIT_H__ +enum drm_mode_status; +struct drm_display_mode; struct drm_connector_state; struct intel_crtc_state; +struct intel_display; +enum intel_output_format; int intel_pfit_compute_config(struct intel_crtc_state *crtc_state, const struct drm_connector_state *conn_state); @@ -17,5 +21,9 @@ void ilk_pfit_get_config(struct intel_crtc_state *crtc_state); void i9xx_pfit_enable(const struct intel_crtc_state *crtc_state); void i9xx_pfit_disable(const struct intel_crtc_state *old_crtc_state); void i9xx_pfit_get_config(struct intel_crtc_state *crtc_state); - +enum drm_mode_status +intel_pfit_mode_valid(struct intel_display *display, + const struct drm_display_mode *mode, + enum intel_output_format output_format, + int num_joined_pipes); #endif /* __INTEL_PFIT_H__ */ diff --git a/drivers/gpu/drm/i915/display/intel_plane.c b/drivers/gpu/drm/i915/display/intel_plane.c index 36fb07471deb..81f05ee9a21a 100644 --- a/drivers/gpu/drm/i915/display/intel_plane.c +++ b/drivers/gpu/drm/i915/display/intel_plane.c @@ -46,7 +46,6 @@ #include "gem/i915_gem_object.h" #include "i915_scheduler_types.h" -#include "i915_vma.h" #include "i9xx_plane_regs.h" #include "intel_bo.h" #include "intel_cdclk.h" @@ -1749,8 +1748,3 @@ int intel_plane_atomic_check(struct intel_atomic_state *state) return 0; } - -u32 intel_plane_ggtt_offset(const struct intel_plane_state *plane_state) -{ - return i915_ggtt_offset(plane_state->ggtt_vma); -} diff --git a/drivers/gpu/drm/i915/display/intel_plane.h b/drivers/gpu/drm/i915/display/intel_plane.h index 4ef012c08fa4..8af41ccc0a69 100644 --- a/drivers/gpu/drm/i915/display/intel_plane.h +++ b/drivers/gpu/drm/i915/display/intel_plane.h @@ -87,7 +87,6 @@ int intel_plane_add_affected(struct intel_atomic_state *state, struct intel_crtc *crtc); int intel_plane_atomic_check(struct intel_atomic_state *state); -u32 intel_plane_ggtt_offset(const struct intel_plane_state *plane_state); bool intel_plane_format_mod_supported_async(struct drm_plane *plane, u32 format, u64 modifier); diff --git a/drivers/gpu/drm/i915/display/intel_plane_initial.c b/drivers/gpu/drm/i915/display/intel_plane_initial.c index 4246173ed311..a9f36b1b50c1 100644 --- a/drivers/gpu/drm/i915/display/intel_plane_initial.c +++ b/drivers/gpu/drm/i915/display/intel_plane_initial.c @@ -360,6 +360,8 @@ valid_fb: i915_vma_pin_fence(vma) == 0 && vma->fence) plane_state->flags |= PLANE_HAS_FENCE; + plane_state->surf = i915_ggtt_offset(plane_state->ggtt_vma); + plane_state->uapi.src_x = 0; plane_state->uapi.src_y = 0; plane_state->uapi.src_w = fb->width << 16; diff --git a/drivers/gpu/drm/i915/display/intel_pps.c b/drivers/gpu/drm/i915/display/intel_pps.c index b64d0b30f5b1..327e0de86f1e 100644 --- a/drivers/gpu/drm/i915/display/intel_pps.c +++ b/drivers/gpu/drm/i915/display/intel_pps.c @@ -4,6 +4,7 @@ */ #include <linux/debugfs.h> +#include <linux/iopoll.h> #include <drm/drm_print.h> @@ -608,6 +609,8 @@ static void wait_panel_status(struct intel_dp *intel_dp, struct intel_display *display = to_intel_display(intel_dp); struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp); i915_reg_t pp_stat_reg, pp_ctrl_reg; + int ret; + u32 val; lockdep_assert_held(&display->pps.mutex); @@ -624,13 +627,18 @@ static void wait_panel_status(struct intel_dp *intel_dp, intel_de_read(display, pp_stat_reg), intel_de_read(display, pp_ctrl_reg)); - if (intel_de_wait(display, pp_stat_reg, mask, value, 5000)) + ret = poll_timeout_us(val = intel_de_read(display, pp_stat_reg), + (val & mask) == value, + 10 * 1000, 5000 * 1000, true); + if (ret) { drm_err(display->drm, "[ENCODER:%d:%s] %s panel status timeout: PP_STATUS: 0x%08x PP_CONTROL: 0x%08x\n", dig_port->base.base.base.id, dig_port->base.base.name, pps_name(intel_dp), intel_de_read(display, pp_stat_reg), intel_de_read(display, pp_ctrl_reg)); + return; + } drm_dbg_kms(display->drm, "Wait complete\n"); } diff --git a/drivers/gpu/drm/i915/display/intel_psr.c b/drivers/gpu/drm/i915/display/intel_psr.c index 41988e193a41..22433fe2ee14 100644 --- a/drivers/gpu/drm/i915/display/intel_psr.c +++ b/drivers/gpu/drm/i915/display/intel_psr.c @@ -233,16 +233,12 @@ bool intel_psr_needs_aux_io_power(struct intel_encoder *encoder, static bool psr_global_enabled(struct intel_dp *intel_dp) { - struct intel_display *display = to_intel_display(intel_dp); struct intel_connector *connector = intel_dp->attached_connector; switch (intel_dp->psr.debug & I915_PSR_DEBUG_MODE_MASK) { case I915_PSR_DEBUG_DEFAULT: - if (display->params.enable_psr == -1) - return intel_dp_is_edp(intel_dp) ? - connector->panel.vbt.psr.enable : - true; - return display->params.enable_psr; + return intel_dp_is_edp(intel_dp) ? + connector->panel.vbt.psr.enable : true; case I915_PSR_DEBUG_DISABLE: return false; default: @@ -250,39 +246,23 @@ static bool psr_global_enabled(struct intel_dp *intel_dp) } } -static bool psr2_global_enabled(struct intel_dp *intel_dp) +static bool sel_update_global_enabled(struct intel_dp *intel_dp) { - struct intel_display *display = to_intel_display(intel_dp); - switch (intel_dp->psr.debug & I915_PSR_DEBUG_MODE_MASK) { case I915_PSR_DEBUG_DISABLE: case I915_PSR_DEBUG_FORCE_PSR1: return false; default: - if (display->params.enable_psr == 1) - return false; return true; } } -static bool psr2_su_region_et_global_enabled(struct intel_dp *intel_dp) -{ - struct intel_display *display = to_intel_display(intel_dp); - - if (display->params.enable_psr != -1) - return false; - - return true; -} - static bool panel_replay_global_enabled(struct intel_dp *intel_dp) { struct intel_display *display = to_intel_display(intel_dp); - if ((display->params.enable_psr != -1) || - (intel_dp->psr.debug & I915_PSR_DEBUG_PANEL_REPLAY_DISABLE)) - return false; - return true; + return !(intel_dp->psr.debug & I915_PSR_DEBUG_PANEL_REPLAY_DISABLE) && + display->params.enable_panel_replay; } static u32 psr_irq_psr_error_bit_get(struct intel_dp *intel_dp) @@ -600,6 +580,16 @@ exit: static void _panel_replay_init_dpcd(struct intel_dp *intel_dp) { struct intel_display *display = to_intel_display(intel_dp); + int ret; + + ret = drm_dp_dpcd_read_data(&intel_dp->aux, DP_PANEL_REPLAY_CAP_SUPPORT, + &intel_dp->pr_dpcd, sizeof(intel_dp->pr_dpcd)); + if (ret < 0) + return; + + if (!(intel_dp->pr_dpcd[INTEL_PR_DPCD_INDEX(DP_PANEL_REPLAY_CAP_SUPPORT)] & + DP_PANEL_REPLAY_SUPPORT)) + return; if (intel_dp_is_edp(intel_dp)) { if (!intel_alpm_aux_less_wake_supported(intel_dp)) { @@ -631,6 +621,15 @@ static void _panel_replay_init_dpcd(struct intel_dp *intel_dp) static void _psr_init_dpcd(struct intel_dp *intel_dp) { struct intel_display *display = to_intel_display(intel_dp); + int ret; + + ret = drm_dp_dpcd_read_data(&intel_dp->aux, DP_PSR_SUPPORT, intel_dp->psr_dpcd, + sizeof(intel_dp->psr_dpcd)); + if (ret < 0) + return; + + if (!intel_dp->psr_dpcd[0]) + return; drm_dbg_kms(display->drm, "eDP panel supports PSR version %x\n", intel_dp->psr_dpcd[0]); @@ -676,18 +675,9 @@ static void _psr_init_dpcd(struct intel_dp *intel_dp) void intel_psr_init_dpcd(struct intel_dp *intel_dp) { - drm_dp_dpcd_read(&intel_dp->aux, DP_PSR_SUPPORT, intel_dp->psr_dpcd, - sizeof(intel_dp->psr_dpcd)); - - drm_dp_dpcd_read(&intel_dp->aux, DP_PANEL_REPLAY_CAP_SUPPORT, - &intel_dp->pr_dpcd, sizeof(intel_dp->pr_dpcd)); - - if (intel_dp->pr_dpcd[INTEL_PR_DPCD_INDEX(DP_PANEL_REPLAY_CAP_SUPPORT)] & - DP_PANEL_REPLAY_SUPPORT) - _panel_replay_init_dpcd(intel_dp); + _psr_init_dpcd(intel_dp); - if (intel_dp->psr_dpcd[0]) - _psr_init_dpcd(intel_dp); + _panel_replay_init_dpcd(intel_dp); if (intel_dp->psr.sink_psr2_support || intel_dp->psr.sink_panel_replay_su_support) @@ -742,8 +732,7 @@ static bool psr2_su_region_et_valid(struct intel_dp *intel_dp, bool panel_replay return panel_replay ? intel_dp->pr_dpcd[INTEL_PR_DPCD_INDEX(DP_PANEL_REPLAY_CAP_SUPPORT)] & DP_PANEL_REPLAY_EARLY_TRANSPORT_SUPPORT : - intel_dp->psr_dpcd[0] == DP_PSR2_WITH_Y_COORD_ET_SUPPORTED && - psr2_su_region_et_global_enabled(intel_dp); + intel_dp->psr_dpcd[0] == DP_PSR2_WITH_Y_COORD_ET_SUPPORTED; } static void _panel_replay_enable_sink(struct intel_dp *intel_dp, @@ -936,7 +925,7 @@ static void hsw_activate_psr1(struct intel_dp *intel_dp) /* Wa_16025596647 */ if ((DISPLAY_VER(display) == 20 || IS_DISPLAY_VERx100_STEP(display, 3000, STEP_A0, STEP_B0)) && - is_dc5_dc6_blocked(intel_dp)) + is_dc5_dc6_blocked(intel_dp) && intel_dp->psr.pkg_c_latency_used) intel_dmc_start_pkgc_exit_at_start_of_undelayed_vblank(display, intel_dp->psr.pipe, true); @@ -1026,7 +1015,7 @@ static void hsw_activate_psr2(struct intel_dp *intel_dp) /* Wa_16025596647 */ if ((DISPLAY_VER(display) == 20 || IS_DISPLAY_VERx100_STEP(display, 3000, STEP_A0, STEP_B0)) && - is_dc5_dc6_blocked(intel_dp)) + is_dc5_dc6_blocked(intel_dp) && intel_dp->psr.pkg_c_latency_used) idle_frames = 0; else idle_frames = psr_compute_idle_frames(intel_dp); @@ -1423,7 +1412,7 @@ static bool intel_psr2_config_valid(struct intel_dp *intel_dp, int crtc_vdisplay = crtc_state->hw.adjusted_mode.crtc_vdisplay; int psr_max_h = 0, psr_max_v = 0, max_bpp = 0; - if (!intel_dp->psr.sink_psr2_support) + if (!intel_dp->psr.sink_psr2_support || display->params.enable_psr == 1) return false; /* JSL and EHL only supports eDP 1.3 */ @@ -1528,7 +1517,7 @@ static bool intel_sel_update_config_valid(struct intel_dp *intel_dp, goto unsupported; } - if (!psr2_global_enabled(intel_dp)) { + if (!sel_update_global_enabled(intel_dp)) { drm_dbg_kms(display->drm, "Selective update disabled by flag\n"); goto unsupported; @@ -1576,7 +1565,7 @@ static bool _psr_compute_config(struct intel_dp *intel_dp, const struct drm_display_mode *adjusted_mode = &crtc_state->hw.adjusted_mode; int entry_setup_frames; - if (!CAN_PSR(intel_dp)) + if (!CAN_PSR(intel_dp) || !display->params.enable_psr) return false; /* @@ -1808,6 +1797,8 @@ static void intel_psr_activate(struct intel_dp *intel_dp) drm_WARN_ON(display->drm, intel_dp->psr.active); + drm_WARN_ON(display->drm, !intel_dp->psr.enabled); + lockdep_assert_held(&intel_dp->psr.lock); /* psr1, psr2 and panel-replay are mutually exclusive.*/ @@ -2027,6 +2018,7 @@ static void intel_psr_enable_locked(struct intel_dp *intel_dp, intel_dp->psr.req_psr2_sdp_prior_scanline = crtc_state->req_psr2_sdp_prior_scanline; intel_dp->psr.active_non_psr_pipes = crtc_state->active_non_psr_pipes; + intel_dp->psr.pkg_c_latency_used = crtc_state->pkg_c_latency_used; if (!psr_interrupt_error_check(intel_dp)) return; @@ -2103,8 +2095,9 @@ static void intel_psr_exit(struct intel_dp *intel_dp) drm_WARN_ON(display->drm, !(val & EDP_PSR2_ENABLE)); } else { - if (DISPLAY_VER(display) == 20 || - IS_DISPLAY_VERx100_STEP(display, 3000, STEP_A0, STEP_B0)) + if ((DISPLAY_VER(display) == 20 || + IS_DISPLAY_VERx100_STEP(display, 3000, STEP_A0, STEP_B0)) && + intel_dp->psr.pkg_c_latency_used) intel_dmc_start_pkgc_exit_at_start_of_undelayed_vblank(display, intel_dp->psr.pipe, false); @@ -2207,6 +2200,7 @@ static void intel_psr_disable_locked(struct intel_dp *intel_dp) intel_dp->psr.su_region_et_enabled = false; intel_dp->psr.psr2_sel_fetch_cff_enabled = false; intel_dp->psr.active_non_psr_pipes = 0; + intel_dp->psr.pkg_c_latency_used = 0; } /** @@ -3099,7 +3093,7 @@ static bool __psr_wait_for_idle_locked(struct intel_dp *intel_dp) /* After the unlocked wait, verify that PSR is still wanted! */ mutex_lock(&intel_dp->psr.lock); - return err == 0 && intel_dp->psr.enabled; + return err == 0 && intel_dp->psr.enabled && !intel_dp->psr.pause_counter; } static int intel_psr_fastset_force(struct intel_display *display) @@ -3228,8 +3222,13 @@ static void intel_psr_work(struct work_struct *work) if (!intel_dp->psr.enabled) goto unlock; - if (READ_ONCE(intel_dp->psr.irq_aux_error)) + if (READ_ONCE(intel_dp->psr.irq_aux_error)) { intel_psr_handle_irq(intel_dp); + goto unlock; + } + + if (intel_dp->psr.pause_counter) + goto unlock; /* * We have to make sure PSR is ready for re-enable @@ -3723,7 +3722,7 @@ static void intel_psr_apply_underrun_on_idle_wa_locked(struct intel_dp *intel_dp struct intel_display *display = to_intel_display(intel_dp); bool dc5_dc6_blocked; - if (!intel_dp->psr.active) + if (!intel_dp->psr.active || !intel_dp->psr.pkg_c_latency_used) return; dc5_dc6_blocked = is_dc5_dc6_blocked(intel_dp); @@ -3748,7 +3747,8 @@ static void psr_dc5_dc6_wa_work(struct work_struct *work) mutex_lock(&intel_dp->psr.lock); - if (intel_dp->psr.enabled && !intel_dp->psr.panel_replay_enabled) + if (intel_dp->psr.enabled && !intel_dp->psr.panel_replay_enabled && + !intel_dp->psr.pkg_c_latency_used) intel_psr_apply_underrun_on_idle_wa_locked(intel_dp); mutex_unlock(&intel_dp->psr.lock); @@ -3826,7 +3826,8 @@ void intel_psr_notify_pipe_change(struct intel_atomic_state *state, goto unlock; if ((enable && intel_dp->psr.active_non_psr_pipes) || - (!enable && !intel_dp->psr.active_non_psr_pipes)) { + (!enable && !intel_dp->psr.active_non_psr_pipes) || + !intel_dp->psr.pkg_c_latency_used) { intel_dp->psr.active_non_psr_pipes = active_non_psr_pipes; goto unlock; } @@ -3861,7 +3862,7 @@ void intel_psr_notify_vblank_enable_disable(struct intel_display *display, break; } - if (intel_dp->psr.enabled) + if (intel_dp->psr.enabled && intel_dp->psr.pkg_c_latency_used) intel_psr_apply_underrun_on_idle_wa_locked(intel_dp); mutex_unlock(&intel_dp->psr.lock); @@ -4157,12 +4158,12 @@ DEFINE_SIMPLE_ATTRIBUTE(i915_edp_psr_debug_fops, void intel_psr_debugfs_register(struct intel_display *display) { - struct drm_minor *minor = display->drm->primary; + struct dentry *debugfs_root = display->drm->debugfs_root; - debugfs_create_file("i915_edp_psr_debug", 0644, minor->debugfs_root, + debugfs_create_file("i915_edp_psr_debug", 0644, debugfs_root, display, &i915_edp_psr_debug_fops); - debugfs_create_file("i915_edp_psr_status", 0444, minor->debugfs_root, + debugfs_create_file("i915_edp_psr_status", 0444, debugfs_root, display, &i915_edp_psr_status_fops); } diff --git a/drivers/gpu/drm/i915/display/intel_quirks.c b/drivers/gpu/drm/i915/display/intel_quirks.c index a32fae510ed2..d2e16b79d6be 100644 --- a/drivers/gpu/drm/i915/display/intel_quirks.c +++ b/drivers/gpu/drm/i915/display/intel_quirks.c @@ -80,6 +80,12 @@ static void quirk_fw_sync_len(struct intel_dp *intel_dp) drm_info(display->drm, "Applying Fast Wake sync pulse count quirk\n"); } +static void quirk_edp_limit_rate_hbr2(struct intel_display *display) +{ + intel_set_quirk(display, QUIRK_EDP_LIMIT_RATE_HBR2); + drm_info(display->drm, "Applying eDP Limit rate to HBR2 quirk\n"); +} + struct intel_quirk { int device; int subsystem_vendor; @@ -231,6 +237,9 @@ static struct intel_quirk intel_quirks[] = { { 0x3184, 0x1019, 0xa94d, quirk_increase_ddi_disabled_time }, /* HP Notebook - 14-r206nv */ { 0x0f31, 0x103c, 0x220f, quirk_invert_brightness }, + + /* Dell XPS 13 7390 2-in-1 */ + { 0x8a12, 0x1028, 0x08b0, quirk_edp_limit_rate_hbr2 }, }; static const struct intel_dpcd_quirk intel_dpcd_quirks[] = { diff --git a/drivers/gpu/drm/i915/display/intel_quirks.h b/drivers/gpu/drm/i915/display/intel_quirks.h index cafdebda7535..06da0e286c67 100644 --- a/drivers/gpu/drm/i915/display/intel_quirks.h +++ b/drivers/gpu/drm/i915/display/intel_quirks.h @@ -20,6 +20,7 @@ enum intel_quirk_id { QUIRK_LVDS_SSC_DISABLE, QUIRK_NO_PPS_BACKLIGHT_POWER_HOOK, QUIRK_FW_SYNC_LEN, + QUIRK_EDP_LIMIT_RATE_HBR2, }; void intel_init_quirks(struct intel_display *display); diff --git a/drivers/gpu/drm/i915/display/intel_sdvo.c b/drivers/gpu/drm/i915/display/intel_sdvo.c index 87aff2754f69..6c032d81e7ee 100644 --- a/drivers/gpu/drm/i915/display/intel_sdvo.c +++ b/drivers/gpu/drm/i915/display/intel_sdvo.c @@ -47,11 +47,11 @@ #include "intel_display_driver.h" #include "intel_display_regs.h" #include "intel_display_types.h" -#include "intel_fdi.h" #include "intel_fifo_underrun.h" #include "intel_gmbus.h" #include "intel_hdmi.h" #include "intel_hotplug.h" +#include "intel_link_bw.h" #include "intel_panel.h" #include "intel_sdvo.h" #include "intel_sdvo_regs.h" @@ -1367,7 +1367,7 @@ static int intel_sdvo_compute_config(struct intel_encoder *encoder, if (HAS_PCH_SPLIT(display)) { pipe_config->has_pch_encoder = true; - if (!intel_fdi_compute_pipe_bpp(pipe_config)) + if (!intel_link_bw_compute_pipe_bpp(pipe_config)) return -EINVAL; } @@ -2052,8 +2052,10 @@ static void intel_sdvo_enable_hotplug(struct intel_encoder *encoder) { struct intel_sdvo *intel_sdvo = to_sdvo(encoder); - intel_sdvo_write_cmd(intel_sdvo, SDVO_CMD_SET_ACTIVE_HOT_PLUG, - &intel_sdvo->hotplug_active, 2); + if (!intel_sdvo_write_cmd(intel_sdvo, SDVO_CMD_SET_ACTIVE_HOT_PLUG, + &intel_sdvo->hotplug_active, 2)) + drm_warn(intel_sdvo->base.base.dev, + "Failed to enable hotplug on SDVO encoder\n"); } static enum intel_hotplug_state diff --git a/drivers/gpu/drm/i915/display/intel_sprite.c b/drivers/gpu/drm/i915/display/intel_sprite.c index e6844df837af..75bbaa923204 100644 --- a/drivers/gpu/drm/i915/display/intel_sprite.c +++ b/drivers/gpu/drm/i915/display/intel_sprite.c @@ -264,8 +264,7 @@ static u32 vlv_sprite_ctl_crtc(const struct intel_crtc_state *crtc_state) return sprctl; } -static u32 vlv_sprite_ctl(const struct intel_crtc_state *crtc_state, - const struct intel_plane_state *plane_state) +static u32 vlv_sprite_ctl(const struct intel_plane_state *plane_state) { const struct drm_framebuffer *fb = plane_state->hw.fb; unsigned int rotation = plane_state->hw.rotation; @@ -395,15 +394,12 @@ vlv_sprite_update_arm(struct intel_dsb *dsb, enum pipe pipe = plane->pipe; enum plane_id plane_id = plane->id; const struct drm_intel_sprite_colorkey *key = &plane_state->ckey; - u32 sprsurf_offset = plane_state->view.color_plane[0].offset; u32 x = plane_state->view.color_plane[0].x; u32 y = plane_state->view.color_plane[0].y; - u32 sprctl, linear_offset; + u32 sprctl; sprctl = plane_state->ctl | vlv_sprite_ctl_crtc(crtc_state); - linear_offset = intel_fb_xy_to_linear(x, y, plane_state, 0); - if (display->platform.cherryview && pipe == PIPE_B) chv_sprite_update_csc(plane_state); @@ -418,7 +414,8 @@ vlv_sprite_update_arm(struct intel_dsb *dsb, intel_de_write_fw(display, SPCONSTALPHA(pipe, plane_id), 0); - intel_de_write_fw(display, SPLINOFF(pipe, plane_id), linear_offset); + intel_de_write_fw(display, SPLINOFF(pipe, plane_id), + intel_fb_xy_to_linear(x, y, plane_state, 0)); intel_de_write_fw(display, SPTILEOFF(pipe, plane_id), SP_OFFSET_Y(y) | SP_OFFSET_X(x)); @@ -428,8 +425,7 @@ vlv_sprite_update_arm(struct intel_dsb *dsb, * the control register just before the surface register. */ intel_de_write_fw(display, SPCNTR(pipe, plane_id), sprctl); - intel_de_write_fw(display, SPSURF(pipe, plane_id), - intel_plane_ggtt_offset(plane_state) + sprsurf_offset); + intel_de_write_fw(display, SPSURF(pipe, plane_id), plane_state->surf); vlv_sprite_update_clrc(plane_state); vlv_sprite_update_gamma(plane_state); @@ -663,8 +659,7 @@ static bool ivb_need_sprite_gamma(const struct intel_plane_state *plane_state) (display->platform.ivybridge || display->platform.haswell); } -static u32 ivb_sprite_ctl(const struct intel_crtc_state *crtc_state, - const struct intel_plane_state *plane_state) +static u32 ivb_sprite_ctl(const struct intel_plane_state *plane_state) { struct intel_display *display = to_intel_display(plane_state); const struct drm_framebuffer *fb = plane_state->hw.fb; @@ -830,15 +825,12 @@ ivb_sprite_update_arm(struct intel_dsb *dsb, struct intel_display *display = to_intel_display(plane); enum pipe pipe = plane->pipe; const struct drm_intel_sprite_colorkey *key = &plane_state->ckey; - u32 sprsurf_offset = plane_state->view.color_plane[0].offset; u32 x = plane_state->view.color_plane[0].x; u32 y = plane_state->view.color_plane[0].y; - u32 sprctl, linear_offset; + u32 sprctl; sprctl = plane_state->ctl | ivb_sprite_ctl_crtc(crtc_state); - linear_offset = intel_fb_xy_to_linear(x, y, plane_state, 0); - if (key->flags) { intel_de_write_fw(display, SPRKEYVAL(pipe), key->min_value); intel_de_write_fw(display, SPRKEYMSK(pipe), @@ -852,7 +844,8 @@ ivb_sprite_update_arm(struct intel_dsb *dsb, intel_de_write_fw(display, SPROFFSET(pipe), SPRITE_OFFSET_Y(y) | SPRITE_OFFSET_X(x)); } else { - intel_de_write_fw(display, SPRLINOFF(pipe), linear_offset); + intel_de_write_fw(display, SPRLINOFF(pipe), + intel_fb_xy_to_linear(x, y, plane_state, 0)); intel_de_write_fw(display, SPRTILEOFF(pipe), SPRITE_OFFSET_Y(y) | SPRITE_OFFSET_X(x)); } @@ -863,8 +856,7 @@ ivb_sprite_update_arm(struct intel_dsb *dsb, * the control register just before the surface register. */ intel_de_write_fw(display, SPRCTL(pipe), sprctl); - intel_de_write_fw(display, SPRSURF(pipe), - intel_plane_ggtt_offset(plane_state) + sprsurf_offset); + intel_de_write_fw(display, SPRSURF(pipe), plane_state->surf); ivb_sprite_update_gamma(plane_state); } @@ -1016,8 +1008,7 @@ static u32 g4x_sprite_ctl_crtc(const struct intel_crtc_state *crtc_state) return dvscntr; } -static u32 g4x_sprite_ctl(const struct intel_crtc_state *crtc_state, - const struct intel_plane_state *plane_state) +static u32 g4x_sprite_ctl(const struct intel_plane_state *plane_state) { struct intel_display *display = to_intel_display(plane_state); const struct drm_framebuffer *fb = plane_state->hw.fb; @@ -1181,15 +1172,12 @@ g4x_sprite_update_arm(struct intel_dsb *dsb, struct intel_display *display = to_intel_display(plane); enum pipe pipe = plane->pipe; const struct drm_intel_sprite_colorkey *key = &plane_state->ckey; - u32 dvssurf_offset = plane_state->view.color_plane[0].offset; u32 x = plane_state->view.color_plane[0].x; u32 y = plane_state->view.color_plane[0].y; - u32 dvscntr, linear_offset; + u32 dvscntr; dvscntr = plane_state->ctl | g4x_sprite_ctl_crtc(crtc_state); - linear_offset = intel_fb_xy_to_linear(x, y, plane_state, 0); - if (key->flags) { intel_de_write_fw(display, DVSKEYVAL(pipe), key->min_value); intel_de_write_fw(display, DVSKEYMSK(pipe), @@ -1197,7 +1185,8 @@ g4x_sprite_update_arm(struct intel_dsb *dsb, intel_de_write_fw(display, DVSKEYMAX(pipe), key->max_value); } - intel_de_write_fw(display, DVSLINOFF(pipe), linear_offset); + intel_de_write_fw(display, DVSLINOFF(pipe), + intel_fb_xy_to_linear(x, y, plane_state, 0)); intel_de_write_fw(display, DVSTILEOFF(pipe), DVS_OFFSET_Y(y) | DVS_OFFSET_X(x)); @@ -1207,8 +1196,7 @@ g4x_sprite_update_arm(struct intel_dsb *dsb, * the control register just before the surface register. */ intel_de_write_fw(display, DVSCNTR(pipe), dvscntr); - intel_de_write_fw(display, DVSSURF(pipe), - intel_plane_ggtt_offset(plane_state) + dvssurf_offset); + intel_de_write_fw(display, DVSSURF(pipe), plane_state->surf); if (display->platform.g4x) g4x_sprite_update_gamma(plane_state); @@ -1387,9 +1375,9 @@ g4x_sprite_check(struct intel_crtc_state *crtc_state, return ret; if (DISPLAY_VER(display) >= 7) - plane_state->ctl = ivb_sprite_ctl(crtc_state, plane_state); + plane_state->ctl = ivb_sprite_ctl(plane_state); else - plane_state->ctl = g4x_sprite_ctl(crtc_state, plane_state); + plane_state->ctl = g4x_sprite_ctl(plane_state); return 0; } @@ -1439,7 +1427,7 @@ vlv_sprite_check(struct intel_crtc_state *crtc_state, if (ret) return ret; - plane_state->ctl = vlv_sprite_ctl(crtc_state, plane_state); + plane_state->ctl = vlv_sprite_ctl(plane_state); return 0; } @@ -1624,6 +1612,7 @@ intel_sprite_plane_create(struct intel_display *display, plane->capture_error = vlv_sprite_capture_error; plane->get_hw_state = vlv_sprite_get_hw_state; plane->check_plane = vlv_sprite_check; + plane->surf_offset = i965_plane_surf_offset; plane->max_stride = i965_plane_max_stride; plane->min_alignment = vlv_plane_min_alignment; plane->min_cdclk = vlv_plane_min_cdclk; @@ -1648,6 +1637,7 @@ intel_sprite_plane_create(struct intel_display *display, plane->capture_error = ivb_sprite_capture_error; plane->get_hw_state = ivb_sprite_get_hw_state; plane->check_plane = g4x_sprite_check; + plane->surf_offset = i965_plane_surf_offset; if (display->platform.broadwell || display->platform.haswell) { plane->max_stride = hsw_sprite_max_stride; @@ -1673,6 +1663,7 @@ intel_sprite_plane_create(struct intel_display *display, plane->capture_error = g4x_sprite_capture_error; plane->get_hw_state = g4x_sprite_get_hw_state; plane->check_plane = g4x_sprite_check; + plane->surf_offset = i965_plane_surf_offset; plane->max_stride = g4x_sprite_max_stride; plane->min_alignment = g4x_sprite_min_alignment; plane->min_cdclk = g4x_sprite_min_cdclk; diff --git a/drivers/gpu/drm/i915/display/intel_tc.c b/drivers/gpu/drm/i915/display/intel_tc.c index 3bc57579fe53..c4a5601c5107 100644 --- a/drivers/gpu/drm/i915/display/intel_tc.c +++ b/drivers/gpu/drm/i915/display/intel_tc.c @@ -3,6 +3,8 @@ * Copyright © 2019 Intel Corporation */ +#include <linux/iopoll.h> + #include <drm/drm_print.h> #include "i915_reg.h" @@ -23,10 +25,6 @@ #include "intel_modeset_lock.h" #include "intel_tc.h" -#define DP_PIN_ASSIGNMENT_C 0x3 -#define DP_PIN_ASSIGNMENT_D 0x4 -#define DP_PIN_ASSIGNMENT_E 0x5 - enum tc_port_mode { TC_PORT_DISCONNECTED, TC_PORT_TBT_ALT, @@ -65,7 +63,9 @@ struct intel_tc_port { enum tc_port_mode mode; enum tc_port_mode init_mode; enum phy_fia phy_fia; + enum intel_tc_pin_assignment pin_assignment; u8 phy_fia_idx; + u8 max_lane_count; }; static enum intel_display_power_domain @@ -251,6 +251,9 @@ tc_port_power_domain(struct intel_tc_port *tc) { enum tc_port tc_port = intel_encoder_to_tc(&tc->dig_port->base); + if (tc_port == TC_PORT_NONE) + return POWER_DOMAIN_INVALID; + return POWER_DOMAIN_PORT_DDI_LANES_TC1 + tc_port - TC_PORT_1; } @@ -263,13 +266,14 @@ assert_tc_port_power_enabled(struct intel_tc_port *tc) !intel_display_power_is_enabled(display, tc_port_power_domain(tc))); } -static u32 intel_tc_port_get_lane_mask(struct intel_digital_port *dig_port) +static u32 get_lane_mask(struct intel_tc_port *tc) { - struct intel_display *display = to_intel_display(dig_port); - struct intel_tc_port *tc = to_tc_port(dig_port); + struct intel_display *display = to_intel_display(tc->dig_port); + intel_wakeref_t wakeref; u32 lane_mask; - lane_mask = intel_de_read(display, PORT_TX_DFLEXDPSP(tc->phy_fia)); + with_intel_display_power(display, POWER_DOMAIN_DISPLAY_CORE, wakeref) + lane_mask = intel_de_read(display, PORT_TX_DFLEXDPSP(tc->phy_fia)); drm_WARN_ON(display->drm, lane_mask == 0xffffffff); assert_tc_cold_blocked(tc); @@ -278,75 +282,87 @@ static u32 intel_tc_port_get_lane_mask(struct intel_digital_port *dig_port) return lane_mask >> DP_LANE_ASSIGNMENT_SHIFT(tc->phy_fia_idx); } -u32 intel_tc_port_get_pin_assignment_mask(struct intel_digital_port *dig_port) +static char pin_assignment_name(enum intel_tc_pin_assignment pin_assignment) { - struct intel_display *display = to_intel_display(dig_port); - struct intel_tc_port *tc = to_tc_port(dig_port); - u32 pin_mask; - - pin_mask = intel_de_read(display, PORT_TX_DFLEXPA1(tc->phy_fia)); - - drm_WARN_ON(display->drm, pin_mask == 0xffffffff); - assert_tc_cold_blocked(tc); + if (pin_assignment == INTEL_TC_PIN_ASSIGNMENT_NONE) + return '-'; - return (pin_mask & DP_PIN_ASSIGNMENT_MASK(tc->phy_fia_idx)) >> - DP_PIN_ASSIGNMENT_SHIFT(tc->phy_fia_idx); + return 'A' + pin_assignment - INTEL_TC_PIN_ASSIGNMENT_A; } -static int lnl_tc_port_get_max_lane_count(struct intel_digital_port *dig_port) +static enum intel_tc_pin_assignment +get_pin_assignment(struct intel_tc_port *tc) { - struct intel_display *display = to_intel_display(dig_port); - enum tc_port tc_port = intel_encoder_to_tc(&dig_port->base); + struct intel_display *display = to_intel_display(tc->dig_port); + enum tc_port tc_port = intel_encoder_to_tc(&tc->dig_port->base); + enum intel_tc_pin_assignment pin_assignment; intel_wakeref_t wakeref; - u32 val, pin_assignment; + i915_reg_t reg; + u32 mask; + u32 val; + + if (tc->mode == TC_PORT_TBT_ALT) + return INTEL_TC_PIN_ASSIGNMENT_NONE; + + if (DISPLAY_VER(display) >= 20) { + reg = TCSS_DDI_STATUS(tc_port); + mask = TCSS_DDI_STATUS_PIN_ASSIGNMENT_MASK; + } else { + reg = PORT_TX_DFLEXPA1(tc->phy_fia); + mask = DP_PIN_ASSIGNMENT_MASK(tc->phy_fia_idx); + } with_intel_display_power(display, POWER_DOMAIN_DISPLAY_CORE, wakeref) - val = intel_de_read(display, TCSS_DDI_STATUS(tc_port)); + val = intel_de_read(display, reg); - pin_assignment = - REG_FIELD_GET(TCSS_DDI_STATUS_PIN_ASSIGNMENT_MASK, val); + drm_WARN_ON(display->drm, val == 0xffffffff); + assert_tc_cold_blocked(tc); + + pin_assignment = (val & mask) >> (ffs(mask) - 1); switch (pin_assignment) { + case INTEL_TC_PIN_ASSIGNMENT_A: + case INTEL_TC_PIN_ASSIGNMENT_B: + case INTEL_TC_PIN_ASSIGNMENT_F: + drm_WARN_ON(display->drm, DISPLAY_VER(display) > 11); + break; + case INTEL_TC_PIN_ASSIGNMENT_NONE: + case INTEL_TC_PIN_ASSIGNMENT_C: + case INTEL_TC_PIN_ASSIGNMENT_D: + case INTEL_TC_PIN_ASSIGNMENT_E: + break; default: MISSING_CASE(pin_assignment); - fallthrough; - case DP_PIN_ASSIGNMENT_D: - return 2; - case DP_PIN_ASSIGNMENT_C: - case DP_PIN_ASSIGNMENT_E: - return 4; } + + return pin_assignment; } -static int mtl_tc_port_get_max_lane_count(struct intel_digital_port *dig_port) +static int mtl_get_max_lane_count(struct intel_tc_port *tc) { - struct intel_display *display = to_intel_display(dig_port); - intel_wakeref_t wakeref; - u32 pin_mask; + enum intel_tc_pin_assignment pin_assignment; - with_intel_display_power(display, POWER_DOMAIN_DISPLAY_CORE, wakeref) - pin_mask = intel_tc_port_get_pin_assignment_mask(dig_port); + pin_assignment = get_pin_assignment(tc); - switch (pin_mask) { + switch (pin_assignment) { + case INTEL_TC_PIN_ASSIGNMENT_NONE: + return 0; default: - MISSING_CASE(pin_mask); + MISSING_CASE(pin_assignment); fallthrough; - case DP_PIN_ASSIGNMENT_D: + case INTEL_TC_PIN_ASSIGNMENT_D: return 2; - case DP_PIN_ASSIGNMENT_C: - case DP_PIN_ASSIGNMENT_E: + case INTEL_TC_PIN_ASSIGNMENT_C: + case INTEL_TC_PIN_ASSIGNMENT_E: return 4; } } -static int intel_tc_port_get_max_lane_count(struct intel_digital_port *dig_port) +static int icl_get_max_lane_count(struct intel_tc_port *tc) { - struct intel_display *display = to_intel_display(dig_port); - intel_wakeref_t wakeref; u32 lane_mask = 0; - with_intel_display_power(display, POWER_DOMAIN_DISPLAY_CORE, wakeref) - lane_mask = intel_tc_port_get_lane_mask(dig_port); + lane_mask = get_lane_mask(tc); switch (lane_mask) { default: @@ -365,23 +381,44 @@ static int intel_tc_port_get_max_lane_count(struct intel_digital_port *dig_port) } } +static int get_max_lane_count(struct intel_tc_port *tc) +{ + struct intel_display *display = to_intel_display(tc->dig_port); + + if (tc->mode != TC_PORT_DP_ALT) + return 4; + + if (DISPLAY_VER(display) >= 14) + return mtl_get_max_lane_count(tc); + + return icl_get_max_lane_count(tc); +} + +static void read_pin_configuration(struct intel_tc_port *tc) +{ + tc->pin_assignment = get_pin_assignment(tc); + tc->max_lane_count = get_max_lane_count(tc); +} + int intel_tc_port_max_lane_count(struct intel_digital_port *dig_port) { - struct intel_display *display = to_intel_display(dig_port); struct intel_tc_port *tc = to_tc_port(dig_port); - if (!intel_encoder_is_tc(&dig_port->base) || tc->mode != TC_PORT_DP_ALT) + if (!intel_encoder_is_tc(&dig_port->base)) return 4; - assert_tc_cold_blocked(tc); + return tc->max_lane_count; +} - if (DISPLAY_VER(display) >= 20) - return lnl_tc_port_get_max_lane_count(dig_port); +enum intel_tc_pin_assignment +intel_tc_port_get_pin_assignment(struct intel_digital_port *dig_port) +{ + struct intel_tc_port *tc = to_tc_port(dig_port); - if (DISPLAY_VER(display) >= 14) - return mtl_tc_port_get_max_lane_count(dig_port); + if (!intel_encoder_is_tc(&dig_port->base)) + return INTEL_TC_PIN_ASSIGNMENT_NONE; - return intel_tc_port_get_max_lane_count(dig_port); + return tc->pin_assignment; } void intel_tc_port_set_fia_lane_count(struct intel_digital_port *dig_port, @@ -596,9 +633,12 @@ static void icl_tc_phy_get_hw_state(struct intel_tc_port *tc) tc_cold_wref = __tc_cold_block(tc, &domain); tc->mode = tc_phy_get_current_mode(tc); - if (tc->mode != TC_PORT_DISCONNECTED) + if (tc->mode != TC_PORT_DISCONNECTED) { tc->lock_wakeref = tc_cold_block(tc); + read_pin_configuration(tc); + } + __tc_cold_unblock(tc, domain, tc_cold_wref); } @@ -656,8 +696,11 @@ static bool icl_tc_phy_connect(struct intel_tc_port *tc, tc->lock_wakeref = tc_cold_block(tc); - if (tc->mode == TC_PORT_TBT_ALT) + if (tc->mode == TC_PORT_TBT_ALT) { + read_pin_configuration(tc); + return true; + } if ((!tc_phy_is_ready(tc) || !icl_tc_phy_take_ownership(tc, true)) && @@ -668,6 +711,7 @@ static bool icl_tc_phy_connect(struct intel_tc_port *tc, goto out_unblock_tc_cold; } + read_pin_configuration(tc); if (!tc_phy_verify_legacy_or_dp_alt_mode(tc, required_lanes)) goto out_release_phy; @@ -858,9 +902,12 @@ static void adlp_tc_phy_get_hw_state(struct intel_tc_port *tc) port_wakeref = intel_display_power_get(display, port_power_domain); tc->mode = tc_phy_get_current_mode(tc); - if (tc->mode != TC_PORT_DISCONNECTED) + if (tc->mode != TC_PORT_DISCONNECTED) { tc->lock_wakeref = tc_cold_block(tc); + read_pin_configuration(tc); + } + intel_display_power_put(display, port_power_domain, port_wakeref); } @@ -873,6 +920,9 @@ static bool adlp_tc_phy_connect(struct intel_tc_port *tc, int required_lanes) if (tc->mode == TC_PORT_TBT_ALT) { tc->lock_wakeref = tc_cold_block(tc); + + read_pin_configuration(tc); + return true; } @@ -894,6 +944,8 @@ static bool adlp_tc_phy_connect(struct intel_tc_port *tc, int required_lanes) tc->lock_wakeref = tc_cold_block(tc); + read_pin_configuration(tc); + if (!tc_phy_verify_legacy_or_dp_alt_mode(tc, required_lanes)) goto out_unblock_tc_cold; @@ -1000,8 +1052,13 @@ static bool xelpdp_tc_phy_wait_for_tcss_power(struct intel_tc_port *tc, bool enabled) { struct intel_display *display = to_intel_display(tc->dig_port); + bool is_enabled; + int ret; - if (wait_for(xelpdp_tc_phy_tcss_power_is_enabled(tc) == enabled, 5)) { + ret = poll_timeout_us(is_enabled = xelpdp_tc_phy_tcss_power_is_enabled(tc), + is_enabled == enabled, + 200, 5000, false); + if (ret) { drm_dbg_kms(display->drm, "Port %s: timeout waiting for TCSS power to get %s\n", str_enabled_disabled(enabled), @@ -1124,9 +1181,18 @@ static void xelpdp_tc_phy_get_hw_state(struct intel_tc_port *tc) tc_cold_wref = __tc_cold_block(tc, &domain); tc->mode = tc_phy_get_current_mode(tc); - if (tc->mode != TC_PORT_DISCONNECTED) + if (tc->mode != TC_PORT_DISCONNECTED) { tc->lock_wakeref = tc_cold_block(tc); + read_pin_configuration(tc); + /* + * Set a valid lane count value for a DP-alt sink which got + * disconnected. The driver can only disable the output on this PHY. + */ + if (tc->max_lane_count == 0) + tc->max_lane_count = 4; + } + drm_WARN_ON(display->drm, (tc->mode == TC_PORT_DP_ALT || tc->mode == TC_PORT_LEGACY) && !xelpdp_tc_phy_tcss_power_is_enabled(tc)); @@ -1138,14 +1204,19 @@ static bool xelpdp_tc_phy_connect(struct intel_tc_port *tc, int required_lanes) { tc->lock_wakeref = tc_cold_block(tc); - if (tc->mode == TC_PORT_TBT_ALT) + if (tc->mode == TC_PORT_TBT_ALT) { + read_pin_configuration(tc); + return true; + } if (!xelpdp_tc_phy_enable_tcss_power(tc, true)) goto out_unblock_tccold; xelpdp_tc_phy_take_ownership(tc, true); + read_pin_configuration(tc); + if (!tc_phy_verify_legacy_or_dp_alt_mode(tc, required_lanes)) goto out_release_phy; @@ -1226,14 +1297,19 @@ static void tc_phy_get_hw_state(struct intel_tc_port *tc) tc->phy_ops->get_hw_state(tc); } -static bool tc_phy_is_ready_and_owned(struct intel_tc_port *tc, - bool phy_is_ready, bool phy_is_owned) +/* Is the PHY owned by display i.e. is it in legacy or DP-alt mode? */ +static bool tc_phy_owned_by_display(struct intel_tc_port *tc, + bool phy_is_ready, bool phy_is_owned) { struct intel_display *display = to_intel_display(tc->dig_port); - drm_WARN_ON(display->drm, phy_is_owned && !phy_is_ready); + if (DISPLAY_VER(display) < 20) { + drm_WARN_ON(display->drm, phy_is_owned && !phy_is_ready); - return phy_is_ready && phy_is_owned; + return phy_is_ready && phy_is_owned; + } else { + return phy_is_owned; + } } static bool tc_phy_is_connected(struct intel_tc_port *tc, @@ -1244,7 +1320,7 @@ static bool tc_phy_is_connected(struct intel_tc_port *tc, bool phy_is_owned = tc_phy_is_owned(tc); bool is_connected; - if (tc_phy_is_ready_and_owned(tc, phy_is_ready, phy_is_owned)) + if (tc_phy_owned_by_display(tc, phy_is_ready, phy_is_owned)) is_connected = port_pll_type == ICL_PORT_DPLL_MG_PHY; else is_connected = port_pll_type == ICL_PORT_DPLL_DEFAULT; @@ -1263,8 +1339,13 @@ static bool tc_phy_is_connected(struct intel_tc_port *tc, static bool tc_phy_wait_for_ready(struct intel_tc_port *tc) { struct intel_display *display = to_intel_display(tc->dig_port); + bool is_ready; + int ret; - if (wait_for(tc_phy_is_ready(tc), 500)) { + ret = poll_timeout_us(is_ready = tc_phy_is_ready(tc), + is_ready, + 1000, 500 * 1000, false); + if (ret) { drm_err(display->drm, "Port %s: timeout waiting for PHY ready\n", tc->port_name); @@ -1352,7 +1433,7 @@ tc_phy_get_current_mode(struct intel_tc_port *tc) phy_is_ready = tc_phy_is_ready(tc); phy_is_owned = tc_phy_is_owned(tc); - if (!tc_phy_is_ready_and_owned(tc, phy_is_ready, phy_is_owned)) { + if (!tc_phy_owned_by_display(tc, phy_is_ready, phy_is_owned)) { mode = get_tc_mode_in_phy_not_owned_state(tc, live_mode); } else { drm_WARN_ON(display->drm, live_mode == TC_PORT_TBT_ALT); @@ -1441,21 +1522,24 @@ static void intel_tc_port_reset_mode(struct intel_tc_port *tc, intel_display_power_flush_work(display); if (!intel_tc_cold_requires_aux_pw(dig_port)) { enum intel_display_power_domain aux_domain; - bool aux_powered; aux_domain = intel_aux_power_domain(dig_port); - aux_powered = intel_display_power_is_enabled(display, aux_domain); - drm_WARN_ON(display->drm, aux_powered); + if (intel_display_power_is_enabled(display, aux_domain)) + drm_dbg_kms(display->drm, "Port %s: AUX unexpectedly powered\n", + tc->port_name); } tc_phy_disconnect(tc); if (!force_disconnect) tc_phy_connect(tc, required_lanes); - drm_dbg_kms(display->drm, "Port %s: TC port mode reset (%s -> %s)\n", + drm_dbg_kms(display->drm, + "Port %s: TC port mode reset (%s -> %s) pin assignment: %c max lanes: %d\n", tc->port_name, tc_port_mode_name(old_tc_mode), - tc_port_mode_name(tc->mode)); + tc_port_mode_name(tc->mode), + pin_assignment_name(tc->pin_assignment), + tc->max_lane_count); } static bool intel_tc_port_needs_reset(struct intel_tc_port *tc) @@ -1610,9 +1694,11 @@ void intel_tc_port_sanitize_mode(struct intel_digital_port *dig_port, __intel_tc_port_put_link(tc); } - drm_dbg_kms(display->drm, "Port %s: sanitize mode (%s)\n", + drm_dbg_kms(display->drm, "Port %s: sanitize mode (%s) pin assignment: %c max lanes: %d\n", tc->port_name, - tc_port_mode_name(tc->mode)); + tc_port_mode_name(tc->mode), + pin_assignment_name(tc->pin_assignment), + tc->max_lane_count); mutex_unlock(&tc->lock); } diff --git a/drivers/gpu/drm/i915/display/intel_tc.h b/drivers/gpu/drm/i915/display/intel_tc.h index 26c4265368c1..fff8b96e4972 100644 --- a/drivers/gpu/drm/i915/display/intel_tc.h +++ b/drivers/gpu/drm/i915/display/intel_tc.h @@ -12,6 +12,75 @@ struct intel_crtc_state; struct intel_digital_port; struct intel_encoder; +/* + * The following enum values must stay fixed, as they match the corresponding + * pin assignment fields in the PORT_TX_DFLEXPA1 and TCSS_DDI_STATUS registers. + */ +enum intel_tc_pin_assignment { /* Lanes (a) Signal/ Cable Notes */ + /* DP USB Rate (b) type */ + INTEL_TC_PIN_ASSIGNMENT_NONE = 0, /* 4 - - - (c) */ + INTEL_TC_PIN_ASSIGNMENT_A, /* 2/4 0 GEN2 TC->TC (d,e) */ + INTEL_TC_PIN_ASSIGNMENT_B, /* 1/2 1 GEN2 TC->TC (d,f,g) */ + INTEL_TC_PIN_ASSIGNMENT_C, /* 4 0 DP2 TC->TC (h) */ + INTEL_TC_PIN_ASSIGNMENT_D, /* 2 1 DP2 TC->TC (h,g) */ + INTEL_TC_PIN_ASSIGNMENT_E, /* 4 0 DP2 TC->DP */ + INTEL_TC_PIN_ASSIGNMENT_F, /* 2 1 GEN1/DP1 TC->DP (d,g,i) */ + /* + * (a) - DP unidirectional lanes, each lane using 1 differential signal + * pair. + * - USB SuperSpeed bidirectional lane, using 2 differential (TX and + * RX) signal pairs. + * - USB 2.0 (HighSpeed) unidirectional lane, using 1 differential + * signal pair. Not indicated, this lane is always present on pin + * assignments A-D and never present on pin assignments E/F. + * (b) - GEN1: USB 3.1 GEN1 bit rate (5 Gbps) and signaling. This + * is used for transferring only a USB stream. + * - GEN2: USB 3.1 GEN2 bit rate (10 Gbps) and signaling. This + * allows transferring an HBR3 (8.1 Gbps) DP stream. + * - DP1: Display Port signaling defined by the DP v1.3 Standard, + * with a maximum bit rate of HBR3. + * - DP2: Display Port signaling defined by the DP v2.1 Standard, + * with a maximum bit rate defined by the DP Alt Mode + * v2.1a Standard depending on the cable type as follows: + * - Passive (Full-Featured) USB 3.2 GEN1 + * TC->TC cables (CC3G1-X) : UHBR10 + * - Passive (Full-Featured) USB 3.2/4 GEN2 and + * Thunderbolt Alt Mode GEN2 + * TC->TC cables (CC3G2-X) all : UHBR10 + * DP54 logo : UHBR13.5 + * - Passive (Full-Featured) USB4 GEN3+ and + * Thunderbolt Alt Mode GEN3+ + * TC->TC cables (CC4G3-X) all : UHBR13.5 + * DP80 logo : UHBR20 + * - Active Re-Timed or + * Active Linear Re-driven (LRD) + * USB3.2 GEN1/2 and USB4 GEN2+ + * TC->TC cables all : HBR3 + * with DP_BR CTS : UHBR10 + * DP54 logo : UHBR13.5 + * DP80 logo : UHBR20 + * - Passive/Active Re-Timed or + * Active Linear Re-driven (LRD) + * TC->DP cables with DP_BR CTS/DP8K logo : HBR3 + * with DP_BR CTS : UHBR10 + * DP54 logo : UHBR13.5 + * DP80 logo : UHBR20 + * (c) Used in TBT-alt/legacy modes and on LNL+ after the sink + * disconnected in DP-alt mode. + * (d) Only defined by the DP Alt Standard v1.0a, deprecated by v1.0b, + * only supported on ICL. + * (e) GEN2 passive 1 m cable: 4 DP lanes, GEN2 active cable: 2 DP lanes. + * (f) GEN2 passive 1 m cable: 2 DP lanes, GEN2 active cable: 1 DP lane. + * (g) These pin assignments are also referred to as (USB/DP) + * multifunction or Multifunction Display Port (MFD) modes. + * (h) Also used where one end of the cable is a captive connector, + * attached to a DP->HDMI/DVI/VGA converter. + * (i) The DP end of the cable is a captive connector attached to a + * (DP/USB) multifunction dock as defined by the DockPort v1.0a + * specification. + */ +}; + bool intel_tc_port_in_tbt_alt_mode(struct intel_digital_port *dig_port); bool intel_tc_port_in_dp_alt_mode(struct intel_digital_port *dig_port); bool intel_tc_port_in_legacy_mode(struct intel_digital_port *dig_port); @@ -19,7 +88,8 @@ bool intel_tc_port_handles_hpd_glitches(struct intel_digital_port *dig_port); bool intel_tc_port_connected(struct intel_encoder *encoder); -u32 intel_tc_port_get_pin_assignment_mask(struct intel_digital_port *dig_port); +enum intel_tc_pin_assignment +intel_tc_port_get_pin_assignment(struct intel_digital_port *dig_port); int intel_tc_port_max_lane_count(struct intel_digital_port *dig_port); void intel_tc_port_set_fia_lane_count(struct intel_digital_port *dig_port, int required_lanes); diff --git a/drivers/gpu/drm/i915/display/intel_vblank.c b/drivers/gpu/drm/i915/display/intel_vblank.c index 70ba7aa26bf4..c15234c1d96e 100644 --- a/drivers/gpu/drm/i915/display/intel_vblank.c +++ b/drivers/gpu/drm/i915/display/intel_vblank.c @@ -3,9 +3,12 @@ * Copyright © 2022-2023 Intel Corporation */ +#include <linux/iopoll.h> + #include <drm/drm_vblank.h> #include "i915_drv.h" +#include "i915_utils.h" #include "intel_color.h" #include "intel_crtc.h" #include "intel_de.h" @@ -492,9 +495,14 @@ static void wait_for_pipe_scanline_moving(struct intel_crtc *crtc, bool state) { struct intel_display *display = to_intel_display(crtc); enum pipe pipe = crtc->pipe; + bool is_moving; + int ret; /* Wait for the display line to settle/start moving */ - if (wait_for(pipe_scanline_is_moving(display, pipe) == state, 100)) + ret = poll_timeout_us(is_moving = pipe_scanline_is_moving(display, pipe), + is_moving == state, + 500, 100 * 1000, false); + if (ret) drm_err(display->drm, "pipe %c scanline %s wait timed out\n", pipe_name(pipe), str_on_off(state)); @@ -724,9 +732,9 @@ int intel_vblank_evade(struct intel_vblank_evade_ctx *evade) break; if (!timeout) { - drm_err(display->drm, - "Potential atomic update failure on pipe %c\n", - pipe_name(crtc->pipe)); + drm_dbg_kms(display->drm, + "Potential atomic update failure on pipe %c\n", + pipe_name(crtc->pipe)); break; } diff --git a/drivers/gpu/drm/i915/display/intel_vbt_defs.h b/drivers/gpu/drm/i915/display/intel_vbt_defs.h index 92c04811aa28..70e31520c560 100644 --- a/drivers/gpu/drm/i915/display/intel_vbt_defs.h +++ b/drivers/gpu/drm/i915/display/intel_vbt_defs.h @@ -37,7 +37,7 @@ #ifndef _INTEL_VBT_DEFS_H_ #define _INTEL_VBT_DEFS_H_ -#include "intel_bios.h" +#include "intel_dsi_vbt_defs.h" /* EDID derived structures */ struct bdb_edid_pnp_id { @@ -437,6 +437,22 @@ enum vbt_gmbus_ddi { #define BDB_230_VBT_DP_MAX_LINK_RATE_UHBR13P5 6 #define BDB_230_VBT_DP_MAX_LINK_RATE_UHBR20 7 +/* EDP link rate 263+ */ +#define BDB_263_VBT_EDP_LINK_RATE_1_62 BIT_U32(0) +#define BDB_263_VBT_EDP_LINK_RATE_2_16 BIT_U32(1) +#define BDB_263_VBT_EDP_LINK_RATE_2_43 BIT_U32(2) +#define BDB_263_VBT_EDP_LINK_RATE_2_7 BIT_U32(3) +#define BDB_263_VBT_EDP_LINK_RATE_3_24 BIT_U32(4) +#define BDB_263_VBT_EDP_LINK_RATE_4_32 BIT_U32(5) +#define BDB_263_VBT_EDP_LINK_RATE_5_4 BIT_U32(6) +#define BDB_263_VBT_EDP_LINK_RATE_6_75 BIT_U32(7) +#define BDB_263_VBT_EDP_LINK_RATE_8_1 BIT_U32(8) +#define BDB_263_VBT_EDP_LINK_RATE_10 BIT_U32(9) +#define BDB_263_VBT_EDP_LINK_RATE_13_5 BIT_U32(10) +#define BDB_263_VBT_EDP_LINK_RATE_20 BIT_U32(11) +#define BDB_263_VBT_EDP_NUM_RATES 12 +#define BDB_263_VBT_EDP_RATES_MASK GENMASK(BDB_263_VBT_EDP_NUM_RATES - 1, 0) + /* * The child device config, aka the display device data structure, provides a * description of a port and its configuration on the platform. @@ -547,6 +563,8 @@ struct child_device_config { u8 dp_max_link_rate:3; /* 216+ */ u8 dp_max_link_rate_reserved:5; /* 216+ */ u8 efp_index; /* 256+ */ + u32 edp_data_rate_override:12; /* 263+ */ + u32 edp_data_rate_override_reserved:20; /* 263+ */ } __packed; struct bdb_general_definitions { diff --git a/drivers/gpu/drm/i915/display/intel_wm.c b/drivers/gpu/drm/i915/display/intel_wm.c index bba82e888db2..f887a664fe22 100644 --- a/drivers/gpu/drm/i915/display/intel_wm.c +++ b/drivers/gpu/drm/i915/display/intel_wm.c @@ -5,7 +5,6 @@ #include <linux/debugfs.h> -#include <drm/drm_file.h> #include <drm/drm_print.h> #include "i9xx_wm.h" @@ -390,15 +389,15 @@ static const struct file_operations i915_cur_wm_latency_fops = { void intel_wm_debugfs_register(struct intel_display *display) { - struct drm_minor *minor = display->drm->primary; + struct dentry *debugfs_root = display->drm->debugfs_root; - debugfs_create_file("i915_pri_wm_latency", 0644, minor->debugfs_root, + debugfs_create_file("i915_pri_wm_latency", 0644, debugfs_root, display, &i915_pri_wm_latency_fops); - debugfs_create_file("i915_spr_wm_latency", 0644, minor->debugfs_root, + debugfs_create_file("i915_spr_wm_latency", 0644, debugfs_root, display, &i915_spr_wm_latency_fops); - debugfs_create_file("i915_cur_wm_latency", 0644, minor->debugfs_root, + debugfs_create_file("i915_cur_wm_latency", 0644, debugfs_root, display, &i915_cur_wm_latency_fops); skl_watermark_debugfs_register(display); diff --git a/drivers/gpu/drm/i915/display/skl_scaler.c b/drivers/gpu/drm/i915/display/skl_scaler.c index d77798499c57..c6cccf170ff1 100644 --- a/drivers/gpu/drm/i915/display/skl_scaler.c +++ b/drivers/gpu/drm/i915/display/skl_scaler.c @@ -10,6 +10,7 @@ #include "intel_display_regs.h" #include "intel_display_trace.h" #include "intel_display_types.h" +#include "intel_display_wa.h" #include "intel_fb.h" #include "skl_scaler.h" #include "skl_universal_plane.h" @@ -91,11 +92,9 @@ static void skl_scaler_min_src_size(const struct drm_format_info *format, } } -static void skl_scaler_max_src_size(struct intel_crtc *crtc, +static void skl_scaler_max_src_size(struct intel_display *display, int *max_w, int *max_h) { - struct intel_display *display = to_intel_display(crtc); - if (DISPLAY_VER(display) >= 14) { *max_w = 4096; *max_h = 8192; @@ -134,6 +133,23 @@ static void skl_scaler_max_dst_size(struct intel_crtc *crtc, } } +enum drm_mode_status +skl_scaler_mode_valid(struct intel_display *display, + const struct drm_display_mode *mode, + enum intel_output_format output_format, + int num_joined_pipes) +{ + int max_h, max_w; + + if (num_joined_pipes < 2 && output_format == INTEL_OUTPUT_FORMAT_YCBCR420) { + skl_scaler_max_src_size(display, &max_w, &max_h); + if (mode->hdisplay > max_h) + return MODE_NO_420; + } + + return MODE_OK; +} + static int skl_update_scaler(struct intel_crtc_state *crtc_state, bool force_detach, unsigned int scaler_user, int *scaler_id, @@ -201,7 +217,7 @@ skl_update_scaler(struct intel_crtc_state *crtc_state, bool force_detach, } skl_scaler_min_src_size(format, modifier, &min_src_w, &min_src_h); - skl_scaler_max_src_size(crtc, &max_src_w, &max_src_h); + skl_scaler_max_src_size(display, &max_src_w, &max_src_h); skl_scaler_min_dst_size(&min_dst_w, &min_dst_h); skl_scaler_max_dst_size(crtc, &max_dst_w, &max_dst_h); @@ -747,6 +763,9 @@ void skl_pfit_enable(const struct intel_crtc_state *crtc_state) crtc_state->scaler_state.scaler_id < 0)) return; + if (intel_display_wa(display, 14011503117)) + adl_scaler_ecc_mask(crtc_state); + drm_rect_init(&src, 0, 0, drm_rect_width(&crtc_state->pipe_src) << 16, drm_rect_height(&crtc_state->pipe_src) << 16); @@ -923,3 +942,29 @@ void skl_scaler_get_config(struct intel_crtc_state *crtc_state) else scaler_state->scaler_users &= ~(1 << SKL_CRTC_INDEX); } + +void adl_scaler_ecc_mask(const struct intel_crtc_state *crtc_state) +{ + struct intel_display *display = to_intel_display(crtc_state); + + if (!crtc_state->pch_pfit.enabled) + return; + + intel_de_write(display, XELPD_DISPLAY_ERR_FATAL_MASK, ~0); +} + +void adl_scaler_ecc_unmask(const struct intel_crtc_state *crtc_state) +{ + struct intel_display *display = to_intel_display(crtc_state); + struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); + const struct intel_crtc_scaler_state *scaler_state = + &crtc_state->scaler_state; + + if (scaler_state->scaler_id < 0) + return; + + intel_de_write_fw(display, + SKL_PS_ECC_STAT(crtc->pipe, scaler_state->scaler_id), + 1); + intel_de_write(display, XELPD_DISPLAY_ERR_FATAL_MASK, 0); +} diff --git a/drivers/gpu/drm/i915/display/skl_scaler.h b/drivers/gpu/drm/i915/display/skl_scaler.h index 355ea15260ca..12a19016c5f6 100644 --- a/drivers/gpu/drm/i915/display/skl_scaler.h +++ b/drivers/gpu/drm/i915/display/skl_scaler.h @@ -5,10 +5,14 @@ #ifndef INTEL_SCALER_H #define INTEL_SCALER_H +enum drm_mode_status; +struct drm_display_mode; struct intel_atomic_state; struct intel_crtc; struct intel_crtc_state; +struct intel_display; struct intel_dsb; +enum intel_output_format; struct intel_plane; struct intel_plane_state; @@ -32,4 +36,13 @@ void skl_scaler_disable(const struct intel_crtc_state *old_crtc_state); void skl_scaler_get_config(struct intel_crtc_state *crtc_state); +enum drm_mode_status +skl_scaler_mode_valid(struct intel_display *display, + const struct drm_display_mode *mode, + enum intel_output_format output_format, + int num_joined_pipes); + +void adl_scaler_ecc_mask(const struct intel_crtc_state *crtc_state); + +void adl_scaler_ecc_unmask(const struct intel_crtc_state *crtc_state); #endif diff --git a/drivers/gpu/drm/i915/display/skl_universal_plane.c b/drivers/gpu/drm/i915/display/skl_universal_plane.c index e20972ddfa09..950dc79dbdd4 100644 --- a/drivers/gpu/drm/i915/display/skl_universal_plane.c +++ b/drivers/gpu/drm/i915/display/skl_universal_plane.c @@ -10,6 +10,7 @@ #include "pxp/intel_pxp.h" #include "i915_drv.h" +#include "i915_utils.h" #include "intel_bo.h" #include "intel_de.h" #include "intel_display_irq.h" @@ -1166,8 +1167,7 @@ static u32 skl_plane_ctl_crtc(const struct intel_crtc_state *crtc_state) return plane_ctl; } -static u32 skl_plane_ctl(const struct intel_crtc_state *crtc_state, - const struct intel_plane_state *plane_state) +static u32 skl_plane_ctl(const struct intel_plane_state *plane_state) { struct intel_display *display = to_intel_display(plane_state); const struct drm_framebuffer *fb = plane_state->hw.fb; @@ -1225,8 +1225,7 @@ static u32 glk_plane_color_ctl_crtc(const struct intel_crtc_state *crtc_state) return plane_color_ctl; } -static u32 glk_plane_color_ctl(const struct intel_crtc_state *crtc_state, - const struct intel_plane_state *plane_state) +static u32 glk_plane_color_ctl(const struct intel_plane_state *plane_state) { struct intel_display *display = to_intel_display(plane_state); const struct drm_framebuffer *fb = plane_state->hw.fb; @@ -1271,12 +1270,6 @@ static u32 skl_surf_address(const struct intel_plane_state *plane_state, u32 offset = plane_state->view.color_plane[color_plane].offset; if (intel_fb_uses_dpt(fb)) { - /* - * The DPT object contains only one vma, so the VMA's offset - * within the DPT is always 0. - */ - drm_WARN_ON(display->drm, plane_state->dpt_vma && - intel_dpt_offset(plane_state->dpt_vma)); drm_WARN_ON(display->drm, offset & 0x1fffff); return offset >> 9; } else { @@ -1285,13 +1278,20 @@ static u32 skl_surf_address(const struct intel_plane_state *plane_state, } } -static u32 skl_plane_surf(const struct intel_plane_state *plane_state, - int color_plane) +static int icl_plane_color_plane(const struct intel_plane_state *plane_state) +{ + if (plane_state->planar_linked_plane && !plane_state->is_y_plane) + return 1; + else + return 0; +} + +static u32 skl_plane_surf_offset(const struct intel_plane_state *plane_state) { + int color_plane = icl_plane_color_plane(plane_state); u32 plane_surf; - plane_surf = intel_plane_ggtt_offset(plane_state) + - skl_surf_address(plane_state, color_plane); + plane_surf = skl_surf_address(plane_state, color_plane); if (plane_state->decrypt) plane_surf |= PLANE_SURF_DECRYPT; @@ -1373,14 +1373,6 @@ static void icl_plane_csc_load_black(struct intel_dsb *dsb, intel_de_write_dsb(display, dsb, PLANE_CSC_POSTOFF(pipe, plane_id, 2), 0); } -static int icl_plane_color_plane(const struct intel_plane_state *plane_state) -{ - if (plane_state->planar_linked_plane && !plane_state->is_y_plane) - return 1; - else - return 0; -} - static void skl_plane_update_noarm(struct intel_dsb *dsb, struct intel_plane *plane, @@ -1476,7 +1468,7 @@ skl_plane_update_arm(struct intel_dsb *dsb, intel_de_write_dsb(display, dsb, PLANE_CTL(pipe, plane_id), plane_ctl); intel_de_write_dsb(display, dsb, PLANE_SURF(pipe, plane_id), - skl_plane_surf(plane_state, 0)); + plane_state->surf); } static void icl_plane_update_sel_fetch_noarm(struct intel_dsb *dsb, @@ -1632,7 +1624,6 @@ icl_plane_update_arm(struct intel_dsb *dsb, struct intel_display *display = to_intel_display(plane); enum plane_id plane_id = plane->id; enum pipe pipe = plane->pipe; - int color_plane = icl_plane_color_plane(plane_state); u32 plane_ctl; plane_ctl = plane_state->ctl | @@ -1658,7 +1649,7 @@ icl_plane_update_arm(struct intel_dsb *dsb, intel_de_write_dsb(display, dsb, PLANE_CTL(pipe, plane_id), plane_ctl); intel_de_write_dsb(display, dsb, PLANE_SURF(pipe, plane_id), - skl_plane_surf(plane_state, color_plane)); + plane_state->surf); } static void skl_plane_capture_error(struct intel_crtc *crtc, @@ -1682,10 +1673,10 @@ skl_plane_async_flip(struct intel_dsb *dsb, struct intel_display *display = to_intel_display(plane); enum plane_id plane_id = plane->id; enum pipe pipe = plane->pipe; - u32 plane_ctl = plane_state->ctl, plane_surf; + u32 plane_ctl = plane_state->ctl; + u32 plane_surf = plane_state->surf; plane_ctl |= skl_plane_ctl_crtc(crtc_state); - plane_surf = skl_plane_surf(plane_state, 0); if (async_flip) { if (DISPLAY_VER(display) >= 30) @@ -2363,11 +2354,10 @@ static int skl_plane_check(struct intel_crtc_state *crtc_state, plane_state->damage = DRM_RECT_INIT(0, 0, 0, 0); } - plane_state->ctl = skl_plane_ctl(crtc_state, plane_state); + plane_state->ctl = skl_plane_ctl(plane_state); if (DISPLAY_VER(display) >= 10) - plane_state->color_ctl = glk_plane_color_ctl(crtc_state, - plane_state); + plane_state->color_ctl = glk_plane_color_ctl(plane_state); if (intel_format_info_is_yuv_semiplanar(fb->format, fb->modifier) && icl_is_hdr_plane(display, plane->id)) @@ -2814,7 +2804,7 @@ static void skl_disable_tiling(struct intel_plane *plane) intel_de_write_fw(display, PLANE_CTL(plane->pipe, plane->id), plane_ctl); intel_de_write_fw(display, PLANE_SURF(plane->pipe, plane->id), - skl_plane_surf(state, 0)); + state->surf); } struct intel_plane * @@ -2865,6 +2855,8 @@ skl_universal_plane_create(struct intel_display *display, } plane->disable_tiling = skl_disable_tiling; + plane->surf_offset = skl_plane_surf_offset; + if (DISPLAY_VER(display) >= 13) plane->max_stride = adl_plane_max_stride; else @@ -3191,21 +3183,18 @@ bool skl_fixup_initial_plane_config(struct intel_crtc *crtc, to_intel_plane_state(plane->base.state); enum plane_id plane_id = plane->id; enum pipe pipe = crtc->pipe; - u32 base; if (!plane_state->uapi.visible) return false; - base = intel_plane_ggtt_offset(plane_state); - /* * We may have moved the surface to a different * part of ggtt, make the plane aware of that. */ - if (plane_config->base == base) + if (plane_config->base == plane_state->surf) return false; - intel_de_write(display, PLANE_SURF(pipe, plane_id), base); + intel_de_write(display, PLANE_SURF(pipe, plane_id), plane_state->surf); return true; } diff --git a/drivers/gpu/drm/i915/display/skl_watermark.c b/drivers/gpu/drm/i915/display/skl_watermark.c index 222c069fdadb..d74cbb43ae6f 100644 --- a/drivers/gpu/drm/i915/display/skl_watermark.c +++ b/drivers/gpu/drm/i915/display/skl_watermark.c @@ -6,7 +6,6 @@ #include <linux/debugfs.h> #include <drm/drm_blend.h> -#include <drm/drm_file.h> #include <drm/drm_print.h> #include "soc/intel_dram.h" @@ -1389,7 +1388,7 @@ skl_allocate_plane_ddb(struct skl_plane_ddb_iter *iter, { u16 size, extra = 0; - if (data_rate) { + if (data_rate && iter->data_rate) { extra = min_t(u16, iter->size, DIV64_U64_ROUND_UP(iter->size * data_rate, iter->data_rate)); @@ -2273,6 +2272,11 @@ static int skl_max_wm0_lines(const struct intel_crtc_state *crtc_state) return wm0_lines; } +/* + * TODO: In case we use PKG_C_LATENCY to allow C-states when the delayed vblank + * size is too small for the package C exit latency we need to notify PSR about + * the scenario to apply Wa_16025596647. + */ static int skl_max_wm_level_for_vblank(struct intel_crtc_state *crtc_state, int wm0_lines) { @@ -3205,12 +3209,12 @@ adjust_wm_latency(struct intel_display *display, } /* - * WA Level-0 adjustment for 16GB DIMMs: SKL+ + * WA Level-0 adjustment for 16Gb DIMMs: SKL+ * If we could not get dimm info enable this WA to prevent from - * any underrun. If not able to get Dimm info assume 16GB dimm + * any underrun. If not able to get DIMM info assume 16Gb DIMM * to avoid any underrun. */ - if (!display->platform.dg2 && dram_info->wm_lv_0_adjust_needed) + if (!display->platform.dg2 && dram_info->has_16gb_dimms) wm[0] += 1; } @@ -4033,14 +4037,14 @@ DEFINE_SHOW_ATTRIBUTE(intel_sagv_status); void skl_watermark_debugfs_register(struct intel_display *display) { - struct drm_minor *minor = display->drm->primary; + struct dentry *debugfs_root = display->drm->debugfs_root; if (HAS_IPC(display)) - debugfs_create_file("i915_ipc_status", 0644, minor->debugfs_root, + debugfs_create_file("i915_ipc_status", 0644, debugfs_root, display, &skl_watermark_ipc_status_fops); if (HAS_SAGV(display)) - debugfs_create_file("i915_sagv_status", 0444, minor->debugfs_root, + debugfs_create_file("i915_sagv_status", 0444, debugfs_root, display, &intel_sagv_status_fops); } diff --git a/drivers/gpu/drm/i915/display/vlv_dsi.c b/drivers/gpu/drm/i915/display/vlv_dsi.c index 6d9f3312de7e..c9a53fde79c4 100644 --- a/drivers/gpu/drm/i915/display/vlv_dsi.c +++ b/drivers/gpu/drm/i915/display/vlv_dsi.c @@ -761,7 +761,7 @@ static void intel_dsi_pre_enable(struct intel_atomic_state *state, if (display->platform.valleyview || display->platform.cherryview) { /* Disable DPOunit clock gating, can stall pipe */ - intel_de_rmw(display, DSPCLK_GATE_D(display), + intel_de_rmw(display, VLV_DSPCLK_GATE_D, 0, DPOUNIT_CLOCK_GATE_DISABLE); } @@ -918,7 +918,7 @@ static void intel_dsi_post_disable(struct intel_atomic_state *state, } else { vlv_dsi_pll_disable(encoder); - intel_de_rmw(display, DSPCLK_GATE_D(display), + intel_de_rmw(display, VLV_DSPCLK_GATE_D, DPOUNIT_CLOCK_GATE_DISABLE, 0); } diff --git a/drivers/gpu/drm/i915/display/vlv_dsi_pll.c b/drivers/gpu/drm/i915/display/vlv_dsi_pll.c index d42b61e6f076..f078b9cda96c 100644 --- a/drivers/gpu/drm/i915/display/vlv_dsi_pll.c +++ b/drivers/gpu/drm/i915/display/vlv_dsi_pll.c @@ -25,12 +25,12 @@ * Yogesh Mohan Marimuthu <yogesh.mohan.marimuthu@intel.com> */ +#include <linux/iopoll.h> #include <linux/kernel.h> #include <linux/string_helpers.h> #include <drm/drm_print.h> -#include "i915_utils.h" #include "intel_de.h" #include "intel_display_types.h" #include "intel_dsi.h" @@ -142,11 +142,9 @@ static int vlv_dsi_pclk(struct intel_encoder *encoder, pll_div &= DSI_PLL_M1_DIV_MASK; pll_div = pll_div >> DSI_PLL_M1_DIV_SHIFT; - while (pll_ctl) { - pll_ctl = pll_ctl >> 1; - p++; - } - p--; + p = fls(pll_ctl); + if (p) + p--; if (!p) { drm_err(display->drm, "wrong P1 divisor\n"); @@ -216,6 +214,8 @@ void vlv_dsi_pll_enable(struct intel_encoder *encoder, const struct intel_crtc_state *config) { struct intel_display *display = to_intel_display(encoder); + u32 val; + int ret; drm_dbg_kms(display->drm, "\n"); @@ -233,9 +233,10 @@ void vlv_dsi_pll_enable(struct intel_encoder *encoder, vlv_cck_write(display->drm, CCK_REG_DSI_PLL_CONTROL, config->dsi_pll.ctrl); - if (wait_for(vlv_cck_read(display->drm, CCK_REG_DSI_PLL_CONTROL) & - DSI_PLL_LOCK, 20)) { - + ret = poll_timeout_us(val = vlv_cck_read(display->drm, CCK_REG_DSI_PLL_CONTROL), + val & DSI_PLL_LOCK, + 500, 20 * 1000, false); + if (ret) { vlv_cck_put(display->drm); drm_err(display->drm, "DSI PLL lock failed\n"); return; @@ -262,6 +263,11 @@ void vlv_dsi_pll_disable(struct intel_encoder *encoder) vlv_cck_put(display->drm); } +static bool has_dsic_clock(struct intel_display *display) +{ + return display->platform.broxton; +} + bool bxt_dsi_pll_is_enabled(struct intel_display *display) { bool enabled; @@ -284,7 +290,7 @@ bool bxt_dsi_pll_is_enabled(struct intel_display *display) * causes a system hang. */ val = intel_de_read(display, BXT_DSI_PLL_CTL); - if (display->platform.geminilake) { + if (!has_dsic_clock(display)) { if (!(val & BXT_DSIA_16X_MASK)) { drm_dbg_kms(display->drm, "Invalid PLL divider (%08x)\n", val); @@ -358,6 +364,8 @@ u32 bxt_dsi_get_pclk(struct intel_encoder *encoder, u32 pclk; config->dsi_pll.ctrl = intel_de_read(display, BXT_DSI_PLL_CTL); + if (!has_dsic_clock(display)) + config->dsi_pll.ctrl &= ~BXT_DSIC_16X_MASK; pclk = bxt_dsi_pclk(encoder, config); @@ -514,7 +522,9 @@ int bxt_dsi_pll_compute(struct intel_encoder *encoder, * Spec says both have to be programmed, even if one is not getting * used. Configure MIPI_CLOCK_CTL dividers in modeset */ - config->dsi_pll.ctrl = dsi_ratio | BXT_DSIA_16X_BY2 | BXT_DSIC_16X_BY2; + config->dsi_pll.ctrl = dsi_ratio | BXT_DSIA_16X_BY2; + if (has_dsic_clock(display)) + config->dsi_pll.ctrl |= BXT_DSIC_16X_BY2; /* As per recommendation from hardware team, * Prog PVD ratio =1 if dsi ratio <= 50 diff --git a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c index ca7e9216934a..f243f8a5215d 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c @@ -1382,8 +1382,9 @@ static void clflush_write32(u32 *addr, u32 value, unsigned int flushes) */ if (flushes & CLFLUSH_AFTER) drm_clflush_virt_range(addr, sizeof(*addr)); - } else + } else { *addr = value; + } } static u64 @@ -1567,36 +1568,36 @@ static int eb_relocate_vma(struct i915_execbuffer *eb, struct eb_vma *ev) do { u64 offset = eb_relocate_entry(eb, ev, r); - if (likely(offset == 0)) { - } else if ((s64)offset < 0) { + if (likely(offset == 0)) + continue; + + if ((s64)offset < 0) { remain = (int)offset; goto out; - } else { - /* - * Note that reporting an error now - * leaves everything in an inconsistent - * state as we have *already* changed - * the relocation value inside the - * object. As we have not changed the - * reloc.presumed_offset or will not - * change the execobject.offset, on the - * call we may not rewrite the value - * inside the object, leaving it - * dangling and causing a GPU hang. Unless - * userspace dynamically rebuilds the - * relocations on each execbuf rather than - * presume a static tree. - * - * We did previously check if the relocations - * were writable (access_ok), an error now - * would be a strange race with mprotect, - * having already demonstrated that we - * can read from this userspace address. - */ - offset = gen8_canonical_addr(offset & ~UPDATE); - __put_user(offset, - &urelocs[r - stack].presumed_offset); } + /* + * Note that reporting an error now + * leaves everything in an inconsistent + * state as we have *already* changed + * the relocation value inside the + * object. As we have not changed the + * reloc.presumed_offset or will not + * change the execobject.offset, on the + * call we may not rewrite the value + * inside the object, leaving it + * dangling and causing a GPU hang. Unless + * userspace dynamically rebuilds the + * relocations on each execbuf rather than + * presume a static tree. + * + * We did previously check if the relocations + * were writable (access_ok), an error now + * would be a strange race with mprotect, + * having already demonstrated that we + * can read from this userspace address. + */ + offset = gen8_canonical_addr(offset & ~UPDATE); + __put_user(offset, &urelocs[r - stack].presumed_offset); } while (r++, --count); urelocs += ARRAY_SIZE(stack); } while (remain); diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_client_blt.c b/drivers/gpu/drm/i915/gem/selftests/i915_gem_client_blt.c index 86d9d2fcb6a6..e747f5ed195e 100644 --- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_client_blt.c +++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_client_blt.c @@ -110,6 +110,7 @@ struct tiled_blits { static bool fastblit_supports_x_tiling(const struct drm_i915_private *i915) { + struct intel_display *display = i915->display; int gen = GRAPHICS_VER(i915); /* XY_FAST_COPY_BLT does not exist on pre-gen9 platforms */ @@ -121,7 +122,7 @@ static bool fastblit_supports_x_tiling(const struct drm_i915_private *i915) if (GRAPHICS_VER_FULL(i915) < IP_VER(12, 55)) return false; - return HAS_DISPLAY(i915); + return HAS_DISPLAY(display); } static bool fast_blit_ok(const struct blit_buffer *buf) diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c b/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c index 9c3f17e51885..78734c404a6d 100644 --- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c +++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c @@ -1096,32 +1096,20 @@ static int ___igt_mmap_migrate(struct drm_i915_private *i915, unsigned long addr, bool unfaultable) { - struct vm_area_struct *area; - int err = 0, i; + int i; pr_info("igt_mmap(%s, %d) @ %lx\n", obj->mm.region->name, I915_MMAP_TYPE_FIXED, addr); - mmap_read_lock(current->mm); - area = vma_lookup(current->mm, addr); - mmap_read_unlock(current->mm); - if (!area) { - pr_err("%s: Did not create a vm_area_struct for the mmap\n", - obj->mm.region->name); - err = -EINVAL; - goto out_unmap; - } - for (i = 0; i < obj->base.size / sizeof(u32); i++) { u32 __user *ux = u64_to_user_ptr((u64)(addr + i * sizeof(*ux))); u32 x; if (get_user(x, ux)) { - err = -EFAULT; if (!unfaultable) { pr_err("%s: Unable to read from mmap, offset:%zd\n", obj->mm.region->name, i * sizeof(x)); - goto out_unmap; + return -EFAULT; } continue; @@ -1130,37 +1118,29 @@ static int ___igt_mmap_migrate(struct drm_i915_private *i915, if (unfaultable) { pr_err("%s: Faulted unmappable memory\n", obj->mm.region->name); - err = -EINVAL; - goto out_unmap; + return -EINVAL; } if (x != expand32(POISON_INUSE)) { pr_err("%s: Read incorrect value from mmap, offset:%zd, found:%x, expected:%x\n", obj->mm.region->name, i * sizeof(x), x, expand32(POISON_INUSE)); - err = -EINVAL; - goto out_unmap; + return -EINVAL; } x = expand32(POISON_FREE); if (put_user(x, ux)) { pr_err("%s: Unable to write to mmap, offset:%zd\n", obj->mm.region->name, i * sizeof(x)); - err = -EFAULT; - goto out_unmap; + return -EFAULT; } } - if (unfaultable) { - if (err == -EFAULT) - err = 0; - } else { - obj->flags &= ~I915_BO_ALLOC_GPU_ONLY; - err = wc_check(obj); - } -out_unmap: - vm_munmap(addr, obj->base.size); - return err; + if (unfaultable) + return 0; + + obj->flags &= ~I915_BO_ALLOC_GPU_ONLY; + return wc_check(obj); } #define IGT_MMAP_MIGRATE_TOPDOWN (1 << 0) @@ -1176,6 +1156,7 @@ static int __igt_mmap_migrate(struct intel_memory_region **placements, struct drm_i915_private *i915 = placements[0]->i915; struct drm_i915_gem_object *obj; struct i915_request *rq = NULL; + struct vm_area_struct *area; unsigned long addr; LIST_HEAD(objects); u64 offset; @@ -1207,20 +1188,30 @@ static int __igt_mmap_migrate(struct intel_memory_region **placements, goto out_put; } + mmap_read_lock(current->mm); + area = vma_lookup(current->mm, addr); + mmap_read_unlock(current->mm); + if (!area) { + pr_err("%s: Did not create a vm_area_struct for the mmap\n", + obj->mm.region->name); + err = -EINVAL; + goto out_addr; + } + if (flags & IGT_MMAP_MIGRATE_FILL) { err = igt_fill_mappable(placements[0], &objects); if (err) - goto out_put; + goto out_addr; } err = i915_gem_object_lock(obj, NULL); if (err) - goto out_put; + goto out_addr; err = i915_gem_object_pin_pages(obj); if (err) { i915_gem_object_unlock(obj); - goto out_put; + goto out_addr; } err = intel_context_migrate_clear(to_gt(i915)->migrate.context, NULL, @@ -1228,7 +1219,7 @@ static int __igt_mmap_migrate(struct intel_memory_region **placements, i915_gem_object_is_lmem(obj), expand32(POISON_INUSE), &rq); i915_gem_object_unpin_pages(obj); - if (rq) { + if (rq && !err) { err = dma_resv_reserve_fences(obj->base.resv, 1); if (!err) dma_resv_add_fence(obj->base.resv, &rq->fence, @@ -1237,7 +1228,7 @@ static int __igt_mmap_migrate(struct intel_memory_region **placements, } i915_gem_object_unlock(obj); if (err) - goto out_put; + goto out_addr; if (flags & IGT_MMAP_MIGRATE_EVICTABLE) igt_make_evictable(&objects); @@ -1245,16 +1236,16 @@ static int __igt_mmap_migrate(struct intel_memory_region **placements, if (flags & IGT_MMAP_MIGRATE_FAIL_GPU) { err = i915_gem_object_lock(obj, NULL); if (err) - goto out_put; + goto out_addr; /* - * Ensure we only simulate the gpu failuire when faulting the + * Ensure we only simulate the gpu failure when faulting the * pages. */ err = i915_gem_object_wait_moving_fence(obj, true); i915_gem_object_unlock(obj); if (err) - goto out_put; + goto out_addr; i915_ttm_migrate_set_failure_modes(true, false); } @@ -1298,6 +1289,9 @@ static int __igt_mmap_migrate(struct intel_memory_region **placements, } } +out_addr: + vm_munmap(addr, obj->base.size); + out_put: i915_gem_object_put(obj); igt_close_objects(i915, &objects); diff --git a/drivers/gpu/drm/i915/gt/intel_gt_clock_utils.c b/drivers/gpu/drm/i915/gt/intel_gt_clock_utils.c index 6c499692d61e..88b147fa5cb1 100644 --- a/drivers/gpu/drm/i915/gt/intel_gt_clock_utils.c +++ b/drivers/gpu/drm/i915/gt/intel_gt_clock_utils.c @@ -148,7 +148,7 @@ static u32 gen4_read_clock_frequency(struct intel_uncore *uncore) * * Testing on actual hardware has shown there is no /16. */ - return DIV_ROUND_CLOSEST(i9xx_fsb_freq(uncore->i915), 4) * 1000; + return DIV_ROUND_CLOSEST(intel_fsb_freq(uncore->i915), 4) * 1000; } static u32 read_clock_frequency(struct intel_uncore *uncore) diff --git a/drivers/gpu/drm/i915/gt/intel_gt_debugfs.c b/drivers/gpu/drm/i915/gt/intel_gt_debugfs.c index 4dc23b8d3aa2..dcd40b30a96b 100644 --- a/drivers/gpu/drm/i915/gt/intel_gt_debugfs.c +++ b/drivers/gpu/drm/i915/gt/intel_gt_debugfs.c @@ -82,14 +82,15 @@ static void gt_debugfs_register(struct intel_gt *gt, struct dentry *root) void intel_gt_debugfs_register(struct intel_gt *gt) { + struct dentry *debugfs_root = gt->i915->drm.debugfs_root; struct dentry *root; char gtname[4]; - if (!gt->i915->drm.primary->debugfs_root) + if (!debugfs_root) return; snprintf(gtname, sizeof(gtname), "gt%u", gt->info.id); - root = debugfs_create_dir(gtname, gt->i915->drm.primary->debugfs_root); + root = debugfs_create_dir(gtname, debugfs_root); if (IS_ERR(root)) return; diff --git a/drivers/gpu/drm/i915/gt/intel_rps.c b/drivers/gpu/drm/i915/gt/intel_rps.c index 0b35fdd461d4..006042e0b229 100644 --- a/drivers/gpu/drm/i915/gt/intel_rps.c +++ b/drivers/gpu/drm/i915/gt/intel_rps.c @@ -9,6 +9,7 @@ #include "display/intel_display.h" #include "display/intel_display_rps.h" +#include "soc/intel_dram.h" #include "i915_drv.h" #include "i915_irq.h" #include "i915_reg.h" @@ -276,20 +277,24 @@ static void gen5_rps_init(struct intel_rps *rps) { struct drm_i915_private *i915 = rps_to_i915(rps); struct intel_uncore *uncore = rps_to_uncore(rps); + unsigned int fsb_freq, mem_freq; u8 fmax, fmin, fstart; u32 rgvmodectl; int c_m, i; - if (i915->fsb_freq <= 3200000) + fsb_freq = intel_fsb_freq(i915); + mem_freq = intel_mem_freq(i915); + + if (fsb_freq <= 3200000) c_m = 0; - else if (i915->fsb_freq <= 4800000) + else if (fsb_freq <= 4800000) c_m = 1; else c_m = 2; for (i = 0; i < ARRAY_SIZE(cparams); i++) { if (cparams[i].i == c_m && - cparams[i].t == DIV_ROUND_CLOSEST(i915->mem_freq, 1000)) { + cparams[i].t == DIV_ROUND_CLOSEST(mem_freq, 1000)) { rps->ips.m = cparams[i].m; rps->ips.c = cparams[i].c; break; diff --git a/drivers/gpu/drm/i915/gt/intel_workarounds.c b/drivers/gpu/drm/i915/gt/intel_workarounds.c index b37e400f74e5..7d486dfa2fc1 100644 --- a/drivers/gpu/drm/i915/gt/intel_workarounds.c +++ b/drivers/gpu/drm/i915/gt/intel_workarounds.c @@ -337,12 +337,26 @@ static void gen6_ctx_workarounds_init(struct intel_engine_cs *engine, struct i915_wa_list *wal) { wa_masked_en(wal, INSTPM, INSTPM_FORCE_ORDERING); + + /* WaDisable_RenderCache_OperationalFlush:snb */ + wa_masked_dis(wal, CACHE_MODE_0, RC_OP_FLUSH_ENABLE); } static void gen7_ctx_workarounds_init(struct intel_engine_cs *engine, struct i915_wa_list *wal) { wa_masked_en(wal, INSTPM, INSTPM_FORCE_ORDERING); + /* WaDisable_RenderCache_OperationalFlush:ivb,vlv,hsw */ + wa_masked_dis(wal, CACHE_MODE_0_GEN7, RC_OP_FLUSH_ENABLE); + + /* + * BSpec says this must be set, even though + * WaDisable4x2SubspanOptimization:ivb,hsw + * WaDisable4x2SubspanOptimization isn't listed for VLV. + */ + wa_masked_en(wal, + CACHE_MODE_1, + PIXEL_SUBSPAN_COLLECT_OPT_DISABLE); } static void gen8_ctx_workarounds_init(struct intel_engine_cs *engine, @@ -634,6 +648,8 @@ static void cfl_ctx_workarounds_init(struct intel_engine_cs *engine, static void icl_ctx_workarounds_init(struct intel_engine_cs *engine, struct i915_wa_list *wal) { + struct drm_i915_private *i915 = engine->i915; + /* Wa_1406697149 (WaDisableBankHangMode:icl) */ wa_write(wal, GEN8_L3CNTLREG, GEN8_ERRDETBCTRL); @@ -669,6 +685,15 @@ static void icl_ctx_workarounds_init(struct intel_engine_cs *engine, /* Wa_1406306137:icl,ehl */ wa_mcr_masked_en(wal, GEN9_ROW_CHICKEN4, GEN11_DIS_PICK_2ND_EU); + + if (IS_JASPERLAKE(i915) || IS_ELKHARTLAKE(i915)) { + /* + * Disable Repacking for Compression (masked R/W access) + * before rendering compressed surfaces for display. + */ + wa_masked_en(wal, CACHE_MODE_0_GEN7, + DISABLE_REPACKING_FOR_COMPRESSION); + } } /* @@ -2306,15 +2331,6 @@ rcs_engine_wa_init(struct intel_engine_cs *engine, struct i915_wa_list *wal) GEN8_RC_SEMA_IDLE_MSG_DISABLE); } - if (IS_JASPERLAKE(i915) || IS_ELKHARTLAKE(i915)) { - /* - * "Disable Repacking for Compression (masked R/W access) - * before rendering compressed surfaces for display." - */ - wa_masked_en(wal, CACHE_MODE_0_GEN7, - DISABLE_REPACKING_FOR_COMPRESSION); - } - if (GRAPHICS_VER(i915) == 11) { /* This is not an Wa. Enable for better image quality */ wa_masked_en(wal, @@ -2565,18 +2581,6 @@ rcs_engine_wa_init(struct intel_engine_cs *engine, struct i915_wa_list *wal) RING_MODE_GEN7(RENDER_RING_BASE), GFX_TLB_INVALIDATE_EXPLICIT | GFX_REPLAY_MODE); - /* WaDisable_RenderCache_OperationalFlush:ivb,vlv,hsw */ - wa_masked_dis(wal, CACHE_MODE_0_GEN7, RC_OP_FLUSH_ENABLE); - - /* - * BSpec says this must be set, even though - * WaDisable4x2SubspanOptimization:ivb,hsw - * WaDisable4x2SubspanOptimization isn't listed for VLV. - */ - wa_masked_en(wal, - CACHE_MODE_1, - PIXEL_SUBSPAN_COLLECT_OPT_DISABLE); - /* * BSpec recommends 8x4 when MSAA is used, * however in practice 16x4 seems fastest. @@ -2643,9 +2647,6 @@ rcs_engine_wa_init(struct intel_engine_cs *engine, struct i915_wa_list *wal) GEN6_WIZ_HASHING_MASK, GEN6_WIZ_HASHING_16x4); - /* WaDisable_RenderCache_OperationalFlush:snb */ - wa_masked_dis(wal, CACHE_MODE_0, RC_OP_FLUSH_ENABLE); - /* * From the Sandybridge PRM, volume 1 part 3, page 24: * "If this bit is set, STCunit will have LRA as replacement diff --git a/drivers/gpu/drm/i915/gt/selftest_hangcheck.c b/drivers/gpu/drm/i915/gt/selftest_hangcheck.c index f057c16410e7..619c70c54ef9 100644 --- a/drivers/gpu/drm/i915/gt/selftest_hangcheck.c +++ b/drivers/gpu/drm/i915/gt/selftest_hangcheck.c @@ -904,6 +904,8 @@ static void active_engine(struct kthread_work *work) arg->result = PTR_ERR(ce[count]); pr_err("[%s] Create context #%ld failed: %d!\n", engine->name, count, arg->result); + if (!count) + return; while (--count) intel_context_put(ce[count]); return; diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c index 0d5197c0824a..380a11c92d63 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c @@ -13,7 +13,7 @@ #include "intel_guc_ct.h" #include "intel_guc_print.h" -#if IS_ENABLED(CONFIG_DRM_I915_DEBUG_GUC) +#if IS_ENABLED(CONFIG_DRM_I915_DEBUG) enum { CT_DEAD_ALIVE = 0, CT_DEAD_SETUP, @@ -144,7 +144,7 @@ void intel_guc_ct_init_early(struct intel_guc_ct *ct) spin_lock_init(&ct->requests.lock); INIT_LIST_HEAD(&ct->requests.pending); INIT_LIST_HEAD(&ct->requests.incoming); -#if IS_ENABLED(CONFIG_DRM_I915_DEBUG_GUC) +#if IS_ENABLED(CONFIG_DRM_I915_DEBUG) INIT_WORK(&ct->dead_ct_worker, ct_dead_ct_worker_func); #endif INIT_WORK(&ct->requests.worker, ct_incoming_request_worker_func); @@ -373,7 +373,7 @@ int intel_guc_ct_enable(struct intel_guc_ct *ct) ct->enabled = true; ct->stall_time = KTIME_MAX; -#if IS_ENABLED(CONFIG_DRM_I915_DEBUG_GUC) +#if IS_ENABLED(CONFIG_DRM_I915_DEBUG) ct->dead_ct_reported = false; ct->dead_ct_reason = CT_DEAD_ALIVE; #endif @@ -1377,7 +1377,7 @@ void intel_guc_ct_print_info(struct intel_guc_ct *ct, ct->ctbs.recv.desc->tail); } -#if IS_ENABLED(CONFIG_DRM_I915_DEBUG_GUC) +#if IS_ENABLED(CONFIG_DRM_I915_DEBUG) static void ct_dead_ct_worker_func(struct work_struct *w) { struct intel_guc_ct *ct = container_of(w, struct intel_guc_ct, dead_ct_worker); @@ -1386,6 +1386,9 @@ static void ct_dead_ct_worker_func(struct work_struct *w) if (ct->dead_ct_reported) return; + if (i915_error_injected()) + return; + ct->dead_ct_reported = true; guc_info(guc, "CTB is dead - reason=0x%X\n", ct->dead_ct_reason); diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.h b/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.h index 2c4bb9a941be..e9a6ec4e6d38 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.h +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.h @@ -97,7 +97,7 @@ struct intel_guc_ct { /** @stall_time: time of first time a CTB submission is stalled */ ktime_t stall_time; -#if IS_ENABLED(CONFIG_DRM_I915_DEBUG_GUC) +#if IS_ENABLED(CONFIG_DRM_I915_DEBUG) int dead_ct_reason; bool dead_ct_reported; struct work_struct dead_ct_worker; diff --git a/drivers/gpu/drm/i915/gvt/debugfs.c b/drivers/gpu/drm/i915/gvt/debugfs.c index 673534f061ef..415422b5943c 100644 --- a/drivers/gpu/drm/i915/gvt/debugfs.c +++ b/drivers/gpu/drm/i915/gvt/debugfs.c @@ -194,9 +194,9 @@ void intel_gvt_debugfs_add_vgpu(struct intel_vgpu *vgpu) void intel_gvt_debugfs_remove_vgpu(struct intel_vgpu *vgpu) { struct intel_gvt *gvt = vgpu->gvt; - struct drm_minor *minor = gvt->gt->i915->drm.primary; + struct dentry *debugfs_root = gvt->gt->i915->drm.debugfs_root; - if (minor->debugfs_root && gvt->debugfs_root) { + if (debugfs_root && gvt->debugfs_root) { debugfs_remove_recursive(vgpu->debugfs); vgpu->debugfs = NULL; } @@ -208,9 +208,9 @@ void intel_gvt_debugfs_remove_vgpu(struct intel_vgpu *vgpu) */ void intel_gvt_debugfs_init(struct intel_gvt *gvt) { - struct drm_minor *minor = gvt->gt->i915->drm.primary; + struct dentry *debugfs_root = gvt->gt->i915->drm.debugfs_root; - gvt->debugfs_root = debugfs_create_dir("gvt", minor->debugfs_root); + gvt->debugfs_root = debugfs_create_dir("gvt", debugfs_root); debugfs_create_ulong("num_tracked_mmio", 0444, gvt->debugfs_root, &gvt->mmio.num_tracked_mmio); @@ -222,9 +222,9 @@ void intel_gvt_debugfs_init(struct intel_gvt *gvt) */ void intel_gvt_debugfs_clean(struct intel_gvt *gvt) { - struct drm_minor *minor = gvt->gt->i915->drm.primary; + struct dentry *debugfs_root = gvt->gt->i915->drm.debugfs_root; - if (minor->debugfs_root) { + if (debugfs_root) { debugfs_remove_recursive(gvt->debugfs_root); gvt->debugfs_root = NULL; } diff --git a/drivers/gpu/drm/i915/i915_active.c b/drivers/gpu/drm/i915/i915_active.c index 0dbc4e289300..6b0c1162505a 100644 --- a/drivers/gpu/drm/i915/i915_active.c +++ b/drivers/gpu/drm/i915/i915_active.c @@ -257,10 +257,9 @@ static struct active_node *__active_lookup(struct i915_active *ref, u64 idx) * claimed the cache and we know that is does not match our * idx. If, and only if, the timeline is currently zero is it * worth competing to claim it atomically for ourselves (for - * only the winner of that race will cmpxchg return the old - * value of 0). + * only the winner of that race will cmpxchg succeed). */ - if (!cached && !cmpxchg64(&it->timeline, 0, idx)) + if (!cached && try_cmpxchg64(&it->timeline, &cached, idx)) return it; } diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c index 967c0501e91e..23fa098c4479 100644 --- a/drivers/gpu/drm/i915/i915_debugfs.c +++ b/drivers/gpu/drm/i915/i915_debugfs.c @@ -720,26 +720,24 @@ static const struct i915_debugfs_files { {"i915_gem_drop_caches", &i915_drop_caches_fops}, }; -void i915_debugfs_register(struct drm_i915_private *dev_priv) +void i915_debugfs_register(struct drm_i915_private *i915) { - struct drm_minor *minor = dev_priv->drm.primary; + struct dentry *debugfs_root = i915->drm.debugfs_root; int i; - i915_debugfs_params(dev_priv); + i915_debugfs_params(i915); - debugfs_create_file("i915_forcewake_user", S_IRUSR, minor->debugfs_root, - to_i915(minor->dev), &i915_forcewake_fops); + debugfs_create_file("i915_forcewake_user", S_IRUSR, debugfs_root, + i915, &i915_forcewake_fops); for (i = 0; i < ARRAY_SIZE(i915_debugfs_files); i++) { - debugfs_create_file(i915_debugfs_files[i].name, - S_IRUGO | S_IWUSR, - minor->debugfs_root, - to_i915(minor->dev), + debugfs_create_file(i915_debugfs_files[i].name, S_IRUGO | S_IWUSR, + debugfs_root, i915, i915_debugfs_files[i].fops); } drm_debugfs_create_files(i915_debugfs_list, ARRAY_SIZE(i915_debugfs_list), - minor->debugfs_root, minor); + debugfs_root, i915->drm.primary); - i915_gpu_error_debugfs_register(dev_priv); + i915_gpu_error_debugfs_register(i915); } diff --git a/drivers/gpu/drm/i915/i915_debugfs_params.c b/drivers/gpu/drm/i915/i915_debugfs_params.c index 33d2dcb0de65..89ab5eb14779 100644 --- a/drivers/gpu/drm/i915/i915_debugfs_params.c +++ b/drivers/gpu/drm/i915/i915_debugfs_params.c @@ -248,11 +248,11 @@ i915_debugfs_create_charp(const char *name, umode_t mode, /* add a subdirectory with files for each i915 param */ struct dentry *i915_debugfs_params(struct drm_i915_private *i915) { - struct drm_minor *minor = i915->drm.primary; + struct dentry *debugfs_root = i915->drm.debugfs_root; struct i915_params *params = &i915->params; struct dentry *dir; - dir = debugfs_create_dir("i915_params", minor->debugfs_root); + dir = debugfs_create_dir("i915_params", debugfs_root); if (IS_ERR(dir)) return dir; diff --git a/drivers/gpu/drm/i915/i915_driver.c b/drivers/gpu/drm/i915/i915_driver.c index c6263c6d3384..70f042ce8705 100644 --- a/drivers/gpu/drm/i915/i915_driver.c +++ b/drivers/gpu/drm/i915/i915_driver.c @@ -977,7 +977,7 @@ void i915_driver_shutdown(struct drm_i915_private *i915) intel_power_domains_disable(display); drm_client_dev_suspend(&i915->drm, false); - if (HAS_DISPLAY(i915)) { + if (HAS_DISPLAY(display)) { drm_kms_helper_poll_disable(&i915->drm); intel_display_driver_disable_user_access(display); @@ -989,7 +989,7 @@ void i915_driver_shutdown(struct drm_i915_private *i915) intel_irq_suspend(i915); intel_hpd_cancel_work(display); - if (HAS_DISPLAY(i915)) + if (HAS_DISPLAY(display)) intel_display_driver_suspend_access(display); intel_encoder_suspend_all(display); @@ -1060,7 +1060,7 @@ static int i915_drm_suspend(struct drm_device *dev) * properly. */ intel_power_domains_disable(display); drm_client_dev_suspend(dev, false); - if (HAS_DISPLAY(dev_priv)) { + if (HAS_DISPLAY(display)) { drm_kms_helper_poll_disable(dev); intel_display_driver_disable_user_access(display); } @@ -1072,7 +1072,7 @@ static int i915_drm_suspend(struct drm_device *dev) intel_irq_suspend(dev_priv); intel_hpd_cancel_work(display); - if (HAS_DISPLAY(dev_priv)) + if (HAS_DISPLAY(display)) intel_display_driver_suspend_access(display); intel_encoder_suspend_all(display); @@ -1219,7 +1219,7 @@ static int i915_drm_resume(struct drm_device *dev) */ intel_irq_resume(dev_priv); - if (HAS_DISPLAY(dev_priv)) + if (HAS_DISPLAY(display)) drm_mode_config_reset(dev); i915_gem_resume(dev_priv); @@ -1228,14 +1228,14 @@ static int i915_drm_resume(struct drm_device *dev) intel_clock_gating_init(dev_priv); - if (HAS_DISPLAY(dev_priv)) + if (HAS_DISPLAY(display)) intel_display_driver_resume_access(display); intel_hpd_init(display); intel_display_driver_resume(display); - if (HAS_DISPLAY(dev_priv)) { + if (HAS_DISPLAY(display)) { intel_display_driver_enable_user_access(display); drm_kms_helper_poll_enable(dev); } diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index 4e4e89746aa6..2f3965feada1 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -237,8 +237,6 @@ struct drm_i915_private { bool preserve_bios_swizzle; - unsigned int fsb_freq, mem_freq, is_ddr3; - unsigned int hpll_freq; unsigned int czclk_freq; diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c index 0e4b832dff84..89f7c04a3330 100644 --- a/drivers/gpu/drm/i915/i915_gpu_error.c +++ b/drivers/gpu/drm/i915/i915_gpu_error.c @@ -2445,11 +2445,11 @@ static const struct file_operations i915_error_state_fops = { void i915_gpu_error_debugfs_register(struct drm_i915_private *i915) { - struct drm_minor *minor = i915->drm.primary; + struct dentry *debugfs_root = i915->drm.debugfs_root; - debugfs_create_file("i915_error_state", 0644, minor->debugfs_root, i915, + debugfs_create_file("i915_error_state", 0644, debugfs_root, i915, &i915_error_state_fops); - debugfs_create_file("i915_gpu_info", 0644, minor->debugfs_root, i915, + debugfs_create_file("i915_gpu_info", 0644, debugfs_root, i915, &i915_gpu_info_fops); } diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c index 191ed8bb1d9c..a5fa40ab5de2 100644 --- a/drivers/gpu/drm/i915/i915_irq.c +++ b/drivers/gpu/drm/i915/i915_irq.c @@ -439,7 +439,7 @@ static irqreturn_t ilk_irq_handler(int irq, void *arg) * able to process them after we restore SDEIER (as soon as we restore * it, we'll get an interrupt if SDEIIR still has something to process * due to its back queue). */ - if (!HAS_PCH_NOP(i915)) { + if (!HAS_PCH_NOP(display)) { sde_ier = raw_reg_read(regs, SDEIER); raw_reg_write(regs, SDEIER, 0); } @@ -459,7 +459,7 @@ static irqreturn_t ilk_irq_handler(int irq, void *arg) de_iir = raw_reg_read(regs, DEIIR); if (de_iir) { raw_reg_write(regs, DEIIR, de_iir); - if (DISPLAY_VER(i915) >= 7) + if (DISPLAY_VER(display) >= 7) ivb_display_irq_handler(display, de_iir); else ilk_display_irq_handler(display, de_iir); @@ -834,6 +834,7 @@ static void cherryview_irq_postinstall(struct drm_i915_private *dev_priv) static u32 i9xx_error_mask(struct drm_i915_private *i915) { + struct intel_display *display = i915->display; /* * On gen2/3 FBC generates (seemingly spurious) * display INVALID_GTT/INVALID_GTT_PTE table errors. @@ -846,7 +847,7 @@ static u32 i9xx_error_mask(struct drm_i915_private *i915) * Unfortunately we can't mask off individual PGTBL_ER bits, * so we just have to mask off all page table errors via EMR. */ - if (HAS_FBC(i915)) + if (HAS_FBC(display)) return I915_ERROR_MEMORY_REFRESH; else return I915_ERROR_PAGE_TABLE | @@ -924,12 +925,12 @@ static void i915_irq_postinstall(struct drm_i915_private *dev_priv) I915_MASTER_ERROR_INTERRUPT | I915_USER_INTERRUPT; - if (DISPLAY_VER(dev_priv) >= 3) { + if (DISPLAY_VER(display) >= 3) { dev_priv->irq_mask &= ~I915_ASLE_INTERRUPT; enable_mask |= I915_ASLE_INTERRUPT; } - if (HAS_HOTPLUG(dev_priv)) { + if (HAS_HOTPLUG(display)) { dev_priv->irq_mask &= ~I915_DISPLAY_PORT_INTERRUPT; enable_mask |= I915_DISPLAY_PORT_INTERRUPT; } @@ -963,7 +964,7 @@ static irqreturn_t i915_irq_handler(int irq, void *arg) ret = IRQ_HANDLED; - if (HAS_HOTPLUG(dev_priv) && + if (HAS_HOTPLUG(display) && iir & I915_DISPLAY_PORT_INTERRUPT) hotplug_status = i9xx_hpd_irq_ack(display); diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h index 03b895897f60..354ef75ef6a5 100644 --- a/drivers/gpu/drm/i915/i915_reg.h +++ b/drivers/gpu/drm/i915/i915_reg.h @@ -412,9 +412,9 @@ #define FW_BLC _MMIO(0x20d8) #define FW_BLC2 _MMIO(0x20dc) #define FW_BLC_SELF _MMIO(0x20e0) /* 915+ only */ -#define FW_BLC_SELF_EN_MASK (1 << 31) -#define FW_BLC_SELF_FIFO_MASK (1 << 16) /* 945 only */ -#define FW_BLC_SELF_EN (1 << 15) /* 945 only */ +#define FW_BLC_SELF_EN_MASK REG_BIT(31) +#define FW_BLC_SELF_FIFO_MASK REG_BIT(16) /* 945 only */ +#define FW_BLC_SELF_EN REG_BIT(15) /* 945 only */ #define MM_BURST_LENGTH 0x00700000 #define MM_FIFO_WATERMARK 0x0001F000 #define LM_BURST_LENGTH 0x00000700 @@ -613,7 +613,8 @@ #define DSTATE_GFX_CLOCK_GATING (1 << 1) #define DSTATE_DOT_CLOCK_GATING (1 << 0) -#define DSPCLK_GATE_D(__i915) _MMIO(DISPLAY_MMIO_BASE(__i915) + 0x6200) +#define DSPCLK_GATE_D _MMIO(0x6200) +#define VLV_DSPCLK_GATE_D _MMIO(VLV_DISPLAY_BASE + 0x6200) # define DPUNIT_B_CLOCK_GATE_DISABLE (1 << 30) /* 965 */ # define VSUNIT_CLOCK_GATE_DISABLE (1 << 29) /* 965 */ # define VRHUNIT_CLOCK_GATE_DISABLE (1 << 28) /* 965 */ diff --git a/drivers/gpu/drm/i915/i915_switcheroo.c b/drivers/gpu/drm/i915/i915_switcheroo.c index 4c02a04be681..3a95a55b2e87 100644 --- a/drivers/gpu/drm/i915/i915_switcheroo.c +++ b/drivers/gpu/drm/i915/i915_switcheroo.c @@ -15,13 +15,14 @@ static void i915_switcheroo_set_state(struct pci_dev *pdev, enum vga_switcheroo_state state) { struct drm_i915_private *i915 = pdev_to_i915(pdev); + struct intel_display *display = i915 ? i915->display : NULL; pm_message_t pmm = { .event = PM_EVENT_SUSPEND }; if (!i915) { dev_err(&pdev->dev, "DRM not initialized, aborting switch.\n"); return; } - if (!HAS_DISPLAY(i915)) { + if (!HAS_DISPLAY(display)) { dev_err(&pdev->dev, "Device state not initialized, aborting switch.\n"); return; } @@ -44,13 +45,14 @@ static void i915_switcheroo_set_state(struct pci_dev *pdev, static bool i915_switcheroo_can_switch(struct pci_dev *pdev) { struct drm_i915_private *i915 = pdev_to_i915(pdev); + struct intel_display *display = i915 ? i915->display : NULL; /* * FIXME: open_count is protected by drm_global_mutex but that would lead to * locking inversion with the driver load path. And the access here is * completely racy anyway. So don't bother with locking for now. */ - return i915 && HAS_DISPLAY(i915) && atomic_read(&i915->drm.open_count) == 0; + return i915 && HAS_DISPLAY(display) && atomic_read(&i915->drm.open_count) == 0; } static const struct vga_switcheroo_client_ops i915_switcheroo_ops = { diff --git a/drivers/gpu/drm/i915/i915_utils.h b/drivers/gpu/drm/i915/i915_utils.h index f7fb40cfdb70..9cb40c2c4b12 100644 --- a/drivers/gpu/drm/i915/i915_utils.h +++ b/drivers/gpu/drm/i915/i915_utils.h @@ -267,8 +267,13 @@ wait_remaining_ms_from_jiffies(unsigned long timestamp_jiffies, int to_wait_ms) (Wmax)) #define wait_for(COND, MS) _wait_for((COND), (MS) * 1000, 10, 1000) -/* If CONFIG_PREEMPT_COUNT is disabled, in_atomic() always reports false. */ -#if IS_ENABLED(CONFIG_DRM_I915_DEBUG) && IS_ENABLED(CONFIG_PREEMPT_COUNT) +/* + * If CONFIG_PREEMPT_COUNT is disabled, in_atomic() always reports false. + * On PREEMPT_RT the context isn't becoming atomic because it is used in an + * interrupt handler or because a spinlock_t is acquired. This leads to + * warnings which don't occur otherwise and therefore the check is disabled. + */ +#if IS_ENABLED(CONFIG_DRM_I915_DEBUG) && IS_ENABLED(CONFIG_PREEMPT_COUNT) && !defined(CONFIG_PREEMPT_RT) # define _WAIT_FOR_ATOMIC_CHECK(ATOMIC) WARN_ON_ONCE((ATOMIC) && !in_atomic()) #else # define _WAIT_FOR_ATOMIC_CHECK(ATOMIC) do { } while (0) diff --git a/drivers/gpu/drm/i915/intel_clock_gating.c b/drivers/gpu/drm/i915/intel_clock_gating.c index f86a3629ae9e..467740969431 100644 --- a/drivers/gpu/drm/i915/intel_clock_gating.c +++ b/drivers/gpu/drm/i915/intel_clock_gating.c @@ -132,16 +132,17 @@ static void ibx_init_clock_gating(struct drm_i915_private *i915) static void g4x_disable_trickle_feed(struct drm_i915_private *dev_priv) { + struct intel_display *display = dev_priv->display; enum pipe pipe; - for_each_pipe(dev_priv, pipe) { - intel_uncore_rmw(&dev_priv->uncore, DSPCNTR(dev_priv, pipe), + for_each_pipe(display, pipe) { + intel_uncore_rmw(&dev_priv->uncore, DSPCNTR(display, pipe), 0, DISP_TRICKLE_FEED_DISABLE); - intel_uncore_rmw(&dev_priv->uncore, DSPSURF(dev_priv, pipe), + intel_uncore_rmw(&dev_priv->uncore, DSPSURF(display, pipe), 0, 0); intel_uncore_posting_read(&dev_priv->uncore, - DSPSURF(dev_priv, pipe)); + DSPSURF(display, pipe)); } } @@ -218,7 +219,7 @@ static void cpt_init_clock_gating(struct drm_i915_private *i915) /* The below fixes the weird display corruption, a few pixels shifted * downward, on (only) LVDS of some HP laptops with IVY. */ - for_each_pipe(i915, pipe) { + for_each_pipe(display, pipe) { val = intel_uncore_read(&i915->uncore, TRANS_CHICKEN2(pipe)); val |= TRANS_CHICKEN2_TIMING_OVERRIDE; val &= ~TRANS_CHICKEN2_FDI_POLARITY_REVERSED; @@ -229,7 +230,7 @@ static void cpt_init_clock_gating(struct drm_i915_private *i915) intel_uncore_write(&i915->uncore, TRANS_CHICKEN2(pipe), val); } /* WADP0ClockGatingDisable */ - for_each_pipe(i915, pipe) { + for_each_pipe(display, pipe) { intel_uncore_write(&i915->uncore, TRANS_CHICKEN1(pipe), TRANS_CHICKEN1_DP0UNIT_GC_DISABLE); } @@ -307,11 +308,13 @@ static void gen6_init_clock_gating(struct drm_i915_private *i915) static void lpt_init_clock_gating(struct drm_i915_private *i915) { + struct intel_display *display = i915->display; + /* * TODO: this bit should only be enabled when really needed, then * disabled when not needed anymore in order to save power. */ - if (HAS_PCH_LPT_LP(i915)) + if (HAS_PCH_LPT_LP(display)) intel_uncore_rmw(&i915->uncore, SOUTH_DSPCLK_GATE_D, 0, PCH_LP_PARTITION_LEVEL_DISABLE); @@ -355,7 +358,9 @@ static void dg2_init_clock_gating(struct drm_i915_private *i915) static void cnp_init_clock_gating(struct drm_i915_private *i915) { - if (!HAS_PCH_CNP(i915)) + struct intel_display *display = i915->display; + + if (!HAS_PCH_CNP(display)) return; /* Display WA #1181 WaSouthDisplayDisablePWMCGEGating: cnp */ @@ -421,6 +426,7 @@ static void skl_init_clock_gating(struct drm_i915_private *i915) static void bdw_init_clock_gating(struct drm_i915_private *i915) { + struct intel_display *display = i915->display; enum pipe pipe; /* WaFbcAsynchFlipDisableFbcQueue:hsw,bdw */ @@ -432,7 +438,7 @@ static void bdw_init_clock_gating(struct drm_i915_private *i915) /* WaPsrDPAMaskVBlankInSRD:bdw */ intel_uncore_rmw(&i915->uncore, CHICKEN_PAR1_1, 0, HSW_MASK_VBL_TO_PIPE_IN_SRD); - for_each_pipe(i915, pipe) { + for_each_pipe(display, pipe) { /* WaPsrDPRSUnmaskVBlankInSRD:bdw */ intel_uncore_rmw(&i915->uncore, CHICKEN_PIPESL_1(pipe), 0, BDW_UNMASK_VBL_TO_REGS_IN_SRD); @@ -468,6 +474,7 @@ static void bdw_init_clock_gating(struct drm_i915_private *i915) static void hsw_init_clock_gating(struct drm_i915_private *i915) { + struct intel_display *display = i915->display; enum pipe pipe; /* WaFbcAsynchFlipDisableFbcQueue:hsw,bdw */ @@ -476,7 +483,7 @@ static void hsw_init_clock_gating(struct drm_i915_private *i915) /* WaPsrDPAMaskVBlankInSRD:hsw */ intel_uncore_rmw(&i915->uncore, CHICKEN_PAR1_1, 0, HSW_MASK_VBL_TO_PIPE_IN_SRD); - for_each_pipe(i915, pipe) { + for_each_pipe(display, pipe) { /* WaPsrDPRSUnmaskVBlankInSRD:hsw */ intel_uncore_rmw(&i915->uncore, CHICKEN_PIPESL_1(pipe), 0, HSW_UNMASK_VBL_TO_REGS_IN_SRD); @@ -494,6 +501,8 @@ static void hsw_init_clock_gating(struct drm_i915_private *i915) static void ivb_init_clock_gating(struct drm_i915_private *i915) { + struct intel_display *display = i915->display; + intel_uncore_write(&i915->uncore, ILK_DSPCLK_GATE_D, ILK_VRHUNIT_CLOCK_GATE_DISABLE); /* WaFbcAsynchFlipDisableFbcQueue:ivb */ @@ -531,7 +540,7 @@ static void ivb_init_clock_gating(struct drm_i915_private *i915) intel_uncore_rmw(&i915->uncore, GEN6_MBCUNIT_SNPCR, GEN6_MBC_SNPCR_MASK, GEN6_MBC_SNPCR_MED); - if (!HAS_PCH_NOP(i915)) + if (!HAS_PCH_NOP(display)) cpt_init_clock_gating(i915); gen6_check_mch_setup(i915); @@ -611,7 +620,7 @@ static void g4x_init_clock_gating(struct drm_i915_private *i915) OVCUNIT_CLOCK_GATE_DISABLE; if (IS_GM45(i915)) dspclk_gate |= DSSUNIT_CLOCK_GATE_DISABLE; - intel_uncore_write(&i915->uncore, DSPCLK_GATE_D(i915), dspclk_gate); + intel_uncore_write(&i915->uncore, DSPCLK_GATE_D, dspclk_gate); g4x_disable_trickle_feed(i915); } @@ -622,7 +631,7 @@ static void i965gm_init_clock_gating(struct drm_i915_private *i915) intel_uncore_write(uncore, RENCLK_GATE_D1, I965_RCC_CLOCK_GATE_DISABLE); intel_uncore_write(uncore, RENCLK_GATE_D2, 0); - intel_uncore_write(uncore, DSPCLK_GATE_D(i915), 0); + intel_uncore_write(uncore, DSPCLK_GATE_D, 0); intel_uncore_write(uncore, RAMCLK_GATE_D, 0); intel_uncore_write16(uncore, DEUC, 0); intel_uncore_write(uncore, diff --git a/drivers/gpu/drm/i915/intel_gvt_mmio_table.c b/drivers/gpu/drm/i915/intel_gvt_mmio_table.c index 87ac4446d306..ca57a3dd3148 100644 --- a/drivers/gpu/drm/i915/intel_gvt_mmio_table.c +++ b/drivers/gpu/drm/i915/intel_gvt_mmio_table.c @@ -62,6 +62,7 @@ static int iterate_generic_mmio(struct intel_gvt_mmio_table_iter *iter) { struct drm_i915_private *dev_priv = iter->i915; + struct intel_display *display = dev_priv->display; MMIO_RING_D(RING_IMR); MMIO_D(SDEIMR); @@ -133,38 +134,38 @@ static int iterate_generic_mmio(struct intel_gvt_mmio_table_iter *iter) MMIO_D(_MMIO(0x650b4)); MMIO_D(_MMIO(0xc4040)); MMIO_D(DERRMR); - MMIO_D(PIPEDSL(dev_priv, PIPE_A)); - MMIO_D(PIPEDSL(dev_priv, PIPE_B)); - MMIO_D(PIPEDSL(dev_priv, PIPE_C)); - MMIO_D(PIPEDSL(dev_priv, _PIPE_EDP)); - MMIO_D(TRANSCONF(dev_priv, TRANSCODER_A)); - MMIO_D(TRANSCONF(dev_priv, TRANSCODER_B)); - MMIO_D(TRANSCONF(dev_priv, TRANSCODER_C)); - MMIO_D(TRANSCONF(dev_priv, TRANSCODER_EDP)); - MMIO_D(PIPESTAT(dev_priv, PIPE_A)); - MMIO_D(PIPESTAT(dev_priv, PIPE_B)); - MMIO_D(PIPESTAT(dev_priv, PIPE_C)); - MMIO_D(PIPESTAT(dev_priv, _PIPE_EDP)); - MMIO_D(PIPE_FLIPCOUNT_G4X(dev_priv, PIPE_A)); - MMIO_D(PIPE_FLIPCOUNT_G4X(dev_priv, PIPE_B)); - MMIO_D(PIPE_FLIPCOUNT_G4X(dev_priv, PIPE_C)); - MMIO_D(PIPE_FLIPCOUNT_G4X(dev_priv, _PIPE_EDP)); - MMIO_D(PIPE_FRMCOUNT_G4X(dev_priv, PIPE_A)); - MMIO_D(PIPE_FRMCOUNT_G4X(dev_priv, PIPE_B)); - MMIO_D(PIPE_FRMCOUNT_G4X(dev_priv, PIPE_C)); - MMIO_D(PIPE_FRMCOUNT_G4X(dev_priv, _PIPE_EDP)); - MMIO_D(CURCNTR(dev_priv, PIPE_A)); - MMIO_D(CURCNTR(dev_priv, PIPE_B)); - MMIO_D(CURCNTR(dev_priv, PIPE_C)); - MMIO_D(CURPOS(dev_priv, PIPE_A)); - MMIO_D(CURPOS(dev_priv, PIPE_B)); - MMIO_D(CURPOS(dev_priv, PIPE_C)); - MMIO_D(CURBASE(dev_priv, PIPE_A)); - MMIO_D(CURBASE(dev_priv, PIPE_B)); - MMIO_D(CURBASE(dev_priv, PIPE_C)); - MMIO_D(CUR_FBC_CTL(dev_priv, PIPE_A)); - MMIO_D(CUR_FBC_CTL(dev_priv, PIPE_B)); - MMIO_D(CUR_FBC_CTL(dev_priv, PIPE_C)); + MMIO_D(PIPEDSL(display, PIPE_A)); + MMIO_D(PIPEDSL(display, PIPE_B)); + MMIO_D(PIPEDSL(display, PIPE_C)); + MMIO_D(PIPEDSL(display, _PIPE_EDP)); + MMIO_D(TRANSCONF(display, TRANSCODER_A)); + MMIO_D(TRANSCONF(display, TRANSCODER_B)); + MMIO_D(TRANSCONF(display, TRANSCODER_C)); + MMIO_D(TRANSCONF(display, TRANSCODER_EDP)); + MMIO_D(PIPESTAT(display, PIPE_A)); + MMIO_D(PIPESTAT(display, PIPE_B)); + MMIO_D(PIPESTAT(display, PIPE_C)); + MMIO_D(PIPESTAT(display, _PIPE_EDP)); + MMIO_D(PIPE_FLIPCOUNT_G4X(display, PIPE_A)); + MMIO_D(PIPE_FLIPCOUNT_G4X(display, PIPE_B)); + MMIO_D(PIPE_FLIPCOUNT_G4X(display, PIPE_C)); + MMIO_D(PIPE_FLIPCOUNT_G4X(display, _PIPE_EDP)); + MMIO_D(PIPE_FRMCOUNT_G4X(display, PIPE_A)); + MMIO_D(PIPE_FRMCOUNT_G4X(display, PIPE_B)); + MMIO_D(PIPE_FRMCOUNT_G4X(display, PIPE_C)); + MMIO_D(PIPE_FRMCOUNT_G4X(display, _PIPE_EDP)); + MMIO_D(CURCNTR(display, PIPE_A)); + MMIO_D(CURCNTR(display, PIPE_B)); + MMIO_D(CURCNTR(display, PIPE_C)); + MMIO_D(CURPOS(display, PIPE_A)); + MMIO_D(CURPOS(display, PIPE_B)); + MMIO_D(CURPOS(display, PIPE_C)); + MMIO_D(CURBASE(display, PIPE_A)); + MMIO_D(CURBASE(display, PIPE_B)); + MMIO_D(CURBASE(display, PIPE_C)); + MMIO_D(CUR_FBC_CTL(display, PIPE_A)); + MMIO_D(CUR_FBC_CTL(display, PIPE_B)); + MMIO_D(CUR_FBC_CTL(display, PIPE_C)); MMIO_D(_MMIO(0x700ac)); MMIO_D(_MMIO(0x710ac)); MMIO_D(_MMIO(0x720ac)); @@ -172,32 +173,32 @@ static int iterate_generic_mmio(struct intel_gvt_mmio_table_iter *iter) MMIO_D(_MMIO(0x70094)); MMIO_D(_MMIO(0x70098)); MMIO_D(_MMIO(0x7009c)); - MMIO_D(DSPCNTR(dev_priv, PIPE_A)); - MMIO_D(DSPADDR(dev_priv, PIPE_A)); - MMIO_D(DSPSTRIDE(dev_priv, PIPE_A)); - MMIO_D(DSPPOS(dev_priv, PIPE_A)); - MMIO_D(DSPSIZE(dev_priv, PIPE_A)); - MMIO_D(DSPSURF(dev_priv, PIPE_A)); - MMIO_D(DSPOFFSET(dev_priv, PIPE_A)); - MMIO_D(DSPSURFLIVE(dev_priv, PIPE_A)); + MMIO_D(DSPCNTR(display, PIPE_A)); + MMIO_D(DSPADDR(display, PIPE_A)); + MMIO_D(DSPSTRIDE(display, PIPE_A)); + MMIO_D(DSPPOS(display, PIPE_A)); + MMIO_D(DSPSIZE(display, PIPE_A)); + MMIO_D(DSPSURF(display, PIPE_A)); + MMIO_D(DSPOFFSET(display, PIPE_A)); + MMIO_D(DSPSURFLIVE(display, PIPE_A)); MMIO_D(REG_50080(PIPE_A, PLANE_PRIMARY)); - MMIO_D(DSPCNTR(dev_priv, PIPE_B)); - MMIO_D(DSPADDR(dev_priv, PIPE_B)); - MMIO_D(DSPSTRIDE(dev_priv, PIPE_B)); - MMIO_D(DSPPOS(dev_priv, PIPE_B)); - MMIO_D(DSPSIZE(dev_priv, PIPE_B)); - MMIO_D(DSPSURF(dev_priv, PIPE_B)); - MMIO_D(DSPOFFSET(dev_priv, PIPE_B)); - MMIO_D(DSPSURFLIVE(dev_priv, PIPE_B)); + MMIO_D(DSPCNTR(display, PIPE_B)); + MMIO_D(DSPADDR(display, PIPE_B)); + MMIO_D(DSPSTRIDE(display, PIPE_B)); + MMIO_D(DSPPOS(display, PIPE_B)); + MMIO_D(DSPSIZE(display, PIPE_B)); + MMIO_D(DSPSURF(display, PIPE_B)); + MMIO_D(DSPOFFSET(display, PIPE_B)); + MMIO_D(DSPSURFLIVE(display, PIPE_B)); MMIO_D(REG_50080(PIPE_B, PLANE_PRIMARY)); - MMIO_D(DSPCNTR(dev_priv, PIPE_C)); - MMIO_D(DSPADDR(dev_priv, PIPE_C)); - MMIO_D(DSPSTRIDE(dev_priv, PIPE_C)); - MMIO_D(DSPPOS(dev_priv, PIPE_C)); - MMIO_D(DSPSIZE(dev_priv, PIPE_C)); - MMIO_D(DSPSURF(dev_priv, PIPE_C)); - MMIO_D(DSPOFFSET(dev_priv, PIPE_C)); - MMIO_D(DSPSURFLIVE(dev_priv, PIPE_C)); + MMIO_D(DSPCNTR(display, PIPE_C)); + MMIO_D(DSPADDR(display, PIPE_C)); + MMIO_D(DSPSTRIDE(display, PIPE_C)); + MMIO_D(DSPPOS(display, PIPE_C)); + MMIO_D(DSPSIZE(display, PIPE_C)); + MMIO_D(DSPSURF(display, PIPE_C)); + MMIO_D(DSPOFFSET(display, PIPE_C)); + MMIO_D(DSPSURFLIVE(display, PIPE_C)); MMIO_D(REG_50080(PIPE_C, PLANE_PRIMARY)); MMIO_D(SPRCTL(PIPE_A)); MMIO_D(SPRLINOFF(PIPE_A)); @@ -238,73 +239,73 @@ static int iterate_generic_mmio(struct intel_gvt_mmio_table_iter *iter) MMIO_D(SPRSCALE(PIPE_C)); MMIO_D(SPRSURFLIVE(PIPE_C)); MMIO_D(REG_50080(PIPE_C, PLANE_SPRITE0)); - MMIO_D(TRANS_HTOTAL(dev_priv, TRANSCODER_A)); - MMIO_D(TRANS_HBLANK(dev_priv, TRANSCODER_A)); - MMIO_D(TRANS_HSYNC(dev_priv, TRANSCODER_A)); - MMIO_D(TRANS_VTOTAL(dev_priv, TRANSCODER_A)); - MMIO_D(TRANS_VBLANK(dev_priv, TRANSCODER_A)); - MMIO_D(TRANS_VSYNC(dev_priv, TRANSCODER_A)); - MMIO_D(BCLRPAT(dev_priv, TRANSCODER_A)); - MMIO_D(TRANS_VSYNCSHIFT(dev_priv, TRANSCODER_A)); - MMIO_D(PIPESRC(dev_priv, TRANSCODER_A)); - MMIO_D(TRANS_HTOTAL(dev_priv, TRANSCODER_B)); - MMIO_D(TRANS_HBLANK(dev_priv, TRANSCODER_B)); - MMIO_D(TRANS_HSYNC(dev_priv, TRANSCODER_B)); - MMIO_D(TRANS_VTOTAL(dev_priv, TRANSCODER_B)); - MMIO_D(TRANS_VBLANK(dev_priv, TRANSCODER_B)); - MMIO_D(TRANS_VSYNC(dev_priv, TRANSCODER_B)); - MMIO_D(BCLRPAT(dev_priv, TRANSCODER_B)); - MMIO_D(TRANS_VSYNCSHIFT(dev_priv, TRANSCODER_B)); - MMIO_D(PIPESRC(dev_priv, TRANSCODER_B)); - MMIO_D(TRANS_HTOTAL(dev_priv, TRANSCODER_C)); - MMIO_D(TRANS_HBLANK(dev_priv, TRANSCODER_C)); - MMIO_D(TRANS_HSYNC(dev_priv, TRANSCODER_C)); - MMIO_D(TRANS_VTOTAL(dev_priv, TRANSCODER_C)); - MMIO_D(TRANS_VBLANK(dev_priv, TRANSCODER_C)); - MMIO_D(TRANS_VSYNC(dev_priv, TRANSCODER_C)); - MMIO_D(BCLRPAT(dev_priv, TRANSCODER_C)); - MMIO_D(TRANS_VSYNCSHIFT(dev_priv, TRANSCODER_C)); - MMIO_D(PIPESRC(dev_priv, TRANSCODER_C)); - MMIO_D(TRANS_HTOTAL(dev_priv, TRANSCODER_EDP)); - MMIO_D(TRANS_HBLANK(dev_priv, TRANSCODER_EDP)); - MMIO_D(TRANS_HSYNC(dev_priv, TRANSCODER_EDP)); - MMIO_D(TRANS_VTOTAL(dev_priv, TRANSCODER_EDP)); - MMIO_D(TRANS_VBLANK(dev_priv, TRANSCODER_EDP)); - MMIO_D(TRANS_VSYNC(dev_priv, TRANSCODER_EDP)); - MMIO_D(BCLRPAT(dev_priv, TRANSCODER_EDP)); - MMIO_D(TRANS_VSYNCSHIFT(dev_priv, TRANSCODER_EDP)); - MMIO_D(PIPE_DATA_M1(dev_priv, TRANSCODER_A)); - MMIO_D(PIPE_DATA_N1(dev_priv, TRANSCODER_A)); - MMIO_D(PIPE_DATA_M2(dev_priv, TRANSCODER_A)); - MMIO_D(PIPE_DATA_N2(dev_priv, TRANSCODER_A)); - MMIO_D(PIPE_LINK_M1(dev_priv, TRANSCODER_A)); - MMIO_D(PIPE_LINK_N1(dev_priv, TRANSCODER_A)); - MMIO_D(PIPE_LINK_M2(dev_priv, TRANSCODER_A)); - MMIO_D(PIPE_LINK_N2(dev_priv, TRANSCODER_A)); - MMIO_D(PIPE_DATA_M1(dev_priv, TRANSCODER_B)); - MMIO_D(PIPE_DATA_N1(dev_priv, TRANSCODER_B)); - MMIO_D(PIPE_DATA_M2(dev_priv, TRANSCODER_B)); - MMIO_D(PIPE_DATA_N2(dev_priv, TRANSCODER_B)); - MMIO_D(PIPE_LINK_M1(dev_priv, TRANSCODER_B)); - MMIO_D(PIPE_LINK_N1(dev_priv, TRANSCODER_B)); - MMIO_D(PIPE_LINK_M2(dev_priv, TRANSCODER_B)); - MMIO_D(PIPE_LINK_N2(dev_priv, TRANSCODER_B)); - MMIO_D(PIPE_DATA_M1(dev_priv, TRANSCODER_C)); - MMIO_D(PIPE_DATA_N1(dev_priv, TRANSCODER_C)); - MMIO_D(PIPE_DATA_M2(dev_priv, TRANSCODER_C)); - MMIO_D(PIPE_DATA_N2(dev_priv, TRANSCODER_C)); - MMIO_D(PIPE_LINK_M1(dev_priv, TRANSCODER_C)); - MMIO_D(PIPE_LINK_N1(dev_priv, TRANSCODER_C)); - MMIO_D(PIPE_LINK_M2(dev_priv, TRANSCODER_C)); - MMIO_D(PIPE_LINK_N2(dev_priv, TRANSCODER_C)); - MMIO_D(PIPE_DATA_M1(dev_priv, TRANSCODER_EDP)); - MMIO_D(PIPE_DATA_N1(dev_priv, TRANSCODER_EDP)); - MMIO_D(PIPE_DATA_M2(dev_priv, TRANSCODER_EDP)); - MMIO_D(PIPE_DATA_N2(dev_priv, TRANSCODER_EDP)); - MMIO_D(PIPE_LINK_M1(dev_priv, TRANSCODER_EDP)); - MMIO_D(PIPE_LINK_N1(dev_priv, TRANSCODER_EDP)); - MMIO_D(PIPE_LINK_M2(dev_priv, TRANSCODER_EDP)); - MMIO_D(PIPE_LINK_N2(dev_priv, TRANSCODER_EDP)); + MMIO_D(TRANS_HTOTAL(display, TRANSCODER_A)); + MMIO_D(TRANS_HBLANK(display, TRANSCODER_A)); + MMIO_D(TRANS_HSYNC(display, TRANSCODER_A)); + MMIO_D(TRANS_VTOTAL(display, TRANSCODER_A)); + MMIO_D(TRANS_VBLANK(display, TRANSCODER_A)); + MMIO_D(TRANS_VSYNC(display, TRANSCODER_A)); + MMIO_D(BCLRPAT(display, TRANSCODER_A)); + MMIO_D(TRANS_VSYNCSHIFT(display, TRANSCODER_A)); + MMIO_D(PIPESRC(display, TRANSCODER_A)); + MMIO_D(TRANS_HTOTAL(display, TRANSCODER_B)); + MMIO_D(TRANS_HBLANK(display, TRANSCODER_B)); + MMIO_D(TRANS_HSYNC(display, TRANSCODER_B)); + MMIO_D(TRANS_VTOTAL(display, TRANSCODER_B)); + MMIO_D(TRANS_VBLANK(display, TRANSCODER_B)); + MMIO_D(TRANS_VSYNC(display, TRANSCODER_B)); + MMIO_D(BCLRPAT(display, TRANSCODER_B)); + MMIO_D(TRANS_VSYNCSHIFT(display, TRANSCODER_B)); + MMIO_D(PIPESRC(display, TRANSCODER_B)); + MMIO_D(TRANS_HTOTAL(display, TRANSCODER_C)); + MMIO_D(TRANS_HBLANK(display, TRANSCODER_C)); + MMIO_D(TRANS_HSYNC(display, TRANSCODER_C)); + MMIO_D(TRANS_VTOTAL(display, TRANSCODER_C)); + MMIO_D(TRANS_VBLANK(display, TRANSCODER_C)); + MMIO_D(TRANS_VSYNC(display, TRANSCODER_C)); + MMIO_D(BCLRPAT(display, TRANSCODER_C)); + MMIO_D(TRANS_VSYNCSHIFT(display, TRANSCODER_C)); + MMIO_D(PIPESRC(display, TRANSCODER_C)); + MMIO_D(TRANS_HTOTAL(display, TRANSCODER_EDP)); + MMIO_D(TRANS_HBLANK(display, TRANSCODER_EDP)); + MMIO_D(TRANS_HSYNC(display, TRANSCODER_EDP)); + MMIO_D(TRANS_VTOTAL(display, TRANSCODER_EDP)); + MMIO_D(TRANS_VBLANK(display, TRANSCODER_EDP)); + MMIO_D(TRANS_VSYNC(display, TRANSCODER_EDP)); + MMIO_D(BCLRPAT(display, TRANSCODER_EDP)); + MMIO_D(TRANS_VSYNCSHIFT(display, TRANSCODER_EDP)); + MMIO_D(PIPE_DATA_M1(display, TRANSCODER_A)); + MMIO_D(PIPE_DATA_N1(display, TRANSCODER_A)); + MMIO_D(PIPE_DATA_M2(display, TRANSCODER_A)); + MMIO_D(PIPE_DATA_N2(display, TRANSCODER_A)); + MMIO_D(PIPE_LINK_M1(display, TRANSCODER_A)); + MMIO_D(PIPE_LINK_N1(display, TRANSCODER_A)); + MMIO_D(PIPE_LINK_M2(display, TRANSCODER_A)); + MMIO_D(PIPE_LINK_N2(display, TRANSCODER_A)); + MMIO_D(PIPE_DATA_M1(display, TRANSCODER_B)); + MMIO_D(PIPE_DATA_N1(display, TRANSCODER_B)); + MMIO_D(PIPE_DATA_M2(display, TRANSCODER_B)); + MMIO_D(PIPE_DATA_N2(display, TRANSCODER_B)); + MMIO_D(PIPE_LINK_M1(display, TRANSCODER_B)); + MMIO_D(PIPE_LINK_N1(display, TRANSCODER_B)); + MMIO_D(PIPE_LINK_M2(display, TRANSCODER_B)); + MMIO_D(PIPE_LINK_N2(display, TRANSCODER_B)); + MMIO_D(PIPE_DATA_M1(display, TRANSCODER_C)); + MMIO_D(PIPE_DATA_N1(display, TRANSCODER_C)); + MMIO_D(PIPE_DATA_M2(display, TRANSCODER_C)); + MMIO_D(PIPE_DATA_N2(display, TRANSCODER_C)); + MMIO_D(PIPE_LINK_M1(display, TRANSCODER_C)); + MMIO_D(PIPE_LINK_N1(display, TRANSCODER_C)); + MMIO_D(PIPE_LINK_M2(display, TRANSCODER_C)); + MMIO_D(PIPE_LINK_N2(display, TRANSCODER_C)); + MMIO_D(PIPE_DATA_M1(display, TRANSCODER_EDP)); + MMIO_D(PIPE_DATA_N1(display, TRANSCODER_EDP)); + MMIO_D(PIPE_DATA_M2(display, TRANSCODER_EDP)); + MMIO_D(PIPE_DATA_N2(display, TRANSCODER_EDP)); + MMIO_D(PIPE_LINK_M1(display, TRANSCODER_EDP)); + MMIO_D(PIPE_LINK_N1(display, TRANSCODER_EDP)); + MMIO_D(PIPE_LINK_M2(display, TRANSCODER_EDP)); + MMIO_D(PIPE_LINK_N2(display, TRANSCODER_EDP)); MMIO_D(PF_CTL(PIPE_A)); MMIO_D(PF_WIN_SZ(PIPE_A)); MMIO_D(PF_WIN_POS(PIPE_A)); @@ -513,12 +514,12 @@ static int iterate_generic_mmio(struct intel_gvt_mmio_table_iter *iter) MMIO_D(GAMMA_MODE(PIPE_A)); MMIO_D(GAMMA_MODE(PIPE_B)); MMIO_D(GAMMA_MODE(PIPE_C)); - MMIO_D(TRANS_MULT(dev_priv, TRANSCODER_A)); - MMIO_D(TRANS_MULT(dev_priv, TRANSCODER_B)); - MMIO_D(TRANS_MULT(dev_priv, TRANSCODER_C)); - MMIO_D(HSW_TVIDEO_DIP_CTL(dev_priv, TRANSCODER_A)); - MMIO_D(HSW_TVIDEO_DIP_CTL(dev_priv, TRANSCODER_B)); - MMIO_D(HSW_TVIDEO_DIP_CTL(dev_priv, TRANSCODER_C)); + MMIO_D(TRANS_MULT(display, TRANSCODER_A)); + MMIO_D(TRANS_MULT(display, TRANSCODER_B)); + MMIO_D(TRANS_MULT(display, TRANSCODER_C)); + MMIO_D(HSW_TVIDEO_DIP_CTL(display, TRANSCODER_A)); + MMIO_D(HSW_TVIDEO_DIP_CTL(display, TRANSCODER_B)); + MMIO_D(HSW_TVIDEO_DIP_CTL(display, TRANSCODER_C)); MMIO_D(SFUSE_STRAP); MMIO_D(SBI_ADDR); MMIO_D(SBI_DATA); @@ -1111,6 +1112,7 @@ static int iterate_skl_plus_mmio(struct intel_gvt_mmio_table_iter *iter) static int iterate_bxt_mmio(struct intel_gvt_mmio_table_iter *iter) { struct drm_i915_private *dev_priv = iter->i915; + struct intel_display *display = dev_priv->display; MMIO_F(_MMIO(0x80000), 0x3000); MMIO_D(GEN7_SAMPLER_INSTDONE); @@ -1242,9 +1244,9 @@ static int iterate_bxt_mmio(struct intel_gvt_mmio_table_iter *iter) MMIO_D(BXT_DSI_PLL_ENABLE); MMIO_D(GEN9_CLKGATE_DIS_0); MMIO_D(GEN9_CLKGATE_DIS_4); - MMIO_D(HSW_TVIDEO_DIP_GCP(dev_priv, TRANSCODER_A)); - MMIO_D(HSW_TVIDEO_DIP_GCP(dev_priv, TRANSCODER_B)); - MMIO_D(HSW_TVIDEO_DIP_GCP(dev_priv, TRANSCODER_C)); + MMIO_D(HSW_TVIDEO_DIP_GCP(display, TRANSCODER_A)); + MMIO_D(HSW_TVIDEO_DIP_GCP(display, TRANSCODER_B)); + MMIO_D(HSW_TVIDEO_DIP_GCP(display, TRANSCODER_C)); MMIO_D(RC6_CTX_BASE); MMIO_D(GEN8_PUSHBUS_CONTROL); MMIO_D(GEN8_PUSHBUS_ENABLE); diff --git a/drivers/gpu/drm/i915/intel_uncore.c b/drivers/gpu/drm/i915/intel_uncore.c index c8e29fd72290..4ccba7c8ffb3 100644 --- a/drivers/gpu/drm/i915/intel_uncore.c +++ b/drivers/gpu/drm/i915/intel_uncore.c @@ -2502,6 +2502,7 @@ static int sanity_check_mmio_access(struct intel_uncore *uncore) int intel_uncore_init_mmio(struct intel_uncore *uncore) { struct drm_i915_private *i915 = uncore->i915; + struct intel_display *display = i915->display; int ret; ret = sanity_check_mmio_access(uncore); @@ -2536,7 +2537,7 @@ int intel_uncore_init_mmio(struct intel_uncore *uncore) GEM_BUG_ON(intel_uncore_has_forcewake(uncore) != !!uncore->funcs.read_fw_domains); GEM_BUG_ON(intel_uncore_has_forcewake(uncore) != !!uncore->funcs.write_fw_domains); - if (HAS_FPGA_DBG_UNCLAIMED(i915)) + if (HAS_FPGA_DBG_UNCLAIMED(display)) uncore->flags |= UNCORE_HAS_FPGA_DBG_UNCLAIMED; if (IS_VALLEYVIEW(i915) || IS_CHERRYVIEW(i915)) diff --git a/drivers/gpu/drm/i915/pxp/intel_pxp_debugfs.c b/drivers/gpu/drm/i915/pxp/intel_pxp_debugfs.c index e07c5b380789..545f79eb0cc5 100644 --- a/drivers/gpu/drm/i915/pxp/intel_pxp_debugfs.c +++ b/drivers/gpu/drm/i915/pxp/intel_pxp_debugfs.c @@ -69,17 +69,17 @@ DEFINE_SIMPLE_ATTRIBUTE(pxp_terminate_fops, pxp_terminate_get, pxp_terminate_set void intel_pxp_debugfs_register(struct intel_pxp *pxp) { - struct drm_minor *minor; + struct dentry *debugfs_root; struct dentry *pxproot; if (!intel_pxp_is_supported(pxp)) return; - minor = pxp->ctrl_gt->i915->drm.primary; - if (!minor->debugfs_root) + debugfs_root = pxp->ctrl_gt->i915->drm.debugfs_root; + if (!debugfs_root) return; - pxproot = debugfs_create_dir("pxp", minor->debugfs_root); + pxproot = debugfs_create_dir("pxp", debugfs_root); if (IS_ERR(pxproot)) return; diff --git a/drivers/gpu/drm/i915/selftests/intel_uncore.c b/drivers/gpu/drm/i915/selftests/intel_uncore.c index 41eaa9b7f67d..58bcbdcef563 100644 --- a/drivers/gpu/drm/i915/selftests/intel_uncore.c +++ b/drivers/gpu/drm/i915/selftests/intel_uncore.c @@ -277,13 +277,15 @@ static int live_forcewake_domains(void *arg) #define FW_RANGE 0x40000 struct intel_gt *gt = arg; struct intel_uncore *uncore = gt->uncore; + struct drm_i915_private *i915 = gt->i915; + struct intel_display *display = i915->display; unsigned long *valid; u32 offset; int err; - if (!HAS_FPGA_DBG_UNCLAIMED(gt->i915) && - !IS_VALLEYVIEW(gt->i915) && - !IS_CHERRYVIEW(gt->i915)) + if (!HAS_FPGA_DBG_UNCLAIMED(display) && + !IS_VALLEYVIEW(i915) && + !IS_CHERRYVIEW(i915)) return 0; /* diff --git a/drivers/gpu/drm/i915/soc/intel_dram.c b/drivers/gpu/drm/i915/soc/intel_dram.c index deb159548a09..149527827624 100644 --- a/drivers/gpu/drm/i915/soc/intel_dram.c +++ b/drivers/gpu/drm/i915/soc/intel_dram.c @@ -11,6 +11,7 @@ #include "i915_drv.h" #include "i915_reg.h" +#include "i915_utils.h" #include "intel_dram.h" #include "intel_mchbar_regs.h" #include "intel_pcode.h" @@ -30,10 +31,11 @@ struct dram_channel_info { #define DRAM_TYPE_STR(type) [INTEL_DRAM_ ## type] = #type -static const char *intel_dram_type_str(enum intel_dram_type type) +const char *intel_dram_type_str(enum intel_dram_type type) { static const char * const str[] = { DRAM_TYPE_STR(UNKNOWN), + DRAM_TYPE_STR(DDR2), DRAM_TYPE_STR(DDR3), DRAM_TYPE_STR(DDR4), DRAM_TYPE_STR(LPDDR3), @@ -54,9 +56,10 @@ static const char *intel_dram_type_str(enum intel_dram_type type) #undef DRAM_TYPE_STR -static bool pnv_is_ddr3(struct drm_i915_private *i915) +static enum intel_dram_type pnv_dram_type(struct drm_i915_private *i915) { - return intel_uncore_read(&i915->uncore, CSHRDDR3CTL) & CSHRDDR3CTL_DDR3; + return intel_uncore_read(&i915->uncore, CSHRDDR3CTL) & CSHRDDR3CTL_DDR3 ? + INTEL_DRAM_DDR3 : INTEL_DRAM_DDR2; } static unsigned int pnv_mem_freq(struct drm_i915_private *dev_priv) @@ -135,25 +138,21 @@ static unsigned int vlv_mem_freq(struct drm_i915_private *i915) return 0; } -static void detect_mem_freq(struct drm_i915_private *i915) +unsigned int intel_mem_freq(struct drm_i915_private *i915) { if (IS_PINEVIEW(i915)) - i915->mem_freq = pnv_mem_freq(i915); + return pnv_mem_freq(i915); else if (GRAPHICS_VER(i915) == 5) - i915->mem_freq = ilk_mem_freq(i915); + return ilk_mem_freq(i915); else if (IS_CHERRYVIEW(i915)) - i915->mem_freq = chv_mem_freq(i915); + return chv_mem_freq(i915); else if (IS_VALLEYVIEW(i915)) - i915->mem_freq = vlv_mem_freq(i915); - - if (IS_PINEVIEW(i915)) - i915->is_ddr3 = pnv_is_ddr3(i915); - - if (i915->mem_freq) - drm_dbg(&i915->drm, "DDR speed: %d kHz\n", i915->mem_freq); + return vlv_mem_freq(i915); + else + return 0; } -unsigned int i9xx_fsb_freq(struct drm_i915_private *i915) +static unsigned int i9xx_fsb_freq(struct drm_i915_private *i915) { u32 fsb; @@ -235,15 +234,30 @@ static unsigned int ilk_fsb_freq(struct drm_i915_private *dev_priv) } } -static void detect_fsb_freq(struct drm_i915_private *i915) +unsigned int intel_fsb_freq(struct drm_i915_private *i915) { if (GRAPHICS_VER(i915) == 5) - i915->fsb_freq = ilk_fsb_freq(i915); + return ilk_fsb_freq(i915); else if (GRAPHICS_VER(i915) == 3 || GRAPHICS_VER(i915) == 4) - i915->fsb_freq = i9xx_fsb_freq(i915); + return i9xx_fsb_freq(i915); + else + return 0; +} - if (i915->fsb_freq) - drm_dbg(&i915->drm, "FSB frequency: %d kHz\n", i915->fsb_freq); +static int i915_get_dram_info(struct drm_i915_private *i915, struct dram_info *dram_info) +{ + dram_info->fsb_freq = intel_fsb_freq(i915); + if (dram_info->fsb_freq) + drm_dbg(&i915->drm, "FSB frequency: %d kHz\n", dram_info->fsb_freq); + + dram_info->mem_freq = intel_mem_freq(i915); + if (dram_info->mem_freq) + drm_dbg(&i915->drm, "DDR speed: %d kHz\n", dram_info->mem_freq); + + if (IS_PINEVIEW(i915)) + dram_info->type = pnv_dram_type(i915); + + return 0; } static int intel_dimm_num_devices(const struct dram_dimm_info *dimm) @@ -392,6 +406,9 @@ skl_dram_get_channels_info(struct drm_i915_private *i915, struct dram_info *dram u32 val; int ret; + /* Assume 16Gb DIMMs are present until proven otherwise */ + dram_info->has_16gb_dimms = true; + val = intel_uncore_read(&i915->uncore, SKL_MAD_DIMM_CH0_0_0_0_MCHBAR_MCMAIN); ret = skl_dram_get_channel_info(i915, &ch0, 0, val); @@ -414,13 +431,16 @@ skl_dram_get_channels_info(struct drm_i915_private *i915, struct dram_info *dram return -EINVAL; } - dram_info->wm_lv_0_adjust_needed = ch0.is_16gb_dimm || ch1.is_16gb_dimm; + dram_info->has_16gb_dimms = ch0.is_16gb_dimm || ch1.is_16gb_dimm; dram_info->symmetric_memory = intel_is_dram_symmetric(&ch0, &ch1); drm_dbg_kms(&i915->drm, "Memory configuration is symmetric? %s\n", str_yes_no(dram_info->symmetric_memory)); + drm_dbg_kms(&i915->drm, "16Gb DIMMs: %s\n", + str_yes_no(dram_info->has_16gb_dimms)); + return 0; } @@ -649,8 +669,9 @@ static int icl_pcode_read_mem_global_info(struct drm_i915_private *dev_priv, static int gen11_get_dram_info(struct drm_i915_private *i915, struct dram_info *dram_info) { - int ret = skl_get_dram_info(i915, dram_info); + int ret; + ret = skl_dram_get_channels_info(i915, dram_info); if (ret) return ret; @@ -659,8 +680,6 @@ static int gen11_get_dram_info(struct drm_i915_private *i915, struct dram_info * static int gen12_get_dram_info(struct drm_i915_private *i915, struct dram_info *dram_info) { - dram_info->wm_lv_0_adjust_needed = false; - return icl_pcode_read_mem_global_info(i915, dram_info); } @@ -709,13 +728,11 @@ static int xelpdp_get_dram_info(struct drm_i915_private *i915, struct dram_info int intel_dram_detect(struct drm_i915_private *i915) { + struct intel_display *display = i915->display; struct dram_info *dram_info; int ret; - detect_fsb_freq(i915); - detect_mem_freq(i915); - - if (GRAPHICS_VER(i915) < 9 || IS_DG2(i915) || !HAS_DISPLAY(i915)) + if (IS_DG2(i915) || !HAS_DISPLAY(display)) return 0; dram_info = drmm_kzalloc(&i915->drm, sizeof(*dram_info), GFP_KERNEL); @@ -724,13 +741,7 @@ int intel_dram_detect(struct drm_i915_private *i915) i915->dram_info = dram_info; - /* - * Assume level 0 watermark latency adjustment is needed until proven - * otherwise, this w/a is not needed by bxt/glk. - */ - dram_info->wm_lv_0_adjust_needed = !IS_BROXTON(i915) && !IS_GEMINILAKE(i915); - - if (DISPLAY_VER(i915) >= 14) + if (DISPLAY_VER(display) >= 14) ret = xelpdp_get_dram_info(i915, dram_info); else if (GRAPHICS_VER(i915) >= 12) ret = gen12_get_dram_info(i915, dram_info); @@ -738,23 +749,23 @@ int intel_dram_detect(struct drm_i915_private *i915) ret = gen11_get_dram_info(i915, dram_info); else if (IS_BROXTON(i915) || IS_GEMINILAKE(i915)) ret = bxt_get_dram_info(i915, dram_info); - else + else if (GRAPHICS_VER(i915) >= 9) ret = skl_get_dram_info(i915, dram_info); + else + ret = i915_get_dram_info(i915, dram_info); drm_dbg_kms(&i915->drm, "DRAM type: %s\n", intel_dram_type_str(dram_info->type)); + drm_dbg_kms(&i915->drm, "DRAM channels: %u\n", dram_info->num_channels); + + drm_dbg_kms(&i915->drm, "Num QGV points %u\n", dram_info->num_qgv_points); + drm_dbg_kms(&i915->drm, "Num PSF GV points %u\n", dram_info->num_psf_gv_points); + /* TODO: Do we want to abort probe on dram detection failures? */ if (ret) return 0; - drm_dbg_kms(&i915->drm, "Num qgv points %u\n", dram_info->num_qgv_points); - - drm_dbg_kms(&i915->drm, "DRAM channels: %u\n", dram_info->num_channels); - - drm_dbg_kms(&i915->drm, "Watermark level 0 adjustment needed: %s\n", - str_yes_no(dram_info->wm_lv_0_adjust_needed)); - return 0; } diff --git a/drivers/gpu/drm/i915/soc/intel_dram.h b/drivers/gpu/drm/i915/soc/intel_dram.h index 2a696e03aad4..03a973f1c941 100644 --- a/drivers/gpu/drm/i915/soc/intel_dram.h +++ b/drivers/gpu/drm/i915/soc/intel_dram.h @@ -12,11 +12,9 @@ struct drm_i915_private; struct drm_device; struct dram_info { - bool wm_lv_0_adjust_needed; - u8 num_channels; - bool symmetric_memory; enum intel_dram_type { INTEL_DRAM_UNKNOWN, + INTEL_DRAM_DDR2, INTEL_DRAM_DDR3, INTEL_DRAM_DDR4, INTEL_DRAM_LPDDR3, @@ -27,13 +25,20 @@ struct dram_info { INTEL_DRAM_GDDR_ECC, __INTEL_DRAM_TYPE_MAX, } type; + unsigned int fsb_freq; + unsigned int mem_freq; + u8 num_channels; u8 num_qgv_points; u8 num_psf_gv_points; + bool symmetric_memory; + bool has_16gb_dimms; }; void intel_dram_edram_detect(struct drm_i915_private *i915); int intel_dram_detect(struct drm_i915_private *i915); -unsigned int i9xx_fsb_freq(struct drm_i915_private *i915); +unsigned int intel_fsb_freq(struct drm_i915_private *i915); +unsigned int intel_mem_freq(struct drm_i915_private *i915); const struct dram_info *intel_dram_info(struct drm_device *drm); +const char *intel_dram_type_str(enum intel_dram_type type); #endif /* __INTEL_DRAM_H__ */ diff --git a/drivers/gpu/drm/i915/soc/intel_gmch.c b/drivers/gpu/drm/i915/soc/intel_gmch.c index 5346b8dda79a..f210c9655b53 100644 --- a/drivers/gpu/drm/i915/soc/intel_gmch.c +++ b/drivers/gpu/drm/i915/soc/intel_gmch.c @@ -148,7 +148,8 @@ void intel_gmch_bar_teardown(struct drm_i915_private *i915) int intel_gmch_vga_set_state(struct drm_i915_private *i915, bool enable_decode) { - unsigned int reg = DISPLAY_VER(i915) >= 6 ? SNB_GMCH_CTRL : INTEL_GMCH_CTRL; + struct intel_display *display = i915->display; + unsigned int reg = DISPLAY_VER(display) >= 6 ? SNB_GMCH_CTRL : INTEL_GMCH_CTRL; u16 gmch_ctrl; if (pci_read_config_word(i915->gmch.pdev, reg, &gmch_ctrl)) { diff --git a/drivers/gpu/drm/radeon/r600_cs.c b/drivers/gpu/drm/radeon/r600_cs.c index ac77d1246b94..811265648a58 100644 --- a/drivers/gpu/drm/radeon/r600_cs.c +++ b/drivers/gpu/drm/radeon/r600_cs.c @@ -1408,7 +1408,7 @@ static void r600_texture_size(unsigned nfaces, unsigned blevel, unsigned llevel, unsigned block_align, unsigned height_align, unsigned base_align, unsigned *l0_size, unsigned *mipmap_size) { - unsigned offset, i, level; + unsigned offset, i; unsigned width, height, depth, size; unsigned blocksize; unsigned nbx, nby; @@ -1420,7 +1420,7 @@ static void r600_texture_size(unsigned nfaces, unsigned blevel, unsigned llevel, w0 = r600_mip_minify(w0, 0); h0 = r600_mip_minify(h0, 0); d0 = r600_mip_minify(d0, 0); - for(i = 0, offset = 0, level = blevel; i < nlevels; i++, level++) { + for (i = 0, offset = 0; i < nlevels; i++) { width = r600_mip_minify(w0, i); nbx = r600_fmt_get_nblocksx(format, width); diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c index 7a3e510327b7..9e35b14e2bf0 100644 --- a/drivers/gpu/drm/radeon/radeon_device.c +++ b/drivers/gpu/drm/radeon/radeon_device.c @@ -554,7 +554,7 @@ int radeon_wb_init(struct radeon_device *rdev) * cover the whole aperture even if VRAM size is inferior to aperture size * Novell bug 204882 + along with lots of ubuntu ones * - * Note 3: when limiting vram it's safe to overwritte real_vram_size because + * Note 3: when limiting vram it's safe to overwrite real_vram_size because * we are not in case where real_vram_size is inferior to mc_vram_size (ie * not affected by bogus hw of Novell bug 204882 + along with lots of ubuntu * ones) @@ -562,7 +562,7 @@ int radeon_wb_init(struct radeon_device *rdev) * Note 4: IGP TOM addr should be the same as the aperture addr, we don't * explicitly check for that thought. * - * FIXME: when reducing VRAM size align new size on power of 2. + * FIXME: when reducing VRAM size, align new size on power of 2. */ void radeon_vram_location(struct radeon_device *rdev, struct radeon_mc *mc, u64 base) { diff --git a/drivers/gpu/drm/radeon/radeon_display.c b/drivers/gpu/drm/radeon/radeon_display.c index 4dc77c398617..351b9dfcdad8 100644 --- a/drivers/gpu/drm/radeon/radeon_display.c +++ b/drivers/gpu/drm/radeon/radeon_display.c @@ -926,10 +926,10 @@ static void avivo_get_fb_ref_div(unsigned nom, unsigned den, unsigned post_div, unsigned *fb_div, unsigned *ref_div) { /* limit reference * post divider to a maximum */ - ref_div_max = max(min(100 / post_div, ref_div_max), 1u); + ref_div_max = clamp(100 / post_div, 1u, ref_div_max); /* get matching reference and feedback divider */ - *ref_div = min(max(den/post_div, 1u), ref_div_max); + *ref_div = clamp(den / post_div, 1u, ref_div_max); *fb_div = DIV_ROUND_CLOSEST(nom * *ref_div * post_div, den); /* limit fb divider to its maximum */ diff --git a/drivers/gpu/drm/radeon/radeon_gart.c b/drivers/gpu/drm/radeon/radeon_gart.c index 4bb242437ff6..acd89a20f272 100644 --- a/drivers/gpu/drm/radeon/radeon_gart.c +++ b/drivers/gpu/drm/radeon/radeon_gart.c @@ -346,14 +346,14 @@ int radeon_gart_init(struct radeon_device *rdev) DRM_INFO("GART: num cpu pages %u, num gpu pages %u\n", rdev->gart.num_cpu_pages, rdev->gart.num_gpu_pages); /* Allocate pages table */ - rdev->gart.pages = vzalloc(array_size(sizeof(void *), - rdev->gart.num_cpu_pages)); + rdev->gart.pages = vcalloc(rdev->gart.num_cpu_pages, + sizeof(void *)); if (rdev->gart.pages == NULL) { radeon_gart_fini(rdev); return -ENOMEM; } - rdev->gart.pages_entry = vmalloc(array_size(sizeof(uint64_t), - rdev->gart.num_gpu_pages)); + rdev->gart.pages_entry = vmalloc_array(rdev->gart.num_gpu_pages, + sizeof(uint64_t)); if (rdev->gart.pages_entry == NULL) { radeon_gart_fini(rdev); return -ENOMEM; diff --git a/drivers/gpu/drm/radeon/radeon_test.c b/drivers/gpu/drm/radeon/radeon_test.c index c9fef9b61ced..818554e60537 100644 --- a/drivers/gpu/drm/radeon/radeon_test.c +++ b/drivers/gpu/drm/radeon/radeon_test.c @@ -455,7 +455,7 @@ static void radeon_test_ring_sync2(struct radeon_device *rdev, r = radeon_ring_lock(rdev, ringC, 64); if (r) { - DRM_ERROR("Failed to lock ring B %p\n", ringC); + DRM_ERROR("Failed to lock ring C %p\n", ringC); goto out_cleanup; } radeon_semaphore_emit_signal(rdev, ringC->idx, semaphore); @@ -481,7 +481,7 @@ static void radeon_test_ring_sync2(struct radeon_device *rdev, r = radeon_ring_lock(rdev, ringC, 64); if (r) { - DRM_ERROR("Failed to lock ring B %p\n", ringC); + DRM_ERROR("Failed to lock ring C %p\n", ringC); goto out_cleanup; } radeon_semaphore_emit_signal(rdev, ringC->idx, semaphore); diff --git a/drivers/gpu/drm/radeon/radeon_vce.c b/drivers/gpu/drm/radeon/radeon_vce.c index 2355a78e1b69..bdbc1bbe8a9b 100644 --- a/drivers/gpu/drm/radeon/radeon_vce.c +++ b/drivers/gpu/drm/radeon/radeon_vce.c @@ -86,7 +86,7 @@ int radeon_vce_init(struct radeon_device *rdev) r = request_firmware(&rdev->vce_fw, fw_name, rdev->dev); if (r) { - dev_err(rdev->dev, "radeon_vce: Can't load firmware \"%s\"\n", + dev_err(rdev->dev, "radeon_vce: can't load firmware \"%s\"\n", fw_name); return r; } @@ -126,7 +126,7 @@ int radeon_vce_init(struct radeon_device *rdev) rdev->vce.fw_version = (start << 24) | (mid << 16) | (end << 8); - /* we can only work with this fw version for now */ + /* we can only work with these fw versions for now */ if ((rdev->vce.fw_version != ((40 << 24) | (2 << 16) | (2 << 8))) && (rdev->vce.fw_version != ((50 << 24) | (0 << 16) | (1 << 8))) && (rdev->vce.fw_version != ((50 << 24) | (1 << 16) | (2 << 8)))) @@ -281,7 +281,7 @@ static void radeon_vce_idle_work_handler(struct work_struct *work) * * @rdev: radeon_device pointer * - * Make sure VCE is powerd up when we want to use it + * Make sure VCE is powered up when we want to use it */ void radeon_vce_note_usage(struct radeon_device *rdev) { diff --git a/drivers/gpu/drm/xe/Makefile b/drivers/gpu/drm/xe/Makefile index 07c71a29963d..987e4fe10538 100644 --- a/drivers/gpu/drm/xe/Makefile +++ b/drivers/gpu/drm/xe/Makefile @@ -35,6 +35,7 @@ $(obj)/generated/%_device_wa_oob.c $(obj)/generated/%_device_wa_oob.h: $(obj)/xe xe-y += xe_bb.o \ xe_bo.o \ xe_bo_evict.o \ + xe_dep_scheduler.o \ xe_devcoredump.o \ xe_device.o \ xe_device_sysfs.o \ @@ -60,7 +61,6 @@ xe-y += xe_bb.o \ xe_gt_pagefault.o \ xe_gt_sysfs.o \ xe_gt_throttle.o \ - xe_gt_tlb_invalidation.o \ xe_gt_topology.o \ xe_guc.o \ xe_guc_ads.o \ @@ -75,16 +75,19 @@ xe-y += xe_bb.o \ xe_guc_log.o \ xe_guc_pc.o \ xe_guc_submit.o \ + xe_guc_tlb_inval.o \ xe_heci_gsc.o \ xe_huc.o \ xe_hw_engine.o \ xe_hw_engine_class_sysfs.o \ xe_hw_engine_group.o \ + xe_hw_error.o \ xe_hw_fence.o \ xe_irq.o \ xe_lrc.o \ xe_migrate.o \ xe_mmio.o \ + xe_mmio_gem.o \ xe_mocs.o \ xe_module.o \ xe_nvm.o \ @@ -95,6 +98,7 @@ xe-y += xe_bb.o \ xe_pcode.o \ xe_pm.o \ xe_preempt_fence.o \ + xe_psmi.o \ xe_pt.o \ xe_pt_walk.o \ xe_pxp.o \ @@ -114,6 +118,8 @@ xe-y += xe_bb.o \ xe_sync.o \ xe_tile.o \ xe_tile_sysfs.o \ + xe_tlb_inval.o \ + xe_tlb_inval_job.o \ xe_trace.o \ xe_trace_bo.o \ xe_trace_guc.o \ @@ -125,6 +131,7 @@ xe-y += xe_bb.o \ xe_uc.o \ xe_uc_fw.o \ xe_vm.o \ + xe_vm_madvise.o \ xe_vram.o \ xe_vram_freq.o \ xe_vsec.o \ @@ -149,6 +156,7 @@ xe-y += \ xe_memirq.o \ xe_sriov.o \ xe_sriov_vf.o \ + xe_sriov_vf_ccs.o \ xe_tile_sriov_vf.o xe-$(CONFIG_PCI_IOV) += \ diff --git a/drivers/gpu/drm/xe/abi/guc_actions_abi.h b/drivers/gpu/drm/xe/abi/guc_actions_abi.h index 81eb046aeebf..d8cf68a0516d 100644 --- a/drivers/gpu/drm/xe/abi/guc_actions_abi.h +++ b/drivers/gpu/drm/xe/abi/guc_actions_abi.h @@ -193,6 +193,14 @@ enum xe_guc_register_context_multi_lrc_param_offsets { XE_GUC_REGISTER_CONTEXT_MULTI_LRC_MSG_MIN_LEN = 11, }; +enum xe_guc_context_wq_item_offsets { + XE_GUC_CONTEXT_WQ_HEADER_DATA_0_TYPE_LEN = 0, + XE_GUC_CONTEXT_WQ_EL_INFO_DATA_1_CTX_DESC_LOW, + XE_GUC_CONTEXT_WQ_EL_INFO_DATA_2_GUCCTX_RINGTAIL_FREEZEPOCS, + XE_GUC_CONTEXT_WQ_EL_INFO_DATA_3_WI_FENCE_ID, + XE_GUC_CONTEXT_WQ_EL_CHILD_LIST_DATA_4_RINGTAIL, +}; + enum xe_guc_report_status { XE_GUC_REPORT_STATUS_UNKNOWN = 0x0, XE_GUC_REPORT_STATUS_ACKED = 0x1, diff --git a/drivers/gpu/drm/xe/abi/guc_errors_abi.h b/drivers/gpu/drm/xe/abi/guc_errors_abi.h index ecf748fd87df..ad76b4baf42e 100644 --- a/drivers/gpu/drm/xe/abi/guc_errors_abi.h +++ b/drivers/gpu/drm/xe/abi/guc_errors_abi.h @@ -63,6 +63,7 @@ enum xe_guc_load_status { XE_GUC_LOAD_STATUS_HWCONFIG_START = 0x05, XE_GUC_LOAD_STATUS_HWCONFIG_DONE = 0x06, XE_GUC_LOAD_STATUS_HWCONFIG_ERROR = 0x07, + XE_GUC_LOAD_STATUS_BOOTROM_VERSION_MISMATCH = 0x08, XE_GUC_LOAD_STATUS_GDT_DONE = 0x10, XE_GUC_LOAD_STATUS_IDT_DONE = 0x20, XE_GUC_LOAD_STATUS_LAPIC_DONE = 0x30, @@ -75,6 +76,8 @@ enum xe_guc_load_status { XE_GUC_LOAD_STATUS_INVALID_INIT_DATA_RANGE_START, XE_GUC_LOAD_STATUS_MPU_DATA_INVALID = 0x73, XE_GUC_LOAD_STATUS_INIT_MMIO_SAVE_RESTORE_INVALID = 0x74, + XE_GUC_LOAD_STATUS_KLV_WORKAROUND_INIT_ERROR = 0x75, + XE_GUC_LOAD_STATUS_INVALID_FTR_FLAG = 0x76, XE_GUC_LOAD_STATUS_INVALID_INIT_DATA_RANGE_END, XE_GUC_LOAD_STATUS_READY = 0xF0, diff --git a/drivers/gpu/drm/xe/abi/guc_klvs_abi.h b/drivers/gpu/drm/xe/abi/guc_klvs_abi.h index 0366a9da5977..0e78351c6ef5 100644 --- a/drivers/gpu/drm/xe/abi/guc_klvs_abi.h +++ b/drivers/gpu/drm/xe/abi/guc_klvs_abi.h @@ -390,12 +390,14 @@ enum { */ enum xe_guc_klv_ids { GUC_WORKAROUND_KLV_BLOCK_INTERRUPTS_WHEN_MGSR_BLOCKED = 0x9002, + GUC_WORKAROUND_KLV_DISABLE_PSMI_INTERRUPTS_AT_C6_ENTRY_RESTORE_AT_EXIT = 0x9004, GUC_WORKAROUND_KLV_ID_GAM_PFQ_SHADOW_TAIL_POLLING = 0x9005, GUC_WORKAROUND_KLV_ID_DISABLE_MTP_DURING_ASYNC_COMPUTE = 0x9007, GUC_WA_KLV_NP_RD_WRITE_TO_CLEAR_RCSM_AT_CGP_LATE_RESTORE = 0x9008, GUC_WORKAROUND_KLV_ID_BACK_TO_BACK_RCS_ENGINE_RESET = 0x9009, GUC_WA_KLV_WAKE_POWER_DOMAINS_FOR_OUTBOUND_MMIO = 0x900a, GUC_WA_KLV_RESET_BB_STACK_PTR_ON_VF_SWITCH = 0x900b, + GUC_WA_KLV_RESTORE_UNSAVED_MEDIA_CONTROL_REG = 0x900c, }; #endif diff --git a/drivers/gpu/drm/xe/compat-i915-headers/i915_drv.h b/drivers/gpu/drm/xe/compat-i915-headers/i915_drv.h index 9b7572e06f34..b8269391bc69 100644 --- a/drivers/gpu/drm/xe/compat-i915-headers/i915_drv.h +++ b/drivers/gpu/drm/xe/compat-i915-headers/i915_drv.h @@ -12,7 +12,6 @@ #include <drm/drm_drv.h> -#include "i915_utils.h" #include "xe_device.h" /* for xe_device_has_flat_ccs() */ #include "xe_device_types.h" @@ -26,34 +25,13 @@ static inline struct drm_i915_private *to_i915(const struct drm_device *dev) #define IS_I915G(dev_priv) (dev_priv && 0) #define IS_I915GM(dev_priv) (dev_priv && 0) #define IS_PINEVIEW(dev_priv) (dev_priv && 0) -#define IS_IVYBRIDGE(dev_priv) (dev_priv && 0) #define IS_VALLEYVIEW(dev_priv) (dev_priv && 0) #define IS_CHERRYVIEW(dev_priv) (dev_priv && 0) #define IS_HASWELL(dev_priv) (dev_priv && 0) #define IS_BROADWELL(dev_priv) (dev_priv && 0) -#define IS_SKYLAKE(dev_priv) (dev_priv && 0) #define IS_BROXTON(dev_priv) (dev_priv && 0) -#define IS_KABYLAKE(dev_priv) (dev_priv && 0) #define IS_GEMINILAKE(dev_priv) (dev_priv && 0) -#define IS_COFFEELAKE(dev_priv) (dev_priv && 0) -#define IS_COMETLAKE(dev_priv) (dev_priv && 0) -#define IS_ICELAKE(dev_priv) (dev_priv && 0) -#define IS_JASPERLAKE(dev_priv) (dev_priv && 0) -#define IS_ELKHARTLAKE(dev_priv) (dev_priv && 0) -#define IS_TIGERLAKE(dev_priv) IS_PLATFORM(dev_priv, XE_TIGERLAKE) -#define IS_ROCKETLAKE(dev_priv) IS_PLATFORM(dev_priv, XE_ROCKETLAKE) -#define IS_DG1(dev_priv) IS_PLATFORM(dev_priv, XE_DG1) -#define IS_ALDERLAKE_S(dev_priv) IS_PLATFORM(dev_priv, XE_ALDERLAKE_S) -#define IS_ALDERLAKE_P(dev_priv) (IS_PLATFORM(dev_priv, XE_ALDERLAKE_P) || \ - IS_PLATFORM(dev_priv, XE_ALDERLAKE_N)) #define IS_DG2(dev_priv) IS_PLATFORM(dev_priv, XE_DG2) -#define IS_METEORLAKE(dev_priv) IS_PLATFORM(dev_priv, XE_METEORLAKE) -#define IS_LUNARLAKE(dev_priv) IS_PLATFORM(dev_priv, XE_LUNARLAKE) -#define IS_BATTLEMAGE(dev_priv) IS_PLATFORM(dev_priv, XE_BATTLEMAGE) -#define IS_PANTHERLAKE(dev_priv) IS_PLATFORM(dev_priv, XE_PANTHERLAKE) - -#define IS_HASWELL_ULT(dev_priv) (dev_priv && 0) -#define IS_BROADWELL_ULT(dev_priv) (dev_priv && 0) #define IS_MOBILE(xe) (xe && 0) diff --git a/drivers/gpu/drm/xe/display/ext/i915_utils.c b/drivers/gpu/drm/xe/display/ext/i915_utils.c index 43b10a2cc508..1421c2a7b64d 100644 --- a/drivers/gpu/drm/xe/display/ext/i915_utils.c +++ b/drivers/gpu/drm/xe/display/ext/i915_utils.c @@ -4,6 +4,7 @@ */ #include "i915_drv.h" +#include "i915_utils.h" bool i915_vtd_active(struct drm_i915_private *i915) { diff --git a/drivers/gpu/drm/xe/display/intel_fbdev_fb.c b/drivers/gpu/drm/xe/display/intel_fbdev_fb.c index fba9617a75a5..d96ba2b51065 100644 --- a/drivers/gpu/drm/xe/display/intel_fbdev_fb.c +++ b/drivers/gpu/drm/xe/display/intel_fbdev_fb.c @@ -41,7 +41,7 @@ struct intel_framebuffer *intel_fbdev_fb_alloc(struct drm_fb_helper *helper, size = PAGE_ALIGN(size); obj = ERR_PTR(-ENODEV); - if (!IS_DGFX(xe) && !XE_WA(xe_root_mmio_gt(xe), 22019338487_display)) { + if (!IS_DGFX(xe) && !XE_GT_WA(xe_root_mmio_gt(xe), 22019338487_display)) { obj = xe_bo_create_pin_map(xe, xe_device_get_root_tile(xe), NULL, size, ttm_bo_type_kernel, XE_BO_FLAG_SCANOUT | diff --git a/drivers/gpu/drm/xe/display/xe_display.c b/drivers/gpu/drm/xe/display/xe_display.c index e2e0771cf274..8b68d70db6c8 100644 --- a/drivers/gpu/drm/xe/display/xe_display.c +++ b/drivers/gpu/drm/xe/display/xe_display.c @@ -96,6 +96,7 @@ static void xe_display_fini_early(void *arg) if (!xe->info.probe_display) return; + intel_hpd_cancel_work(display); intel_display_driver_remove_nogem(display); intel_display_driver_remove_noirq(display); intel_opregion_cleanup(display); @@ -340,6 +341,8 @@ void xe_display_pm_suspend(struct xe_device *xe) xe_display_flush_cleanup_work(xe); + intel_encoder_block_all_hpds(display); + intel_hpd_cancel_work(display); if (has_display(xe)) { @@ -370,6 +373,7 @@ void xe_display_pm_shutdown(struct xe_device *xe) xe_display_flush_cleanup_work(xe); intel_dp_mst_suspend(display); + intel_encoder_block_all_hpds(display); intel_hpd_cancel_work(display); if (has_display(xe)) @@ -471,6 +475,8 @@ void xe_display_pm_resume(struct xe_device *xe) intel_hpd_init(display); + intel_encoder_unblock_all_hpds(display); + if (has_display(xe)) { intel_display_driver_resume(display); drm_kms_helper_poll_enable(&xe->drm); diff --git a/drivers/gpu/drm/xe/display/xe_display_wa.c b/drivers/gpu/drm/xe/display/xe_display_wa.c index 68d1387d81a0..8ada1cbcb16c 100644 --- a/drivers/gpu/drm/xe/display/xe_display_wa.c +++ b/drivers/gpu/drm/xe/display/xe_display_wa.c @@ -14,5 +14,5 @@ bool intel_display_needs_wa_16023588340(struct intel_display *display) { struct xe_device *xe = to_xe_device(display->drm); - return XE_WA(xe_root_mmio_gt(xe), 16023588340); + return XE_GT_WA(xe_root_mmio_gt(xe), 16023588340); } diff --git a/drivers/gpu/drm/xe/display/xe_fb_pin.c b/drivers/gpu/drm/xe/display/xe_fb_pin.c index c38fba18effe..f1f8b5ab53ef 100644 --- a/drivers/gpu/drm/xe/display/xe_fb_pin.c +++ b/drivers/gpu/drm/xe/display/xe_fb_pin.c @@ -16,6 +16,7 @@ #include "xe_device.h" #include "xe_ggtt.h" #include "xe_pm.h" +#include "xe_vram_types.h" static void write_dpt_rotated(struct xe_bo *bo, struct iosys_map *map, u32 *dpt_ofs, u32 bo_ofs, @@ -289,7 +290,7 @@ static struct i915_vma *__xe_pin_fb_vma(const struct intel_framebuffer *fb, if (IS_DGFX(to_xe_device(bo->ttm.base.dev)) && intel_fb_rc_ccs_cc_plane(&fb->base) >= 0 && !(bo->flags & XE_BO_FLAG_NEEDS_CPU_ACCESS)) { - struct xe_tile *tile = xe_device_get_root_tile(xe); + struct xe_vram_region *vram = xe_device_get_root_tile(xe)->mem.vram; /* * If we need to able to access the clear-color value stored in @@ -297,7 +298,7 @@ static struct i915_vma *__xe_pin_fb_vma(const struct intel_framebuffer *fb, * accessible. This is important on small-bar systems where * only some subset of VRAM is CPU accessible. */ - if (tile->mem.vram.io_size < tile->mem.vram.usable_size) { + if (xe_vram_region_io_size(vram) < xe_vram_region_usable_size(vram)) { ret = -EINVAL; goto err; } @@ -382,6 +383,7 @@ static bool reuse_vma(struct intel_plane_state *new_plane_state, const struct intel_plane_state *old_plane_state) { struct intel_framebuffer *fb = to_intel_framebuffer(new_plane_state->hw.fb); + struct intel_plane *plane = to_intel_plane(new_plane_state->uapi.plane); struct xe_device *xe = to_xe_device(fb->base.dev); struct intel_display *display = xe->display; struct i915_vma *vma; @@ -405,6 +407,10 @@ static bool reuse_vma(struct intel_plane_state *new_plane_state, found: refcount_inc(&vma->ref); new_plane_state->ggtt_vma = vma; + + new_plane_state->surf = i915_ggtt_offset(new_plane_state->ggtt_vma) + + plane->surf_offset(new_plane_state); + return true; } @@ -431,6 +437,10 @@ int intel_plane_pin_fb(struct intel_plane_state *new_plane_state, return PTR_ERR(vma); new_plane_state->ggtt_vma = vma; + + new_plane_state->surf = i915_ggtt_offset(new_plane_state->ggtt_vma) + + plane->surf_offset(new_plane_state); + return 0; } diff --git a/drivers/gpu/drm/xe/display/xe_plane_initial.c b/drivers/gpu/drm/xe/display/xe_plane_initial.c index dcbc4b2d3fd9..826ac3d578b7 100644 --- a/drivers/gpu/drm/xe/display/xe_plane_initial.c +++ b/drivers/gpu/drm/xe/display/xe_plane_initial.c @@ -10,6 +10,7 @@ #include "xe_ggtt.h" #include "xe_mmio.h" +#include "i915_vma.h" #include "intel_crtc.h" #include "intel_display.h" #include "intel_display_core.h" @@ -21,6 +22,7 @@ #include "intel_plane.h" #include "intel_plane_initial.h" #include "xe_bo.h" +#include "xe_vram_types.h" #include "xe_wa.h" #include <generated/xe_wa_oob.h> @@ -103,7 +105,7 @@ initial_plane_bo(struct xe_device *xe, * We don't currently expect this to ever be placed in the * stolen portion. */ - if (phys_base >= tile0->mem.vram.usable_size) { + if (phys_base >= xe_vram_region_usable_size(tile0->mem.vram)) { drm_err(&xe->drm, "Initial plane programming using invalid range, phys_base=%pa\n", &phys_base); @@ -121,7 +123,7 @@ initial_plane_bo(struct xe_device *xe, phys_base = base; flags |= XE_BO_FLAG_STOLEN; - if (XE_WA(xe_root_mmio_gt(xe), 22019338487_display)) + if (XE_GT_WA(xe_root_mmio_gt(xe), 22019338487_display)) return NULL; /* @@ -234,6 +236,9 @@ intel_find_initial_plane_obj(struct intel_crtc *crtc, goto nofb; plane_state->ggtt_vma = vma; + + plane_state->surf = i915_ggtt_offset(plane_state->ggtt_vma); + plane_state->uapi.src_x = 0; plane_state->uapi.src_y = 0; plane_state->uapi.src_w = fb->width << 16; diff --git a/drivers/gpu/drm/xe/instructions/xe_mi_commands.h b/drivers/gpu/drm/xe/instructions/xe_mi_commands.h index e3f5e8bb3ebc..c47b290e0e9f 100644 --- a/drivers/gpu/drm/xe/instructions/xe_mi_commands.h +++ b/drivers/gpu/drm/xe/instructions/xe_mi_commands.h @@ -65,6 +65,7 @@ #define MI_LOAD_REGISTER_MEM (__MI_INSTR(0x29) | XE_INSTR_NUM_DW(4)) #define MI_LRM_USE_GGTT REG_BIT(22) +#define MI_LRM_ASYNC REG_BIT(21) #define MI_LOAD_REGISTER_REG (__MI_INSTR(0x2a) | XE_INSTR_NUM_DW(3)) #define MI_LRR_DST_CS_MMIO REG_BIT(19) diff --git a/drivers/gpu/drm/xe/regs/xe_engine_regs.h b/drivers/gpu/drm/xe/regs/xe_engine_regs.h index 7ade41e2b7b3..f4c3e1187a00 100644 --- a/drivers/gpu/drm/xe/regs/xe_engine_regs.h +++ b/drivers/gpu/drm/xe/regs/xe_engine_regs.h @@ -111,6 +111,9 @@ #define PPHWSP_CSB_AND_TIMESTAMP_REPORT_DIS REG_BIT(14) #define CS_PRIORITY_MEM_READ REG_BIT(7) +#define CS_DEBUG_MODE2(base) XE_REG((base) + 0xd8, XE_REG_OPTION_MASKED) +#define INSTRUCTION_STATE_CACHE_INVALIDATE REG_BIT(6) + #define FF_SLICE_CS_CHICKEN1(base) XE_REG((base) + 0xe0, XE_REG_OPTION_MASKED) #define FFSC_PERCTX_PREEMPT_CTRL REG_BIT(14) diff --git a/drivers/gpu/drm/xe/regs/xe_gsc_regs.h b/drivers/gpu/drm/xe/regs/xe_gsc_regs.h index 9b66cc972a63..180be82672ab 100644 --- a/drivers/gpu/drm/xe/regs/xe_gsc_regs.h +++ b/drivers/gpu/drm/xe/regs/xe_gsc_regs.h @@ -13,6 +13,8 @@ /* Definitions of GSC H/W registers, bits, etc */ +#define BMG_GSC_HECI1_BASE 0x373000 + #define MTL_GSC_HECI1_BASE 0x00116000 #define MTL_GSC_HECI2_BASE 0x00117000 diff --git a/drivers/gpu/drm/xe/regs/xe_gt_regs.h b/drivers/gpu/drm/xe/regs/xe_gt_regs.h index 5cd5ab8529c5..f96b2e2b3064 100644 --- a/drivers/gpu/drm/xe/regs/xe_gt_regs.h +++ b/drivers/gpu/drm/xe/regs/xe_gt_regs.h @@ -42,7 +42,7 @@ #define FORCEWAKE_ACK_GSC XE_REG(0xdf8) #define FORCEWAKE_ACK_GT_MTL XE_REG(0xdfc) -#define MCFG_MCR_SELECTOR XE_REG(0xfd0) +#define STEER_SEMAPHORE XE_REG(0xfd0) #define MTL_MCR_SELECTOR XE_REG(0xfd4) #define SF_MCR_SELECTOR XE_REG(0xfd8) #define MCR_SELECTOR XE_REG(0xfdc) diff --git a/drivers/gpu/drm/xe/regs/xe_hw_error_regs.h b/drivers/gpu/drm/xe/regs/xe_hw_error_regs.h new file mode 100644 index 000000000000..c146b9ef44eb --- /dev/null +++ b/drivers/gpu/drm/xe/regs/xe_hw_error_regs.h @@ -0,0 +1,20 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2025 Intel Corporation + */ + +#ifndef _XE_HW_ERROR_REGS_H_ +#define _XE_HW_ERROR_REGS_H_ + +#define HEC_UNCORR_ERR_STATUS(base) XE_REG((base) + 0x118) +#define UNCORR_FW_REPORTED_ERR BIT(6) + +#define HEC_UNCORR_FW_ERR_DW0(base) XE_REG((base) + 0x124) + +#define DEV_ERR_STAT_NONFATAL 0x100178 +#define DEV_ERR_STAT_CORRECTABLE 0x10017c +#define DEV_ERR_STAT_REG(x) XE_REG(_PICK_EVEN((x), \ + DEV_ERR_STAT_CORRECTABLE, \ + DEV_ERR_STAT_NONFATAL)) +#define XE_CSC_ERROR BIT(17) +#endif diff --git a/drivers/gpu/drm/xe/regs/xe_irq_regs.h b/drivers/gpu/drm/xe/regs/xe_irq_regs.h index 13635e4331d4..7c2a3a140142 100644 --- a/drivers/gpu/drm/xe/regs/xe_irq_regs.h +++ b/drivers/gpu/drm/xe/regs/xe_irq_regs.h @@ -18,6 +18,7 @@ #define GFX_MSTR_IRQ XE_REG(0x190010, XE_REG_OPTION_VF) #define MASTER_IRQ REG_BIT(31) #define GU_MISC_IRQ REG_BIT(29) +#define ERROR_IRQ(x) REG_BIT(26 + (x)) #define DISPLAY_IRQ REG_BIT(16) #define I2C_IRQ REG_BIT(12) #define GT_DW_IRQ(x) REG_BIT(x) diff --git a/drivers/gpu/drm/xe/regs/xe_pmt.h b/drivers/gpu/drm/xe/regs/xe_pmt.h index 2995d72c3f78..264e9baf949c 100644 --- a/drivers/gpu/drm/xe/regs/xe_pmt.h +++ b/drivers/gpu/drm/xe/regs/xe_pmt.h @@ -21,4 +21,14 @@ #define SG_REMAP_INDEX1 XE_REG(SOC_BASE + 0x08) #define SG_REMAP_BITS REG_GENMASK(31, 24) +#define BMG_MODS_RESIDENCY_OFFSET (0x4D0) +#define BMG_G2_RESIDENCY_OFFSET (0x530) +#define BMG_G6_RESIDENCY_OFFSET (0x538) +#define BMG_G8_RESIDENCY_OFFSET (0x540) +#define BMG_G10_RESIDENCY_OFFSET (0x548) + +#define BMG_PCIE_LINK_L0_RESIDENCY_OFFSET (0x570) +#define BMG_PCIE_LINK_L1_RESIDENCY_OFFSET (0x578) +#define BMG_PCIE_LINK_L1_2_RESIDENCY_OFFSET (0x580) + #endif diff --git a/drivers/gpu/drm/xe/tests/xe_dma_buf.c b/drivers/gpu/drm/xe/tests/xe_dma_buf.c index c53f67ce4b0a..3c5ad8cc65a0 100644 --- a/drivers/gpu/drm/xe/tests/xe_dma_buf.c +++ b/drivers/gpu/drm/xe/tests/xe_dma_buf.c @@ -57,16 +57,12 @@ static void check_residency(struct kunit *test, struct xe_bo *exported, return; /* - * Evict exporter. Note that the gem object dma_buf member isn't - * set from xe_gem_prime_export(), and it's needed for the move_notify() - * functionality, so hack that up here. Evicting the exported bo will + * Evict exporter. Evicting the exported bo will * evict also the imported bo through the move_notify() functionality if * importer is on a different device. If they're on the same device, * the exporter and the importer should be the same bo. */ - swap(exported->ttm.base.dma_buf, dmabuf); ret = xe_bo_evict(exported); - swap(exported->ttm.base.dma_buf, dmabuf); if (ret) { if (ret != -EINTR && ret != -ERESTARTSYS) KUNIT_FAIL(test, "Evicting exporter failed with err=%d.\n", @@ -139,6 +135,7 @@ static void xe_test_dmabuf_import_same_driver(struct xe_device *xe) PTR_ERR(dmabuf)); goto out; } + bo->ttm.base.dma_buf = dmabuf; import = xe_gem_prime_import(&xe->drm, dmabuf); if (!IS_ERR(import)) { @@ -186,6 +183,7 @@ static void xe_test_dmabuf_import_same_driver(struct xe_device *xe) KUNIT_FAIL(test, "dynamic p2p attachment failed with err=%ld\n", PTR_ERR(import)); } + bo->ttm.base.dma_buf = NULL; dma_buf_put(dmabuf); out: drm_gem_object_put(&bo->ttm.base); @@ -206,7 +204,7 @@ static const struct dma_buf_attach_ops nop2p_attach_ops = { static const struct dma_buf_test_params test_params[] = { {.mem_mask = XE_BO_FLAG_VRAM0, .attach_ops = &xe_dma_buf_attach_ops}, - {.mem_mask = XE_BO_FLAG_VRAM0, + {.mem_mask = XE_BO_FLAG_VRAM0 | XE_BO_FLAG_NEEDS_CPU_ACCESS, .attach_ops = &xe_dma_buf_attach_ops, .force_different_devices = true}, @@ -238,7 +236,8 @@ static const struct dma_buf_test_params test_params[] = { {.mem_mask = XE_BO_FLAG_SYSTEM | XE_BO_FLAG_VRAM0, .attach_ops = &xe_dma_buf_attach_ops}, - {.mem_mask = XE_BO_FLAG_SYSTEM | XE_BO_FLAG_VRAM0, + {.mem_mask = XE_BO_FLAG_SYSTEM | XE_BO_FLAG_VRAM0 | + XE_BO_FLAG_NEEDS_CPU_ACCESS, .attach_ops = &xe_dma_buf_attach_ops, .force_different_devices = true}, diff --git a/drivers/gpu/drm/xe/tests/xe_pci.c b/drivers/gpu/drm/xe/tests/xe_pci.c index 9c715e59f030..db30c5156d0c 100644 --- a/drivers/gpu/drm/xe/tests/xe_pci.c +++ b/drivers/gpu/drm/xe/tests/xe_pci.c @@ -101,6 +101,11 @@ static void fake_read_gmdid(struct xe_device *xe, enum xe_gmdid_type type, } } +static void fake_xe_info_probe_tile_count(struct xe_device *xe) +{ + /* Nothing to do, just use the statically defined value. */ +} + int xe_pci_fake_device_init(struct xe_device *xe) { struct kunit *test = kunit_get_current_test(); @@ -138,6 +143,8 @@ done: data->sriov_mode : XE_SRIOV_MODE_NONE; kunit_activate_static_stub(test, read_gmdid, fake_read_gmdid); + kunit_activate_static_stub(test, xe_info_probe_tile_count, + fake_xe_info_probe_tile_count); xe_info_init_early(xe, desc, subplatform_desc); xe_info_init(xe, desc); diff --git a/drivers/gpu/drm/xe/tests/xe_wa_test.c b/drivers/gpu/drm/xe/tests/xe_wa_test.c index c96d1fe34151..416258c193f6 100644 --- a/drivers/gpu/drm/xe/tests/xe_wa_test.c +++ b/drivers/gpu/drm/xe/tests/xe_wa_test.c @@ -75,6 +75,7 @@ static const struct platform_test_case cases[] = { GMDID_CASE(LUNARLAKE, 2004, A0, 2000, A0), GMDID_CASE(LUNARLAKE, 2004, B0, 2000, A0), GMDID_CASE(BATTLEMAGE, 2001, A0, 1301, A1), + GMDID_CASE(PANTHERLAKE, 3000, A0, 3000, A0), }; static void platform_desc(const struct platform_test_case *t, char *desc) diff --git a/drivers/gpu/drm/xe/xe_assert.h b/drivers/gpu/drm/xe/xe_assert.h index 68fe70ce2be3..a818eaa05b7d 100644 --- a/drivers/gpu/drm/xe/xe_assert.h +++ b/drivers/gpu/drm/xe/xe_assert.h @@ -12,6 +12,7 @@ #include "xe_gt_types.h" #include "xe_step.h" +#include "xe_vram.h" /** * DOC: Xe Asserts @@ -145,7 +146,8 @@ const struct xe_tile *__tile = (tile); \ char __buf[10] __maybe_unused; \ xe_assert_msg(tile_to_xe(__tile), condition, "tile: %u VRAM %s\n" msg, \ - __tile->id, ({ string_get_size(__tile->mem.vram.actual_physical_size, 1, \ + __tile->id, ({ string_get_size( \ + xe_vram_region_actual_physical_size(__tile->mem.vram), 1, \ STRING_UNITS_2, __buf, sizeof(__buf)); __buf; }), ## arg); \ }) diff --git a/drivers/gpu/drm/xe/xe_bb.c b/drivers/gpu/drm/xe/xe_bb.c index 5ce0e26822f2..feb6e013dc38 100644 --- a/drivers/gpu/drm/xe/xe_bb.c +++ b/drivers/gpu/drm/xe/xe_bb.c @@ -60,6 +60,41 @@ err: return ERR_PTR(err); } +struct xe_bb *xe_bb_ccs_new(struct xe_gt *gt, u32 dwords, + enum xe_sriov_vf_ccs_rw_ctxs ctx_id) +{ + struct xe_bb *bb = kmalloc(sizeof(*bb), GFP_KERNEL); + struct xe_tile *tile = gt_to_tile(gt); + struct xe_sa_manager *bb_pool; + int err; + + if (!bb) + return ERR_PTR(-ENOMEM); + /* + * We need to allocate space for the requested number of dwords & + * one additional MI_BATCH_BUFFER_END dword. Since the whole SA + * is submitted to HW, we need to make sure that the last instruction + * is not over written when the last chunk of SA is allocated for BB. + * So, this extra DW acts as a guard here. + */ + + bb_pool = tile->sriov.vf.ccs[ctx_id].mem.ccs_bb_pool; + bb->bo = xe_sa_bo_new(bb_pool, 4 * (dwords + 1)); + + if (IS_ERR(bb->bo)) { + err = PTR_ERR(bb->bo); + goto err; + } + + bb->cs = xe_sa_bo_cpu_addr(bb->bo); + bb->len = 0; + + return bb; +err: + kfree(bb); + return ERR_PTR(err); +} + static struct xe_sched_job * __xe_bb_create_job(struct xe_exec_queue *q, struct xe_bb *bb, u64 *addr) { diff --git a/drivers/gpu/drm/xe/xe_bb.h b/drivers/gpu/drm/xe/xe_bb.h index b5cc65506696..2a8adc9a6dee 100644 --- a/drivers/gpu/drm/xe/xe_bb.h +++ b/drivers/gpu/drm/xe/xe_bb.h @@ -13,8 +13,11 @@ struct dma_fence; struct xe_gt; struct xe_exec_queue; struct xe_sched_job; +enum xe_sriov_vf_ccs_rw_ctxs; struct xe_bb *xe_bb_new(struct xe_gt *gt, u32 dwords, bool usm); +struct xe_bb *xe_bb_ccs_new(struct xe_gt *gt, u32 dwords, + enum xe_sriov_vf_ccs_rw_ctxs ctx_id); struct xe_sched_job *xe_bb_create_job(struct xe_exec_queue *q, struct xe_bb *bb); struct xe_sched_job *xe_bb_create_migration_job(struct xe_exec_queue *q, diff --git a/drivers/gpu/drm/xe/xe_bo.c b/drivers/gpu/drm/xe/xe_bo.c index 18f27da47a36..4faf15d5fa6d 100644 --- a/drivers/gpu/drm/xe/xe_bo.c +++ b/drivers/gpu/drm/xe/xe_bo.c @@ -33,9 +33,11 @@ #include "xe_pxp.h" #include "xe_res_cursor.h" #include "xe_shrinker.h" +#include "xe_sriov_vf_ccs.h" #include "xe_trace_bo.h" #include "xe_ttm_stolen_mgr.h" #include "xe_vm.h" +#include "xe_vram_types.h" const char *const xe_mem_type_to_name[TTM_NUM_MEM_TYPES] = { [XE_PL_SYSTEM] = "system", @@ -198,6 +200,8 @@ static bool force_contiguous(u32 bo_flags) else if (bo_flags & XE_BO_FLAG_PINNED && !(bo_flags & XE_BO_FLAG_PINNED_LATE_RESTORE)) return true; /* needs vmap */ + else if (bo_flags & XE_BO_FLAG_CPU_ADDR_MIRROR) + return true; /* * For eviction / restore on suspend / resume objects pinned in VRAM @@ -812,14 +816,14 @@ static int xe_bo_move(struct ttm_buffer_object *ttm_bo, bool evict, } if (ttm_bo->type == ttm_bo_type_sg) { - ret = xe_bo_move_notify(bo, ctx); + if (new_mem->mem_type == XE_PL_SYSTEM) + ret = xe_bo_move_notify(bo, ctx); if (!ret) ret = xe_bo_move_dmabuf(ttm_bo, new_mem); return ret; } - tt_has_data = ttm && (ttm_tt_is_populated(ttm) || - (ttm->page_flags & TTM_TT_FLAG_SWAPPED)); + tt_has_data = ttm && (ttm_tt_is_populated(ttm) || ttm_tt_is_swapped(ttm)); move_lacks_source = !old_mem || (handle_system_ccs ? (!bo->ccs_cleared) : (!mem_type_is_vram(old_mem_type) && !tt_has_data)); @@ -964,6 +968,20 @@ static int xe_bo_move(struct ttm_buffer_object *ttm_bo, bool evict, dma_fence_put(fence); xe_pm_runtime_put(xe); + /* + * CCS meta data is migrated from TT -> SMEM. So, let us detach the + * BBs from BO as it is no longer needed. + */ + if (IS_VF_CCS_BB_VALID(xe, bo) && old_mem_type == XE_PL_TT && + new_mem->mem_type == XE_PL_SYSTEM) + xe_sriov_vf_ccs_detach_bo(bo); + + if (IS_SRIOV_VF(xe) && + ((move_lacks_source && new_mem->mem_type == XE_PL_TT) || + (old_mem_type == XE_PL_SYSTEM && new_mem->mem_type == XE_PL_TT)) && + handle_system_ccs) + ret = xe_sriov_vf_ccs_attach_bo(bo); + out: if ((!ttm_bo->resource || ttm_bo->resource->mem_type == XE_PL_SYSTEM) && ttm_bo->ttm) { @@ -974,6 +992,9 @@ out: if (timeout < 0) ret = timeout; + if (IS_VF_CCS_BB_VALID(xe, bo)) + xe_sriov_vf_ccs_detach_bo(bo); + xe_tt_unmap_sg(xe, ttm_bo->ttm); } @@ -1501,9 +1522,14 @@ static void xe_ttm_bo_release_notify(struct ttm_buffer_object *ttm_bo) static void xe_ttm_bo_delete_mem_notify(struct ttm_buffer_object *ttm_bo) { + struct xe_bo *bo = ttm_to_xe_bo(ttm_bo); + if (!xe_bo_is_xe_bo(ttm_bo)) return; + if (IS_VF_CCS_BB_VALID(ttm_to_xe_device(ttm_bo->bdev), bo)) + xe_sriov_vf_ccs_detach_bo(bo); + /* * Object is idle and about to be destroyed. Release the * dma-buf attachment. @@ -1685,6 +1711,18 @@ static void xe_gem_object_close(struct drm_gem_object *obj, } } +static bool should_migrate_to_smem(struct xe_bo *bo) +{ + /* + * NOTE: The following atomic checks are platform-specific. For example, + * if a device supports CXL atomics, these may not be necessary or + * may behave differently. + */ + + return bo->attr.atomic_access == DRM_XE_ATOMIC_GLOBAL || + bo->attr.atomic_access == DRM_XE_ATOMIC_CPU; +} + static vm_fault_t xe_gem_fault(struct vm_fault *vmf) { struct ttm_buffer_object *tbo = vmf->vma->vm_private_data; @@ -1693,7 +1731,7 @@ static vm_fault_t xe_gem_fault(struct vm_fault *vmf) struct xe_bo *bo = ttm_to_xe_bo(tbo); bool needs_rpm = bo->flags & XE_BO_FLAG_VRAM_MASK; vm_fault_t ret; - int idx; + int idx, r = 0; if (needs_rpm) xe_pm_runtime_get(xe); @@ -1705,25 +1743,40 @@ static vm_fault_t xe_gem_fault(struct vm_fault *vmf) if (drm_dev_enter(ddev, &idx)) { trace_xe_bo_cpu_fault(bo); - ret = ttm_bo_vm_fault_reserved(vmf, vmf->vma->vm_page_prot, - TTM_BO_VM_NUM_PREFAULT); + if (should_migrate_to_smem(bo)) { + xe_assert(xe, bo->flags & XE_BO_FLAG_SYSTEM); + + r = xe_bo_migrate(bo, XE_PL_TT); + if (r == -EBUSY || r == -ERESTARTSYS || r == -EINTR) + ret = VM_FAULT_NOPAGE; + else if (r) + ret = VM_FAULT_SIGBUS; + } + if (!ret) + ret = ttm_bo_vm_fault_reserved(vmf, + vmf->vma->vm_page_prot, + TTM_BO_VM_NUM_PREFAULT); drm_dev_exit(idx); + + if (ret == VM_FAULT_RETRY && + !(vmf->flags & FAULT_FLAG_RETRY_NOWAIT)) + goto out; + + /* + * ttm_bo_vm_reserve() already has dma_resv_lock. + */ + if (ret == VM_FAULT_NOPAGE && + mem_type_is_vram(tbo->resource->mem_type)) { + mutex_lock(&xe->mem_access.vram_userfault.lock); + if (list_empty(&bo->vram_userfault_link)) + list_add(&bo->vram_userfault_link, + &xe->mem_access.vram_userfault.list); + mutex_unlock(&xe->mem_access.vram_userfault.lock); + } } else { ret = ttm_bo_vm_dummy_page(vmf, vmf->vma->vm_page_prot); } - if (ret == VM_FAULT_RETRY && !(vmf->flags & FAULT_FLAG_RETRY_NOWAIT)) - goto out; - /* - * ttm_bo_vm_reserve() already has dma_resv_lock. - */ - if (ret == VM_FAULT_NOPAGE && mem_type_is_vram(tbo->resource->mem_type)) { - mutex_lock(&xe->mem_access.vram_userfault.lock); - if (list_empty(&bo->vram_userfault_link)) - list_add(&bo->vram_userfault_link, &xe->mem_access.vram_userfault.list); - mutex_unlock(&xe->mem_access.vram_userfault.lock); - } - dma_resv_unlock(tbo->base.resv); out: if (needs_rpm) @@ -2438,7 +2491,6 @@ int xe_bo_validate(struct xe_bo *bo, struct xe_vm *vm, bool allow_res_evict) .no_wait_gpu = false, .gfp_retry_mayfail = true, }; - struct pin_cookie cookie; int ret; if (vm) { @@ -2449,10 +2501,10 @@ int xe_bo_validate(struct xe_bo *bo, struct xe_vm *vm, bool allow_res_evict) ctx.resv = xe_vm_resv(vm); } - cookie = xe_vm_set_validating(vm, allow_res_evict); + xe_vm_set_validating(vm, allow_res_evict); trace_xe_bo_validate(bo); ret = ttm_bo_validate(&bo->ttm, &bo->placement, &ctx); - xe_vm_clear_validating(vm, allow_res_evict, cookie); + xe_vm_clear_validating(vm, allow_res_evict); return ret; } diff --git a/drivers/gpu/drm/xe/xe_bo.h b/drivers/gpu/drm/xe/xe_bo.h index 02e8cde4c6b2..8cce413b5235 100644 --- a/drivers/gpu/drm/xe/xe_bo.h +++ b/drivers/gpu/drm/xe/xe_bo.h @@ -12,6 +12,7 @@ #include "xe_macros.h" #include "xe_vm_types.h" #include "xe_vm.h" +#include "xe_vram_types.h" #define XE_DEFAULT_GTT_SIZE_MB 3072ULL /* 3GB by default */ @@ -23,8 +24,9 @@ #define XE_BO_FLAG_VRAM_MASK (XE_BO_FLAG_VRAM0 | XE_BO_FLAG_VRAM1) /* -- */ #define XE_BO_FLAG_STOLEN BIT(4) +#define XE_BO_FLAG_VRAM(vram) (XE_BO_FLAG_VRAM0 << ((vram)->id)) #define XE_BO_FLAG_VRAM_IF_DGFX(tile) (IS_DGFX(tile_to_xe(tile)) ? \ - XE_BO_FLAG_VRAM0 << (tile)->id : \ + XE_BO_FLAG_VRAM((tile)->mem.vram) : \ XE_BO_FLAG_SYSTEM) #define XE_BO_FLAG_GGTT BIT(5) #define XE_BO_FLAG_IGNORE_MIN_PAGE_SIZE BIT(6) diff --git a/drivers/gpu/drm/xe/xe_bo_types.h b/drivers/gpu/drm/xe/xe_bo_types.h index ff560d82496f..314652afdca7 100644 --- a/drivers/gpu/drm/xe/xe_bo_types.h +++ b/drivers/gpu/drm/xe/xe_bo_types.h @@ -9,6 +9,7 @@ #include <linux/iosys-map.h> #include <drm/drm_gpusvm.h> +#include <drm/drm_pagemap.h> #include <drm/ttm/ttm_bo.h> #include <drm/ttm/ttm_device.h> #include <drm/ttm/ttm_placement.h> @@ -60,6 +61,14 @@ struct xe_bo { */ struct list_head client_link; #endif + /** @attr: User controlled attributes for bo */ + struct { + /** + * @atomic_access: type of atomic access bo needs + * protected by bo dma-resv lock + */ + u32 atomic_access; + } attr; /** * @pxp_key_instance: PXP key instance this BO was created against. A * 0 in this variable indicates that the BO does not use PXP encryption. @@ -76,6 +85,9 @@ struct xe_bo { /** @ccs_cleared */ bool ccs_cleared; + /** @bb_ccs_rw: BB instructions of CCS read/write. Valid only for VF */ + struct xe_bb *bb_ccs[XE_SRIOV_VF_CCS_CTX_COUNT]; + /** * @cpu_caching: CPU caching mode. Currently only used for userspace * objects. Exceptions are system memory on DGFX, which is always diff --git a/drivers/gpu/drm/xe/xe_configfs.c b/drivers/gpu/drm/xe/xe_configfs.c index e9b46a2d0019..1025d3979b06 100644 --- a/drivers/gpu/drm/xe/xe_configfs.c +++ b/drivers/gpu/drm/xe/xe_configfs.c @@ -5,6 +5,7 @@ #include <linux/bitops.h> #include <linux/configfs.h> +#include <linux/cleanup.h> #include <linux/find.h> #include <linux/init.h> #include <linux/module.h> @@ -12,9 +13,9 @@ #include <linux/string.h> #include "xe_configfs.h" -#include "xe_module.h" - #include "xe_hw_engine_types.h" +#include "xe_module.h" +#include "xe_pci_types.h" /** * DOC: Xe Configfs @@ -23,23 +24,45 @@ * ========= * * Configfs is a filesystem-based manager of kernel objects. XE KMD registers a - * configfs subsystem called ``'xe'`` that creates a directory in the mounted configfs directory - * The user can create devices under this directory and configure them as necessary - * See Documentation/filesystems/configfs.rst for more information about how configfs works. + * configfs subsystem called ``xe`` that creates a directory in the mounted + * configfs directory. The user can create devices under this directory and + * configure them as necessary. See Documentation/filesystems/configfs.rst for + * more information about how configfs works. * * Create devices - * =============== + * ============== * - * In order to create a device, the user has to create a directory inside ``'xe'``:: + * To create a device, the ``xe`` module should already be loaded, but some + * attributes can only be set before binding the device. It can be accomplished + * by blocking the driver autoprobe: * - * mkdir /sys/kernel/config/xe/0000:03:00.0/ + * # echo 0 > /sys/bus/pci/drivers_autoprobe + * # modprobe xe + * + * In order to create a device, the user has to create a directory inside ``xe``:: + * + * # mkdir /sys/kernel/config/xe/0000:03:00.0/ * * Every device created is populated by the driver with entries that can be * used to configure it:: * * /sys/kernel/config/xe/ - * .. 0000:03:00.0/ - * ... survivability_mode + * ├── 0000:00:02.0 + * │ └── ... + * ├── 0000:00:02.1 + * │ └── ... + * : + * └── 0000:03:00.0 + * ├── survivability_mode + * ├── engines_allowed + * └── enable_psmi + * + * After configuring the attributes as per next section, the device can be + * probed with:: + * + * # echo 0000:03:00.0 > /sys/bus/pci/drivers/xe/bind + * # # or + * # echo 0000:03:00.0 > /sys/bus/pci/drivers_probe * * Configure Attributes * ==================== @@ -51,7 +74,8 @@ * effect when probing the device. Example to enable it:: * * # echo 1 > /sys/kernel/config/xe/0000:03:00.0/survivability_mode - * # echo 0000:03:00.0 > /sys/bus/pci/drivers/xe/bind (Enters survivability mode if supported) + * + * This attribute can only be set before binding to the device. * * Allowed engines: * ---------------- @@ -77,24 +101,52 @@ * available for migrations, but it's disabled. This is intended for debugging * purposes only. * + * This attribute can only be set before binding to the device. + * + * PSMI + * ---- + * + * Enable extra debugging capabilities to trace engine execution. Only useful + * during early platform enabling and requires additional hardware connected. + * Once it's enabled, additionals WAs are added and runtime configuration is + * done via debugfs. Example to enable it:: + * + * # echo 1 > /sys/kernel/config/xe/0000:03:00.0/enable_psmi + * + * This attribute can only be set before binding to the device. + * * Remove devices * ============== * * The created device directories can be removed using ``rmdir``:: * - * rmdir /sys/kernel/config/xe/0000:03:00.0/ + * # rmdir /sys/kernel/config/xe/0000:03:00.0/ */ -struct xe_config_device { +struct xe_config_group_device { struct config_group group; - bool survivability_mode; - u64 engines_allowed; + struct xe_config_device { + u64 engines_allowed; + bool survivability_mode; + bool enable_psmi; + } config; /* protects attributes */ struct mutex lock; }; +static const struct xe_config_device device_defaults = { + .engines_allowed = U64_MAX, + .survivability_mode = false, + .enable_psmi = false, +}; + +static void set_device_defaults(struct xe_config_device *config) +{ + *config = device_defaults; +} + struct engine_info { const char *cls; u64 mask; @@ -113,9 +165,40 @@ static const struct engine_info engine_info[] = { { .cls = "gsccs", .mask = XE_HW_ENGINE_GSCCS_MASK }, }; +static struct xe_config_group_device *to_xe_config_group_device(struct config_item *item) +{ + return container_of(to_config_group(item), struct xe_config_group_device, group); +} + static struct xe_config_device *to_xe_config_device(struct config_item *item) { - return container_of(to_config_group(item), struct xe_config_device, group); + return &to_xe_config_group_device(item)->config; +} + +static bool is_bound(struct xe_config_group_device *dev) +{ + unsigned int domain, bus, slot, function; + struct pci_dev *pdev; + const char *name; + bool ret; + + lockdep_assert_held(&dev->lock); + + name = dev->group.cg_item.ci_name; + if (sscanf(name, "%x:%x:%x.%x", &domain, &bus, &slot, &function) != 4) + return false; + + pdev = pci_get_domain_bus_and_slot(domain, bus, PCI_DEVFN(slot, function)); + if (!pdev) + return false; + + ret = pci_get_drvdata(pdev); + pci_dev_put(pdev); + + if (ret) + pci_dbg(pdev, "Already bound to driver\n"); + + return ret; } static ssize_t survivability_mode_show(struct config_item *item, char *page) @@ -127,7 +210,7 @@ static ssize_t survivability_mode_show(struct config_item *item, char *page) static ssize_t survivability_mode_store(struct config_item *item, const char *page, size_t len) { - struct xe_config_device *dev = to_xe_config_device(item); + struct xe_config_group_device *dev = to_xe_config_group_device(item); bool survivability_mode; int ret; @@ -135,9 +218,11 @@ static ssize_t survivability_mode_store(struct config_item *item, const char *pa if (ret) return ret; - mutex_lock(&dev->lock); - dev->survivability_mode = survivability_mode; - mutex_unlock(&dev->lock); + guard(mutex)(&dev->lock); + if (is_bound(dev)) + return -EBUSY; + + dev->config.survivability_mode = survivability_mode; return len; } @@ -199,7 +284,7 @@ static bool lookup_engine_mask(const char *pattern, u64 *mask) static ssize_t engines_allowed_store(struct config_item *item, const char *page, size_t len) { - struct xe_config_device *dev = to_xe_config_device(item); + struct xe_config_group_device *dev = to_xe_config_group_device(item); size_t patternlen, p; u64 mask, val = 0; @@ -219,25 +304,55 @@ static ssize_t engines_allowed_store(struct config_item *item, const char *page, val |= mask; } - mutex_lock(&dev->lock); - dev->engines_allowed = val; - mutex_unlock(&dev->lock); + guard(mutex)(&dev->lock); + if (is_bound(dev)) + return -EBUSY; + + dev->config.engines_allowed = val; return len; } -CONFIGFS_ATTR(, survivability_mode); +static ssize_t enable_psmi_show(struct config_item *item, char *page) +{ + struct xe_config_device *dev = to_xe_config_device(item); + + return sprintf(page, "%d\n", dev->enable_psmi); +} + +static ssize_t enable_psmi_store(struct config_item *item, const char *page, size_t len) +{ + struct xe_config_group_device *dev = to_xe_config_group_device(item); + bool val; + int ret; + + ret = kstrtobool(page, &val); + if (ret) + return ret; + + guard(mutex)(&dev->lock); + if (is_bound(dev)) + return -EBUSY; + + dev->config.enable_psmi = val; + + return len; +} + +CONFIGFS_ATTR(, enable_psmi); CONFIGFS_ATTR(, engines_allowed); +CONFIGFS_ATTR(, survivability_mode); static struct configfs_attribute *xe_config_device_attrs[] = { - &attr_survivability_mode, + &attr_enable_psmi, &attr_engines_allowed, + &attr_survivability_mode, NULL, }; static void xe_config_device_release(struct config_item *item) { - struct xe_config_device *dev = to_xe_config_device(item); + struct xe_config_group_device *dev = to_xe_config_group_device(item); mutex_destroy(&dev->lock); kfree(dev); @@ -253,29 +368,81 @@ static const struct config_item_type xe_config_device_type = { .ct_owner = THIS_MODULE, }; +static const struct xe_device_desc *xe_match_desc(struct pci_dev *pdev) +{ + struct device_driver *driver = driver_find("xe", &pci_bus_type); + struct pci_driver *drv = to_pci_driver(driver); + const struct pci_device_id *ids = drv ? drv->id_table : NULL; + const struct pci_device_id *found = pci_match_id(ids, pdev); + + return found ? (const void *)found->driver_data : NULL; +} + +static struct pci_dev *get_physfn_instead(struct pci_dev *virtfn) +{ + struct pci_dev *physfn = pci_physfn(virtfn); + + pci_dev_get(physfn); + pci_dev_put(virtfn); + return physfn; +} + static struct config_group *xe_config_make_device_group(struct config_group *group, const char *name) { unsigned int domain, bus, slot, function; - struct xe_config_device *dev; + struct xe_config_group_device *dev; + const struct xe_device_desc *match; struct pci_dev *pdev; + char canonical[16]; + int vfnumber = 0; int ret; - ret = sscanf(name, "%04x:%02x:%02x.%x", &domain, &bus, &slot, &function); + ret = sscanf(name, "%x:%x:%x.%x", &domain, &bus, &slot, &function); if (ret != 4) return ERR_PTR(-EINVAL); + ret = scnprintf(canonical, sizeof(canonical), "%04x:%02x:%02x.%d", domain, bus, + PCI_SLOT(PCI_DEVFN(slot, function)), + PCI_FUNC(PCI_DEVFN(slot, function))); + if (ret != 12 || strcmp(name, canonical)) + return ERR_PTR(-EINVAL); + pdev = pci_get_domain_bus_and_slot(domain, bus, PCI_DEVFN(slot, function)); + if (!pdev && function) + pdev = pci_get_domain_bus_and_slot(domain, bus, PCI_DEVFN(slot, 0)); + if (!pdev && slot) + pdev = pci_get_domain_bus_and_slot(domain, bus, PCI_DEVFN(0, 0)); if (!pdev) return ERR_PTR(-ENODEV); + + if (PCI_DEVFN(slot, function) != pdev->devfn) { + pdev = get_physfn_instead(pdev); + vfnumber = PCI_DEVFN(slot, function) - pdev->devfn; + if (!dev_is_pf(&pdev->dev) || vfnumber > pci_sriov_get_totalvfs(pdev)) { + pci_dev_put(pdev); + return ERR_PTR(-ENODEV); + } + } + + match = xe_match_desc(pdev); + if (match && vfnumber && !match->has_sriov) { + pci_info(pdev, "xe driver does not support VFs on this device\n"); + match = NULL; + } else if (!match) { + pci_info(pdev, "xe driver does not support configuration of this device\n"); + } + pci_dev_put(pdev); + if (!match) + return ERR_PTR(-ENOENT); + dev = kzalloc(sizeof(*dev), GFP_KERNEL); if (!dev) return ERR_PTR(-ENOMEM); - /* Default values */ - dev->engines_allowed = U64_MAX; + set_device_defaults(&dev->config); config_group_init_type_name(&dev->group, name, &xe_config_device_type); @@ -302,102 +469,145 @@ static struct configfs_subsystem xe_configfs = { }, }; -static struct xe_config_device *configfs_find_group(struct pci_dev *pdev) +static struct xe_config_group_device *find_xe_config_group_device(struct pci_dev *pdev) { struct config_item *item; - char name[64]; - - snprintf(name, sizeof(name), "%04x:%02x:%02x.%x", pci_domain_nr(pdev->bus), - pdev->bus->number, PCI_SLOT(pdev->devfn), PCI_FUNC(pdev->devfn)); mutex_lock(&xe_configfs.su_mutex); - item = config_group_find_item(&xe_configfs.su_group, name); + item = config_group_find_item(&xe_configfs.su_group, pci_name(pdev)); mutex_unlock(&xe_configfs.su_mutex); if (!item) return NULL; - return to_xe_config_device(item); + return to_xe_config_group_device(item); +} + +static void dump_custom_dev_config(struct pci_dev *pdev, + struct xe_config_group_device *dev) +{ +#define PRI_CUSTOM_ATTR(fmt_, attr_) do { \ + if (dev->config.attr_ != device_defaults.attr_) \ + pci_info(pdev, "configfs: " __stringify(attr_) " = " fmt_ "\n", \ + dev->config.attr_); \ + } while (0) + + PRI_CUSTOM_ATTR("%llx", engines_allowed); + PRI_CUSTOM_ATTR("%d", enable_psmi); + PRI_CUSTOM_ATTR("%d", survivability_mode); + +#undef PRI_CUSTOM_ATTR +} + +/** + * xe_configfs_check_device() - Test if device was configured by configfs + * @pdev: the &pci_dev device to test + * + * Try to find the configfs group that belongs to the specified pci device + * and print a diagnostic message if different than the default value. + */ +void xe_configfs_check_device(struct pci_dev *pdev) +{ + struct xe_config_group_device *dev = find_xe_config_group_device(pdev); + + if (!dev) + return; + + /* memcmp here is safe as both are zero-initialized */ + if (memcmp(&dev->config, &device_defaults, sizeof(dev->config))) { + pci_info(pdev, "Found custom settings in configfs\n"); + dump_custom_dev_config(pdev, dev); + } + + config_group_put(&dev->group); } /** * xe_configfs_get_survivability_mode - get configfs survivability mode attribute * @pdev: pci device * - * find the configfs group that belongs to the pci device and return - * the survivability mode attribute - * - * Return: survivability mode if config group is found, false otherwise + * Return: survivability_mode attribute in configfs */ bool xe_configfs_get_survivability_mode(struct pci_dev *pdev) { - struct xe_config_device *dev = configfs_find_group(pdev); + struct xe_config_group_device *dev = find_xe_config_group_device(pdev); bool mode; if (!dev) - return false; + return device_defaults.survivability_mode; - mode = dev->survivability_mode; - config_item_put(&dev->group.cg_item); + mode = dev->config.survivability_mode; + config_group_put(&dev->group); return mode; } /** - * xe_configfs_clear_survivability_mode - clear configfs survivability mode attribute + * xe_configfs_clear_survivability_mode - clear configfs survivability mode * @pdev: pci device - * - * find the configfs group that belongs to the pci device and clear survivability - * mode attribute */ void xe_configfs_clear_survivability_mode(struct pci_dev *pdev) { - struct xe_config_device *dev = configfs_find_group(pdev); + struct xe_config_group_device *dev = find_xe_config_group_device(pdev); if (!dev) return; - mutex_lock(&dev->lock); - dev->survivability_mode = 0; - mutex_unlock(&dev->lock); + guard(mutex)(&dev->lock); + dev->config.survivability_mode = 0; - config_item_put(&dev->group.cg_item); + config_group_put(&dev->group); } /** * xe_configfs_get_engines_allowed - get engine allowed mask from configfs * @pdev: pci device * - * Find the configfs group that belongs to the pci device and return - * the mask of engines allowed to be used. - * - * Return: engine mask with allowed engines + * Return: engine mask with allowed engines set in configfs */ u64 xe_configfs_get_engines_allowed(struct pci_dev *pdev) { - struct xe_config_device *dev = configfs_find_group(pdev); + struct xe_config_group_device *dev = find_xe_config_group_device(pdev); u64 engines_allowed; if (!dev) - return U64_MAX; + return device_defaults.engines_allowed; - engines_allowed = dev->engines_allowed; - config_item_put(&dev->group.cg_item); + engines_allowed = dev->config.engines_allowed; + config_group_put(&dev->group); return engines_allowed; } +/** + * xe_configfs_get_psmi_enabled - get configfs enable_psmi setting + * @pdev: pci device + * + * Return: enable_psmi setting in configfs + */ +bool xe_configfs_get_psmi_enabled(struct pci_dev *pdev) +{ + struct xe_config_group_device *dev = find_xe_config_group_device(pdev); + bool ret; + + if (!dev) + return false; + + ret = dev->config.enable_psmi; + config_item_put(&dev->group.cg_item); + + return ret; +} + int __init xe_configfs_init(void) { - struct config_group *root = &xe_configfs.su_group; int ret; - config_group_init(root); + config_group_init(&xe_configfs.su_group); mutex_init(&xe_configfs.su_mutex); ret = configfs_register_subsystem(&xe_configfs); if (ret) { - pr_err("Error %d while registering %s subsystem\n", - ret, root->cg_item.ci_namebuf); + mutex_destroy(&xe_configfs.su_mutex); return ret; } @@ -407,5 +617,5 @@ int __init xe_configfs_init(void) void __exit xe_configfs_exit(void) { configfs_unregister_subsystem(&xe_configfs); + mutex_destroy(&xe_configfs.su_mutex); } - diff --git a/drivers/gpu/drm/xe/xe_configfs.h b/drivers/gpu/drm/xe/xe_configfs.h index fb8764008089..58c8c3164000 100644 --- a/drivers/gpu/drm/xe/xe_configfs.h +++ b/drivers/gpu/drm/xe/xe_configfs.h @@ -13,15 +13,19 @@ struct pci_dev; #if IS_ENABLED(CONFIG_CONFIGFS_FS) int xe_configfs_init(void); void xe_configfs_exit(void); +void xe_configfs_check_device(struct pci_dev *pdev); bool xe_configfs_get_survivability_mode(struct pci_dev *pdev); void xe_configfs_clear_survivability_mode(struct pci_dev *pdev); u64 xe_configfs_get_engines_allowed(struct pci_dev *pdev); +bool xe_configfs_get_psmi_enabled(struct pci_dev *pdev); #else static inline int xe_configfs_init(void) { return 0; } static inline void xe_configfs_exit(void) { } +static inline void xe_configfs_check_device(struct pci_dev *pdev) { } static inline bool xe_configfs_get_survivability_mode(struct pci_dev *pdev) { return false; } static inline void xe_configfs_clear_survivability_mode(struct pci_dev *pdev) { } static inline u64 xe_configfs_get_engines_allowed(struct pci_dev *pdev) { return U64_MAX; } +static inline bool xe_configfs_get_psmi_enabled(struct pci_dev *pdev) { return false; } #endif #endif diff --git a/drivers/gpu/drm/xe/xe_debugfs.c b/drivers/gpu/drm/xe/xe_debugfs.c index 26e9d146ccbf..8d6df6bd885e 100644 --- a/drivers/gpu/drm/xe/xe_debugfs.c +++ b/drivers/gpu/drm/xe/xe_debugfs.c @@ -11,18 +11,22 @@ #include <drm/drm_debugfs.h> +#include "regs/xe_pmt.h" #include "xe_bo.h" #include "xe_device.h" #include "xe_force_wake.h" #include "xe_gt_debugfs.h" #include "xe_gt_printk.h" #include "xe_guc_ads.h" +#include "xe_mmio.h" #include "xe_pm.h" +#include "xe_psmi.h" #include "xe_pxp_debugfs.h" #include "xe_sriov.h" #include "xe_sriov_pf.h" #include "xe_step.h" #include "xe_wa.h" +#include "xe_vsec.h" #ifdef CONFIG_DRM_XE_DEBUG #include "xe_bo_evict.h" @@ -31,6 +35,24 @@ #endif DECLARE_FAULT_ATTR(gt_reset_failure); +DECLARE_FAULT_ATTR(inject_csc_hw_error); + +static void read_residency_counter(struct xe_device *xe, struct xe_mmio *mmio, + u32 offset, char *name, struct drm_printer *p) +{ + u64 residency = 0; + int ret; + + ret = xe_pmt_telem_read(to_pci_dev(xe->drm.dev), + xe_mmio_read32(mmio, PUNIT_TELEMETRY_GUID), + &residency, offset, sizeof(residency)); + if (ret != sizeof(residency)) { + drm_warn(&xe->drm, "%s counter failed to read, ret %d\n", name, ret); + return; + } + + drm_printf(p, "%s : %llu\n", name, residency); +} static struct xe_device *node_to_xe(struct drm_info_node *node) { @@ -102,12 +124,72 @@ static int workaround_info(struct seq_file *m, void *data) return 0; } +static int dgfx_pkg_residencies_show(struct seq_file *m, void *data) +{ + struct xe_device *xe; + struct xe_mmio *mmio; + struct drm_printer p; + + xe = node_to_xe(m->private); + p = drm_seq_file_printer(m); + xe_pm_runtime_get(xe); + mmio = xe_root_tile_mmio(xe); + struct { + u32 offset; + char *name; + } residencies[] = { + {BMG_G2_RESIDENCY_OFFSET, "Package G2"}, + {BMG_G6_RESIDENCY_OFFSET, "Package G6"}, + {BMG_G8_RESIDENCY_OFFSET, "Package G8"}, + {BMG_G10_RESIDENCY_OFFSET, "Package G10"}, + {BMG_MODS_RESIDENCY_OFFSET, "Package ModS"} + }; + + for (int i = 0; i < ARRAY_SIZE(residencies); i++) + read_residency_counter(xe, mmio, residencies[i].offset, residencies[i].name, &p); + + xe_pm_runtime_put(xe); + return 0; +} + +static int dgfx_pcie_link_residencies_show(struct seq_file *m, void *data) +{ + struct xe_device *xe; + struct xe_mmio *mmio; + struct drm_printer p; + + xe = node_to_xe(m->private); + p = drm_seq_file_printer(m); + xe_pm_runtime_get(xe); + mmio = xe_root_tile_mmio(xe); + + struct { + u32 offset; + char *name; + } residencies[] = { + {BMG_PCIE_LINK_L0_RESIDENCY_OFFSET, "PCIE LINK L0 RESIDENCY"}, + {BMG_PCIE_LINK_L1_RESIDENCY_OFFSET, "PCIE LINK L1 RESIDENCY"}, + {BMG_PCIE_LINK_L1_2_RESIDENCY_OFFSET, "PCIE LINK L1.2 RESIDENCY"} + }; + + for (int i = 0; i < ARRAY_SIZE(residencies); i++) + read_residency_counter(xe, mmio, residencies[i].offset, residencies[i].name, &p); + + xe_pm_runtime_put(xe); + return 0; +} + static const struct drm_info_list debugfs_list[] = { {"info", info, 0}, { .name = "sriov_info", .show = sriov_info, }, { .name = "workarounds", .show = workaround_info, }, }; +static const struct drm_info_list debugfs_residencies[] = { + { .name = "dgfx_pkg_residencies", .show = dgfx_pkg_residencies_show, }, + { .name = "dgfx_pcie_link_residencies", .show = dgfx_pcie_link_residencies_show, }, +}; + static int forcewake_open(struct inode *inode, struct file *file) { struct xe_device *xe = inode->i_private; @@ -247,20 +329,47 @@ static const struct file_operations atomic_svm_timeslice_ms_fops = { .write = atomic_svm_timeslice_ms_set, }; +static void create_tile_debugfs(struct xe_tile *tile, struct dentry *root) +{ + char name[8]; + + snprintf(name, sizeof(name), "tile%u", tile->id); + tile->debugfs = debugfs_create_dir(name, root); + if (IS_ERR(tile->debugfs)) + return; + + /* + * Store the xe_tile pointer as private data of the tile/ directory + * node so other tile specific attributes under that directory may + * refer to it by looking at its parent node private data. + */ + tile->debugfs->d_inode->i_private = tile; +} + void xe_debugfs_register(struct xe_device *xe) { struct ttm_device *bdev = &xe->ttm; struct drm_minor *minor = xe->drm.primary; struct dentry *root = minor->debugfs_root; struct ttm_resource_manager *man; + struct xe_tile *tile; struct xe_gt *gt; u32 mem_type; + u8 tile_id; u8 id; drm_debugfs_create_files(debugfs_list, ARRAY_SIZE(debugfs_list), root, minor); + if (xe->info.platform == XE_BATTLEMAGE) { + drm_debugfs_create_files(debugfs_residencies, + ARRAY_SIZE(debugfs_residencies), + root, minor); + fault_create_debugfs_attr("inject_csc_hw_error", root, + &inject_csc_hw_error); + } + debugfs_create_file("forcewake_all", 0400, root, xe, &forcewake_all_fops); @@ -288,11 +397,16 @@ void xe_debugfs_register(struct xe_device *xe) if (man) ttm_resource_manager_create_debugfs(man, root, "stolen_mm"); + for_each_tile(tile, xe, tile_id) + create_tile_debugfs(tile, root); + for_each_gt(gt, xe, id) xe_gt_debugfs_register(gt); xe_pxp_debugfs_register(xe->pxp); + xe_psmi_debugfs_register(xe); + fault_create_debugfs_attr("fail_gt_reset", root, >_reset_failure); if (IS_SRIOV_PF(xe)) diff --git a/drivers/gpu/drm/xe/xe_dep_job_types.h b/drivers/gpu/drm/xe/xe_dep_job_types.h new file mode 100644 index 000000000000..c6a484f24c8c --- /dev/null +++ b/drivers/gpu/drm/xe/xe_dep_job_types.h @@ -0,0 +1,29 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2025 Intel Corporation + */ + +#ifndef _XE_DEP_JOB_TYPES_H_ +#define _XE_DEP_JOB_TYPES_H_ + +#include <drm/gpu_scheduler.h> + +struct xe_dep_job; + +/** struct xe_dep_job_ops - Generic Xe dependency job operations */ +struct xe_dep_job_ops { + /** @run_job: Run generic Xe dependency job */ + struct dma_fence *(*run_job)(struct xe_dep_job *job); + /** @free_job: Free generic Xe dependency job */ + void (*free_job)(struct xe_dep_job *job); +}; + +/** struct xe_dep_job - Generic dependency Xe job */ +struct xe_dep_job { + /** @drm: base DRM scheduler job */ + struct drm_sched_job drm; + /** @ops: dependency job operations */ + const struct xe_dep_job_ops *ops; +}; + +#endif diff --git a/drivers/gpu/drm/xe/xe_dep_scheduler.c b/drivers/gpu/drm/xe/xe_dep_scheduler.c new file mode 100644 index 000000000000..9bd3bfd2e526 --- /dev/null +++ b/drivers/gpu/drm/xe/xe_dep_scheduler.c @@ -0,0 +1,143 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright © 2025 Intel Corporation + */ + +#include <linux/slab.h> + +#include <drm/gpu_scheduler.h> + +#include "xe_dep_job_types.h" +#include "xe_dep_scheduler.h" +#include "xe_device_types.h" + +/** + * DOC: Xe Dependency Scheduler + * + * The Xe dependency scheduler is a simple wrapper built around the DRM + * scheduler to execute jobs once their dependencies are resolved (i.e., all + * input fences specified as dependencies are signaled). The jobs that are + * executed contain virtual functions to run (execute) and free the job, + * allowing a single dependency scheduler to handle jobs performing different + * operations. + * + * Example use cases include deferred resource freeing, TLB invalidations after + * bind jobs, etc. + */ + +/** struct xe_dep_scheduler - Generic Xe dependency scheduler */ +struct xe_dep_scheduler { + /** @sched: DRM GPU scheduler */ + struct drm_gpu_scheduler sched; + /** @entity: DRM scheduler entity */ + struct drm_sched_entity entity; + /** @rcu: For safe freeing of exported dma fences */ + struct rcu_head rcu; +}; + +static struct dma_fence *xe_dep_scheduler_run_job(struct drm_sched_job *drm_job) +{ + struct xe_dep_job *dep_job = + container_of(drm_job, typeof(*dep_job), drm); + + return dep_job->ops->run_job(dep_job); +} + +static void xe_dep_scheduler_free_job(struct drm_sched_job *drm_job) +{ + struct xe_dep_job *dep_job = + container_of(drm_job, typeof(*dep_job), drm); + + dep_job->ops->free_job(dep_job); +} + +static const struct drm_sched_backend_ops sched_ops = { + .run_job = xe_dep_scheduler_run_job, + .free_job = xe_dep_scheduler_free_job, +}; + +/** + * xe_dep_scheduler_create() - Generic Xe dependency scheduler create + * @xe: Xe device + * @submit_wq: Submit workqueue struct (can be NULL) + * @name: Name of dependency scheduler + * @job_limit: Max dependency jobs that can be scheduled + * + * Create a generic Xe dependency scheduler and initialize internal DRM + * scheduler objects. + * + * Return: Generic Xe dependency scheduler object on success, ERR_PTR failure + */ +struct xe_dep_scheduler * +xe_dep_scheduler_create(struct xe_device *xe, + struct workqueue_struct *submit_wq, + const char *name, u32 job_limit) +{ + struct xe_dep_scheduler *dep_scheduler; + struct drm_gpu_scheduler *sched; + const struct drm_sched_init_args args = { + .ops = &sched_ops, + .submit_wq = submit_wq, + .num_rqs = 1, + .credit_limit = job_limit, + .timeout = MAX_SCHEDULE_TIMEOUT, + .name = name, + .dev = xe->drm.dev, + }; + int err; + + dep_scheduler = kzalloc(sizeof(*dep_scheduler), GFP_KERNEL); + if (!dep_scheduler) + return ERR_PTR(-ENOMEM); + + err = drm_sched_init(&dep_scheduler->sched, &args); + if (err) + goto err_free; + + sched = &dep_scheduler->sched; + err = drm_sched_entity_init(&dep_scheduler->entity, 0, &sched, 1, NULL); + if (err) + goto err_sched; + + init_rcu_head(&dep_scheduler->rcu); + + return dep_scheduler; + +err_sched: + drm_sched_fini(&dep_scheduler->sched); +err_free: + kfree(dep_scheduler); + + return ERR_PTR(err); +} + +/** + * xe_dep_scheduler_fini() - Generic Xe dependency scheduler finalize + * @dep_scheduler: Generic Xe dependency scheduler object + * + * Finalize internal DRM scheduler objects and free generic Xe dependency + * scheduler object + */ +void xe_dep_scheduler_fini(struct xe_dep_scheduler *dep_scheduler) +{ + drm_sched_entity_fini(&dep_scheduler->entity); + drm_sched_fini(&dep_scheduler->sched); + /* + * RCU free due sched being exported via DRM scheduler fences + * (timeline name). + */ + kfree_rcu(dep_scheduler, rcu); +} + +/** + * xe_dep_scheduler_entity() - Retrieve a generic Xe dependency scheduler + * DRM scheduler entity + * @dep_scheduler: Generic Xe dependency scheduler object + * + * Return: The generic Xe dependency scheduler's DRM scheduler entity + */ +struct drm_sched_entity * +xe_dep_scheduler_entity(struct xe_dep_scheduler *dep_scheduler) +{ + return &dep_scheduler->entity; +} diff --git a/drivers/gpu/drm/xe/xe_dep_scheduler.h b/drivers/gpu/drm/xe/xe_dep_scheduler.h new file mode 100644 index 000000000000..853961eec64b --- /dev/null +++ b/drivers/gpu/drm/xe/xe_dep_scheduler.h @@ -0,0 +1,21 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2025 Intel Corporation + */ + +#include <linux/types.h> + +struct drm_sched_entity; +struct workqueue_struct; +struct xe_dep_scheduler; +struct xe_device; + +struct xe_dep_scheduler * +xe_dep_scheduler_create(struct xe_device *xe, + struct workqueue_struct *submit_wq, + const char *name, u32 job_limit); + +void xe_dep_scheduler_fini(struct xe_dep_scheduler *dep_scheduler); + +struct drm_sched_entity * +xe_dep_scheduler_entity(struct xe_dep_scheduler *dep_scheduler); diff --git a/drivers/gpu/drm/xe/xe_device.c b/drivers/gpu/drm/xe/xe_device.c index 6ece4defa9df..9e4773a17ef8 100644 --- a/drivers/gpu/drm/xe/xe_device.c +++ b/drivers/gpu/drm/xe/xe_device.c @@ -54,6 +54,7 @@ #include "xe_pcode.h" #include "xe_pm.h" #include "xe_pmu.h" +#include "xe_psmi.h" #include "xe_pxp.h" #include "xe_query.h" #include "xe_shrinker.h" @@ -63,7 +64,9 @@ #include "xe_ttm_stolen_mgr.h" #include "xe_ttm_sys_mgr.h" #include "xe_vm.h" +#include "xe_vm_madvise.h" #include "xe_vram.h" +#include "xe_vram_types.h" #include "xe_vsec.h" #include "xe_wait_user_fence.h" #include "xe_wa.h" @@ -200,6 +203,9 @@ static const struct drm_ioctl_desc xe_ioctls[] = { DRM_IOCTL_DEF_DRV(XE_WAIT_USER_FENCE, xe_wait_user_fence_ioctl, DRM_RENDER_ALLOW), DRM_IOCTL_DEF_DRV(XE_OBSERVATION, xe_observation_ioctl, DRM_RENDER_ALLOW), + DRM_IOCTL_DEF_DRV(XE_MADVISE, xe_vm_madvise_ioctl, DRM_RENDER_ALLOW), + DRM_IOCTL_DEF_DRV(XE_VM_QUERY_MEM_RANGE_ATTRS, xe_vm_query_vmas_attrs_ioctl, + DRM_RENDER_ALLOW), }; static long xe_drm_ioctl(struct file *file, unsigned int cmd, unsigned long arg) @@ -688,6 +694,21 @@ static void sriov_update_device_info(struct xe_device *xe) } } +static int xe_device_vram_alloc(struct xe_device *xe) +{ + struct xe_vram_region *vram; + + if (!IS_DGFX(xe)) + return 0; + + vram = drmm_kzalloc(&xe->drm, sizeof(*vram), GFP_KERNEL); + if (!vram) + return -ENOMEM; + + xe->mem.vram = vram; + return 0; +} + /** * xe_device_probe_early: Device early probe * @xe: xe device instance @@ -722,7 +743,7 @@ int xe_device_probe_early(struct xe_device *xe) * possible, but still return the previous error for error * propagation */ - err = xe_survivability_mode_enable(xe); + err = xe_survivability_mode_boot_enable(xe); if (err) return err; @@ -735,6 +756,10 @@ int xe_device_probe_early(struct xe_device *xe) xe->wedged.mode = xe_modparam.wedged_mode; + err = xe_device_vram_alloc(xe); + if (err) + return err; + return 0; } ALLOW_ERROR_INJECTION(xe_device_probe_early, ERRNO); /* See xe_pci_probe() */ @@ -863,7 +888,7 @@ int xe_device_probe(struct xe_device *xe) } if (xe->tiles->media_gt && - XE_WA(xe->tiles->media_gt, 15015404425_disable)) + XE_GT_WA(xe->tiles->media_gt, 15015404425_disable)) XE_DEVICE_WA_DISABLE(xe, 15015404425); err = xe_devcoredump_init(xe); @@ -888,6 +913,10 @@ int xe_device_probe(struct xe_device *xe) if (err) return err; + err = xe_psmi_init(xe); + if (err) + return err; + err = drm_dev_register(&xe->drm, 0); if (err) return err; @@ -921,6 +950,10 @@ int xe_device_probe(struct xe_device *xe) xe_vsec_init(xe); + err = xe_sriov_late_init(xe); + if (err) + goto err_unregister_display; + return devm_add_action_or_reset(xe->drm.dev, xe_device_sanitize, xe); err_unregister_display: @@ -1019,7 +1052,7 @@ void xe_device_l2_flush(struct xe_device *xe) gt = xe_root_mmio_gt(xe); - if (!XE_WA(gt, 16023588340)) + if (!XE_GT_WA(gt, 16023588340)) return; fw_ref = xe_force_wake_get(gt_to_fw(gt), XE_FW_GT); @@ -1063,7 +1096,7 @@ void xe_device_td_flush(struct xe_device *xe) return; root_gt = xe_root_mmio_gt(xe); - if (XE_WA(root_gt, 16023588340)) { + if (XE_GT_WA(root_gt, 16023588340)) { /* A transient flush is not sufficient: flush the L2 */ xe_device_l2_flush(xe); } else { @@ -1134,11 +1167,63 @@ static void xe_device_wedged_fini(struct drm_device *drm, void *arg) } /** + * DOC: Xe Device Wedging + * + * Xe driver uses drm device wedged uevent as documented in Documentation/gpu/drm-uapi.rst. + * When device is in wedged state, every IOCTL will be blocked and GT cannot be + * used. Certain critical errors like gt reset failure, firmware failures can cause + * the device to be wedged. The default recovery method for a wedged state + * is rebind/bus-reset. + * + * Another recovery method is vendor-specific. Below are the cases that send + * ``WEDGED=vendor-specific`` recovery method in drm device wedged uevent. + * + * Case: Firmware Flash + * -------------------- + * + * Identification Hint + * +++++++++++++++++++ + * + * ``WEDGED=vendor-specific`` drm device wedged uevent with + * :ref:`Runtime Survivability mode <xe-survivability-mode>` is used to notify + * admin/userspace consumer about the need for a firmware flash. + * + * Recovery Procedure + * ++++++++++++++++++ + * + * Once ``WEDGED=vendor-specific`` drm device wedged uevent is received, follow + * the below steps + * + * - Check Runtime Survivability mode sysfs. + * If enabled, firmware flash is required to recover the device. + * + * /sys/bus/pci/devices/<device>/survivability_mode + * + * - Admin/userpsace consumer can use firmware flashing tools like fwupd to flash + * firmware and restore device to normal operation. + */ + +/** + * xe_device_set_wedged_method - Set wedged recovery method + * @xe: xe device instance + * @method: recovery method to set + * + * Set wedged recovery method to be sent in drm wedged uevent. + */ +void xe_device_set_wedged_method(struct xe_device *xe, unsigned long method) +{ + xe->wedged.method = method; +} + +/** * xe_device_declare_wedged - Declare device wedged * @xe: xe device instance * - * This is a final state that can only be cleared with a module - * re-probe (unbind + bind). + * This is a final state that can only be cleared with the recovery method + * specified in the drm wedged uevent. The method can be set using + * xe_device_set_wedged_method before declaring the device as wedged. If no method + * is set, reprobe (unbind/re-bind) will be sent by default. + * * In this state every IOCTL will be blocked so the GT cannot be used. * In general it will be called upon any critical error such as gt reset * failure or guc loading failure. Userspace will be notified of this state @@ -1172,13 +1257,18 @@ void xe_device_declare_wedged(struct xe_device *xe) "IOCTLs and executions are blocked. Only a rebind may clear the failure\n" "Please file a _new_ bug report at https://gitlab.freedesktop.org/drm/xe/kernel/issues/new\n", dev_name(xe->drm.dev)); - - /* Notify userspace of wedged device */ - drm_dev_wedged_event(&xe->drm, - DRM_WEDGE_RECOVERY_REBIND | DRM_WEDGE_RECOVERY_BUS_RESET, - NULL); } for_each_gt(gt, xe, id) xe_gt_declare_wedged(gt); + + if (xe_device_wedged(xe)) { + /* If no wedge recovery method is set, use default */ + if (!xe->wedged.method) + xe_device_set_wedged_method(xe, DRM_WEDGE_RECOVERY_REBIND | + DRM_WEDGE_RECOVERY_BUS_RESET); + + /* Notify userspace of wedged device */ + drm_dev_wedged_event(&xe->drm, xe->wedged.method, NULL); + } } diff --git a/drivers/gpu/drm/xe/xe_device.h b/drivers/gpu/drm/xe/xe_device.h index bc802e066a7d..32cc6323b7f6 100644 --- a/drivers/gpu/drm/xe/xe_device.h +++ b/drivers/gpu/drm/xe/xe_device.h @@ -187,6 +187,7 @@ static inline bool xe_device_wedged(struct xe_device *xe) return atomic_read(&xe->wedged.flag); } +void xe_device_set_wedged_method(struct xe_device *xe, unsigned long method); void xe_device_declare_wedged(struct xe_device *xe); struct xe_file *xe_file_get(struct xe_file *xef); diff --git a/drivers/gpu/drm/xe/xe_device_sysfs.c b/drivers/gpu/drm/xe/xe_device_sysfs.c index bd9015761aa0..6ee422594b56 100644 --- a/drivers/gpu/drm/xe/xe_device_sysfs.c +++ b/drivers/gpu/drm/xe/xe_device_sysfs.c @@ -76,7 +76,7 @@ lb_fan_control_version_show(struct device *dev, struct device_attribute *attr, c { struct xe_device *xe = pdev_to_xe_device(to_pci_dev(dev)); struct xe_tile *root = xe_device_get_root_tile(xe); - u32 cap, ver_low = FAN_TABLE, ver_high = FAN_TABLE; + u32 cap = 0, ver_low = FAN_TABLE, ver_high = FAN_TABLE; u16 major = 0, minor = 0, hotfix = 0, build = 0; int ret; @@ -115,7 +115,7 @@ lb_voltage_regulator_version_show(struct device *dev, struct device_attribute *a { struct xe_device *xe = pdev_to_xe_device(to_pci_dev(dev)); struct xe_tile *root = xe_device_get_root_tile(xe); - u32 cap, ver_low = VR_CONFIG, ver_high = VR_CONFIG; + u32 cap = 0, ver_low = VR_CONFIG, ver_high = VR_CONFIG; u16 major = 0, minor = 0, hotfix = 0, build = 0; int ret; @@ -153,7 +153,7 @@ static int late_bind_create_files(struct device *dev) { struct xe_device *xe = pdev_to_xe_device(to_pci_dev(dev)); struct xe_tile *root = xe_device_get_root_tile(xe); - u32 cap; + u32 cap = 0; int ret; xe_pm_runtime_get(xe); @@ -186,7 +186,7 @@ static void late_bind_remove_files(struct device *dev) { struct xe_device *xe = pdev_to_xe_device(to_pci_dev(dev)); struct xe_tile *root = xe_device_get_root_tile(xe); - u32 cap; + u32 cap = 0; int ret; xe_pm_runtime_get(xe); diff --git a/drivers/gpu/drm/xe/xe_device_types.h b/drivers/gpu/drm/xe/xe_device_types.h index d4d2c6854790..092004d14db2 100644 --- a/drivers/gpu/drm/xe/xe_device_types.h +++ b/drivers/gpu/drm/xe/xe_device_types.h @@ -10,7 +10,6 @@ #include <drm/drm_device.h> #include <drm/drm_file.h> -#include <drm/drm_pagemap.h> #include <drm/ttm/ttm_device.h> #include "xe_devcoredump_types.h" @@ -24,9 +23,9 @@ #include "xe_sriov_pf_types.h" #include "xe_sriov_types.h" #include "xe_sriov_vf_types.h" +#include "xe_sriov_vf_ccs_types.h" #include "xe_step_types.h" #include "xe_survivability_mode_types.h" -#include "xe_ttm_vram_mgr_types.h" #if IS_ENABLED(CONFIG_DRM_XE_DEBUG) #define TEST_VM_OPS_ERROR @@ -39,6 +38,7 @@ struct xe_ggtt; struct xe_i2c; struct xe_pat_ops; struct xe_pxp; +struct xe_vram_region; #define XE_BO_INVALID_OFFSET LONG_MAX @@ -72,61 +72,6 @@ struct xe_pxp; struct xe_tile * : (tile__)->xe) /** - * struct xe_vram_region - memory region structure - * This is used to describe a memory region in xe - * device, such as HBM memory or CXL extension memory. - */ -struct xe_vram_region { - /** @io_start: IO start address of this VRAM instance */ - resource_size_t io_start; - /** - * @io_size: IO size of this VRAM instance - * - * This represents how much of this VRAM we can access - * via the CPU through the VRAM BAR. This can be smaller - * than @usable_size, in which case only part of VRAM is CPU - * accessible (typically the first 256M). This - * configuration is known as small-bar. - */ - resource_size_t io_size; - /** @dpa_base: This memory regions's DPA (device physical address) base */ - resource_size_t dpa_base; - /** - * @usable_size: usable size of VRAM - * - * Usable size of VRAM excluding reserved portions - * (e.g stolen mem) - */ - resource_size_t usable_size; - /** - * @actual_physical_size: Actual VRAM size - * - * Actual VRAM size including reserved portions - * (e.g stolen mem) - */ - resource_size_t actual_physical_size; - /** @mapping: pointer to VRAM mappable space */ - void __iomem *mapping; - /** @ttm: VRAM TTM manager */ - struct xe_ttm_vram_mgr ttm; -#if IS_ENABLED(CONFIG_DRM_XE_PAGEMAP) - /** @pagemap: Used to remap device memory as ZONE_DEVICE */ - struct dev_pagemap pagemap; - /** - * @dpagemap: The struct drm_pagemap of the ZONE_DEVICE memory - * pages of this tile. - */ - struct drm_pagemap dpagemap; - /** - * @hpa_base: base host physical address - * - * This is generated when remap device memory as ZONE_DEVICE - */ - resource_size_t hpa_base; -#endif -}; - -/** * struct xe_mmio - register mmio structure * * Represents an MMIO region that the CPU may use to access registers. A @@ -216,7 +161,7 @@ struct xe_tile { * Although VRAM is associated with a specific tile, it can * still be accessed by all tiles' GTs. */ - struct xe_vram_region vram; + struct xe_vram_region *vram; /** @mem.ggtt: Global graphics translation table */ struct xe_ggtt *ggtt; @@ -238,12 +183,18 @@ struct xe_tile { struct { /** @sriov.vf.ggtt_balloon: GGTT regions excluded from use. */ struct xe_ggtt_node *ggtt_balloon[2]; + + /** @sriov.vf.ccs: CCS read and write contexts for VF. */ + struct xe_tile_vf_ccs ccs[XE_SRIOV_VF_CCS_CTX_COUNT]; } vf; } sriov; /** @memirq: Memory Based Interrupts. */ struct xe_memirq memirq; + /** @csc_hw_error_work: worker to report CSC HW errors */ + struct work_struct csc_hw_error_work; + /** @pcode: tile's PCODE */ struct { /** @pcode.lock: protecting tile's PCODE mailbox data */ @@ -255,6 +206,9 @@ struct xe_tile { /** @sysfs: sysfs' kobj used by xe_tile_sysfs */ struct kobject *sysfs; + + /** @debugfs: debugfs directory associated with this tile */ + struct dentry *debugfs; }; /** @@ -336,8 +290,8 @@ struct xe_device { u8 has_mbx_power_limits:1; /** @info.has_pxp: Device has PXP support */ u8 has_pxp:1; - /** @info.has_range_tlb_invalidation: Has range based TLB invalidations */ - u8 has_range_tlb_invalidation:1; + /** @info.has_range_tlb_inval: Has range based TLB invalidations */ + u8 has_range_tlb_inval:1; /** @info.has_sriov: Supports SR-IOV */ u8 has_sriov:1; /** @info.has_usm: Device has unified shared memory support */ @@ -412,7 +366,7 @@ struct xe_device { /** @mem: memory info for device */ struct { /** @mem.vram: VRAM info for device */ - struct xe_vram_region vram; + struct xe_vram_region *vram; /** @mem.sys_mgr: system TTM manager */ struct ttm_resource_manager sys_mgr; /** @mem.sys_mgr: system memory shrinker. */ @@ -476,7 +430,7 @@ struct xe_device { /** @ordered_wq: used to serialize compute mode resume */ struct workqueue_struct *ordered_wq; - /** @unordered_wq: used to serialize unordered work, mostly display */ + /** @unordered_wq: used to serialize unordered work */ struct workqueue_struct *unordered_wq; /** @destroy_wq: used to serialize user destroy work, like queue */ @@ -590,6 +544,8 @@ struct xe_device { atomic_t flag; /** @wedged.mode: Mode controlled by kernel parameter and debugfs */ int mode; + /** @wedged.method: Recovery method to be sent in the drm device wedged uevent */ + unsigned long method; } wedged; /** @bo_device: Struct to control async free of BOs */ @@ -625,6 +581,14 @@ struct xe_device { atomic64_t global_total_pages; #endif + /** @psmi: GPU debugging via additional validation HW */ + struct { + /** @psmi.capture_obj: PSMI buffer for VRAM */ + struct xe_bo *capture_obj[XE_MAX_TILES_PER_DEVICE + 1]; + /** @psmi.region_mask: Mask of valid memory regions */ + u8 region_mask; + } psmi; + /* private: */ #if IS_ENABLED(CONFIG_DRM_XE_DISPLAY) @@ -658,7 +622,6 @@ struct xe_device { struct { unsigned int hpll_freq; unsigned int czclk_freq; - unsigned int fsb_freq, mem_freq, is_ddr3; }; #endif }; diff --git a/drivers/gpu/drm/xe/xe_eu_stall.c b/drivers/gpu/drm/xe/xe_eu_stall.c index af7916315ac6..fdd514fec5ef 100644 --- a/drivers/gpu/drm/xe/xe_eu_stall.c +++ b/drivers/gpu/drm/xe/xe_eu_stall.c @@ -649,7 +649,7 @@ static int xe_eu_stall_stream_enable(struct xe_eu_stall_data_stream *stream) return -ETIMEDOUT; } - if (XE_WA(gt, 22016596838)) + if (XE_GT_WA(gt, 22016596838)) xe_gt_mcr_multicast_write(gt, ROW_CHICKEN2, _MASKED_BIT_ENABLE(DISABLE_DOP_GATING)); @@ -805,7 +805,7 @@ static int xe_eu_stall_disable_locked(struct xe_eu_stall_data_stream *stream) cancel_delayed_work_sync(&stream->buf_poll_work); - if (XE_WA(gt, 22016596838)) + if (XE_GT_WA(gt, 22016596838)) xe_gt_mcr_multicast_write(gt, ROW_CHICKEN2, _MASKED_BIT_DISABLE(DISABLE_DOP_GATING)); diff --git a/drivers/gpu/drm/xe/xe_exec_queue.c b/drivers/gpu/drm/xe/xe_exec_queue.c index 8991b4aed440..063c89d981e5 100644 --- a/drivers/gpu/drm/xe/xe_exec_queue.c +++ b/drivers/gpu/drm/xe/xe_exec_queue.c @@ -12,6 +12,7 @@ #include <drm/drm_file.h> #include <uapi/drm/xe_drm.h> +#include "xe_dep_scheduler.h" #include "xe_device.h" #include "xe_gt.h" #include "xe_hw_engine_class_sysfs.h" @@ -39,6 +40,12 @@ static int exec_queue_user_extensions(struct xe_device *xe, struct xe_exec_queue static void __xe_exec_queue_free(struct xe_exec_queue *q) { + int i; + + for (i = 0; i < XE_EXEC_QUEUE_TLB_INVAL_COUNT; ++i) + if (q->tlb_inval[i].dep_scheduler) + xe_dep_scheduler_fini(q->tlb_inval[i].dep_scheduler); + if (xe_exec_queue_uses_pxp(q)) xe_pxp_exec_queue_remove(gt_to_xe(q->gt)->pxp, q); if (q->vm) @@ -50,6 +57,39 @@ static void __xe_exec_queue_free(struct xe_exec_queue *q) kfree(q); } +static int alloc_dep_schedulers(struct xe_device *xe, struct xe_exec_queue *q) +{ + struct xe_tile *tile = gt_to_tile(q->gt); + int i; + + for (i = 0; i < XE_EXEC_QUEUE_TLB_INVAL_COUNT; ++i) { + struct xe_dep_scheduler *dep_scheduler; + struct xe_gt *gt; + struct workqueue_struct *wq; + + if (i == XE_EXEC_QUEUE_TLB_INVAL_PRIMARY_GT) + gt = tile->primary_gt; + else + gt = tile->media_gt; + + if (!gt) + continue; + + wq = gt->tlb_inval.job_wq; + +#define MAX_TLB_INVAL_JOBS 16 /* Picking a reasonable value */ + dep_scheduler = xe_dep_scheduler_create(xe, wq, q->name, + MAX_TLB_INVAL_JOBS); + if (IS_ERR(dep_scheduler)) + return PTR_ERR(dep_scheduler); + + q->tlb_inval[i].dep_scheduler = dep_scheduler; + } +#undef MAX_TLB_INVAL_JOBS + + return 0; +} + static struct xe_exec_queue *__xe_exec_queue_alloc(struct xe_device *xe, struct xe_vm *vm, u32 logical_mask, @@ -94,6 +134,14 @@ static struct xe_exec_queue *__xe_exec_queue_alloc(struct xe_device *xe, else q->sched_props.priority = XE_EXEC_QUEUE_PRIORITY_NORMAL; + if (q->flags & (EXEC_QUEUE_FLAG_MIGRATE | EXEC_QUEUE_FLAG_VM)) { + err = alloc_dep_schedulers(xe, q); + if (err) { + __xe_exec_queue_free(q); + return ERR_PTR(err); + } + } + if (vm) q->vm = xe_vm_get(vm); @@ -742,6 +790,21 @@ int xe_exec_queue_get_property_ioctl(struct drm_device *dev, void *data, } /** + * xe_exec_queue_lrc() - Get the LRC from exec queue. + * @q: The exec_queue. + * + * Retrieves the primary LRC for the exec queue. Note that this function + * returns only the first LRC instance, even when multiple parallel LRCs + * are configured. + * + * Return: Pointer to LRC on success, error on failure + */ +struct xe_lrc *xe_exec_queue_lrc(struct xe_exec_queue *q) +{ + return q->lrc[0]; +} + +/** * xe_exec_queue_is_lr() - Whether an exec_queue is long-running * @q: The exec_queue * @@ -1028,3 +1091,51 @@ int xe_exec_queue_last_fence_test_dep(struct xe_exec_queue *q, struct xe_vm *vm) return err; } + +/** + * xe_exec_queue_contexts_hwsp_rebase - Re-compute GGTT references + * within all LRCs of a queue. + * @q: the &xe_exec_queue struct instance containing target LRCs + * @scratch: scratch buffer to be used as temporary storage + * + * Returns: zero on success, negative error code on failure + */ +int xe_exec_queue_contexts_hwsp_rebase(struct xe_exec_queue *q, void *scratch) +{ + int i; + int err = 0; + + for (i = 0; i < q->width; ++i) { + xe_lrc_update_memirq_regs_with_address(q->lrc[i], q->hwe, scratch); + xe_lrc_update_hwctx_regs_with_address(q->lrc[i]); + err = xe_lrc_setup_wa_bb_with_scratch(q->lrc[i], q->hwe, scratch); + if (err) + break; + } + + return err; +} + +/** + * xe_exec_queue_jobs_ring_restore - Re-emit ring commands of requests pending on given queue. + * @q: the &xe_exec_queue struct instance + */ +void xe_exec_queue_jobs_ring_restore(struct xe_exec_queue *q) +{ + struct xe_gpu_scheduler *sched = &q->guc->sched; + struct xe_sched_job *job; + + /* + * This routine is used within VF migration recovery. This means + * using the lock here introduces a restriction: we cannot wait + * for any GFX HW response while the lock is taken. + */ + spin_lock(&sched->base.job_list_lock); + list_for_each_entry(job, &sched->base.pending_list, drm.list) { + if (xe_sched_job_is_error(job)) + continue; + + q->ring_ops->emit_job(job); + } + spin_unlock(&sched->base.job_list_lock); +} diff --git a/drivers/gpu/drm/xe/xe_exec_queue.h b/drivers/gpu/drm/xe/xe_exec_queue.h index 17bc50a7f05a..15ec852e7f7e 100644 --- a/drivers/gpu/drm/xe/xe_exec_queue.h +++ b/drivers/gpu/drm/xe/xe_exec_queue.h @@ -90,4 +90,9 @@ int xe_exec_queue_last_fence_test_dep(struct xe_exec_queue *q, struct xe_vm *vm); void xe_exec_queue_update_run_ticks(struct xe_exec_queue *q); +int xe_exec_queue_contexts_hwsp_rebase(struct xe_exec_queue *q, void *scratch); + +void xe_exec_queue_jobs_ring_restore(struct xe_exec_queue *q); + +struct xe_lrc *xe_exec_queue_lrc(struct xe_exec_queue *q); #endif diff --git a/drivers/gpu/drm/xe/xe_exec_queue_types.h b/drivers/gpu/drm/xe/xe_exec_queue_types.h index cc1cffb5c87f..ba443a497b38 100644 --- a/drivers/gpu/drm/xe/xe_exec_queue_types.h +++ b/drivers/gpu/drm/xe/xe_exec_queue_types.h @@ -87,6 +87,8 @@ struct xe_exec_queue { #define EXEC_QUEUE_FLAG_HIGH_PRIORITY BIT(4) /* flag to indicate low latency hint to guc */ #define EXEC_QUEUE_FLAG_LOW_LATENCY BIT(5) +/* for migration (kernel copy, clear, bind) jobs */ +#define EXEC_QUEUE_FLAG_MIGRATE BIT(6) /** * @flags: flags for this exec queue, should statically setup aside from ban @@ -132,6 +134,19 @@ struct xe_exec_queue { struct list_head link; } lr; +#define XE_EXEC_QUEUE_TLB_INVAL_PRIMARY_GT 0 +#define XE_EXEC_QUEUE_TLB_INVAL_MEDIA_GT 1 +#define XE_EXEC_QUEUE_TLB_INVAL_COUNT (XE_EXEC_QUEUE_TLB_INVAL_MEDIA_GT + 1) + + /** @tlb_inval: TLB invalidations exec queue state */ + struct { + /** + * @tlb_inval.dep_scheduler: The TLB invalidation + * dependency scheduler + */ + struct xe_dep_scheduler *dep_scheduler; + } tlb_inval[XE_EXEC_QUEUE_TLB_INVAL_COUNT]; + /** @pxp: PXP info tracking */ struct { /** @pxp.type: PXP session type used by this queue */ diff --git a/drivers/gpu/drm/xe/xe_gen_wa_oob.c b/drivers/gpu/drm/xe/xe_gen_wa_oob.c index 6581cb0f0e59..247e41c1c48d 100644 --- a/drivers/gpu/drm/xe/xe_gen_wa_oob.c +++ b/drivers/gpu/drm/xe/xe_gen_wa_oob.c @@ -123,11 +123,19 @@ static int parse(FILE *input, FILE *csource, FILE *cheader, char *prefix) return 0; } +/* Avoid GNU vs POSIX basename() discrepancy, just use our own */ +static const char *xbasename(const char *s) +{ + const char *p = strrchr(s, '/'); + + return p ? p + 1 : s; +} + static int fn_to_prefix(const char *fn, char *prefix, size_t size) { size_t len; - fn = basename(fn); + fn = xbasename(fn); len = strlen(fn); if (len > size - 1) diff --git a/drivers/gpu/drm/xe/xe_ggtt.c b/drivers/gpu/drm/xe/xe_ggtt.c index 29d4d3f51da1..71c7690a92b3 100644 --- a/drivers/gpu/drm/xe/xe_ggtt.c +++ b/drivers/gpu/drm/xe/xe_ggtt.c @@ -23,13 +23,13 @@ #include "xe_device.h" #include "xe_gt.h" #include "xe_gt_printk.h" -#include "xe_gt_tlb_invalidation.h" #include "xe_map.h" #include "xe_mmio.h" #include "xe_pm.h" #include "xe_res_cursor.h" #include "xe_sriov.h" #include "xe_tile_sriov_vf.h" +#include "xe_tlb_inval.h" #include "xe_wa.h" #include "xe_wopcm.h" @@ -106,10 +106,10 @@ static unsigned int probe_gsm_size(struct pci_dev *pdev) static void ggtt_update_access_counter(struct xe_ggtt *ggtt) { struct xe_tile *tile = ggtt->tile; - struct xe_gt *affected_gt = XE_WA(tile->primary_gt, 22019338487) ? + struct xe_gt *affected_gt = XE_GT_WA(tile->primary_gt, 22019338487) ? tile->primary_gt : tile->media_gt; struct xe_mmio *mmio = &affected_gt->mmio; - u32 max_gtt_writes = XE_WA(ggtt->tile->primary_gt, 22019338487) ? 1100 : 63; + u32 max_gtt_writes = XE_GT_WA(ggtt->tile->primary_gt, 22019338487) ? 1100 : 63; /* * Wa_22019338487: GMD_ID is a RO register, a dummy write forces gunit * to wait for completion of prior GTT writes before letting this through. @@ -284,8 +284,8 @@ int xe_ggtt_init_early(struct xe_ggtt *ggtt) if (GRAPHICS_VERx100(xe) >= 1270) ggtt->pt_ops = (ggtt->tile->media_gt && - XE_WA(ggtt->tile->media_gt, 22019338487)) || - XE_WA(ggtt->tile->primary_gt, 22019338487) ? + XE_GT_WA(ggtt->tile->media_gt, 22019338487)) || + XE_GT_WA(ggtt->tile->primary_gt, 22019338487) ? &xelpg_pt_wa_ops : &xelpg_pt_ops; else ggtt->pt_ops = &xelp_pt_ops; @@ -438,9 +438,8 @@ static void ggtt_invalidate_gt_tlb(struct xe_gt *gt) if (!gt) return; - err = xe_gt_tlb_invalidation_ggtt(gt); - if (err) - drm_warn(>_to_xe(gt)->drm, "xe_gt_tlb_invalidation_ggtt error=%d", err); + err = xe_tlb_inval_ggtt(>->tlb_inval); + xe_gt_WARN(gt, err, "Failed to invalidate GGTT (%pe)", ERR_PTR(err)); } static void xe_ggtt_invalidate(struct xe_ggtt *ggtt) diff --git a/drivers/gpu/drm/xe/xe_gpu_scheduler.c b/drivers/gpu/drm/xe/xe_gpu_scheduler.c index 869b43a4151d..455ccaf17314 100644 --- a/drivers/gpu/drm/xe/xe_gpu_scheduler.c +++ b/drivers/gpu/drm/xe/xe_gpu_scheduler.c @@ -101,6 +101,19 @@ void xe_sched_submission_stop(struct xe_gpu_scheduler *sched) cancel_work_sync(&sched->work_process_msg); } +/** + * xe_sched_submission_stop_async - Stop further runs of submission tasks on a scheduler. + * @sched: the &xe_gpu_scheduler struct instance + * + * This call disables further runs of scheduling work queue. It does not wait + * for any in-progress runs to finish, only makes sure no further runs happen + * afterwards. + */ +void xe_sched_submission_stop_async(struct xe_gpu_scheduler *sched) +{ + drm_sched_wqueue_stop(&sched->base); +} + void xe_sched_submission_resume_tdr(struct xe_gpu_scheduler *sched) { drm_sched_resume_timeout(&sched->base, sched->base.timeout); diff --git a/drivers/gpu/drm/xe/xe_gpu_scheduler.h b/drivers/gpu/drm/xe/xe_gpu_scheduler.h index 308061f0cf37..e548b2aed95a 100644 --- a/drivers/gpu/drm/xe/xe_gpu_scheduler.h +++ b/drivers/gpu/drm/xe/xe_gpu_scheduler.h @@ -21,6 +21,7 @@ void xe_sched_fini(struct xe_gpu_scheduler *sched); void xe_sched_submission_start(struct xe_gpu_scheduler *sched); void xe_sched_submission_stop(struct xe_gpu_scheduler *sched); +void xe_sched_submission_stop_async(struct xe_gpu_scheduler *sched); void xe_sched_submission_resume_tdr(struct xe_gpu_scheduler *sched); diff --git a/drivers/gpu/drm/xe/xe_gsc.c b/drivers/gpu/drm/xe/xe_gsc.c index 1d84bf2f2cef..f5ae28af60d4 100644 --- a/drivers/gpu/drm/xe/xe_gsc.c +++ b/drivers/gpu/drm/xe/xe_gsc.c @@ -266,7 +266,7 @@ static int gsc_upload_and_init(struct xe_gsc *gsc) unsigned int fw_ref; int ret; - if (XE_WA(tile->primary_gt, 14018094691)) { + if (XE_GT_WA(tile->primary_gt, 14018094691)) { fw_ref = xe_force_wake_get(gt_to_fw(tile->primary_gt), XE_FORCEWAKE_ALL); /* @@ -281,7 +281,7 @@ static int gsc_upload_and_init(struct xe_gsc *gsc) ret = gsc_upload(gsc); - if (XE_WA(tile->primary_gt, 14018094691)) + if (XE_GT_WA(tile->primary_gt, 14018094691)) xe_force_wake_put(gt_to_fw(tile->primary_gt), fw_ref); if (ret) @@ -593,7 +593,7 @@ void xe_gsc_wa_14015076503(struct xe_gt *gt, bool prep) u32 gs1_clr = prep ? 0 : HECI_H_GS1_ER_PREP; /* WA only applies if the GSC is loaded */ - if (!XE_WA(gt, 14015076503) || !gsc_fw_is_loaded(gt)) + if (!XE_GT_WA(gt, 14015076503) || !gsc_fw_is_loaded(gt)) return; xe_mmio_rmw32(>->mmio, HECI_H_GS1(MTL_GSC_HECI2_BASE), gs1_clr, gs1_set); diff --git a/drivers/gpu/drm/xe/xe_gt.c b/drivers/gpu/drm/xe/xe_gt.c index c8eda36546d3..34505a6d93ed 100644 --- a/drivers/gpu/drm/xe/xe_gt.c +++ b/drivers/gpu/drm/xe/xe_gt.c @@ -37,10 +37,10 @@ #include "xe_gt_sriov_pf.h" #include "xe_gt_sriov_vf.h" #include "xe_gt_sysfs.h" -#include "xe_gt_tlb_invalidation.h" #include "xe_gt_topology.h" #include "xe_guc_exec_queue_types.h" #include "xe_guc_pc.h" +#include "xe_guc_submit.h" #include "xe_hw_fence.h" #include "xe_hw_engine_class_sysfs.h" #include "xe_irq.h" @@ -57,6 +57,7 @@ #include "xe_sa.h" #include "xe_sched_job.h" #include "xe_sriov.h" +#include "xe_tlb_inval.h" #include "xe_tuning.h" #include "xe_uc.h" #include "xe_uc_fw.h" @@ -105,7 +106,7 @@ static void xe_gt_enable_host_l2_vram(struct xe_gt *gt) unsigned int fw_ref; u32 reg; - if (!XE_WA(gt, 16023588340)) + if (!XE_GT_WA(gt, 16023588340)) return; fw_ref = xe_force_wake_get(gt_to_fw(gt), XE_FW_GT); @@ -127,7 +128,7 @@ static void xe_gt_disable_host_l2_vram(struct xe_gt *gt) unsigned int fw_ref; u32 reg; - if (!XE_WA(gt, 16023588340)) + if (!XE_GT_WA(gt, 16023588340)) return; if (xe_gt_is_media_type(gt)) @@ -399,7 +400,7 @@ int xe_gt_init_early(struct xe_gt *gt) xe_reg_sr_init(>->reg_sr, "GT", gt_to_xe(gt)); - err = xe_wa_init(gt); + err = xe_wa_gt_init(gt); if (err) return err; @@ -407,12 +408,12 @@ int xe_gt_init_early(struct xe_gt *gt) if (err) return err; - xe_wa_process_oob(gt); + xe_wa_process_gt_oob(gt); xe_force_wake_init_gt(gt, gt_to_fw(gt)); spin_lock_init(>->global_invl_lock); - err = xe_gt_tlb_invalidation_init_early(gt); + err = xe_gt_tlb_inval_init_early(gt); if (err) return err; @@ -564,11 +565,9 @@ static int gt_init_with_all_forcewake(struct xe_gt *gt) if (xe_gt_is_main_type(gt)) { struct xe_tile *tile = gt_to_tile(gt); - tile->migrate = xe_migrate_init(tile); - if (IS_ERR(tile->migrate)) { - err = PTR_ERR(tile->migrate); + err = xe_migrate_init(tile->migrate); + if (err) goto err_force_wake; - } } err = xe_uc_load_hw(>->uc); @@ -804,6 +803,11 @@ static int do_gt_restart(struct xe_gt *gt) return 0; } +static int gt_wait_reset_unblock(struct xe_gt *gt) +{ + return xe_guc_wait_reset_unblock(>->uc.guc); +} + static int gt_reset(struct xe_gt *gt) { unsigned int fw_ref; @@ -818,6 +822,10 @@ static int gt_reset(struct xe_gt *gt) xe_gt_info(gt, "reset started\n"); + err = gt_wait_reset_unblock(gt); + if (!err) + xe_gt_warn(gt, "reset block failed to get lifted"); + xe_pm_runtime_get(gt_to_xe(gt)); if (xe_fault_inject_gt_reset()) { @@ -842,7 +850,7 @@ static int gt_reset(struct xe_gt *gt) xe_uc_stop(>->uc); - xe_gt_tlb_invalidation_reset(gt); + xe_tlb_inval_reset(>->tlb_inval); err = do_gt_reset(gt); if (err) @@ -958,7 +966,7 @@ int xe_gt_sanitize_freq(struct xe_gt *gt) if ((!xe_uc_fw_is_available(>->uc.gsc.fw) || xe_uc_fw_is_loaded(>->uc.gsc.fw) || xe_uc_fw_is_in_error_state(>->uc.gsc.fw)) && - XE_WA(gt, 22019338487)) + XE_GT_WA(gt, 22019338487)) ret = xe_guc_pc_restore_stashed_freq(>->uc.guc.pc); return ret; @@ -1056,5 +1064,5 @@ void xe_gt_declare_wedged(struct xe_gt *gt) xe_gt_assert(gt, gt_to_xe(gt)->wedged.mode); xe_uc_declare_wedged(>->uc); - xe_gt_tlb_invalidation_reset(gt); + xe_tlb_inval_reset(>->tlb_inval); } diff --git a/drivers/gpu/drm/xe/xe_gt_debugfs.c b/drivers/gpu/drm/xe/xe_gt_debugfs.c index 848618acdca8..bf3a67b5951c 100644 --- a/drivers/gpu/drm/xe/xe_gt_debugfs.c +++ b/drivers/gpu/drm/xe/xe_gt_debugfs.c @@ -29,6 +29,7 @@ #include "xe_pm.h" #include "xe_reg_sr.h" #include "xe_reg_whitelist.h" +#include "xe_sa.h" #include "xe_sriov.h" #include "xe_tuning.h" #include "xe_uc_debugfs.h" @@ -128,7 +129,34 @@ static int sa_info(struct xe_gt *gt, struct drm_printer *p) xe_pm_runtime_get(gt_to_xe(gt)); drm_suballoc_dump_debug_info(&tile->mem.kernel_bb_pool->base, p, - tile->mem.kernel_bb_pool->gpu_addr); + xe_sa_manager_gpu_addr(tile->mem.kernel_bb_pool)); + xe_pm_runtime_put(gt_to_xe(gt)); + + return 0; +} + +static int sa_info_vf_ccs(struct xe_gt *gt, struct drm_printer *p) +{ + struct xe_tile *tile = gt_to_tile(gt); + struct xe_sa_manager *bb_pool; + enum xe_sriov_vf_ccs_rw_ctxs ctx_id; + + if (!IS_VF_CCS_READY(gt_to_xe(gt))) + return 0; + + xe_pm_runtime_get(gt_to_xe(gt)); + + for_each_ccs_rw_ctx(ctx_id) { + bb_pool = tile->sriov.vf.ccs[ctx_id].mem.ccs_bb_pool; + if (!bb_pool) + break; + + drm_printf(p, "ccs %s bb suballoc info\n", ctx_id ? "write" : "read"); + drm_printf(p, "-------------------------\n"); + drm_suballoc_dump_debug_info(&bb_pool->base, p, xe_sa_manager_gpu_addr(bb_pool)); + drm_puts(p, "\n"); + } + xe_pm_runtime_put(gt_to_xe(gt)); return 0; @@ -303,6 +331,13 @@ static const struct drm_info_list vf_safe_debugfs_list[] = { {"hwconfig", .show = xe_gt_debugfs_simple_show, .data = hwconfig}, }; +/* + * only for GT debugfs files which are valid on VF. Not valid on PF. + */ +static const struct drm_info_list vf_only_debugfs_list[] = { + {"sa_info_vf_ccs", .show = xe_gt_debugfs_simple_show, .data = sa_info_vf_ccs}, +}; + /* everything else should be added here */ static const struct drm_info_list pf_only_debugfs_list[] = { {"hw_engines", .show = xe_gt_debugfs_simple_show, .data = hw_engines}, @@ -388,13 +423,18 @@ void xe_gt_debugfs_register(struct xe_gt *gt) { struct xe_device *xe = gt_to_xe(gt); struct drm_minor *minor = gt_to_xe(gt)->drm.primary; + struct dentry *parent = gt->tile->debugfs; struct dentry *root; + char symlink[16]; char name[8]; xe_gt_assert(gt, minor->debugfs_root); + if (IS_ERR(parent)) + return; + snprintf(name, sizeof(name), "gt%d", gt->info.id); - root = debugfs_create_dir(name, minor->debugfs_root); + root = debugfs_create_dir(name, parent); if (IS_ERR(root)) { drm_warn(&xe->drm, "Create GT directory failed"); return; @@ -419,6 +459,11 @@ void xe_gt_debugfs_register(struct xe_gt *gt) drm_debugfs_create_files(pf_only_debugfs_list, ARRAY_SIZE(pf_only_debugfs_list), root, minor); + else + drm_debugfs_create_files(vf_only_debugfs_list, + ARRAY_SIZE(vf_only_debugfs_list), + root, minor); + xe_uc_debugfs_register(>->uc, root); @@ -426,4 +471,11 @@ void xe_gt_debugfs_register(struct xe_gt *gt) xe_gt_sriov_pf_debugfs_register(gt, root); else if (IS_SRIOV_VF(xe)) xe_gt_sriov_vf_debugfs_register(gt, root); + + /* + * Backwards compatibility only: create a link for the legacy clients + * who may expect gt/ directory at the root level, not the tile level. + */ + snprintf(symlink, sizeof(symlink), "tile%u/%s", gt->tile->id, name); + debugfs_create_symlink(name, minor->debugfs_root, symlink); } diff --git a/drivers/gpu/drm/xe/xe_gt_idle.c b/drivers/gpu/drm/xe/xe_gt_idle.c index ffb210216aa9..f8950a52d0a4 100644 --- a/drivers/gpu/drm/xe/xe_gt_idle.c +++ b/drivers/gpu/drm/xe/xe_gt_idle.c @@ -322,15 +322,11 @@ static void gt_idle_fini(void *arg) { struct kobject *kobj = arg; struct xe_gt *gt = kobj_to_gt(kobj->parent); - unsigned int fw_ref; xe_gt_idle_disable_pg(gt); - if (gt_to_xe(gt)->info.skip_guc_pc) { - fw_ref = xe_force_wake_get(gt_to_fw(gt), XE_FW_GT); + if (gt_to_xe(gt)->info.skip_guc_pc) xe_gt_idle_disable_c6(gt); - xe_force_wake_put(gt_to_fw(gt), fw_ref); - } sysfs_remove_files(kobj, gt_idle_attrs); kobject_put(kobj); @@ -390,14 +386,23 @@ void xe_gt_idle_enable_c6(struct xe_gt *gt) RC_CTL_HW_ENABLE | RC_CTL_TO_MODE | RC_CTL_RC6_ENABLE); } -void xe_gt_idle_disable_c6(struct xe_gt *gt) +int xe_gt_idle_disable_c6(struct xe_gt *gt) { + unsigned int fw_ref; + xe_device_assert_mem_access(gt_to_xe(gt)); - xe_force_wake_assert_held(gt_to_fw(gt), XE_FW_GT); if (IS_SRIOV_VF(gt_to_xe(gt))) - return; + return 0; + + fw_ref = xe_force_wake_get(gt_to_fw(gt), XE_FW_GT); + if (!fw_ref) + return -ETIMEDOUT; xe_mmio_write32(>->mmio, RC_CONTROL, 0); xe_mmio_write32(>->mmio, RC_STATE, 0); + + xe_force_wake_put(gt_to_fw(gt), fw_ref); + + return 0; } diff --git a/drivers/gpu/drm/xe/xe_gt_idle.h b/drivers/gpu/drm/xe/xe_gt_idle.h index 591a01e181bc..9c34a155e102 100644 --- a/drivers/gpu/drm/xe/xe_gt_idle.h +++ b/drivers/gpu/drm/xe/xe_gt_idle.h @@ -13,7 +13,7 @@ struct xe_gt; int xe_gt_idle_init(struct xe_gt_idle *gtidle); void xe_gt_idle_enable_c6(struct xe_gt *gt); -void xe_gt_idle_disable_c6(struct xe_gt *gt); +int xe_gt_idle_disable_c6(struct xe_gt *gt); void xe_gt_idle_enable_pg(struct xe_gt *gt); void xe_gt_idle_disable_pg(struct xe_gt *gt); int xe_gt_idle_pg_print(struct xe_gt *gt, struct drm_printer *p); diff --git a/drivers/gpu/drm/xe/xe_gt_mcr.c b/drivers/gpu/drm/xe/xe_gt_mcr.c index 64a2f0d6aaf9..683ac021a06d 100644 --- a/drivers/gpu/drm/xe/xe_gt_mcr.c +++ b/drivers/gpu/drm/xe/xe_gt_mcr.c @@ -46,8 +46,6 @@ * MCR registers are not available on Virtual Function (VF). */ -#define STEER_SEMAPHORE XE_REG(0xFD0) - static inline struct xe_reg to_xe_reg(struct xe_reg_mcr reg_mcr) { return reg_mcr.__reg; @@ -533,7 +531,7 @@ void xe_gt_mcr_set_implicit_defaults(struct xe_gt *gt) u32 steer_val = REG_FIELD_PREP(MCR_SLICE_MASK, 0) | REG_FIELD_PREP(MCR_SUBSLICE_MASK, 2); - xe_mmio_write32(>->mmio, MCFG_MCR_SELECTOR, steer_val); + xe_mmio_write32(>->mmio, STEER_SEMAPHORE, steer_val); xe_mmio_write32(>->mmio, SF_MCR_SELECTOR, steer_val); /* * For GAM registers, all reads should be directed to instance 1 diff --git a/drivers/gpu/drm/xe/xe_gt_pagefault.c b/drivers/gpu/drm/xe/xe_gt_pagefault.c index 5a75d56d8558..d02d22fb3659 100644 --- a/drivers/gpu/drm/xe/xe_gt_pagefault.c +++ b/drivers/gpu/drm/xe/xe_gt_pagefault.c @@ -16,13 +16,13 @@ #include "xe_gt.h" #include "xe_gt_printk.h" #include "xe_gt_stats.h" -#include "xe_gt_tlb_invalidation.h" #include "xe_guc.h" #include "xe_guc_ct.h" #include "xe_migrate.h" #include "xe_svm.h" #include "xe_trace_bo.h" #include "xe_vm.h" +#include "xe_vram_types.h" struct pagefault { u64 page_addr; @@ -74,7 +74,7 @@ static bool vma_is_valid(struct xe_tile *tile, struct xe_vma *vma) } static int xe_pf_begin(struct drm_exec *exec, struct xe_vma *vma, - bool atomic, unsigned int id) + bool need_vram_move, struct xe_vram_region *vram) { struct xe_bo *bo = xe_vma_bo(vma); struct xe_vm *vm = xe_vma_vm(vma); @@ -84,24 +84,13 @@ static int xe_pf_begin(struct drm_exec *exec, struct xe_vma *vma, if (err) return err; - if (atomic && IS_DGFX(vm->xe)) { - if (xe_vma_is_userptr(vma)) { - err = -EACCES; - return err; - } + if (!bo) + return 0; - /* Migrate to VRAM, move should invalidate the VMA first */ - err = xe_bo_migrate(bo, XE_PL_VRAM0 + id); - if (err) - return err; - } else if (bo) { - /* Create backing store if needed */ - err = xe_bo_validate(bo, vm, true); - if (err) - return err; - } + err = need_vram_move ? xe_bo_migrate(bo, vram->placement) : + xe_bo_validate(bo, vm, true); - return 0; + return err; } static int handle_vma_pagefault(struct xe_gt *gt, struct xe_vma *vma, @@ -112,10 +101,14 @@ static int handle_vma_pagefault(struct xe_gt *gt, struct xe_vma *vma, struct drm_exec exec; struct dma_fence *fence; ktime_t end = 0; - int err; + int err, needs_vram; lockdep_assert_held_write(&vm->lock); + needs_vram = xe_vma_need_vram_for_atomic(vm->xe, vma, atomic); + if (needs_vram < 0 || (needs_vram && xe_vma_is_userptr(vma))) + return needs_vram < 0 ? needs_vram : -EACCES; + xe_gt_stats_incr(gt, XE_GT_STATS_ID_VMA_PAGEFAULT_COUNT, 1); xe_gt_stats_incr(gt, XE_GT_STATS_ID_VMA_PAGEFAULT_KB, xe_vma_size(vma) / 1024); @@ -138,7 +131,7 @@ retry_userptr: /* Lock VM and BOs dma-resv */ drm_exec_init(&exec, 0, 0); drm_exec_until_all_locked(&exec) { - err = xe_pf_begin(&exec, vma, atomic, tile->id); + err = xe_pf_begin(&exec, vma, needs_vram == 1, tile->mem.vram); drm_exec_retry_on_contention(&exec); if (xe_vm_validate_should_retry(&exec, err, &end)) err = -EAGAIN; @@ -573,7 +566,7 @@ static int handle_acc(struct xe_gt *gt, struct acc *acc) /* Lock VM and BOs dma-resv */ drm_exec_init(&exec, 0, 0); drm_exec_until_all_locked(&exec) { - ret = xe_pf_begin(&exec, vma, true, tile->id); + ret = xe_pf_begin(&exec, vma, IS_DGFX(vm->xe), tile->mem.vram); drm_exec_retry_on_contention(&exec); if (ret) break; diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_pf.c b/drivers/gpu/drm/xe/xe_gt_sriov_pf.c index bdbd15f3afe3..c4dda87b47cc 100644 --- a/drivers/gpu/drm/xe/xe_gt_sriov_pf.c +++ b/drivers/gpu/drm/xe/xe_gt_sriov_pf.c @@ -55,7 +55,12 @@ static void pf_init_workers(struct xe_gt *gt) static void pf_fini_workers(struct xe_gt *gt) { xe_gt_assert(gt, IS_SRIOV_PF(gt_to_xe(gt))); - disable_work_sync(>->sriov.pf.workers.restart); + + if (disable_work_sync(>->sriov.pf.workers.restart)) { + xe_gt_sriov_dbg_verbose(gt, "pending restart disabled!\n"); + /* release an rpm reference taken on the worker's behalf */ + xe_pm_runtime_put(gt_to_xe(gt)); + } } /** @@ -207,8 +212,11 @@ static void pf_cancel_restart(struct xe_gt *gt) { xe_gt_assert(gt, IS_SRIOV_PF(gt_to_xe(gt))); - if (cancel_work_sync(>->sriov.pf.workers.restart)) + if (cancel_work_sync(>->sriov.pf.workers.restart)) { xe_gt_sriov_dbg_verbose(gt, "pending restart canceled!\n"); + /* release an rpm reference taken on the worker's behalf */ + xe_pm_runtime_put(gt_to_xe(gt)); + } } /** @@ -226,9 +234,12 @@ static void pf_restart(struct xe_gt *gt) { struct xe_device *xe = gt_to_xe(gt); - xe_pm_runtime_get(xe); + xe_gt_assert(gt, !xe_pm_runtime_suspended(xe)); + xe_gt_sriov_pf_config_restart(gt); xe_gt_sriov_pf_control_restart(gt); + + /* release an rpm reference taken on our behalf */ xe_pm_runtime_put(xe); xe_gt_sriov_dbg(gt, "restart completed\n"); @@ -247,8 +258,13 @@ static void pf_queue_restart(struct xe_gt *gt) xe_gt_assert(gt, IS_SRIOV_PF(xe)); - if (!queue_work(xe->sriov.wq, >->sriov.pf.workers.restart)) + /* take an rpm reference on behalf of the worker */ + xe_pm_runtime_get_noresume(xe); + + if (!queue_work(xe->sriov.wq, >->sriov.pf.workers.restart)) { xe_gt_sriov_dbg(gt, "restart already in queue!\n"); + xe_pm_runtime_put(xe); + } } /** diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c b/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c index 494909f74eb2..c8f0320d032f 100644 --- a/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c +++ b/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c @@ -33,6 +33,7 @@ #include "xe_migrate.h" #include "xe_sriov.h" #include "xe_ttm_vram_mgr.h" +#include "xe_vram_types.h" #include "xe_wopcm.h" #define make_u64_from_u32(hi, lo) ((u64)((u64)(u32)(hi) << 32 | (u32)(lo))) @@ -1433,7 +1434,8 @@ fail: return err; } -static void pf_release_vf_config_lmem(struct xe_gt *gt, struct xe_gt_sriov_config *config) +/* Return: %true if there was an LMEM provisioned, %false otherwise */ +static bool pf_release_vf_config_lmem(struct xe_gt *gt, struct xe_gt_sriov_config *config) { xe_gt_assert(gt, IS_DGFX(gt_to_xe(gt))); xe_gt_assert(gt, xe_gt_is_main_type(gt)); @@ -1442,7 +1444,9 @@ static void pf_release_vf_config_lmem(struct xe_gt *gt, struct xe_gt_sriov_confi if (config->lmem_obj) { xe_bo_unpin_map_no_vm(config->lmem_obj); config->lmem_obj = NULL; + return true; } + return false; } static int pf_provision_vf_lmem(struct xe_gt *gt, unsigned int vfid, u64 size) @@ -1604,7 +1608,7 @@ static u64 pf_query_free_lmem(struct xe_gt *gt) { struct xe_tile *tile = gt->tile; - return xe_ttm_vram_get_avail(&tile->mem.vram.ttm.manager); + return xe_ttm_vram_get_avail(&tile->mem.vram->ttm.manager); } static u64 pf_query_max_lmem(struct xe_gt *gt) @@ -2020,12 +2024,13 @@ static void pf_release_vf_config(struct xe_gt *gt, unsigned int vfid) { struct xe_gt_sriov_config *config = pf_pick_vf_config(gt, vfid); struct xe_device *xe = gt_to_xe(gt); + bool released; if (xe_gt_is_main_type(gt)) { pf_release_vf_config_ggtt(gt, config); if (IS_DGFX(xe)) { - pf_release_vf_config_lmem(gt, config); - if (xe_device_has_lmtt(xe)) + released = pf_release_vf_config_lmem(gt, config); + if (released && xe_device_has_lmtt(xe)) pf_update_vf_lmtt(xe, vfid); } } diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c index b282838d59e6..0461d5513487 100644 --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c @@ -25,6 +25,7 @@ #include "xe_guc.h" #include "xe_guc_hxg_helpers.h" #include "xe_guc_relay.h" +#include "xe_lrc.h" #include "xe_mmio.h" #include "xe_sriov.h" #include "xe_sriov_vf.h" @@ -751,6 +752,19 @@ failed: } /** + * xe_gt_sriov_vf_default_lrcs_hwsp_rebase - Update GGTT references in HWSP of default LRCs. + * @gt: the &xe_gt struct instance + */ +void xe_gt_sriov_vf_default_lrcs_hwsp_rebase(struct xe_gt *gt) +{ + struct xe_hw_engine *hwe; + enum xe_hw_engine_id id; + + for_each_hw_engine(hwe, gt, id) + xe_default_lrc_update_memirq_regs_with_address(hwe); +} + +/** * xe_gt_sriov_vf_migrated_event_handler - Start a VF migration recovery, * or just mark that a GuC is ready for it. * @gt: the &xe_gt struct instance linked to target GuC diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf.h index e0357f341a2d..0af1dc769fe0 100644 --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.h +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.h @@ -21,6 +21,7 @@ void xe_gt_sriov_vf_guc_versions(struct xe_gt *gt, int xe_gt_sriov_vf_query_config(struct xe_gt *gt); int xe_gt_sriov_vf_connect(struct xe_gt *gt); int xe_gt_sriov_vf_query_runtime(struct xe_gt *gt); +void xe_gt_sriov_vf_default_lrcs_hwsp_rebase(struct xe_gt *gt); int xe_gt_sriov_vf_notify_resfix_done(struct xe_gt *gt); void xe_gt_sriov_vf_migrated_event_handler(struct xe_gt *gt); diff --git a/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c b/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c deleted file mode 100644 index 086c12ee3d9d..000000000000 --- a/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c +++ /dev/null @@ -1,596 +0,0 @@ -// SPDX-License-Identifier: MIT -/* - * Copyright © 2023 Intel Corporation - */ - -#include "xe_gt_tlb_invalidation.h" - -#include "abi/guc_actions_abi.h" -#include "xe_device.h" -#include "xe_force_wake.h" -#include "xe_gt.h" -#include "xe_gt_printk.h" -#include "xe_guc.h" -#include "xe_guc_ct.h" -#include "xe_gt_stats.h" -#include "xe_mmio.h" -#include "xe_pm.h" -#include "xe_sriov.h" -#include "xe_trace.h" -#include "regs/xe_guc_regs.h" - -#define FENCE_STACK_BIT DMA_FENCE_FLAG_USER_BITS - -/* - * TLB inval depends on pending commands in the CT queue and then the real - * invalidation time. Double up the time to process full CT queue - * just to be on the safe side. - */ -static long tlb_timeout_jiffies(struct xe_gt *gt) -{ - /* this reflects what HW/GuC needs to process TLB inv request */ - const long hw_tlb_timeout = HZ / 4; - - /* this estimates actual delay caused by the CTB transport */ - long delay = xe_guc_ct_queue_proc_time_jiffies(>->uc.guc.ct); - - return hw_tlb_timeout + 2 * delay; -} - -static void xe_gt_tlb_invalidation_fence_fini(struct xe_gt_tlb_invalidation_fence *fence) -{ - if (WARN_ON_ONCE(!fence->gt)) - return; - - xe_pm_runtime_put(gt_to_xe(fence->gt)); - fence->gt = NULL; /* fini() should be called once */ -} - -static void -__invalidation_fence_signal(struct xe_device *xe, struct xe_gt_tlb_invalidation_fence *fence) -{ - bool stack = test_bit(FENCE_STACK_BIT, &fence->base.flags); - - trace_xe_gt_tlb_invalidation_fence_signal(xe, fence); - xe_gt_tlb_invalidation_fence_fini(fence); - dma_fence_signal(&fence->base); - if (!stack) - dma_fence_put(&fence->base); -} - -static void -invalidation_fence_signal(struct xe_device *xe, struct xe_gt_tlb_invalidation_fence *fence) -{ - list_del(&fence->link); - __invalidation_fence_signal(xe, fence); -} - -void xe_gt_tlb_invalidation_fence_signal(struct xe_gt_tlb_invalidation_fence *fence) -{ - if (WARN_ON_ONCE(!fence->gt)) - return; - - __invalidation_fence_signal(gt_to_xe(fence->gt), fence); -} - -static void xe_gt_tlb_fence_timeout(struct work_struct *work) -{ - struct xe_gt *gt = container_of(work, struct xe_gt, - tlb_invalidation.fence_tdr.work); - struct xe_device *xe = gt_to_xe(gt); - struct xe_gt_tlb_invalidation_fence *fence, *next; - - LNL_FLUSH_WORK(>->uc.guc.ct.g2h_worker); - - spin_lock_irq(>->tlb_invalidation.pending_lock); - list_for_each_entry_safe(fence, next, - >->tlb_invalidation.pending_fences, link) { - s64 since_inval_ms = ktime_ms_delta(ktime_get(), - fence->invalidation_time); - - if (msecs_to_jiffies(since_inval_ms) < tlb_timeout_jiffies(gt)) - break; - - trace_xe_gt_tlb_invalidation_fence_timeout(xe, fence); - xe_gt_err(gt, "TLB invalidation fence timeout, seqno=%d recv=%d", - fence->seqno, gt->tlb_invalidation.seqno_recv); - - fence->base.error = -ETIME; - invalidation_fence_signal(xe, fence); - } - if (!list_empty(>->tlb_invalidation.pending_fences)) - queue_delayed_work(system_wq, - >->tlb_invalidation.fence_tdr, - tlb_timeout_jiffies(gt)); - spin_unlock_irq(>->tlb_invalidation.pending_lock); -} - -/** - * xe_gt_tlb_invalidation_init_early - Initialize GT TLB invalidation state - * @gt: GT structure - * - * Initialize GT TLB invalidation state, purely software initialization, should - * be called once during driver load. - * - * Return: 0 on success, negative error code on error. - */ -int xe_gt_tlb_invalidation_init_early(struct xe_gt *gt) -{ - gt->tlb_invalidation.seqno = 1; - INIT_LIST_HEAD(>->tlb_invalidation.pending_fences); - spin_lock_init(>->tlb_invalidation.pending_lock); - spin_lock_init(>->tlb_invalidation.lock); - INIT_DELAYED_WORK(>->tlb_invalidation.fence_tdr, - xe_gt_tlb_fence_timeout); - - return 0; -} - -/** - * xe_gt_tlb_invalidation_reset - Initialize GT TLB invalidation reset - * @gt: GT structure - * - * Signal any pending invalidation fences, should be called during a GT reset - */ -void xe_gt_tlb_invalidation_reset(struct xe_gt *gt) -{ - struct xe_gt_tlb_invalidation_fence *fence, *next; - int pending_seqno; - - /* - * we can get here before the CTs are even initialized if we're wedging - * very early, in which case there are not going to be any pending - * fences so we can bail immediately. - */ - if (!xe_guc_ct_initialized(>->uc.guc.ct)) - return; - - /* - * CT channel is already disabled at this point. No new TLB requests can - * appear. - */ - - mutex_lock(>->uc.guc.ct.lock); - spin_lock_irq(>->tlb_invalidation.pending_lock); - cancel_delayed_work(>->tlb_invalidation.fence_tdr); - /* - * We might have various kworkers waiting for TLB flushes to complete - * which are not tracked with an explicit TLB fence, however at this - * stage that will never happen since the CT is already disabled, so - * make sure we signal them here under the assumption that we have - * completed a full GT reset. - */ - if (gt->tlb_invalidation.seqno == 1) - pending_seqno = TLB_INVALIDATION_SEQNO_MAX - 1; - else - pending_seqno = gt->tlb_invalidation.seqno - 1; - WRITE_ONCE(gt->tlb_invalidation.seqno_recv, pending_seqno); - - list_for_each_entry_safe(fence, next, - >->tlb_invalidation.pending_fences, link) - invalidation_fence_signal(gt_to_xe(gt), fence); - spin_unlock_irq(>->tlb_invalidation.pending_lock); - mutex_unlock(>->uc.guc.ct.lock); -} - -static bool tlb_invalidation_seqno_past(struct xe_gt *gt, int seqno) -{ - int seqno_recv = READ_ONCE(gt->tlb_invalidation.seqno_recv); - - if (seqno - seqno_recv < -(TLB_INVALIDATION_SEQNO_MAX / 2)) - return false; - - if (seqno - seqno_recv > (TLB_INVALIDATION_SEQNO_MAX / 2)) - return true; - - return seqno_recv >= seqno; -} - -static int send_tlb_invalidation(struct xe_guc *guc, - struct xe_gt_tlb_invalidation_fence *fence, - u32 *action, int len) -{ - struct xe_gt *gt = guc_to_gt(guc); - struct xe_device *xe = gt_to_xe(gt); - int seqno; - int ret; - - xe_gt_assert(gt, fence); - - /* - * XXX: The seqno algorithm relies on TLB invalidation being processed - * in order which they currently are, if that changes the algorithm will - * need to be updated. - */ - - mutex_lock(&guc->ct.lock); - seqno = gt->tlb_invalidation.seqno; - fence->seqno = seqno; - trace_xe_gt_tlb_invalidation_fence_send(xe, fence); - action[1] = seqno; - ret = xe_guc_ct_send_locked(&guc->ct, action, len, - G2H_LEN_DW_TLB_INVALIDATE, 1); - if (!ret) { - spin_lock_irq(>->tlb_invalidation.pending_lock); - /* - * We haven't actually published the TLB fence as per - * pending_fences, but in theory our seqno could have already - * been written as we acquired the pending_lock. In such a case - * we can just go ahead and signal the fence here. - */ - if (tlb_invalidation_seqno_past(gt, seqno)) { - __invalidation_fence_signal(xe, fence); - } else { - fence->invalidation_time = ktime_get(); - list_add_tail(&fence->link, - >->tlb_invalidation.pending_fences); - - if (list_is_singular(>->tlb_invalidation.pending_fences)) - queue_delayed_work(system_wq, - >->tlb_invalidation.fence_tdr, - tlb_timeout_jiffies(gt)); - } - spin_unlock_irq(>->tlb_invalidation.pending_lock); - } else { - __invalidation_fence_signal(xe, fence); - } - if (!ret) { - gt->tlb_invalidation.seqno = (gt->tlb_invalidation.seqno + 1) % - TLB_INVALIDATION_SEQNO_MAX; - if (!gt->tlb_invalidation.seqno) - gt->tlb_invalidation.seqno = 1; - } - mutex_unlock(&guc->ct.lock); - xe_gt_stats_incr(gt, XE_GT_STATS_ID_TLB_INVAL, 1); - - return ret; -} - -#define MAKE_INVAL_OP(type) ((type << XE_GUC_TLB_INVAL_TYPE_SHIFT) | \ - XE_GUC_TLB_INVAL_MODE_HEAVY << XE_GUC_TLB_INVAL_MODE_SHIFT | \ - XE_GUC_TLB_INVAL_FLUSH_CACHE) - -/** - * xe_gt_tlb_invalidation_guc - Issue a TLB invalidation on this GT for the GuC - * @gt: GT structure - * @fence: invalidation fence which will be signal on TLB invalidation - * completion - * - * Issue a TLB invalidation for the GuC. Completion of TLB is asynchronous and - * caller can use the invalidation fence to wait for completion. - * - * Return: 0 on success, negative error code on error - */ -static int xe_gt_tlb_invalidation_guc(struct xe_gt *gt, - struct xe_gt_tlb_invalidation_fence *fence) -{ - u32 action[] = { - XE_GUC_ACTION_TLB_INVALIDATION, - 0, /* seqno, replaced in send_tlb_invalidation */ - MAKE_INVAL_OP(XE_GUC_TLB_INVAL_GUC), - }; - int ret; - - ret = send_tlb_invalidation(>->uc.guc, fence, action, - ARRAY_SIZE(action)); - /* - * -ECANCELED indicates the CT is stopped for a GT reset. TLB caches - * should be nuked on a GT reset so this error can be ignored. - */ - if (ret == -ECANCELED) - return 0; - - return ret; -} - -/** - * xe_gt_tlb_invalidation_ggtt - Issue a TLB invalidation on this GT for the GGTT - * @gt: GT structure - * - * Issue a TLB invalidation for the GGTT. Completion of TLB invalidation is - * synchronous. - * - * Return: 0 on success, negative error code on error - */ -int xe_gt_tlb_invalidation_ggtt(struct xe_gt *gt) -{ - struct xe_device *xe = gt_to_xe(gt); - unsigned int fw_ref; - - if (xe_guc_ct_enabled(>->uc.guc.ct) && - gt->uc.guc.submission_state.enabled) { - struct xe_gt_tlb_invalidation_fence fence; - int ret; - - xe_gt_tlb_invalidation_fence_init(gt, &fence, true); - ret = xe_gt_tlb_invalidation_guc(gt, &fence); - if (ret) - return ret; - - xe_gt_tlb_invalidation_fence_wait(&fence); - } else if (xe_device_uc_enabled(xe) && !xe_device_wedged(xe)) { - struct xe_mmio *mmio = >->mmio; - - if (IS_SRIOV_VF(xe)) - return 0; - - fw_ref = xe_force_wake_get(gt_to_fw(gt), XE_FW_GT); - if (xe->info.platform == XE_PVC || GRAPHICS_VER(xe) >= 20) { - xe_mmio_write32(mmio, PVC_GUC_TLB_INV_DESC1, - PVC_GUC_TLB_INV_DESC1_INVALIDATE); - xe_mmio_write32(mmio, PVC_GUC_TLB_INV_DESC0, - PVC_GUC_TLB_INV_DESC0_VALID); - } else { - xe_mmio_write32(mmio, GUC_TLB_INV_CR, - GUC_TLB_INV_CR_INVALIDATE); - } - xe_force_wake_put(gt_to_fw(gt), fw_ref); - } - - return 0; -} - -static int send_tlb_invalidation_all(struct xe_gt *gt, - struct xe_gt_tlb_invalidation_fence *fence) -{ - u32 action[] = { - XE_GUC_ACTION_TLB_INVALIDATION_ALL, - 0, /* seqno, replaced in send_tlb_invalidation */ - MAKE_INVAL_OP(XE_GUC_TLB_INVAL_FULL), - }; - - return send_tlb_invalidation(>->uc.guc, fence, action, ARRAY_SIZE(action)); -} - -/** - * xe_gt_tlb_invalidation_all - Invalidate all TLBs across PF and all VFs. - * @gt: the &xe_gt structure - * @fence: the &xe_gt_tlb_invalidation_fence to be signaled on completion - * - * Send a request to invalidate all TLBs across PF and all VFs. - * - * Return: 0 on success, negative error code on error - */ -int xe_gt_tlb_invalidation_all(struct xe_gt *gt, struct xe_gt_tlb_invalidation_fence *fence) -{ - int err; - - xe_gt_assert(gt, gt == fence->gt); - - err = send_tlb_invalidation_all(gt, fence); - if (err) - xe_gt_err(gt, "TLB invalidation request failed (%pe)", ERR_PTR(err)); - - return err; -} - -/* - * Ensure that roundup_pow_of_two(length) doesn't overflow. - * Note that roundup_pow_of_two() operates on unsigned long, - * not on u64. - */ -#define MAX_RANGE_TLB_INVALIDATION_LENGTH (rounddown_pow_of_two(ULONG_MAX)) - -/** - * xe_gt_tlb_invalidation_range - Issue a TLB invalidation on this GT for an - * address range - * - * @gt: GT structure - * @fence: invalidation fence which will be signal on TLB invalidation - * completion - * @start: start address - * @end: end address - * @asid: address space id - * - * Issue a range based TLB invalidation if supported, if not fallback to a full - * TLB invalidation. Completion of TLB is asynchronous and caller can use - * the invalidation fence to wait for completion. - * - * Return: Negative error code on error, 0 on success - */ -int xe_gt_tlb_invalidation_range(struct xe_gt *gt, - struct xe_gt_tlb_invalidation_fence *fence, - u64 start, u64 end, u32 asid) -{ - struct xe_device *xe = gt_to_xe(gt); -#define MAX_TLB_INVALIDATION_LEN 7 - u32 action[MAX_TLB_INVALIDATION_LEN]; - u64 length = end - start; - int len = 0; - - xe_gt_assert(gt, fence); - - /* Execlists not supported */ - if (gt_to_xe(gt)->info.force_execlist) { - __invalidation_fence_signal(xe, fence); - return 0; - } - - action[len++] = XE_GUC_ACTION_TLB_INVALIDATION; - action[len++] = 0; /* seqno, replaced in send_tlb_invalidation */ - if (!xe->info.has_range_tlb_invalidation || - length > MAX_RANGE_TLB_INVALIDATION_LENGTH) { - action[len++] = MAKE_INVAL_OP(XE_GUC_TLB_INVAL_FULL); - } else { - u64 orig_start = start; - u64 align; - - if (length < SZ_4K) - length = SZ_4K; - - /* - * We need to invalidate a higher granularity if start address - * is not aligned to length. When start is not aligned with - * length we need to find the length large enough to create an - * address mask covering the required range. - */ - align = roundup_pow_of_two(length); - start = ALIGN_DOWN(start, align); - end = ALIGN(end, align); - length = align; - while (start + length < end) { - length <<= 1; - start = ALIGN_DOWN(orig_start, length); - } - - /* - * Minimum invalidation size for a 2MB page that the hardware - * expects is 16MB - */ - if (length >= SZ_2M) { - length = max_t(u64, SZ_16M, length); - start = ALIGN_DOWN(orig_start, length); - } - - xe_gt_assert(gt, length >= SZ_4K); - xe_gt_assert(gt, is_power_of_2(length)); - xe_gt_assert(gt, !(length & GENMASK(ilog2(SZ_16M) - 1, - ilog2(SZ_2M) + 1))); - xe_gt_assert(gt, IS_ALIGNED(start, length)); - - action[len++] = MAKE_INVAL_OP(XE_GUC_TLB_INVAL_PAGE_SELECTIVE); - action[len++] = asid; - action[len++] = lower_32_bits(start); - action[len++] = upper_32_bits(start); - action[len++] = ilog2(length) - ilog2(SZ_4K); - } - - xe_gt_assert(gt, len <= MAX_TLB_INVALIDATION_LEN); - - return send_tlb_invalidation(>->uc.guc, fence, action, len); -} - -/** - * xe_gt_tlb_invalidation_vm - Issue a TLB invalidation on this GT for a VM - * @gt: graphics tile - * @vm: VM to invalidate - * - * Invalidate entire VM's address space - */ -void xe_gt_tlb_invalidation_vm(struct xe_gt *gt, struct xe_vm *vm) -{ - struct xe_gt_tlb_invalidation_fence fence; - u64 range = 1ull << vm->xe->info.va_bits; - int ret; - - xe_gt_tlb_invalidation_fence_init(gt, &fence, true); - - ret = xe_gt_tlb_invalidation_range(gt, &fence, 0, range, vm->usm.asid); - if (ret < 0) - return; - - xe_gt_tlb_invalidation_fence_wait(&fence); -} - -/** - * xe_guc_tlb_invalidation_done_handler - TLB invalidation done handler - * @guc: guc - * @msg: message indicating TLB invalidation done - * @len: length of message - * - * Parse seqno of TLB invalidation, wake any waiters for seqno, and signal any - * invalidation fences for seqno. Algorithm for this depends on seqno being - * received in-order and asserts this assumption. - * - * Return: 0 on success, -EPROTO for malformed messages. - */ -int xe_guc_tlb_invalidation_done_handler(struct xe_guc *guc, u32 *msg, u32 len) -{ - struct xe_gt *gt = guc_to_gt(guc); - struct xe_device *xe = gt_to_xe(gt); - struct xe_gt_tlb_invalidation_fence *fence, *next; - unsigned long flags; - - if (unlikely(len != 1)) - return -EPROTO; - - /* - * This can also be run both directly from the IRQ handler and also in - * process_g2h_msg(). Only one may process any individual CT message, - * however the order they are processed here could result in skipping a - * seqno. To handle that we just process all the seqnos from the last - * seqno_recv up to and including the one in msg[0]. The delta should be - * very small so there shouldn't be much of pending_fences we actually - * need to iterate over here. - * - * From GuC POV we expect the seqnos to always appear in-order, so if we - * see something later in the timeline we can be sure that anything - * appearing earlier has already signalled, just that we have yet to - * officially process the CT message like if racing against - * process_g2h_msg(). - */ - spin_lock_irqsave(>->tlb_invalidation.pending_lock, flags); - if (tlb_invalidation_seqno_past(gt, msg[0])) { - spin_unlock_irqrestore(>->tlb_invalidation.pending_lock, flags); - return 0; - } - - WRITE_ONCE(gt->tlb_invalidation.seqno_recv, msg[0]); - - list_for_each_entry_safe(fence, next, - >->tlb_invalidation.pending_fences, link) { - trace_xe_gt_tlb_invalidation_fence_recv(xe, fence); - - if (!tlb_invalidation_seqno_past(gt, fence->seqno)) - break; - - invalidation_fence_signal(xe, fence); - } - - if (!list_empty(>->tlb_invalidation.pending_fences)) - mod_delayed_work(system_wq, - >->tlb_invalidation.fence_tdr, - tlb_timeout_jiffies(gt)); - else - cancel_delayed_work(>->tlb_invalidation.fence_tdr); - - spin_unlock_irqrestore(>->tlb_invalidation.pending_lock, flags); - - return 0; -} - -static const char * -invalidation_fence_get_driver_name(struct dma_fence *dma_fence) -{ - return "xe"; -} - -static const char * -invalidation_fence_get_timeline_name(struct dma_fence *dma_fence) -{ - return "invalidation_fence"; -} - -static const struct dma_fence_ops invalidation_fence_ops = { - .get_driver_name = invalidation_fence_get_driver_name, - .get_timeline_name = invalidation_fence_get_timeline_name, -}; - -/** - * xe_gt_tlb_invalidation_fence_init - Initialize TLB invalidation fence - * @gt: GT - * @fence: TLB invalidation fence to initialize - * @stack: fence is stack variable - * - * Initialize TLB invalidation fence for use. xe_gt_tlb_invalidation_fence_fini - * will be automatically called when fence is signalled (all fences must signal), - * even on error. - */ -void xe_gt_tlb_invalidation_fence_init(struct xe_gt *gt, - struct xe_gt_tlb_invalidation_fence *fence, - bool stack) -{ - xe_pm_runtime_get_noresume(gt_to_xe(gt)); - - spin_lock_irq(>->tlb_invalidation.lock); - dma_fence_init(&fence->base, &invalidation_fence_ops, - >->tlb_invalidation.lock, - dma_fence_context_alloc(1), 1); - spin_unlock_irq(>->tlb_invalidation.lock); - INIT_LIST_HEAD(&fence->link); - if (stack) - set_bit(FENCE_STACK_BIT, &fence->base.flags); - else - dma_fence_get(&fence->base); - fence->gt = gt; -} diff --git a/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.h b/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.h deleted file mode 100644 index f7f0f2eaf4b5..000000000000 --- a/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.h +++ /dev/null @@ -1,40 +0,0 @@ -/* SPDX-License-Identifier: MIT */ -/* - * Copyright © 2023 Intel Corporation - */ - -#ifndef _XE_GT_TLB_INVALIDATION_H_ -#define _XE_GT_TLB_INVALIDATION_H_ - -#include <linux/types.h> - -#include "xe_gt_tlb_invalidation_types.h" - -struct xe_gt; -struct xe_guc; -struct xe_vm; -struct xe_vma; - -int xe_gt_tlb_invalidation_init_early(struct xe_gt *gt); - -void xe_gt_tlb_invalidation_reset(struct xe_gt *gt); -int xe_gt_tlb_invalidation_ggtt(struct xe_gt *gt); -void xe_gt_tlb_invalidation_vm(struct xe_gt *gt, struct xe_vm *vm); -int xe_gt_tlb_invalidation_all(struct xe_gt *gt, struct xe_gt_tlb_invalidation_fence *fence); -int xe_gt_tlb_invalidation_range(struct xe_gt *gt, - struct xe_gt_tlb_invalidation_fence *fence, - u64 start, u64 end, u32 asid); -int xe_guc_tlb_invalidation_done_handler(struct xe_guc *guc, u32 *msg, u32 len); - -void xe_gt_tlb_invalidation_fence_init(struct xe_gt *gt, - struct xe_gt_tlb_invalidation_fence *fence, - bool stack); -void xe_gt_tlb_invalidation_fence_signal(struct xe_gt_tlb_invalidation_fence *fence); - -static inline void -xe_gt_tlb_invalidation_fence_wait(struct xe_gt_tlb_invalidation_fence *fence) -{ - dma_fence_wait(&fence->base, false); -} - -#endif /* _XE_GT_TLB_INVALIDATION_ */ diff --git a/drivers/gpu/drm/xe/xe_gt_tlb_invalidation_types.h b/drivers/gpu/drm/xe/xe_gt_tlb_invalidation_types.h deleted file mode 100644 index de6e825e0851..000000000000 --- a/drivers/gpu/drm/xe/xe_gt_tlb_invalidation_types.h +++ /dev/null @@ -1,32 +0,0 @@ -/* SPDX-License-Identifier: MIT */ -/* - * Copyright © 2023 Intel Corporation - */ - -#ifndef _XE_GT_TLB_INVALIDATION_TYPES_H_ -#define _XE_GT_TLB_INVALIDATION_TYPES_H_ - -#include <linux/dma-fence.h> - -struct xe_gt; - -/** - * struct xe_gt_tlb_invalidation_fence - XE GT TLB invalidation fence - * - * Optionally passed to xe_gt_tlb_invalidation and will be signaled upon TLB - * invalidation completion. - */ -struct xe_gt_tlb_invalidation_fence { - /** @base: dma fence base */ - struct dma_fence base; - /** @gt: GT which fence belong to */ - struct xe_gt *gt; - /** @link: link into list of pending tlb fences */ - struct list_head link; - /** @seqno: seqno of TLB invalidation to signal fence one */ - int seqno; - /** @invalidation_time: time of TLB invalidation */ - ktime_t invalidation_time; -}; - -#endif diff --git a/drivers/gpu/drm/xe/xe_gt_topology.c b/drivers/gpu/drm/xe/xe_gt_topology.c index 8c63e3263643..a0baa560dd71 100644 --- a/drivers/gpu/drm/xe/xe_gt_topology.c +++ b/drivers/gpu/drm/xe/xe_gt_topology.c @@ -138,7 +138,7 @@ load_l3_bank_mask(struct xe_gt *gt, xe_l3_bank_mask_t l3_bank_mask) * but there's no tracking number assigned yet so we use a custom * OOB workaround descriptor. */ - if (XE_WA(gt, no_media_l3)) + if (XE_GT_WA(gt, no_media_l3)) return; if (GRAPHICS_VER(xe) >= 30) { diff --git a/drivers/gpu/drm/xe/xe_gt_types.h b/drivers/gpu/drm/xe/xe_gt_types.h index 96344c604726..66158105aca5 100644 --- a/drivers/gpu/drm/xe/xe_gt_types.h +++ b/drivers/gpu/drm/xe/xe_gt_types.h @@ -17,6 +17,7 @@ #include "xe_oa_types.h" #include "xe_reg_sr_types.h" #include "xe_sa_types.h" +#include "xe_tlb_inval_types.h" #include "xe_uc_types.h" struct xe_exec_queue_ops; @@ -185,34 +186,8 @@ struct xe_gt { struct work_struct worker; } reset; - /** @tlb_invalidation: TLB invalidation state */ - struct { - /** @tlb_invalidation.seqno: TLB invalidation seqno, protected by CT lock */ -#define TLB_INVALIDATION_SEQNO_MAX 0x100000 - int seqno; - /** - * @tlb_invalidation.seqno_recv: last received TLB invalidation seqno, - * protected by CT lock - */ - int seqno_recv; - /** - * @tlb_invalidation.pending_fences: list of pending fences waiting TLB - * invaliations, protected by CT lock - */ - struct list_head pending_fences; - /** - * @tlb_invalidation.pending_lock: protects @tlb_invalidation.pending_fences - * and updating @tlb_invalidation.seqno_recv. - */ - spinlock_t pending_lock; - /** - * @tlb_invalidation.fence_tdr: schedules a delayed call to - * xe_gt_tlb_fence_timeout after the timeut interval is over. - */ - struct delayed_work fence_tdr; - /** @tlb_invalidation.lock: protects TLB invalidation fences */ - spinlock_t lock; - } tlb_invalidation; + /** @tlb_inval: TLB invalidation state */ + struct xe_tlb_inval tlb_inval; /** * @ccs_mode: Number of compute engines enabled. @@ -411,7 +386,7 @@ struct xe_gt { unsigned long *oob; /** * @wa_active.oob_initialized: mark oob as initialized to help - * detecting misuse of XE_WA() - it can only be called on + * detecting misuse of XE_GT_WA() - it can only be called on * initialization after OOB WAs have being processed */ bool oob_initialized; diff --git a/drivers/gpu/drm/xe/xe_guc.c b/drivers/gpu/drm/xe/xe_guc.c index b1d1d6da3758..37d06c51180c 100644 --- a/drivers/gpu/drm/xe/xe_guc.c +++ b/drivers/gpu/drm/xe/xe_guc.c @@ -16,6 +16,7 @@ #include "regs/xe_guc_regs.h" #include "regs/xe_irq_regs.h" #include "xe_bo.h" +#include "xe_configfs.h" #include "xe_device.h" #include "xe_force_wake.h" #include "xe_gt.h" @@ -81,11 +82,15 @@ static u32 guc_ctl_debug_flags(struct xe_guc *guc) static u32 guc_ctl_feature_flags(struct xe_guc *guc) { + struct xe_device *xe = guc_to_xe(guc); u32 flags = GUC_CTL_ENABLE_LITE_RESTORE; - if (!guc_to_xe(guc)->info.skip_guc_pc) + if (!xe->info.skip_guc_pc) flags |= GUC_CTL_ENABLE_SLPC; + if (xe_configfs_get_psmi_enabled(to_pci_dev(xe->drm.dev))) + flags |= GUC_CTL_ENABLE_PSMI_LOGGING; + return flags; } @@ -157,7 +162,7 @@ static bool needs_wa_dual_queue(struct xe_gt *gt) * on RCS and CCSes with different address spaces, which on DG2 is * required as a WA for an HW bug. */ - if (XE_WA(gt, 22011391025)) + if (XE_GT_WA(gt, 22011391025)) return true; /* @@ -184,10 +189,10 @@ static u32 guc_ctl_wa_flags(struct xe_guc *guc) struct xe_gt *gt = guc_to_gt(guc); u32 flags = 0; - if (XE_WA(gt, 22012773006)) + if (XE_GT_WA(gt, 22012773006)) flags |= GUC_WA_POLLCS; - if (XE_WA(gt, 14014475959)) + if (XE_GT_WA(gt, 14014475959)) flags |= GUC_WA_HOLD_CCS_SWITCHOUT; if (needs_wa_dual_queue(gt)) @@ -201,19 +206,22 @@ static u32 guc_ctl_wa_flags(struct xe_guc *guc) if (GRAPHICS_VERx100(xe) < 1270) flags |= GUC_WA_PRE_PARSER; - if (XE_WA(gt, 22012727170) || XE_WA(gt, 22012727685)) + if (XE_GT_WA(gt, 22012727170) || XE_GT_WA(gt, 22012727685)) flags |= GUC_WA_CONTEXT_ISOLATION; - if (XE_WA(gt, 18020744125) && + if (XE_GT_WA(gt, 18020744125) && !xe_hw_engine_mask_per_class(gt, XE_ENGINE_CLASS_RENDER)) flags |= GUC_WA_RCS_REGS_IN_CCS_REGS_LIST; - if (XE_WA(gt, 1509372804)) + if (XE_GT_WA(gt, 1509372804)) flags |= GUC_WA_RENDER_RST_RC6_EXIT; - if (XE_WA(gt, 14018913170)) + if (XE_GT_WA(gt, 14018913170)) flags |= GUC_WA_ENABLE_TSC_CHECK_ON_RC6; + if (XE_GT_WA(gt, 16023683509)) + flags |= GUC_WA_SAVE_RESTORE_MCFG_REG_AT_MC6; + return flags; } @@ -992,11 +1000,14 @@ static int guc_load_done(u32 status) case XE_GUC_LOAD_STATUS_GUC_PREPROD_BUILD_MISMATCH: case XE_GUC_LOAD_STATUS_ERROR_DEVID_INVALID_GUCTYPE: case XE_GUC_LOAD_STATUS_HWCONFIG_ERROR: + case XE_GUC_LOAD_STATUS_BOOTROM_VERSION_MISMATCH: case XE_GUC_LOAD_STATUS_DPC_ERROR: case XE_GUC_LOAD_STATUS_EXCEPTION: case XE_GUC_LOAD_STATUS_INIT_DATA_INVALID: case XE_GUC_LOAD_STATUS_MPU_DATA_INVALID: case XE_GUC_LOAD_STATUS_INIT_MMIO_SAVE_RESTORE_INVALID: + case XE_GUC_LOAD_STATUS_KLV_WORKAROUND_INIT_ERROR: + case XE_GUC_LOAD_STATUS_INVALID_FTR_FLAG: return -1; } @@ -1136,17 +1147,29 @@ static void guc_wait_ucode(struct xe_guc *guc) } switch (ukernel) { + case XE_GUC_LOAD_STATUS_HWCONFIG_START: + xe_gt_err(gt, "still extracting hwconfig table.\n"); + break; + case XE_GUC_LOAD_STATUS_EXCEPTION: xe_gt_err(gt, "firmware exception. EIP: %#x\n", xe_mmio_read32(mmio, SOFT_SCRATCH(13))); break; + case XE_GUC_LOAD_STATUS_INIT_DATA_INVALID: + xe_gt_err(gt, "illegal init/ADS data\n"); + break; + case XE_GUC_LOAD_STATUS_INIT_MMIO_SAVE_RESTORE_INVALID: xe_gt_err(gt, "illegal register in save/restore workaround list\n"); break; - case XE_GUC_LOAD_STATUS_HWCONFIG_START: - xe_gt_err(gt, "still extracting hwconfig table.\n"); + case XE_GUC_LOAD_STATUS_KLV_WORKAROUND_INIT_ERROR: + xe_gt_err(gt, "illegal workaround KLV data\n"); + break; + + case XE_GUC_LOAD_STATUS_INVALID_FTR_FLAG: + xe_gt_err(gt, "illegal feature flag specified\n"); break; } diff --git a/drivers/gpu/drm/xe/xe_guc_ads.c b/drivers/gpu/drm/xe/xe_guc_ads.c index 131cfc56be00..5631722f34f5 100644 --- a/drivers/gpu/drm/xe/xe_guc_ads.c +++ b/drivers/gpu/drm/xe/xe_guc_ads.c @@ -247,7 +247,7 @@ static size_t calculate_regset_size(struct xe_gt *gt) count += ADS_REGSET_EXTRA_MAX * XE_NUM_HW_ENGINES; - if (XE_WA(gt, 1607983814)) + if (XE_GT_WA(gt, 1607983814)) count += LNCFCMOCS_REG_COUNT; return count * sizeof(struct guc_mmio_reg); @@ -284,52 +284,26 @@ static size_t calculate_golden_lrc_size(struct xe_guc_ads *ads) return total_size; } -static void guc_waklv_enable_one_word(struct xe_guc_ads *ads, - enum xe_guc_klv_ids klv_id, - u32 value, - u32 *offset, u32 *remain) +static void guc_waklv_enable(struct xe_guc_ads *ads, + u32 data[], u32 data_len_dw, + u32 *offset, u32 *remain, + enum xe_guc_klv_ids klv_id) { - u32 size; - u32 klv_entry[] = { - /* 16:16 key/length */ - FIELD_PREP(GUC_KLV_0_KEY, klv_id) | - FIELD_PREP(GUC_KLV_0_LEN, 1), - value, - /* 1 dword data */ - }; - - size = sizeof(klv_entry); + size_t size = sizeof(u32) * (1 + data_len_dw); if (*remain < size) { drm_warn(&ads_to_xe(ads)->drm, - "w/a klv buffer too small to add klv id %d\n", klv_id); - } else { - xe_map_memcpy_to(ads_to_xe(ads), ads_to_map(ads), *offset, - klv_entry, size); - *offset += size; - *remain -= size; + "w/a klv buffer too small to add klv id 0x%04X\n", klv_id); + return; } -} -static void guc_waklv_enable_simple(struct xe_guc_ads *ads, - enum xe_guc_klv_ids klv_id, u32 *offset, u32 *remain) -{ - u32 klv_entry[] = { - /* 16:16 key/length */ - FIELD_PREP(GUC_KLV_0_KEY, klv_id) | - FIELD_PREP(GUC_KLV_0_LEN, 0), - /* 0 dwords data */ - }; - u32 size; + /* 16:16 key/length */ + xe_map_wr(ads_to_xe(ads), ads_to_map(ads), *offset, u32, + FIELD_PREP(GUC_KLV_0_KEY, klv_id) | FIELD_PREP(GUC_KLV_0_LEN, data_len_dw)); + /* data_len_dw dwords of data */ + xe_map_memcpy_to(ads_to_xe(ads), ads_to_map(ads), + *offset + sizeof(u32), data, data_len_dw * sizeof(u32)); - size = sizeof(klv_entry); - - if (xe_gt_WARN(ads_to_gt(ads), *remain < size, - "w/a klv buffer too small to add klv id %d\n", klv_id)) - return; - - xe_map_memcpy_to(ads_to_xe(ads), ads_to_map(ads), *offset, - klv_entry, size); *offset += size; *remain -= size; } @@ -343,44 +317,51 @@ static void guc_waklv_init(struct xe_guc_ads *ads) offset = guc_ads_waklv_offset(ads); remain = guc_ads_waklv_size(ads); - if (XE_WA(gt, 14019882105) || XE_WA(gt, 16021333562)) - guc_waklv_enable_simple(ads, - GUC_WORKAROUND_KLV_BLOCK_INTERRUPTS_WHEN_MGSR_BLOCKED, - &offset, &remain); - if (XE_WA(gt, 18024947630)) - guc_waklv_enable_simple(ads, - GUC_WORKAROUND_KLV_ID_GAM_PFQ_SHADOW_TAIL_POLLING, - &offset, &remain); - if (XE_WA(gt, 16022287689)) - guc_waklv_enable_simple(ads, - GUC_WORKAROUND_KLV_ID_DISABLE_MTP_DURING_ASYNC_COMPUTE, - &offset, &remain); - - if (XE_WA(gt, 14022866841)) - guc_waklv_enable_simple(ads, - GUC_WA_KLV_WAKE_POWER_DOMAINS_FOR_OUTBOUND_MMIO, - &offset, &remain); + if (XE_GT_WA(gt, 14019882105) || XE_GT_WA(gt, 16021333562)) + guc_waklv_enable(ads, NULL, 0, &offset, &remain, + GUC_WORKAROUND_KLV_BLOCK_INTERRUPTS_WHEN_MGSR_BLOCKED); + if (XE_GT_WA(gt, 18024947630)) + guc_waklv_enable(ads, NULL, 0, &offset, &remain, + GUC_WORKAROUND_KLV_ID_GAM_PFQ_SHADOW_TAIL_POLLING); + if (XE_GT_WA(gt, 16022287689)) + guc_waklv_enable(ads, NULL, 0, &offset, &remain, + GUC_WORKAROUND_KLV_ID_DISABLE_MTP_DURING_ASYNC_COMPUTE); + + if (XE_GT_WA(gt, 14022866841)) + guc_waklv_enable(ads, NULL, 0, &offset, &remain, + GUC_WA_KLV_WAKE_POWER_DOMAINS_FOR_OUTBOUND_MMIO); /* * On RC6 exit, GuC will write register 0xB04 with the default value provided. As of now, * the default value for this register is determined to be 0xC40. This could change in the * future, so GuC depends on KMD to send it the correct value. */ - if (XE_WA(gt, 13011645652)) - guc_waklv_enable_one_word(ads, - GUC_WA_KLV_NP_RD_WRITE_TO_CLEAR_RCSM_AT_CGP_LATE_RESTORE, - 0xC40, - &offset, &remain); - - if (XE_WA(gt, 14022293748) || XE_WA(gt, 22019794406)) - guc_waklv_enable_simple(ads, - GUC_WORKAROUND_KLV_ID_BACK_TO_BACK_RCS_ENGINE_RESET, - &offset, &remain); - - if (GUC_FIRMWARE_VER(>->uc.guc) >= MAKE_GUC_VER(70, 44, 0) && XE_WA(gt, 16026508708)) - guc_waklv_enable_simple(ads, - GUC_WA_KLV_RESET_BB_STACK_PTR_ON_VF_SWITCH, - &offset, &remain); + if (XE_GT_WA(gt, 13011645652)) { + u32 data = 0xC40; + + guc_waklv_enable(ads, &data, sizeof(data) / sizeof(u32), &offset, &remain, + GUC_WA_KLV_NP_RD_WRITE_TO_CLEAR_RCSM_AT_CGP_LATE_RESTORE); + } + + if (XE_GT_WA(gt, 14022293748) || XE_GT_WA(gt, 22019794406)) + guc_waklv_enable(ads, NULL, 0, &offset, &remain, + GUC_WORKAROUND_KLV_ID_BACK_TO_BACK_RCS_ENGINE_RESET); + + if (GUC_FIRMWARE_VER(>->uc.guc) >= MAKE_GUC_VER(70, 44, 0) && XE_GT_WA(gt, 16026508708)) + guc_waklv_enable(ads, NULL, 0, &offset, &remain, + GUC_WA_KLV_RESET_BB_STACK_PTR_ON_VF_SWITCH); + if (GUC_FIRMWARE_VER(>->uc.guc) >= MAKE_GUC_VER(70, 47, 0) && XE_GT_WA(gt, 16026007364)) { + u32 data[] = { + 0x0, + 0xF, + }; + guc_waklv_enable(ads, data, sizeof(data) / sizeof(u32), &offset, &remain, + GUC_WA_KLV_RESTORE_UNSAVED_MEDIA_CONTROL_REG); + } + + if (XE_GT_WA(gt, 14020001231)) + guc_waklv_enable(ads, NULL, 0, &offset, &remain, + GUC_WORKAROUND_KLV_DISABLE_PSMI_INTERRUPTS_AT_C6_ENTRY_RESTORE_AT_EXIT); size = guc_ads_waklv_size(ads) - remain; if (!size) @@ -784,7 +765,7 @@ static unsigned int guc_mmio_regset_write(struct xe_guc_ads *ads, guc_mmio_regset_write_one(ads, regset_map, e->reg, count++); } - if (XE_WA(hwe->gt, 1607983814) && hwe->class == XE_ENGINE_CLASS_RENDER) { + if (XE_GT_WA(hwe->gt, 1607983814) && hwe->class == XE_ENGINE_CLASS_RENDER) { for (i = 0; i < LNCFCMOCS_REG_COUNT; i++) { guc_mmio_regset_write_one(ads, regset_map, XELP_LNCFCMOCS(i), count++); diff --git a/drivers/gpu/drm/xe/xe_guc_buf.c b/drivers/gpu/drm/xe/xe_guc_buf.c index 14a07dca48e7..502ca3a4ee60 100644 --- a/drivers/gpu/drm/xe/xe_guc_buf.c +++ b/drivers/gpu/drm/xe/xe_guc_buf.c @@ -164,7 +164,7 @@ u64 xe_guc_cache_gpu_addr_from_ptr(struct xe_guc_buf_cache *cache, const void *p if (offset < 0 || offset + size > cache->sam->base.size) return 0; - return cache->sam->gpu_addr + offset; + return xe_sa_manager_gpu_addr(cache->sam) + offset; } #if IS_BUILTIN(CONFIG_DRM_XE_KUNIT_TEST) diff --git a/drivers/gpu/drm/xe/xe_guc_ct.c b/drivers/gpu/drm/xe/xe_guc_ct.c index 3f4e6a46ff16..848065a25c44 100644 --- a/drivers/gpu/drm/xe/xe_guc_ct.c +++ b/drivers/gpu/drm/xe/xe_guc_ct.c @@ -26,11 +26,11 @@ #include "xe_gt_sriov_pf_control.h" #include "xe_gt_sriov_pf_monitor.h" #include "xe_gt_sriov_printk.h" -#include "xe_gt_tlb_invalidation.h" #include "xe_guc.h" #include "xe_guc_log.h" #include "xe_guc_relay.h" #include "xe_guc_submit.h" +#include "xe_guc_tlb_inval.h" #include "xe_map.h" #include "xe_pm.h" #include "xe_trace_guc.h" @@ -1416,8 +1416,7 @@ static int process_g2h_msg(struct xe_guc_ct *ct, u32 *msg, u32 len) ret = xe_guc_pagefault_handler(guc, payload, adj_len); break; case XE_GUC_ACTION_TLB_INVALIDATION_DONE: - ret = xe_guc_tlb_invalidation_done_handler(guc, payload, - adj_len); + ret = xe_guc_tlb_inval_done_handler(guc, payload, adj_len); break; case XE_GUC_ACTION_ACCESS_COUNTER_NOTIFY: ret = xe_guc_access_counter_notify_handler(guc, payload, @@ -1618,8 +1617,7 @@ static void g2h_fast_path(struct xe_guc_ct *ct, u32 *msg, u32 len) break; case XE_GUC_ACTION_TLB_INVALIDATION_DONE: __g2h_release_space(ct, len); - ret = xe_guc_tlb_invalidation_done_handler(guc, payload, - adj_len); + ret = xe_guc_tlb_inval_done_handler(guc, payload, adj_len); break; default: xe_gt_warn(gt, "NOT_POSSIBLE"); diff --git a/drivers/gpu/drm/xe/xe_guc_fwif.h b/drivers/gpu/drm/xe/xe_guc_fwif.h index 6f57578b07cb..0508f1064178 100644 --- a/drivers/gpu/drm/xe/xe_guc_fwif.h +++ b/drivers/gpu/drm/xe/xe_guc_fwif.h @@ -45,6 +45,11 @@ #define GUC_MAX_ENGINE_CLASSES 16 #define GUC_MAX_INSTANCES_PER_CLASS 32 +#define GUC_CONTEXT_NORMAL 0 +#define GUC_CONTEXT_COMPRESSION_SAVE 1 +#define GUC_CONTEXT_COMPRESSION_RESTORE 2 +#define GUC_CONTEXT_COUNT (GUC_CONTEXT_COMPRESSION_RESTORE + 1) + /* Helper for context registration H2G */ struct guc_ctxt_registration_info { u32 flags; @@ -103,10 +108,12 @@ struct guc_update_exec_queue_policy { #define GUC_WA_RENDER_RST_RC6_EXIT BIT(19) #define GUC_WA_RCS_REGS_IN_CCS_REGS_LIST BIT(21) #define GUC_WA_ENABLE_TSC_CHECK_ON_RC6 BIT(22) +#define GUC_WA_SAVE_RESTORE_MCFG_REG_AT_MC6 BIT(25) #define GUC_CTL_FEATURE 2 #define GUC_CTL_ENABLE_SLPC BIT(2) #define GUC_CTL_ENABLE_LITE_RESTORE BIT(4) +#define GUC_CTL_ENABLE_PSMI_LOGGING BIT(7) #define GUC_CTL_DISABLE_SCHEDULER BIT(14) #define GUC_CTL_DEBUG 3 diff --git a/drivers/gpu/drm/xe/xe_guc_pc.c b/drivers/gpu/drm/xe/xe_guc_pc.c index 68b192fe3b32..88557e86d637 100644 --- a/drivers/gpu/drm/xe/xe_guc_pc.c +++ b/drivers/gpu/drm/xe/xe_guc_pc.c @@ -722,7 +722,7 @@ static int xe_guc_pc_set_max_freq_locked(struct xe_guc_pc *pc, u32 freq) */ int xe_guc_pc_set_max_freq(struct xe_guc_pc *pc, u32 freq) { - if (XE_WA(pc_to_gt(pc), 22019338487)) { + if (XE_GT_WA(pc_to_gt(pc), 22019338487)) { if (wait_for_flush_complete(pc) != 0) return -EAGAIN; } @@ -835,7 +835,7 @@ static u32 pc_max_freq_cap(struct xe_guc_pc *pc) { struct xe_gt *gt = pc_to_gt(pc); - if (XE_WA(gt, 22019338487)) { + if (XE_GT_WA(gt, 22019338487)) { if (xe_gt_is_media_type(gt)) return min(LNL_MERT_FREQ_CAP, pc->rp0_freq); else @@ -899,7 +899,7 @@ static int pc_adjust_freq_bounds(struct xe_guc_pc *pc) if (pc_get_min_freq(pc) > pc->rp0_freq) ret = pc_set_min_freq(pc, pc->rp0_freq); - if (XE_WA(tile->primary_gt, 14022085890)) + if (XE_GT_WA(tile->primary_gt, 14022085890)) ret = pc_set_min_freq(pc, max(BMG_MIN_FREQ, pc_get_min_freq(pc))); out: @@ -931,7 +931,7 @@ static bool needs_flush_freq_limit(struct xe_guc_pc *pc) { struct xe_gt *gt = pc_to_gt(pc); - return XE_WA(gt, 22019338487) && + return XE_GT_WA(gt, 22019338487) && pc->rp0_freq > BMG_MERT_FLUSH_FREQ_CAP; } @@ -1017,7 +1017,7 @@ static int pc_set_mert_freq_cap(struct xe_guc_pc *pc) { int ret; - if (!XE_WA(pc_to_gt(pc), 22019338487)) + if (!XE_GT_WA(pc_to_gt(pc), 22019338487)) return 0; guard(mutex)(&pc->freq_lock); @@ -1076,7 +1076,6 @@ int xe_guc_pc_gucrc_disable(struct xe_guc_pc *pc) { struct xe_device *xe = pc_to_xe(pc); struct xe_gt *gt = pc_to_gt(pc); - unsigned int fw_ref; int ret = 0; if (xe->info.skip_guc_pc) @@ -1086,17 +1085,7 @@ int xe_guc_pc_gucrc_disable(struct xe_guc_pc *pc) if (ret) return ret; - fw_ref = xe_force_wake_get(gt_to_fw(gt), XE_FORCEWAKE_ALL); - if (!xe_force_wake_ref_has_domain(fw_ref, XE_FORCEWAKE_ALL)) { - xe_force_wake_put(gt_to_fw(gt), fw_ref); - return -ETIMEDOUT; - } - - xe_gt_idle_disable_c6(gt); - - xe_force_wake_put(gt_to_fw(gt), fw_ref); - - return 0; + return xe_gt_idle_disable_c6(gt); } /** diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c index cafb47711e9b..1185b23b1384 100644 --- a/drivers/gpu/drm/xe/xe_guc_submit.c +++ b/drivers/gpu/drm/xe/xe_guc_submit.c @@ -542,7 +542,7 @@ static void __register_exec_queue(struct xe_guc *guc, xe_guc_ct_send(&guc->ct, action, ARRAY_SIZE(action), 0, 0); } -static void register_exec_queue(struct xe_exec_queue *q) +static void register_exec_queue(struct xe_exec_queue *q, int ctx_type) { struct xe_guc *guc = exec_queue_to_guc(q); struct xe_device *xe = guc_to_xe(guc); @@ -550,6 +550,7 @@ static void register_exec_queue(struct xe_exec_queue *q) struct guc_ctxt_registration_info info; xe_gt_assert(guc_to_gt(guc), !exec_queue_registered(q)); + xe_gt_assert(guc_to_gt(guc), ctx_type < GUC_CONTEXT_COUNT); memset(&info, 0, sizeof(info)); info.context_idx = q->guc->id; @@ -559,6 +560,9 @@ static void register_exec_queue(struct xe_exec_queue *q) info.hwlrca_hi = upper_32_bits(xe_lrc_descriptor(lrc)); info.flags = CONTEXT_REGISTRATION_FLAG_KMD; + if (ctx_type != GUC_CONTEXT_NORMAL) + info.flags |= BIT(ctx_type); + if (xe_exec_queue_is_parallel(q)) { u64 ggtt_addr = xe_lrc_parallel_ggtt_addr(lrc); struct iosys_map map = xe_lrc_parallel_map(lrc); @@ -667,12 +671,18 @@ static void wq_item_append(struct xe_exec_queue *q) if (wq_wait_for_space(q, wqi_size)) return; + xe_gt_assert(guc_to_gt(guc), i == XE_GUC_CONTEXT_WQ_HEADER_DATA_0_TYPE_LEN); wqi[i++] = FIELD_PREP(WQ_TYPE_MASK, WQ_TYPE_MULTI_LRC) | FIELD_PREP(WQ_LEN_MASK, len_dw); + xe_gt_assert(guc_to_gt(guc), i == XE_GUC_CONTEXT_WQ_EL_INFO_DATA_1_CTX_DESC_LOW); wqi[i++] = xe_lrc_descriptor(q->lrc[0]); + xe_gt_assert(guc_to_gt(guc), i == + XE_GUC_CONTEXT_WQ_EL_INFO_DATA_2_GUCCTX_RINGTAIL_FREEZEPOCS); wqi[i++] = FIELD_PREP(WQ_GUC_ID_MASK, q->guc->id) | FIELD_PREP(WQ_RING_TAIL_MASK, q->lrc[0]->ring.tail / sizeof(u64)); + xe_gt_assert(guc_to_gt(guc), i == XE_GUC_CONTEXT_WQ_EL_INFO_DATA_3_WI_FENCE_ID); wqi[i++] = 0; + xe_gt_assert(guc_to_gt(guc), i == XE_GUC_CONTEXT_WQ_EL_CHILD_LIST_DATA_4_RINGTAIL); for (j = 1; j < q->width; ++j) { struct xe_lrc *lrc = q->lrc[j]; @@ -693,6 +703,50 @@ static void wq_item_append(struct xe_exec_queue *q) parallel_write(xe, map, wq_desc.tail, q->guc->wqi_tail); } +static int wq_items_rebase(struct xe_exec_queue *q) +{ + struct xe_guc *guc = exec_queue_to_guc(q); + struct xe_device *xe = guc_to_xe(guc); + struct iosys_map map = xe_lrc_parallel_map(q->lrc[0]); + int i = q->guc->wqi_head; + + /* the ring starts after a header struct */ + iosys_map_incr(&map, offsetof(struct guc_submit_parallel_scratch, wq[0])); + + while ((i % WQ_SIZE) != (q->guc->wqi_tail % WQ_SIZE)) { + u32 len_dw, type, val; + + if (drm_WARN_ON_ONCE(&xe->drm, i < 0 || i > 2 * WQ_SIZE)) + break; + + val = xe_map_rd_ring_u32(xe, &map, i / sizeof(u32) + + XE_GUC_CONTEXT_WQ_HEADER_DATA_0_TYPE_LEN, + WQ_SIZE / sizeof(u32)); + len_dw = FIELD_GET(WQ_LEN_MASK, val); + type = FIELD_GET(WQ_TYPE_MASK, val); + + if (drm_WARN_ON_ONCE(&xe->drm, len_dw >= WQ_SIZE / sizeof(u32))) + break; + + if (type == WQ_TYPE_MULTI_LRC) { + val = xe_lrc_descriptor(q->lrc[0]); + xe_map_wr_ring_u32(xe, &map, i / sizeof(u32) + + XE_GUC_CONTEXT_WQ_EL_INFO_DATA_1_CTX_DESC_LOW, + WQ_SIZE / sizeof(u32), val); + } else if (drm_WARN_ON_ONCE(&xe->drm, type != WQ_TYPE_NOOP)) { + break; + } + + i += (len_dw + 1) * sizeof(u32); + } + + if ((i % WQ_SIZE) != (q->guc->wqi_tail % WQ_SIZE)) { + xe_gt_err(q->gt, "Exec queue fixups incomplete - wqi parse failed\n"); + return -EBADMSG; + } + return 0; +} + #define RESUME_PENDING ~0x0ull static void submit_exec_queue(struct xe_exec_queue *q) { @@ -761,7 +815,7 @@ guc_exec_queue_run_job(struct drm_sched_job *drm_job) if (!exec_queue_killed_or_banned_or_wedged(q) && !xe_sched_job_is_error(job)) { if (!exec_queue_registered(q)) - register_exec_queue(q); + register_exec_queue(q, GUC_CONTEXT_NORMAL); if (!lr) /* LR jobs are emitted in the exec IOCTL */ q->ring_ops->emit_job(job); submit_exec_queue(q); @@ -777,6 +831,30 @@ guc_exec_queue_run_job(struct drm_sched_job *drm_job) return fence; } +/** + * xe_guc_jobs_ring_rebase - Re-emit ring commands of requests pending + * on all queues under a guc. + * @guc: the &xe_guc struct instance + */ +void xe_guc_jobs_ring_rebase(struct xe_guc *guc) +{ + struct xe_exec_queue *q; + unsigned long index; + + /* + * This routine is used within VF migration recovery. This means + * using the lock here introduces a restriction: we cannot wait + * for any GFX HW response while the lock is taken. + */ + mutex_lock(&guc->submission_state.lock); + xa_for_each(&guc->submission_state.exec_queue_lookup, index, q) { + if (exec_queue_killed_or_banned_or_wedged(q)) + continue; + xe_exec_queue_jobs_ring_restore(q); + } + mutex_unlock(&guc->submission_state.lock); +} + static void guc_exec_queue_free_job(struct drm_sched_job *drm_job) { struct xe_sched_job *job = to_xe_sched_job(drm_job); @@ -1773,6 +1851,43 @@ static void guc_exec_queue_stop(struct xe_guc *guc, struct xe_exec_queue *q) } } +/** + * xe_guc_submit_reset_block - Disallow reset calls on given GuC. + * @guc: the &xe_guc struct instance + */ +int xe_guc_submit_reset_block(struct xe_guc *guc) +{ + return atomic_fetch_or(1, &guc->submission_state.reset_blocked); +} + +/** + * xe_guc_submit_reset_unblock - Allow back reset calls on given GuC. + * @guc: the &xe_guc struct instance + */ +void xe_guc_submit_reset_unblock(struct xe_guc *guc) +{ + atomic_set_release(&guc->submission_state.reset_blocked, 0); + wake_up_all(&guc->ct.wq); +} + +static int guc_submit_reset_is_blocked(struct xe_guc *guc) +{ + return atomic_read_acquire(&guc->submission_state.reset_blocked); +} + +/* Maximum time of blocking reset */ +#define RESET_BLOCK_PERIOD_MAX (HZ * 5) + +/** + * xe_guc_wait_reset_unblock - Wait until reset blocking flag is lifted, or timeout. + * @guc: the &xe_guc struct instance + */ +int xe_guc_wait_reset_unblock(struct xe_guc *guc) +{ + return wait_event_timeout(guc->ct.wq, + !guc_submit_reset_is_blocked(guc), RESET_BLOCK_PERIOD_MAX); +} + int xe_guc_submit_reset_prepare(struct xe_guc *guc) { int ret; @@ -1826,6 +1941,19 @@ void xe_guc_submit_stop(struct xe_guc *guc) } +/** + * xe_guc_submit_pause - Stop further runs of submission tasks on given GuC. + * @guc: the &xe_guc struct instance whose scheduler is to be disabled + */ +void xe_guc_submit_pause(struct xe_guc *guc) +{ + struct xe_exec_queue *q; + unsigned long index; + + xa_for_each(&guc->submission_state.exec_queue_lookup, index, q) + xe_sched_submission_stop_async(&q->guc->sched); +} + static void guc_exec_queue_start(struct xe_exec_queue *q) { struct xe_gpu_scheduler *sched = &q->guc->sched; @@ -1866,6 +1994,28 @@ int xe_guc_submit_start(struct xe_guc *guc) return 0; } +static void guc_exec_queue_unpause(struct xe_exec_queue *q) +{ + struct xe_gpu_scheduler *sched = &q->guc->sched; + + xe_sched_submission_start(sched); +} + +/** + * xe_guc_submit_unpause - Allow further runs of submission tasks on given GuC. + * @guc: the &xe_guc struct instance whose scheduler is to be enabled + */ +void xe_guc_submit_unpause(struct xe_guc *guc) +{ + struct xe_exec_queue *q; + unsigned long index; + + xa_for_each(&guc->submission_state.exec_queue_lookup, index, q) + guc_exec_queue_unpause(q); + + wake_up_all(&guc->ct.wq); +} + static struct xe_exec_queue * g2h_exec_queue_lookup(struct xe_guc *guc, u32 guc_id) { @@ -2378,6 +2528,32 @@ static void guc_exec_queue_print(struct xe_exec_queue *q, struct drm_printer *p) } /** + * xe_guc_register_exec_queue - Register exec queue for a given context type. + * @q: Execution queue + * @ctx_type: Type of the context + * + * This function registers the execution queue with the guc. Special context + * types like GUC_CONTEXT_COMPRESSION_SAVE and GUC_CONTEXT_COMPRESSION_RESTORE + * are only applicable for IGPU and in the VF. + * Submits the execution queue to GUC after registering it. + * + * Returns - None. + */ +void xe_guc_register_exec_queue(struct xe_exec_queue *q, int ctx_type) +{ + struct xe_guc *guc = exec_queue_to_guc(q); + struct xe_device *xe = guc_to_xe(guc); + + xe_assert(xe, IS_SRIOV_VF(xe)); + xe_assert(xe, !IS_DGFX(xe)); + xe_assert(xe, (ctx_type > GUC_CONTEXT_NORMAL && + ctx_type < GUC_CONTEXT_COUNT)); + + register_exec_queue(q, ctx_type); + enable_scheduling(q); +} + +/** * xe_guc_submit_print - GuC Submit Print. * @guc: GuC. * @p: drm_printer where it will be printed out. @@ -2397,3 +2573,32 @@ void xe_guc_submit_print(struct xe_guc *guc, struct drm_printer *p) guc_exec_queue_print(q, p); mutex_unlock(&guc->submission_state.lock); } + +/** + * xe_guc_contexts_hwsp_rebase - Re-compute GGTT references within all + * exec queues registered to given GuC. + * @guc: the &xe_guc struct instance + * @scratch: scratch buffer to be used as temporary storage + * + * Returns: zero on success, negative error code on failure. + */ +int xe_guc_contexts_hwsp_rebase(struct xe_guc *guc, void *scratch) +{ + struct xe_exec_queue *q; + unsigned long index; + int err = 0; + + mutex_lock(&guc->submission_state.lock); + xa_for_each(&guc->submission_state.exec_queue_lookup, index, q) { + err = xe_exec_queue_contexts_hwsp_rebase(q, scratch); + if (err) + break; + if (xe_exec_queue_is_parallel(q)) + err = wq_items_rebase(q); + if (err) + break; + } + mutex_unlock(&guc->submission_state.lock); + + return err; +} diff --git a/drivers/gpu/drm/xe/xe_guc_submit.h b/drivers/gpu/drm/xe/xe_guc_submit.h index 9b71a986c6ca..6b5df5d0956b 100644 --- a/drivers/gpu/drm/xe/xe_guc_submit.h +++ b/drivers/gpu/drm/xe/xe_guc_submit.h @@ -18,6 +18,11 @@ int xe_guc_submit_reset_prepare(struct xe_guc *guc); void xe_guc_submit_reset_wait(struct xe_guc *guc); void xe_guc_submit_stop(struct xe_guc *guc); int xe_guc_submit_start(struct xe_guc *guc); +void xe_guc_submit_pause(struct xe_guc *guc); +void xe_guc_submit_unpause(struct xe_guc *guc); +int xe_guc_submit_reset_block(struct xe_guc *guc); +void xe_guc_submit_reset_unblock(struct xe_guc *guc); +int xe_guc_wait_reset_unblock(struct xe_guc *guc); void xe_guc_submit_wedge(struct xe_guc *guc); int xe_guc_read_stopped(struct xe_guc *guc); @@ -29,6 +34,8 @@ int xe_guc_exec_queue_memory_cat_error_handler(struct xe_guc *guc, u32 *msg, int xe_guc_exec_queue_reset_failure_handler(struct xe_guc *guc, u32 *msg, u32 len); int xe_guc_error_capture_handler(struct xe_guc *guc, u32 *msg, u32 len); +void xe_guc_jobs_ring_rebase(struct xe_guc *guc); + struct xe_guc_submit_exec_queue_snapshot * xe_guc_exec_queue_snapshot_capture(struct xe_exec_queue *q); void @@ -39,5 +46,8 @@ xe_guc_exec_queue_snapshot_print(struct xe_guc_submit_exec_queue_snapshot *snaps void xe_guc_exec_queue_snapshot_free(struct xe_guc_submit_exec_queue_snapshot *snapshot); void xe_guc_submit_print(struct xe_guc *guc, struct drm_printer *p); +void xe_guc_register_exec_queue(struct xe_exec_queue *q, int ctx_type); + +int xe_guc_contexts_hwsp_rebase(struct xe_guc *guc, void *scratch); #endif diff --git a/drivers/gpu/drm/xe/xe_guc_tlb_inval.c b/drivers/gpu/drm/xe/xe_guc_tlb_inval.c new file mode 100644 index 000000000000..6bf2103602f8 --- /dev/null +++ b/drivers/gpu/drm/xe/xe_guc_tlb_inval.c @@ -0,0 +1,242 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright © 2025 Intel Corporation + */ + +#include "abi/guc_actions_abi.h" + +#include "xe_device.h" +#include "xe_gt_stats.h" +#include "xe_gt_types.h" +#include "xe_guc.h" +#include "xe_guc_ct.h" +#include "xe_guc_tlb_inval.h" +#include "xe_force_wake.h" +#include "xe_mmio.h" +#include "xe_tlb_inval.h" + +#include "regs/xe_guc_regs.h" + +/* + * XXX: The seqno algorithm relies on TLB invalidation being processed in order + * which they currently are by the GuC, if that changes the algorithm will need + * to be updated. + */ + +static int send_tlb_inval(struct xe_guc *guc, const u32 *action, int len) +{ + struct xe_gt *gt = guc_to_gt(guc); + + xe_gt_assert(gt, action[1]); /* Seqno */ + + xe_gt_stats_incr(gt, XE_GT_STATS_ID_TLB_INVAL, 1); + return xe_guc_ct_send(&guc->ct, action, len, + G2H_LEN_DW_TLB_INVALIDATE, 1); +} + +#define MAKE_INVAL_OP(type) ((type << XE_GUC_TLB_INVAL_TYPE_SHIFT) | \ + XE_GUC_TLB_INVAL_MODE_HEAVY << XE_GUC_TLB_INVAL_MODE_SHIFT | \ + XE_GUC_TLB_INVAL_FLUSH_CACHE) + +static int send_tlb_inval_all(struct xe_tlb_inval *tlb_inval, u32 seqno) +{ + struct xe_guc *guc = tlb_inval->private; + u32 action[] = { + XE_GUC_ACTION_TLB_INVALIDATION_ALL, + seqno, + MAKE_INVAL_OP(XE_GUC_TLB_INVAL_FULL), + }; + + return send_tlb_inval(guc, action, ARRAY_SIZE(action)); +} + +static int send_tlb_inval_ggtt(struct xe_tlb_inval *tlb_inval, u32 seqno) +{ + struct xe_guc *guc = tlb_inval->private; + struct xe_gt *gt = guc_to_gt(guc); + struct xe_device *xe = guc_to_xe(guc); + + /* + * Returning -ECANCELED in this function is squashed at the caller and + * signals waiters. + */ + + if (xe_guc_ct_enabled(&guc->ct) && guc->submission_state.enabled) { + u32 action[] = { + XE_GUC_ACTION_TLB_INVALIDATION, + seqno, + MAKE_INVAL_OP(XE_GUC_TLB_INVAL_GUC), + }; + + return send_tlb_inval(guc, action, ARRAY_SIZE(action)); + } else if (xe_device_uc_enabled(xe) && !xe_device_wedged(xe)) { + struct xe_mmio *mmio = >->mmio; + unsigned int fw_ref; + + if (IS_SRIOV_VF(xe)) + return -ECANCELED; + + fw_ref = xe_force_wake_get(gt_to_fw(gt), XE_FW_GT); + if (xe->info.platform == XE_PVC || GRAPHICS_VER(xe) >= 20) { + xe_mmio_write32(mmio, PVC_GUC_TLB_INV_DESC1, + PVC_GUC_TLB_INV_DESC1_INVALIDATE); + xe_mmio_write32(mmio, PVC_GUC_TLB_INV_DESC0, + PVC_GUC_TLB_INV_DESC0_VALID); + } else { + xe_mmio_write32(mmio, GUC_TLB_INV_CR, + GUC_TLB_INV_CR_INVALIDATE); + } + xe_force_wake_put(gt_to_fw(gt), fw_ref); + } + + return -ECANCELED; +} + +/* + * Ensure that roundup_pow_of_two(length) doesn't overflow. + * Note that roundup_pow_of_two() operates on unsigned long, + * not on u64. + */ +#define MAX_RANGE_TLB_INVALIDATION_LENGTH (rounddown_pow_of_two(ULONG_MAX)) + +static int send_tlb_inval_ppgtt(struct xe_tlb_inval *tlb_inval, u32 seqno, + u64 start, u64 end, u32 asid) +{ +#define MAX_TLB_INVALIDATION_LEN 7 + struct xe_guc *guc = tlb_inval->private; + struct xe_gt *gt = guc_to_gt(guc); + u32 action[MAX_TLB_INVALIDATION_LEN]; + u64 length = end - start; + int len = 0; + + if (guc_to_xe(guc)->info.force_execlist) + return -ECANCELED; + + action[len++] = XE_GUC_ACTION_TLB_INVALIDATION; + action[len++] = seqno; + if (!gt_to_xe(gt)->info.has_range_tlb_inval || + length > MAX_RANGE_TLB_INVALIDATION_LENGTH) { + action[len++] = MAKE_INVAL_OP(XE_GUC_TLB_INVAL_FULL); + } else { + u64 orig_start = start; + u64 align; + + if (length < SZ_4K) + length = SZ_4K; + + /* + * We need to invalidate a higher granularity if start address + * is not aligned to length. When start is not aligned with + * length we need to find the length large enough to create an + * address mask covering the required range. + */ + align = roundup_pow_of_two(length); + start = ALIGN_DOWN(start, align); + end = ALIGN(end, align); + length = align; + while (start + length < end) { + length <<= 1; + start = ALIGN_DOWN(orig_start, length); + } + + /* + * Minimum invalidation size for a 2MB page that the hardware + * expects is 16MB + */ + if (length >= SZ_2M) { + length = max_t(u64, SZ_16M, length); + start = ALIGN_DOWN(orig_start, length); + } + + xe_gt_assert(gt, length >= SZ_4K); + xe_gt_assert(gt, is_power_of_2(length)); + xe_gt_assert(gt, !(length & GENMASK(ilog2(SZ_16M) - 1, + ilog2(SZ_2M) + 1))); + xe_gt_assert(gt, IS_ALIGNED(start, length)); + + action[len++] = MAKE_INVAL_OP(XE_GUC_TLB_INVAL_PAGE_SELECTIVE); + action[len++] = asid; + action[len++] = lower_32_bits(start); + action[len++] = upper_32_bits(start); + action[len++] = ilog2(length) - ilog2(SZ_4K); + } + + xe_gt_assert(gt, len <= MAX_TLB_INVALIDATION_LEN); + + return send_tlb_inval(guc, action, len); +} + +static bool tlb_inval_initialized(struct xe_tlb_inval *tlb_inval) +{ + struct xe_guc *guc = tlb_inval->private; + + return xe_guc_ct_initialized(&guc->ct); +} + +static void tlb_inval_flush(struct xe_tlb_inval *tlb_inval) +{ + struct xe_guc *guc = tlb_inval->private; + + LNL_FLUSH_WORK(&guc->ct.g2h_worker); +} + +static long tlb_inval_timeout_delay(struct xe_tlb_inval *tlb_inval) +{ + struct xe_guc *guc = tlb_inval->private; + + /* this reflects what HW/GuC needs to process TLB inv request */ + const long hw_tlb_timeout = HZ / 4; + + /* this estimates actual delay caused by the CTB transport */ + long delay = xe_guc_ct_queue_proc_time_jiffies(&guc->ct); + + return hw_tlb_timeout + 2 * delay; +} + +static const struct xe_tlb_inval_ops guc_tlb_inval_ops = { + .all = send_tlb_inval_all, + .ggtt = send_tlb_inval_ggtt, + .ppgtt = send_tlb_inval_ppgtt, + .initialized = tlb_inval_initialized, + .flush = tlb_inval_flush, + .timeout_delay = tlb_inval_timeout_delay, +}; + +/** + * xe_guc_tlb_inval_init_early() - Init GuC TLB invalidation early + * @guc: GuC object + * @tlb_inval: TLB invalidation client + * + * Inititialize GuC TLB invalidation by setting back pointer in TLB invalidation + * client to the GuC and setting GuC backend ops. + */ +void xe_guc_tlb_inval_init_early(struct xe_guc *guc, + struct xe_tlb_inval *tlb_inval) +{ + tlb_inval->private = guc; + tlb_inval->ops = &guc_tlb_inval_ops; +} + +/** + * xe_guc_tlb_inval_done_handler() - TLB invalidation done handler + * @guc: guc + * @msg: message indicating TLB invalidation done + * @len: length of message + * + * Parse seqno of TLB invalidation, wake any waiters for seqno, and signal any + * invalidation fences for seqno. Algorithm for this depends on seqno being + * received in-order and asserts this assumption. + * + * Return: 0 on success, -EPROTO for malformed messages. + */ +int xe_guc_tlb_inval_done_handler(struct xe_guc *guc, u32 *msg, u32 len) +{ + struct xe_gt *gt = guc_to_gt(guc); + + if (unlikely(len != 1)) + return -EPROTO; + + xe_tlb_inval_done_handler(>->tlb_inval, msg[0]); + + return 0; +} diff --git a/drivers/gpu/drm/xe/xe_guc_tlb_inval.h b/drivers/gpu/drm/xe/xe_guc_tlb_inval.h new file mode 100644 index 000000000000..07d668b02e3d --- /dev/null +++ b/drivers/gpu/drm/xe/xe_guc_tlb_inval.h @@ -0,0 +1,19 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2025 Intel Corporation + */ + +#ifndef _XE_GUC_TLB_INVAL_H_ +#define _XE_GUC_TLB_INVAL_H_ + +#include <linux/types.h> + +struct xe_guc; +struct xe_tlb_inval; + +void xe_guc_tlb_inval_init_early(struct xe_guc *guc, + struct xe_tlb_inval *tlb_inval); + +int xe_guc_tlb_inval_done_handler(struct xe_guc *guc, u32 *msg, u32 len); + +#endif diff --git a/drivers/gpu/drm/xe/xe_guc_types.h b/drivers/gpu/drm/xe/xe_guc_types.h index 1fde7614fcc5..c7b9642b41ba 100644 --- a/drivers/gpu/drm/xe/xe_guc_types.h +++ b/drivers/gpu/drm/xe/xe_guc_types.h @@ -85,6 +85,12 @@ struct xe_guc { struct xarray exec_queue_lookup; /** @submission_state.stopped: submissions are stopped */ atomic_t stopped; + /** + * @submission_state.reset_blocked: reset attempts are blocked; + * blocking reset in order to delay it may be required if running + * an operation which is sensitive to resets. + */ + atomic_t reset_blocked; /** @submission_state.lock: protects submission state */ struct mutex lock; /** @submission_state.enabled: submission is enabled */ diff --git a/drivers/gpu/drm/xe/xe_heci_gsc.c b/drivers/gpu/drm/xe/xe_heci_gsc.c index 6d7b62724126..a415ca488791 100644 --- a/drivers/gpu/drm/xe/xe_heci_gsc.c +++ b/drivers/gpu/drm/xe/xe_heci_gsc.c @@ -197,7 +197,7 @@ int xe_heci_gsc_init(struct xe_device *xe) if (ret) return ret; - if (!def->use_polling && !xe_survivability_mode_is_enabled(xe)) { + if (!def->use_polling && !xe_survivability_mode_is_boot_enabled(xe)) { ret = heci_gsc_irq_setup(xe); if (ret) return ret; diff --git a/drivers/gpu/drm/xe/xe_hw_engine.c b/drivers/gpu/drm/xe/xe_hw_engine.c index 796ba8c34a16..1cf623b4a5bc 100644 --- a/drivers/gpu/drm/xe/xe_hw_engine.c +++ b/drivers/gpu/drm/xe/xe_hw_engine.c @@ -576,7 +576,7 @@ static void adjust_idledly(struct xe_hw_engine *hwe) u32 maxcnt_units_ns = 640; bool inhibit_switch = 0; - if (!IS_SRIOV_VF(gt_to_xe(hwe->gt)) && XE_WA(gt, 16023105232)) { + if (!IS_SRIOV_VF(gt_to_xe(hwe->gt)) && XE_GT_WA(gt, 16023105232)) { idledly = xe_mmio_read32(>->mmio, RING_IDLEDLY(hwe->mmio_base)); maxcnt = xe_mmio_read32(>->mmio, RING_PWRCTX_MAXCNT(hwe->mmio_base)); diff --git a/drivers/gpu/drm/xe/xe_hw_engine_group.c b/drivers/gpu/drm/xe/xe_hw_engine_group.c index c926f840c87b..58bee3ffe881 100644 --- a/drivers/gpu/drm/xe/xe_hw_engine_group.c +++ b/drivers/gpu/drm/xe/xe_hw_engine_group.c @@ -103,8 +103,8 @@ int xe_hw_engine_setup_groups(struct xe_gt *gt) break; case XE_ENGINE_CLASS_OTHER: break; - default: - drm_warn(&xe->drm, "NOT POSSIBLE"); + case XE_ENGINE_CLASS_MAX: + xe_gt_assert(gt, false); } } diff --git a/drivers/gpu/drm/xe/xe_hw_error.c b/drivers/gpu/drm/xe/xe_hw_error.c new file mode 100644 index 000000000000..8c65291f36fc --- /dev/null +++ b/drivers/gpu/drm/xe/xe_hw_error.c @@ -0,0 +1,182 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright © 2025 Intel Corporation + */ + +#include <linux/fault-inject.h> + +#include "regs/xe_gsc_regs.h" +#include "regs/xe_hw_error_regs.h" +#include "regs/xe_irq_regs.h" + +#include "xe_device.h" +#include "xe_hw_error.h" +#include "xe_mmio.h" +#include "xe_survivability_mode.h" + +#define HEC_UNCORR_FW_ERR_BITS 4 +extern struct fault_attr inject_csc_hw_error; + +/* Error categories reported by hardware */ +enum hardware_error { + HARDWARE_ERROR_CORRECTABLE = 0, + HARDWARE_ERROR_NONFATAL = 1, + HARDWARE_ERROR_FATAL = 2, + HARDWARE_ERROR_MAX, +}; + +static const char * const hec_uncorrected_fw_errors[] = { + "Fatal", + "CSE Disabled", + "FD Corruption", + "Data Corruption" +}; + +static const char *hw_error_to_str(const enum hardware_error hw_err) +{ + switch (hw_err) { + case HARDWARE_ERROR_CORRECTABLE: + return "CORRECTABLE"; + case HARDWARE_ERROR_NONFATAL: + return "NONFATAL"; + case HARDWARE_ERROR_FATAL: + return "FATAL"; + default: + return "UNKNOWN"; + } +} + +static bool fault_inject_csc_hw_error(void) +{ + return IS_ENABLED(CONFIG_DEBUG_FS) && should_fail(&inject_csc_hw_error, 1); +} + +static void csc_hw_error_work(struct work_struct *work) +{ + struct xe_tile *tile = container_of(work, typeof(*tile), csc_hw_error_work); + struct xe_device *xe = tile_to_xe(tile); + int ret; + + ret = xe_survivability_mode_runtime_enable(xe); + if (ret) + drm_err(&xe->drm, "Failed to enable runtime survivability mode\n"); +} + +static void csc_hw_error_handler(struct xe_tile *tile, const enum hardware_error hw_err) +{ + const char *hw_err_str = hw_error_to_str(hw_err); + struct xe_device *xe = tile_to_xe(tile); + struct xe_mmio *mmio = &tile->mmio; + u32 base, err_bit, err_src; + unsigned long fw_err; + + if (xe->info.platform != XE_BATTLEMAGE) + return; + + base = BMG_GSC_HECI1_BASE; + lockdep_assert_held(&xe->irq.lock); + err_src = xe_mmio_read32(mmio, HEC_UNCORR_ERR_STATUS(base)); + if (!err_src) { + drm_err_ratelimited(&xe->drm, HW_ERR "Tile%d reported HEC_ERR_STATUS_%s blank\n", + tile->id, hw_err_str); + return; + } + + if (err_src & UNCORR_FW_REPORTED_ERR) { + fw_err = xe_mmio_read32(mmio, HEC_UNCORR_FW_ERR_DW0(base)); + for_each_set_bit(err_bit, &fw_err, HEC_UNCORR_FW_ERR_BITS) { + drm_err_ratelimited(&xe->drm, HW_ERR + "%s: HEC Uncorrected FW %s error reported, bit[%d] is set\n", + hw_err_str, hec_uncorrected_fw_errors[err_bit], + err_bit); + + schedule_work(&tile->csc_hw_error_work); + } + } + + xe_mmio_write32(mmio, HEC_UNCORR_ERR_STATUS(base), err_src); +} + +static void hw_error_source_handler(struct xe_tile *tile, const enum hardware_error hw_err) +{ + const char *hw_err_str = hw_error_to_str(hw_err); + struct xe_device *xe = tile_to_xe(tile); + unsigned long flags; + u32 err_src; + + if (xe->info.platform != XE_BATTLEMAGE) + return; + + spin_lock_irqsave(&xe->irq.lock, flags); + err_src = xe_mmio_read32(&tile->mmio, DEV_ERR_STAT_REG(hw_err)); + if (!err_src) { + drm_err_ratelimited(&xe->drm, HW_ERR "Tile%d reported DEV_ERR_STAT_%s blank!\n", + tile->id, hw_err_str); + goto unlock; + } + + if (err_src & XE_CSC_ERROR) + csc_hw_error_handler(tile, hw_err); + + xe_mmio_write32(&tile->mmio, DEV_ERR_STAT_REG(hw_err), err_src); + +unlock: + spin_unlock_irqrestore(&xe->irq.lock, flags); +} + +/** + * xe_hw_error_irq_handler - irq handling for hw errors + * @tile: tile instance + * @master_ctl: value read from master interrupt register + * + * Xe platforms add three error bits to the master interrupt register to support error handling. + * These three bits are used to convey the class of error FATAL, NONFATAL, or CORRECTABLE. + * To process the interrupt, determine the source of error by reading the Device Error Source + * Register that corresponds to the class of error being serviced. + */ +void xe_hw_error_irq_handler(struct xe_tile *tile, const u32 master_ctl) +{ + enum hardware_error hw_err; + + if (fault_inject_csc_hw_error()) + schedule_work(&tile->csc_hw_error_work); + + for (hw_err = 0; hw_err < HARDWARE_ERROR_MAX; hw_err++) + if (master_ctl & ERROR_IRQ(hw_err)) + hw_error_source_handler(tile, hw_err); +} + +/* + * Process hardware errors during boot + */ +static void process_hw_errors(struct xe_device *xe) +{ + struct xe_tile *tile; + u32 master_ctl; + u8 id; + + for_each_tile(tile, xe, id) { + master_ctl = xe_mmio_read32(&tile->mmio, GFX_MSTR_IRQ); + xe_hw_error_irq_handler(tile, master_ctl); + xe_mmio_write32(&tile->mmio, GFX_MSTR_IRQ, master_ctl); + } +} + +/** + * xe_hw_error_init - Initialize hw errors + * @xe: xe device instance + * + * Initialize and check for errors that occurred during boot + * prior to driver load + */ +void xe_hw_error_init(struct xe_device *xe) +{ + struct xe_tile *tile = xe_device_get_root_tile(xe); + + if (!IS_DGFX(xe) || IS_SRIOV_VF(xe)) + return; + + INIT_WORK(&tile->csc_hw_error_work, csc_hw_error_work); + + process_hw_errors(xe); +} diff --git a/drivers/gpu/drm/xe/xe_hw_error.h b/drivers/gpu/drm/xe/xe_hw_error.h new file mode 100644 index 000000000000..d86e28c5180c --- /dev/null +++ b/drivers/gpu/drm/xe/xe_hw_error.h @@ -0,0 +1,15 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2025 Intel Corporation + */ +#ifndef XE_HW_ERROR_H_ +#define XE_HW_ERROR_H_ + +#include <linux/types.h> + +struct xe_tile; +struct xe_device; + +void xe_hw_error_irq_handler(struct xe_tile *tile, const u32 master_ctl); +void xe_hw_error_init(struct xe_device *xe); +#endif diff --git a/drivers/gpu/drm/xe/xe_hwmon.c b/drivers/gpu/drm/xe/xe_hwmon.c index c17ed1ae8649..32a76ae6e9dc 100644 --- a/drivers/gpu/drm/xe/xe_hwmon.c +++ b/drivers/gpu/drm/xe/xe_hwmon.c @@ -179,7 +179,7 @@ static int xe_hwmon_pcode_rmw_power_limit(const struct xe_hwmon *hwmon, u32 attr u32 clr, u32 set) { struct xe_tile *root_tile = xe_device_get_root_tile(hwmon->xe); - u32 val0, val1; + u32 val0 = 0, val1 = 0; int ret = 0; ret = xe_pcode_read(root_tile, PCODE_MBOX(PCODE_POWER_SETUP, @@ -734,7 +734,7 @@ static int xe_hwmon_power_curr_crit_read(struct xe_hwmon *hwmon, int channel, long *value, u32 scale_factor) { int ret; - u32 uval; + u32 uval = 0; mutex_lock(&hwmon->hwmon_lock); @@ -918,7 +918,7 @@ xe_hwmon_power_write(struct xe_hwmon *hwmon, u32 attr, int channel, long val) static umode_t xe_hwmon_curr_is_visible(const struct xe_hwmon *hwmon, u32 attr, int channel) { - u32 uval; + u32 uval = 0; /* hwmon sysfs attribute of current available only for package */ if (channel != CHANNEL_PKG) @@ -1020,7 +1020,7 @@ xe_hwmon_energy_read(struct xe_hwmon *hwmon, u32 attr, int channel, long *val) static umode_t xe_hwmon_fan_is_visible(struct xe_hwmon *hwmon, u32 attr, int channel) { - u32 uval; + u32 uval = 0; if (!hwmon->xe->info.has_fan_control) return 0; diff --git a/drivers/gpu/drm/xe/xe_i2c.c b/drivers/gpu/drm/xe/xe_i2c.c index bc7dc2099470..044dda517b7c 100644 --- a/drivers/gpu/drm/xe/xe_i2c.c +++ b/drivers/gpu/drm/xe/xe_i2c.c @@ -147,6 +147,20 @@ static void xe_i2c_unregister_adapter(struct xe_i2c *i2c) } /** + * xe_i2c_present - I2C controller is present and functional + * @xe: xe device instance + * + * Check whether the I2C controller is present and functioning with valid + * endpoint cookie. + * + * Return: %true if present, %false otherwise. + */ +bool xe_i2c_present(struct xe_device *xe) +{ + return xe->i2c && xe->i2c->ep.cookie == XE_I2C_EP_COOKIE_DEVICE; +} + +/** * xe_i2c_irq_handler: Handler for I2C interrupts * @xe: xe device instance * @master_ctl: interrupt register @@ -230,7 +244,7 @@ void xe_i2c_pm_suspend(struct xe_device *xe) { struct xe_mmio *mmio = xe_root_tile_mmio(xe); - if (!xe->i2c || xe->i2c->ep.cookie != XE_I2C_EP_COOKIE_DEVICE) + if (!xe_i2c_present(xe)) return; xe_mmio_rmw32(mmio, I2C_CONFIG_PMCSR, PCI_PM_CTRL_STATE_MASK, (__force u32)PCI_D3hot); @@ -241,7 +255,7 @@ void xe_i2c_pm_resume(struct xe_device *xe, bool d3cold) { struct xe_mmio *mmio = xe_root_tile_mmio(xe); - if (!xe->i2c || xe->i2c->ep.cookie != XE_I2C_EP_COOKIE_DEVICE) + if (!xe_i2c_present(xe)) return; if (d3cold) diff --git a/drivers/gpu/drm/xe/xe_i2c.h b/drivers/gpu/drm/xe/xe_i2c.h index b767ed8ce52b..ecd5f10358e2 100644 --- a/drivers/gpu/drm/xe/xe_i2c.h +++ b/drivers/gpu/drm/xe/xe_i2c.h @@ -49,11 +49,13 @@ struct xe_i2c { #if IS_ENABLED(CONFIG_I2C) int xe_i2c_probe(struct xe_device *xe); +bool xe_i2c_present(struct xe_device *xe); void xe_i2c_irq_handler(struct xe_device *xe, u32 master_ctl); void xe_i2c_pm_suspend(struct xe_device *xe); void xe_i2c_pm_resume(struct xe_device *xe, bool d3cold); #else static inline int xe_i2c_probe(struct xe_device *xe) { return 0; } +static inline bool xe_i2c_present(struct xe_device *xe) { return false; } static inline void xe_i2c_irq_handler(struct xe_device *xe, u32 master_ctl) { } static inline void xe_i2c_pm_suspend(struct xe_device *xe) { } static inline void xe_i2c_pm_resume(struct xe_device *xe, bool d3cold) { } diff --git a/drivers/gpu/drm/xe/xe_irq.c b/drivers/gpu/drm/xe/xe_irq.c index 5df5b8c2a3e4..870edaf69388 100644 --- a/drivers/gpu/drm/xe/xe_irq.c +++ b/drivers/gpu/drm/xe/xe_irq.c @@ -18,6 +18,7 @@ #include "xe_gt.h" #include "xe_guc.h" #include "xe_hw_engine.h" +#include "xe_hw_error.h" #include "xe_i2c.h" #include "xe_memirq.h" #include "xe_mmio.h" @@ -468,6 +469,7 @@ static irqreturn_t dg1_irq_handler(int irq, void *arg) xe_mmio_write32(mmio, GFX_MSTR_IRQ, master_ctl); gt_irq_handler(tile, master_ctl, intr_dw, identity); + xe_hw_error_irq_handler(tile, master_ctl); /* * Display interrupts (including display backlight operations @@ -756,6 +758,8 @@ int xe_irq_install(struct xe_device *xe) int nvec = 1; int err; + xe_hw_error_init(xe); + xe_irq_reset(xe); if (xe_device_has_msix(xe)) { diff --git a/drivers/gpu/drm/xe/xe_lmtt.c b/drivers/gpu/drm/xe/xe_lmtt.c index a2000307d5bf..f2bfbfa3efa1 100644 --- a/drivers/gpu/drm/xe/xe_lmtt.c +++ b/drivers/gpu/drm/xe/xe_lmtt.c @@ -11,7 +11,7 @@ #include "xe_assert.h" #include "xe_bo.h" -#include "xe_gt_tlb_invalidation.h" +#include "xe_tlb_inval.h" #include "xe_lmtt.h" #include "xe_map.h" #include "xe_mmio.h" @@ -195,14 +195,17 @@ static void lmtt_setup_dir_ptr(struct xe_lmtt *lmtt) struct xe_tile *tile = lmtt_to_tile(lmtt); struct xe_device *xe = tile_to_xe(tile); dma_addr_t offset = xe_bo_main_addr(lmtt->pd->bo, XE_PAGE_SIZE); + struct xe_gt *gt; + u8 id; lmtt_debug(lmtt, "DIR offset %pad\n", &offset); lmtt_assert(lmtt, xe_bo_is_vram(lmtt->pd->bo)); lmtt_assert(lmtt, IS_ALIGNED(offset, SZ_64K)); - xe_mmio_write32(&tile->mmio, - GRAPHICS_VER(xe) >= 20 ? XE2_LMEM_CFG : LMEM_CFG, - LMEM_EN | REG_FIELD_PREP(LMTT_DIR_PTR, offset / SZ_64K)); + for_each_gt_on_tile(gt, tile, id) + xe_mmio_write32(>->mmio, + GRAPHICS_VER(xe) >= 20 ? XE2_LMEM_CFG : LMEM_CFG, + LMEM_EN | REG_FIELD_PREP(LMTT_DIR_PTR, offset / SZ_64K)); } /** @@ -225,8 +228,8 @@ void xe_lmtt_init_hw(struct xe_lmtt *lmtt) static int lmtt_invalidate_hw(struct xe_lmtt *lmtt) { - struct xe_gt_tlb_invalidation_fence fences[XE_MAX_GT_PER_TILE]; - struct xe_gt_tlb_invalidation_fence *fence = fences; + struct xe_tlb_inval_fence fences[XE_MAX_GT_PER_TILE]; + struct xe_tlb_inval_fence *fence = fences; struct xe_tile *tile = lmtt_to_tile(lmtt); struct xe_gt *gt; int result = 0; @@ -234,8 +237,8 @@ static int lmtt_invalidate_hw(struct xe_lmtt *lmtt) u8 id; for_each_gt_on_tile(gt, tile, id) { - xe_gt_tlb_invalidation_fence_init(gt, fence, true); - err = xe_gt_tlb_invalidation_all(gt, fence); + xe_tlb_inval_fence_init(>->tlb_inval, fence, true); + err = xe_tlb_inval_all(>->tlb_inval, fence); result = result ?: err; fence++; } @@ -249,7 +252,7 @@ static int lmtt_invalidate_hw(struct xe_lmtt *lmtt) */ fence = fences; for_each_gt_on_tile(gt, tile, id) - xe_gt_tlb_invalidation_fence_wait(fence++); + xe_tlb_inval_fence_wait(fence++); return result; } diff --git a/drivers/gpu/drm/xe/xe_lrc.c b/drivers/gpu/drm/xe/xe_lrc.c index 6d38411bdeba..8f6c3ba47882 100644 --- a/drivers/gpu/drm/xe/xe_lrc.c +++ b/drivers/gpu/drm/xe/xe_lrc.c @@ -41,7 +41,6 @@ #define LRC_PPHWSP_SIZE SZ_4K #define LRC_INDIRECT_CTX_BO_SIZE SZ_4K #define LRC_INDIRECT_RING_STATE_SIZE SZ_4K -#define LRC_WA_BB_SIZE SZ_4K /* * Layout of the LRC and associated data allocated as @@ -76,6 +75,11 @@ lrc_to_xe(struct xe_lrc *lrc) static bool gt_engine_needs_indirect_ctx(struct xe_gt *gt, enum xe_engine_class class) { + if (XE_GT_WA(gt, 16010904313) && + (class == XE_ENGINE_CLASS_RENDER || + class == XE_ENGINE_CLASS_COMPUTE)) + return true; + return false; } @@ -692,7 +696,13 @@ u32 xe_lrc_regs_offset(struct xe_lrc *lrc) return xe_lrc_pphwsp_offset(lrc) + LRC_PPHWSP_SIZE; } -static size_t lrc_reg_size(struct xe_device *xe) +/** + * xe_lrc_reg_size() - Get size of the LRC registers area within queues + * @xe: the &xe_device struct instance + * + * Returns: Size of the LRC registers area for current platform + */ +size_t xe_lrc_reg_size(struct xe_device *xe) { if (GRAPHICS_VERx100(xe) >= 1250) return 96 * sizeof(u32); @@ -702,7 +712,7 @@ static size_t lrc_reg_size(struct xe_device *xe) size_t xe_lrc_skip_size(struct xe_device *xe) { - return LRC_PPHWSP_SIZE + lrc_reg_size(xe); + return LRC_PPHWSP_SIZE + xe_lrc_reg_size(xe); } static inline u32 __xe_lrc_seqno_offset(struct xe_lrc *lrc) @@ -943,6 +953,47 @@ static void *empty_lrc_data(struct xe_hw_engine *hwe) return data; } +/** + * xe_default_lrc_update_memirq_regs_with_address - Re-compute GGTT references in default LRC + * of given engine. + * @hwe: the &xe_hw_engine struct instance + */ +void xe_default_lrc_update_memirq_regs_with_address(struct xe_hw_engine *hwe) +{ + struct xe_gt *gt = hwe->gt; + u32 *regs; + + if (!gt->default_lrc[hwe->class]) + return; + + regs = gt->default_lrc[hwe->class] + LRC_PPHWSP_SIZE; + set_memory_based_intr(regs, hwe); +} + +/** + * xe_lrc_update_memirq_regs_with_address - Re-compute GGTT references in mem interrupt data + * for given LRC. + * @lrc: the &xe_lrc struct instance + * @hwe: the &xe_hw_engine struct instance + * @regs: scratch buffer to be used as temporary storage + */ +void xe_lrc_update_memirq_regs_with_address(struct xe_lrc *lrc, struct xe_hw_engine *hwe, + u32 *regs) +{ + struct xe_gt *gt = hwe->gt; + struct iosys_map map; + size_t regs_len; + + if (!xe_device_uses_memirq(gt_to_xe(gt))) + return; + + map = __xe_lrc_regs_map(lrc); + regs_len = xe_lrc_reg_size(gt_to_xe(gt)); + xe_map_memcpy_from(gt_to_xe(gt), regs, &map, 0, regs_len); + set_memory_based_intr(regs, hwe); + xe_map_memcpy_to(gt_to_xe(gt), &map, 0, regs, regs_len); +} + static void xe_lrc_set_ppgtt(struct xe_lrc *lrc, struct xe_vm *vm) { u64 desc = xe_vm_pdp4_descriptor(vm, gt_to_tile(lrc->gt)); @@ -1014,6 +1065,63 @@ static ssize_t setup_utilization_wa(struct xe_lrc *lrc, return cmd - batch; } +static ssize_t setup_timestamp_wa(struct xe_lrc *lrc, struct xe_hw_engine *hwe, + u32 *batch, size_t max_len) +{ + const u32 ts_addr = __xe_lrc_ctx_timestamp_ggtt_addr(lrc); + u32 *cmd = batch; + + if (!XE_GT_WA(lrc->gt, 16010904313) || + !(hwe->class == XE_ENGINE_CLASS_RENDER || + hwe->class == XE_ENGINE_CLASS_COMPUTE || + hwe->class == XE_ENGINE_CLASS_COPY || + hwe->class == XE_ENGINE_CLASS_VIDEO_DECODE || + hwe->class == XE_ENGINE_CLASS_VIDEO_ENHANCE)) + return 0; + + if (xe_gt_WARN_ON(lrc->gt, max_len < 12)) + return -ENOSPC; + + *cmd++ = MI_LOAD_REGISTER_MEM | MI_LRM_USE_GGTT | MI_LRI_LRM_CS_MMIO | + MI_LRM_ASYNC; + *cmd++ = RING_CTX_TIMESTAMP(0).addr; + *cmd++ = ts_addr; + *cmd++ = 0; + + *cmd++ = MI_LOAD_REGISTER_MEM | MI_LRM_USE_GGTT | MI_LRI_LRM_CS_MMIO | + MI_LRM_ASYNC; + *cmd++ = RING_CTX_TIMESTAMP(0).addr; + *cmd++ = ts_addr; + *cmd++ = 0; + + *cmd++ = MI_LOAD_REGISTER_MEM | MI_LRM_USE_GGTT | MI_LRI_LRM_CS_MMIO; + *cmd++ = RING_CTX_TIMESTAMP(0).addr; + *cmd++ = ts_addr; + *cmd++ = 0; + + return cmd - batch; +} + +static ssize_t setup_invalidate_state_cache_wa(struct xe_lrc *lrc, + struct xe_hw_engine *hwe, + u32 *batch, size_t max_len) +{ + u32 *cmd = batch; + + if (!XE_GT_WA(lrc->gt, 18022495364) || + hwe->class != XE_ENGINE_CLASS_RENDER) + return 0; + + if (xe_gt_WARN_ON(lrc->gt, max_len < 3)) + return -ENOSPC; + + *cmd++ = MI_LOAD_REGISTER_IMM | MI_LRI_NUM_REGS(1); + *cmd++ = CS_DEBUG_MODE1(0).addr; + *cmd++ = _MASKED_BIT_ENABLE(INSTRUCTION_STATE_CACHE_INVALIDATE); + + return cmd - batch; +} + struct bo_setup { ssize_t (*setup)(struct xe_lrc *lrc, struct xe_hw_engine *hwe, u32 *batch, size_t max_size); @@ -1040,13 +1148,11 @@ static int setup_bo(struct bo_setup_state *state) ssize_t remain; if (state->lrc->bo->vmap.is_iomem) { - state->buffer = kmalloc(state->max_size, GFP_KERNEL); if (!state->buffer) return -ENOMEM; state->ptr = state->buffer; } else { state->ptr = state->lrc->bo->vmap.vaddr + state->offset; - state->buffer = NULL; } remain = state->max_size / sizeof(u32); @@ -1071,7 +1177,6 @@ static int setup_bo(struct bo_setup_state *state) return 0; fail: - kfree(state->buffer); return -ENOSPC; } @@ -1083,18 +1188,27 @@ static void finish_bo(struct bo_setup_state *state) xe_map_memcpy_to(gt_to_xe(state->lrc->gt), &state->lrc->bo->vmap, state->offset, state->buffer, state->written * sizeof(u32)); - kfree(state->buffer); } -static int setup_wa_bb(struct xe_lrc *lrc, struct xe_hw_engine *hwe) +/** + * xe_lrc_setup_wa_bb_with_scratch - Execute all wa bb setup callbacks. + * @lrc: the &xe_lrc struct instance + * @hwe: the &xe_hw_engine struct instance + * @scratch: preallocated scratch buffer for temporary storage + * Return: 0 on success, negative error code on failure + */ +int xe_lrc_setup_wa_bb_with_scratch(struct xe_lrc *lrc, struct xe_hw_engine *hwe, u32 *scratch) { static const struct bo_setup funcs[] = { + { .setup = setup_timestamp_wa }, + { .setup = setup_invalidate_state_cache_wa }, { .setup = setup_utilization_wa }, }; struct bo_setup_state state = { .lrc = lrc, .hwe = hwe, .max_size = LRC_WA_BB_SIZE, + .buffer = scratch, .reserve_dw = 1, .offset = __xe_lrc_wa_bb_offset(lrc), .funcs = funcs, @@ -1117,15 +1231,32 @@ static int setup_wa_bb(struct xe_lrc *lrc, struct xe_hw_engine *hwe) return 0; } +static int setup_wa_bb(struct xe_lrc *lrc, struct xe_hw_engine *hwe) +{ + u32 *buf = NULL; + int ret; + + if (lrc->bo->vmap.is_iomem) + buf = kmalloc(LRC_WA_BB_SIZE, GFP_KERNEL); + + ret = xe_lrc_setup_wa_bb_with_scratch(lrc, hwe, buf); + + kfree(buf); + + return ret; +} + static int setup_indirect_ctx(struct xe_lrc *lrc, struct xe_hw_engine *hwe) { static struct bo_setup rcs_funcs[] = { + { .setup = setup_timestamp_wa }, }; struct bo_setup_state state = { .lrc = lrc, .hwe = hwe, .max_size = (63 * 64) /* max 63 cachelines */, + .buffer = NULL, .offset = __xe_lrc_indirect_ctx_offset(lrc), }; int ret; @@ -1142,9 +1273,14 @@ setup_indirect_ctx(struct xe_lrc *lrc, struct xe_hw_engine *hwe) if (xe_gt_WARN_ON(lrc->gt, !state.funcs)) return 0; + if (lrc->bo->vmap.is_iomem) + state.buffer = kmalloc(state.max_size, GFP_KERNEL); + ret = setup_bo(&state); - if (ret) + if (ret) { + kfree(state.buffer); return ret; + } /* * Align to 64B cacheline so there's no garbage at the end for CS to @@ -1156,6 +1292,7 @@ setup_indirect_ctx(struct xe_lrc *lrc, struct xe_hw_engine *hwe) } finish_bo(&state); + kfree(state.buffer); xe_lrc_write_ctx_reg(lrc, CTX_CS_INDIRECT_CTX, @@ -1374,6 +1511,23 @@ void xe_lrc_destroy(struct kref *ref) kfree(lrc); } +/** + * xe_lrc_update_hwctx_regs_with_address - Re-compute GGTT references within given LRC. + * @lrc: the &xe_lrc struct instance + */ +void xe_lrc_update_hwctx_regs_with_address(struct xe_lrc *lrc) +{ + if (xe_lrc_has_indirect_ring_state(lrc)) { + xe_lrc_write_ctx_reg(lrc, CTX_INDIRECT_RING_STATE, + __xe_lrc_indirect_ring_ggtt_addr(lrc)); + + xe_lrc_write_indirect_ctx_reg(lrc, INDIRECT_CTX_RING_START, + __xe_lrc_ring_ggtt_addr(lrc)); + } else { + xe_lrc_write_ctx_reg(lrc, CTX_RING_START, __xe_lrc_ring_ggtt_addr(lrc)); + } +} + void xe_lrc_set_ring_tail(struct xe_lrc *lrc, u32 tail) { if (xe_lrc_has_indirect_ring_state(lrc)) @@ -1939,7 +2093,7 @@ u32 *xe_lrc_emit_hwe_state_instructions(struct xe_exec_queue *q, u32 *cs) * continue to emit all of the SVG state since it's best not to leak * any of the state between contexts, even if that leakage is harmless. */ - if (XE_WA(gt, 14019789679) && q->hwe->class == XE_ENGINE_CLASS_RENDER) { + if (XE_GT_WA(gt, 14019789679) && q->hwe->class == XE_ENGINE_CLASS_RENDER) { state_table = xe_hpg_svg_state; state_table_size = ARRAY_SIZE(xe_hpg_svg_state); } diff --git a/drivers/gpu/drm/xe/xe_lrc.h b/drivers/gpu/drm/xe/xe_lrc.h index b6c8053c581b..188565465779 100644 --- a/drivers/gpu/drm/xe/xe_lrc.h +++ b/drivers/gpu/drm/xe/xe_lrc.h @@ -42,6 +42,8 @@ struct xe_lrc_snapshot { #define LRC_PPHWSP_FLUSH_INVAL_SCRATCH_ADDR (0x34 * 4) #define LRC_PPHWSP_PXP_INVAL_SCRATCH_ADDR (0x40 * 4) +#define LRC_WA_BB_SIZE SZ_4K + #define XE_LRC_CREATE_RUNALONE 0x1 #define XE_LRC_CREATE_PXP 0x2 struct xe_lrc *xe_lrc_create(struct xe_hw_engine *hwe, struct xe_vm *vm, @@ -88,6 +90,10 @@ bool xe_lrc_ring_is_idle(struct xe_lrc *lrc); u32 xe_lrc_indirect_ring_ggtt_addr(struct xe_lrc *lrc); u32 xe_lrc_ggtt_addr(struct xe_lrc *lrc); u32 *xe_lrc_regs(struct xe_lrc *lrc); +void xe_lrc_update_hwctx_regs_with_address(struct xe_lrc *lrc); +void xe_default_lrc_update_memirq_regs_with_address(struct xe_hw_engine *hwe); +void xe_lrc_update_memirq_regs_with_address(struct xe_lrc *lrc, struct xe_hw_engine *hwe, + u32 *regs); u32 xe_lrc_read_ctx_reg(struct xe_lrc *lrc, int reg_nr); void xe_lrc_write_ctx_reg(struct xe_lrc *lrc, int reg_nr, u32 val); @@ -106,6 +112,7 @@ s32 xe_lrc_start_seqno(struct xe_lrc *lrc); u32 xe_lrc_parallel_ggtt_addr(struct xe_lrc *lrc); struct iosys_map xe_lrc_parallel_map(struct xe_lrc *lrc); +size_t xe_lrc_reg_size(struct xe_device *xe); size_t xe_lrc_skip_size(struct xe_device *xe); void xe_lrc_dump_default(struct drm_printer *p, @@ -124,6 +131,8 @@ u32 xe_lrc_ctx_timestamp_udw_ggtt_addr(struct xe_lrc *lrc); u64 xe_lrc_ctx_timestamp(struct xe_lrc *lrc); u32 xe_lrc_ctx_job_timestamp_ggtt_addr(struct xe_lrc *lrc); u32 xe_lrc_ctx_job_timestamp(struct xe_lrc *lrc); +int xe_lrc_setup_wa_bb_with_scratch(struct xe_lrc *lrc, struct xe_hw_engine *hwe, + u32 *scratch); /** * xe_lrc_update_timestamp - readout LRC timestamp and update cached value diff --git a/drivers/gpu/drm/xe/xe_migrate.c b/drivers/gpu/drm/xe/xe_migrate.c index 7d20ac4bb633..9643442ef101 100644 --- a/drivers/gpu/drm/xe/xe_migrate.c +++ b/drivers/gpu/drm/xe/xe_migrate.c @@ -9,6 +9,7 @@ #include <linux/sizes.h> #include <drm/drm_managed.h> +#include <drm/drm_pagemap.h> #include <drm/ttm/ttm_tt.h> #include <uapi/drm/xe_drm.h> @@ -30,10 +31,12 @@ #include "xe_mocs.h" #include "xe_pt.h" #include "xe_res_cursor.h" +#include "xe_sa.h" #include "xe_sched_job.h" #include "xe_sync.h" #include "xe_trace_bo.h" #include "xe_vm.h" +#include "xe_vram.h" /** * struct xe_migrate - migrate context. @@ -84,19 +87,6 @@ struct xe_migrate { */ #define MAX_PTE_PER_SDI 0x1FEU -/** - * xe_tile_migrate_exec_queue() - Get this tile's migrate exec queue. - * @tile: The tile. - * - * Returns the default migrate exec queue of this tile. - * - * Return: The default migrate exec queue - */ -struct xe_exec_queue *xe_tile_migrate_exec_queue(struct xe_tile *tile) -{ - return tile->migrate->q; -} - static void xe_migrate_fini(void *arg) { struct xe_migrate *m = arg; @@ -130,38 +120,39 @@ static u64 xe_migrate_vram_ofs(struct xe_device *xe, u64 addr, bool is_comp_pte) u64 identity_offset = IDENTITY_OFFSET; if (GRAPHICS_VER(xe) >= 20 && is_comp_pte) - identity_offset += DIV_ROUND_UP_ULL(xe->mem.vram.actual_physical_size, SZ_1G); + identity_offset += DIV_ROUND_UP_ULL(xe_vram_region_actual_physical_size + (xe->mem.vram), SZ_1G); - addr -= xe->mem.vram.dpa_base; + addr -= xe_vram_region_dpa_base(xe->mem.vram); return addr + (identity_offset << xe_pt_shift(2)); } static void xe_migrate_program_identity(struct xe_device *xe, struct xe_vm *vm, struct xe_bo *bo, u64 map_ofs, u64 vram_offset, u16 pat_index, u64 pt_2m_ofs) { + struct xe_vram_region *vram = xe->mem.vram; + resource_size_t dpa_base = xe_vram_region_dpa_base(vram); u64 pos, ofs, flags; u64 entry; /* XXX: Unclear if this should be usable_size? */ - u64 vram_limit = xe->mem.vram.actual_physical_size + - xe->mem.vram.dpa_base; + u64 vram_limit = xe_vram_region_actual_physical_size(vram) + dpa_base; u32 level = 2; ofs = map_ofs + XE_PAGE_SIZE * level + vram_offset * 8; flags = vm->pt_ops->pte_encode_addr(xe, 0, pat_index, level, true, 0); - xe_assert(xe, IS_ALIGNED(xe->mem.vram.usable_size, SZ_2M)); + xe_assert(xe, IS_ALIGNED(xe_vram_region_usable_size(vram), SZ_2M)); /* * Use 1GB pages when possible, last chunk always use 2M * pages as mixing reserved memory (stolen, WOCPM) with a single * mapping is not allowed on certain platforms. */ - for (pos = xe->mem.vram.dpa_base; pos < vram_limit; + for (pos = dpa_base; pos < vram_limit; pos += SZ_1G, ofs += 8) { if (pos + SZ_1G >= vram_limit) { - entry = vm->pt_ops->pde_encode_bo(bo, pt_2m_ofs, - pat_index); + entry = vm->pt_ops->pde_encode_bo(bo, pt_2m_ofs); xe_map_wr(xe, &bo->vmap, ofs, u64, entry); flags = vm->pt_ops->pte_encode_addr(xe, 0, @@ -215,7 +206,7 @@ static int xe_migrate_prepare_vm(struct xe_tile *tile, struct xe_migrate *m, /* PT30 & PT31 reserved for 2M identity map */ pt29_ofs = xe_bo_size(bo) - 3 * XE_PAGE_SIZE; - entry = vm->pt_ops->pde_encode_bo(bo, pt29_ofs, pat_index); + entry = vm->pt_ops->pde_encode_bo(bo, pt29_ofs); xe_pt_write(xe, &vm->pt_root[id]->bo->vmap, 0, entry); map_ofs = (num_entries - num_setup) * XE_PAGE_SIZE; @@ -283,15 +274,14 @@ static int xe_migrate_prepare_vm(struct xe_tile *tile, struct xe_migrate *m, flags = XE_PDE_64K; entry = vm->pt_ops->pde_encode_bo(bo, map_ofs + (u64)(level - 1) * - XE_PAGE_SIZE, pat_index); + XE_PAGE_SIZE); xe_map_wr(xe, &bo->vmap, map_ofs + XE_PAGE_SIZE * level, u64, entry | flags); } /* Write PDE's that point to our BO. */ - for (i = 0; i < map_ofs / PAGE_SIZE; i++) { - entry = vm->pt_ops->pde_encode_bo(bo, (u64)i * XE_PAGE_SIZE, - pat_index); + for (i = 0; i < map_ofs / XE_PAGE_SIZE; i++) { + entry = vm->pt_ops->pde_encode_bo(bo, (u64)i * XE_PAGE_SIZE); xe_map_wr(xe, &bo->vmap, map_ofs + XE_PAGE_SIZE + (i + 1) * 8, u64, entry); @@ -307,11 +297,11 @@ static int xe_migrate_prepare_vm(struct xe_tile *tile, struct xe_migrate *m, /* Identity map the entire vram at 256GiB offset */ if (IS_DGFX(xe)) { u64 pt30_ofs = xe_bo_size(bo) - 2 * XE_PAGE_SIZE; + resource_size_t actual_phy_size = xe_vram_region_actual_physical_size(xe->mem.vram); xe_migrate_program_identity(xe, vm, bo, map_ofs, IDENTITY_OFFSET, pat_index, pt30_ofs); - xe_assert(xe, xe->mem.vram.actual_physical_size <= - (MAX_NUM_PTE - IDENTITY_OFFSET) * SZ_1G); + xe_assert(xe, actual_phy_size <= (MAX_NUM_PTE - IDENTITY_OFFSET) * SZ_1G); /* * Identity map the entire vram for compressed pat_index for xe2+ @@ -320,11 +310,11 @@ static int xe_migrate_prepare_vm(struct xe_tile *tile, struct xe_migrate *m, if (GRAPHICS_VER(xe) >= 20 && xe_device_has_flat_ccs(xe)) { u16 comp_pat_index = xe->pat.idx[XE_CACHE_NONE_COMPRESSION]; u64 vram_offset = IDENTITY_OFFSET + - DIV_ROUND_UP_ULL(xe->mem.vram.actual_physical_size, SZ_1G); + DIV_ROUND_UP_ULL(actual_phy_size, SZ_1G); u64 pt31_ofs = xe_bo_size(bo) - XE_PAGE_SIZE; - xe_assert(xe, xe->mem.vram.actual_physical_size <= (MAX_NUM_PTE - - IDENTITY_OFFSET - IDENTITY_OFFSET / 2) * SZ_1G); + xe_assert(xe, actual_phy_size <= (MAX_NUM_PTE - IDENTITY_OFFSET - + IDENTITY_OFFSET / 2) * SZ_1G); xe_migrate_program_identity(xe, vm, bo, map_ofs, vram_offset, comp_pat_index, pt31_ofs); } @@ -387,38 +377,47 @@ static bool xe_migrate_needs_ccs_emit(struct xe_device *xe) } /** + * xe_migrate_alloc - Allocate a migrate struct for a given &xe_tile + * @tile: &xe_tile + * + * Allocates a &xe_migrate for a given tile. + * + * Return: &xe_migrate on success, or NULL when out of memory. + */ +struct xe_migrate *xe_migrate_alloc(struct xe_tile *tile) +{ + struct xe_migrate *m = drmm_kzalloc(&tile_to_xe(tile)->drm, sizeof(*m), GFP_KERNEL); + + if (m) + m->tile = tile; + return m; +} + +/** * xe_migrate_init() - Initialize a migrate context - * @tile: Back-pointer to the tile we're initializing for. + * @m: The migration context * - * Return: Pointer to a migrate context on success. Error pointer on error. + * Return: 0 if successful, negative error code on failure */ -struct xe_migrate *xe_migrate_init(struct xe_tile *tile) +int xe_migrate_init(struct xe_migrate *m) { - struct xe_device *xe = tile_to_xe(tile); + struct xe_tile *tile = m->tile; struct xe_gt *primary_gt = tile->primary_gt; - struct xe_migrate *m; + struct xe_device *xe = tile_to_xe(tile); struct xe_vm *vm; int err; - m = devm_kzalloc(xe->drm.dev, sizeof(*m), GFP_KERNEL); - if (!m) - return ERR_PTR(-ENOMEM); - - m->tile = tile; - /* Special layout, prepared below.. */ vm = xe_vm_create(xe, XE_VM_FLAG_MIGRATION | - XE_VM_FLAG_SET_TILE_ID(tile)); + XE_VM_FLAG_SET_TILE_ID(tile), NULL); if (IS_ERR(vm)) - return ERR_CAST(vm); + return PTR_ERR(vm); xe_vm_lock(vm, false); err = xe_migrate_prepare_vm(tile, m, vm); xe_vm_unlock(vm); - if (err) { - xe_vm_close_and_put(vm); - return ERR_PTR(err); - } + if (err) + goto err_out; if (xe->info.has_usm) { struct xe_hw_engine *hwe = xe_gt_hw_engine(primary_gt, @@ -427,8 +426,10 @@ struct xe_migrate *xe_migrate_init(struct xe_tile *tile) false); u32 logical_mask = xe_migrate_usm_logical_mask(primary_gt); - if (!hwe || !logical_mask) - return ERR_PTR(-EINVAL); + if (!hwe || !logical_mask) { + err = -EINVAL; + goto err_out; + } /* * XXX: Currently only reserving 1 (likely slow) BCS instance on @@ -437,16 +438,18 @@ struct xe_migrate *xe_migrate_init(struct xe_tile *tile) m->q = xe_exec_queue_create(xe, vm, logical_mask, 1, hwe, EXEC_QUEUE_FLAG_KERNEL | EXEC_QUEUE_FLAG_PERMANENT | - EXEC_QUEUE_FLAG_HIGH_PRIORITY, 0); + EXEC_QUEUE_FLAG_HIGH_PRIORITY | + EXEC_QUEUE_FLAG_MIGRATE, 0); } else { m->q = xe_exec_queue_create_class(xe, primary_gt, vm, XE_ENGINE_CLASS_COPY, EXEC_QUEUE_FLAG_KERNEL | - EXEC_QUEUE_FLAG_PERMANENT, 0); + EXEC_QUEUE_FLAG_PERMANENT | + EXEC_QUEUE_FLAG_MIGRATE, 0); } if (IS_ERR(m->q)) { - xe_vm_close_and_put(vm); - return ERR_CAST(m->q); + err = PTR_ERR(m->q); + goto err_out; } mutex_init(&m->job_mutex); @@ -456,7 +459,7 @@ struct xe_migrate *xe_migrate_init(struct xe_tile *tile) err = devm_add_action_or_reset(xe->drm.dev, xe_migrate_fini, m); if (err) - return ERR_PTR(err); + return err; if (IS_DGFX(xe)) { if (xe_migrate_needs_ccs_emit(xe)) @@ -471,7 +474,12 @@ struct xe_migrate *xe_migrate_init(struct xe_tile *tile) (unsigned long long)m->min_chunk_size); } - return m; + return err; + +err_out: + xe_vm_close_and_put(vm); + return err; + } static u64 max_mem_transfer_per_pass(struct xe_device *xe) @@ -896,7 +904,7 @@ struct dma_fence *xe_migrate_copy(struct xe_migrate *m, goto err; } - xe_sched_job_add_migrate_flush(job, flush_flags); + xe_sched_job_add_migrate_flush(job, flush_flags | MI_INVALIDATE_TLB); if (!fence) { err = xe_sched_job_add_deps(job, src_bo->ttm.base.resv, DMA_RESV_USAGE_BOOKKEEP); @@ -940,6 +948,167 @@ err_sync: return fence; } +/** + * xe_migrate_lrc() - Get the LRC from migrate context. + * @migrate: Migrate context. + * + * Return: Pointer to LRC on success, error on failure + */ +struct xe_lrc *xe_migrate_lrc(struct xe_migrate *migrate) +{ + return migrate->q->lrc[0]; +} + +static int emit_flush_invalidate(struct xe_exec_queue *q, u32 *dw, int i, + u32 flags) +{ + struct xe_lrc *lrc = xe_exec_queue_lrc(q); + dw[i++] = MI_FLUSH_DW | MI_INVALIDATE_TLB | MI_FLUSH_DW_OP_STOREDW | + MI_FLUSH_IMM_DW | flags; + dw[i++] = lower_32_bits(xe_lrc_start_seqno_ggtt_addr(lrc)) | + MI_FLUSH_DW_USE_GTT; + dw[i++] = upper_32_bits(xe_lrc_start_seqno_ggtt_addr(lrc)); + dw[i++] = MI_NOOP; + dw[i++] = MI_NOOP; + + return i; +} + +/** + * xe_migrate_ccs_rw_copy() - Copy content of TTM resources. + * @tile: Tile whose migration context to be used. + * @q : Execution to be used along with migration context. + * @src_bo: The buffer object @src is currently bound to. + * @read_write : Creates BB commands for CCS read/write. + * + * Creates batch buffer instructions to copy CCS metadata from CCS pool to + * memory and vice versa. + * + * This function should only be called for IGPU. + * + * Return: 0 if successful, negative error code on failure. + */ +int xe_migrate_ccs_rw_copy(struct xe_tile *tile, struct xe_exec_queue *q, + struct xe_bo *src_bo, + enum xe_sriov_vf_ccs_rw_ctxs read_write) + +{ + bool src_is_pltt = read_write == XE_SRIOV_VF_CCS_READ_CTX; + bool dst_is_pltt = read_write == XE_SRIOV_VF_CCS_WRITE_CTX; + struct ttm_resource *src = src_bo->ttm.resource; + struct xe_migrate *m = tile->migrate; + struct xe_gt *gt = tile->primary_gt; + u32 batch_size, batch_size_allocated; + struct xe_device *xe = gt_to_xe(gt); + struct xe_res_cursor src_it, ccs_it; + u64 size = xe_bo_size(src_bo); + struct xe_bb *bb = NULL; + u64 src_L0, src_L0_ofs; + u32 src_L0_pt; + int err; + + xe_res_first_sg(xe_bo_sg(src_bo), 0, size, &src_it); + + xe_res_first_sg(xe_bo_sg(src_bo), xe_bo_ccs_pages_start(src_bo), + PAGE_ALIGN(xe_device_ccs_bytes(xe, size)), + &ccs_it); + + /* Calculate Batch buffer size */ + batch_size = 0; + while (size) { + batch_size += 10; /* Flush + ggtt addr + 2 NOP */ + u64 ccs_ofs, ccs_size; + u32 ccs_pt; + + u32 avail_pts = max_mem_transfer_per_pass(xe) / LEVEL0_PAGE_TABLE_ENCODE_SIZE; + + src_L0 = min_t(u64, max_mem_transfer_per_pass(xe), size); + + batch_size += pte_update_size(m, false, src, &src_it, &src_L0, + &src_L0_ofs, &src_L0_pt, 0, 0, + avail_pts); + + ccs_size = xe_device_ccs_bytes(xe, src_L0); + batch_size += pte_update_size(m, 0, NULL, &ccs_it, &ccs_size, &ccs_ofs, + &ccs_pt, 0, avail_pts, avail_pts); + xe_assert(xe, IS_ALIGNED(ccs_it.start, PAGE_SIZE)); + + /* Add copy commands size here */ + batch_size += EMIT_COPY_CCS_DW; + + size -= src_L0; + } + + bb = xe_bb_ccs_new(gt, batch_size, read_write); + if (IS_ERR(bb)) { + drm_err(&xe->drm, "BB allocation failed.\n"); + err = PTR_ERR(bb); + goto err_ret; + } + + batch_size_allocated = batch_size; + size = xe_bo_size(src_bo); + batch_size = 0; + + /* + * Emit PTE and copy commands here. + * The CCS copy command can only support limited size. If the size to be + * copied is more than the limit, divide copy into chunks. So, calculate + * sizes here again before copy command is emitted. + */ + while (size) { + batch_size += 10; /* Flush + ggtt addr + 2 NOP */ + u32 flush_flags = 0; + u64 ccs_ofs, ccs_size; + u32 ccs_pt; + + u32 avail_pts = max_mem_transfer_per_pass(xe) / LEVEL0_PAGE_TABLE_ENCODE_SIZE; + + src_L0 = xe_migrate_res_sizes(m, &src_it); + + batch_size += pte_update_size(m, false, src, &src_it, &src_L0, + &src_L0_ofs, &src_L0_pt, 0, 0, + avail_pts); + + ccs_size = xe_device_ccs_bytes(xe, src_L0); + batch_size += pte_update_size(m, 0, NULL, &ccs_it, &ccs_size, &ccs_ofs, + &ccs_pt, 0, avail_pts, avail_pts); + xe_assert(xe, IS_ALIGNED(ccs_it.start, PAGE_SIZE)); + batch_size += EMIT_COPY_CCS_DW; + + emit_pte(m, bb, src_L0_pt, false, true, &src_it, src_L0, src); + + emit_pte(m, bb, ccs_pt, false, false, &ccs_it, ccs_size, src); + + bb->len = emit_flush_invalidate(q, bb->cs, bb->len, flush_flags); + flush_flags = xe_migrate_ccs_copy(m, bb, src_L0_ofs, src_is_pltt, + src_L0_ofs, dst_is_pltt, + src_L0, ccs_ofs, true); + bb->len = emit_flush_invalidate(q, bb->cs, bb->len, flush_flags); + + size -= src_L0; + } + + xe_assert(xe, (batch_size_allocated == bb->len)); + src_bo->bb_ccs[read_write] = bb; + + return 0; + +err_ret: + return err; +} + +/** + * xe_get_migrate_exec_queue() - Get the execution queue from migrate context. + * @migrate: Migrate context. + * + * Return: Pointer to execution queue on success, error on failure + */ +struct xe_exec_queue *xe_migrate_exec_queue(struct xe_migrate *migrate) +{ + return migrate->q; +} + static void emit_clear_link_copy(struct xe_gt *gt, struct xe_bb *bb, u64 src_ofs, u32 size, u32 pitch) { @@ -1119,11 +1288,13 @@ struct dma_fence *xe_migrate_clear(struct xe_migrate *m, size -= clear_L0; /* Preemption is enabled again by the ring ops. */ - if (clear_vram && xe_migrate_allow_identity(clear_L0, &src_it)) + if (clear_vram && xe_migrate_allow_identity(clear_L0, &src_it)) { xe_res_next(&src_it, clear_L0); - else - emit_pte(m, bb, clear_L0_pt, clear_vram, clear_only_system_ccs, - &src_it, clear_L0, dst); + } else { + emit_pte(m, bb, clear_L0_pt, clear_vram, + clear_only_system_ccs, &src_it, clear_L0, dst); + flush_flags |= MI_INVALIDATE_TLB; + } bb->cs[bb->len++] = MI_BATCH_BUFFER_END; update_idx = bb->len; @@ -1134,7 +1305,7 @@ struct dma_fence *xe_migrate_clear(struct xe_migrate *m, if (xe_migrate_needs_ccs_emit(xe)) { emit_copy_ccs(gt, bb, clear_L0_ofs, true, m->cleared_mem_ofs, false, clear_L0); - flush_flags = MI_FLUSH_DW_CCS; + flush_flags |= MI_FLUSH_DW_CCS; } job = xe_bb_create_migration_job(m->q, bb, @@ -1469,6 +1640,8 @@ next_cmd: goto err_sa; } + xe_sched_job_add_migrate_flush(job, MI_INVALIDATE_TLB); + if (ops->pre_commit) { pt_update->job = job; err = ops->pre_commit(pt_update); @@ -1571,7 +1744,8 @@ static u32 pte_update_cmd_size(u64 size) static void build_pt_update_batch_sram(struct xe_migrate *m, struct xe_bb *bb, u32 pt_offset, - dma_addr_t *sram_addr, u32 size) + struct drm_pagemap_addr *sram_addr, + u32 size) { u16 pat_index = tile_to_xe(m->tile)->pat.idx[XE_CACHE_WB]; u32 ptes; @@ -1589,14 +1763,18 @@ static void build_pt_update_batch_sram(struct xe_migrate *m, ptes -= chunk; while (chunk--) { - u64 addr = sram_addr[i++] & PAGE_MASK; + u64 addr = sram_addr[i].addr & PAGE_MASK; + xe_tile_assert(m->tile, sram_addr[i].proto == + DRM_INTERCONNECT_SYSTEM); xe_tile_assert(m->tile, addr); addr = m->q->vm->pt_ops->pte_encode_addr(m->tile->xe, addr, pat_index, 0, false, 0); bb->cs[bb->len++] = lower_32_bits(addr); bb->cs[bb->len++] = upper_32_bits(addr); + + i++; } } } @@ -1612,7 +1790,8 @@ enum xe_migrate_copy_dir { static struct dma_fence *xe_migrate_vram(struct xe_migrate *m, unsigned long len, unsigned long sram_offset, - dma_addr_t *sram_addr, u64 vram_addr, + struct drm_pagemap_addr *sram_addr, + u64 vram_addr, const enum xe_migrate_copy_dir dir) { struct xe_gt *gt = m->tile->primary_gt; @@ -1628,6 +1807,7 @@ static struct dma_fence *xe_migrate_vram(struct xe_migrate *m, unsigned int pitch = len >= PAGE_SIZE && !(len & ~PAGE_MASK) ? PAGE_SIZE : 4; int err; + unsigned long i, j; if (drm_WARN_ON(&xe->drm, (len & XE_CACHELINE_MASK) || (sram_offset | vram_addr) & XE_CACHELINE_MASK)) @@ -1644,6 +1824,24 @@ static struct dma_fence *xe_migrate_vram(struct xe_migrate *m, return ERR_PTR(err); } + /* + * If the order of a struct drm_pagemap_addr entry is greater than 0, + * the entry is populated by GPU pagemap but subsequent entries within + * the range of that order are not populated. + * build_pt_update_batch_sram() expects a fully populated array of + * struct drm_pagemap_addr. Ensure this is the case even with higher + * orders. + */ + for (i = 0; i < npages;) { + unsigned int order = sram_addr[i].order; + + for (j = 1; j < NR_PAGES(order) && i + j < npages; j++) + if (!sram_addr[i + j].addr) + sram_addr[i + j].addr = sram_addr[i].addr + j * PAGE_SIZE; + + i += NR_PAGES(order); + } + build_pt_update_batch_sram(m, bb, pt_slot * XE_PAGE_SIZE, sram_addr, len + sram_offset); @@ -1669,7 +1867,7 @@ static struct dma_fence *xe_migrate_vram(struct xe_migrate *m, goto err; } - xe_sched_job_add_migrate_flush(job, 0); + xe_sched_job_add_migrate_flush(job, MI_INVALIDATE_TLB); mutex_lock(&m->job_mutex); xe_sched_job_arm(job); @@ -1694,7 +1892,7 @@ err: * xe_migrate_to_vram() - Migrate to VRAM * @m: The migration context. * @npages: Number of pages to migrate. - * @src_addr: Array of dma addresses (source of migrate) + * @src_addr: Array of DMA information (source of migrate) * @dst_addr: Device physical address of VRAM (destination of migrate) * * Copy from an array dma addresses to a VRAM device physical address @@ -1704,7 +1902,7 @@ err: */ struct dma_fence *xe_migrate_to_vram(struct xe_migrate *m, unsigned long npages, - dma_addr_t *src_addr, + struct drm_pagemap_addr *src_addr, u64 dst_addr) { return xe_migrate_vram(m, npages * PAGE_SIZE, 0, src_addr, dst_addr, @@ -1716,7 +1914,7 @@ struct dma_fence *xe_migrate_to_vram(struct xe_migrate *m, * @m: The migration context. * @npages: Number of pages to migrate. * @src_addr: Device physical address of VRAM (source of migrate) - * @dst_addr: Array of dma addresses (destination of migrate) + * @dst_addr: Array of DMA information (destination of migrate) * * Copy from a VRAM device physical address to an array dma addresses * @@ -1726,61 +1924,65 @@ struct dma_fence *xe_migrate_to_vram(struct xe_migrate *m, struct dma_fence *xe_migrate_from_vram(struct xe_migrate *m, unsigned long npages, u64 src_addr, - dma_addr_t *dst_addr) + struct drm_pagemap_addr *dst_addr) { return xe_migrate_vram(m, npages * PAGE_SIZE, 0, dst_addr, src_addr, XE_MIGRATE_COPY_TO_SRAM); } -static void xe_migrate_dma_unmap(struct xe_device *xe, dma_addr_t *dma_addr, +static void xe_migrate_dma_unmap(struct xe_device *xe, + struct drm_pagemap_addr *pagemap_addr, int len, int write) { unsigned long i, npages = DIV_ROUND_UP(len, PAGE_SIZE); for (i = 0; i < npages; ++i) { - if (!dma_addr[i]) + if (!pagemap_addr[i].addr) break; - dma_unmap_page(xe->drm.dev, dma_addr[i], PAGE_SIZE, + dma_unmap_page(xe->drm.dev, pagemap_addr[i].addr, PAGE_SIZE, write ? DMA_TO_DEVICE : DMA_FROM_DEVICE); } - kfree(dma_addr); + kfree(pagemap_addr); } -static dma_addr_t *xe_migrate_dma_map(struct xe_device *xe, - void *buf, int len, int write) +static struct drm_pagemap_addr *xe_migrate_dma_map(struct xe_device *xe, + void *buf, int len, + int write) { - dma_addr_t *dma_addr; + struct drm_pagemap_addr *pagemap_addr; unsigned long i, npages = DIV_ROUND_UP(len, PAGE_SIZE); - dma_addr = kcalloc(npages, sizeof(*dma_addr), GFP_KERNEL); - if (!dma_addr) + pagemap_addr = kcalloc(npages, sizeof(*pagemap_addr), GFP_KERNEL); + if (!pagemap_addr) return ERR_PTR(-ENOMEM); for (i = 0; i < npages; ++i) { dma_addr_t addr; struct page *page; + enum dma_data_direction dir = write ? DMA_TO_DEVICE : + DMA_FROM_DEVICE; if (is_vmalloc_addr(buf)) page = vmalloc_to_page(buf); else page = virt_to_page(buf); - addr = dma_map_page(xe->drm.dev, - page, 0, PAGE_SIZE, - write ? DMA_TO_DEVICE : - DMA_FROM_DEVICE); + addr = dma_map_page(xe->drm.dev, page, 0, PAGE_SIZE, dir); if (dma_mapping_error(xe->drm.dev, addr)) goto err_fault; - dma_addr[i] = addr; + pagemap_addr[i] = + drm_pagemap_addr_encode(addr, + DRM_INTERCONNECT_SYSTEM, + 0, dir); buf += PAGE_SIZE; } - return dma_addr; + return pagemap_addr; err_fault: - xe_migrate_dma_unmap(xe, dma_addr, len, write); + xe_migrate_dma_unmap(xe, pagemap_addr, len, write); return ERR_PTR(-EFAULT); } @@ -1809,7 +2011,7 @@ int xe_migrate_access_memory(struct xe_migrate *m, struct xe_bo *bo, struct xe_device *xe = tile_to_xe(tile); struct xe_res_cursor cursor; struct dma_fence *fence = NULL; - dma_addr_t *dma_addr; + struct drm_pagemap_addr *pagemap_addr; unsigned long page_offset = (unsigned long)buf & ~PAGE_MASK; int bytes_left = len, current_page = 0; void *orig_buf = buf; @@ -1869,9 +2071,9 @@ int xe_migrate_access_memory(struct xe_migrate *m, struct xe_bo *bo, return err; } - dma_addr = xe_migrate_dma_map(xe, buf, len + page_offset, write); - if (IS_ERR(dma_addr)) - return PTR_ERR(dma_addr); + pagemap_addr = xe_migrate_dma_map(xe, buf, len + page_offset, write); + if (IS_ERR(pagemap_addr)) + return PTR_ERR(pagemap_addr); xe_res_first(bo->ttm.resource, offset, xe_bo_size(bo) - offset, &cursor); @@ -1895,7 +2097,7 @@ int xe_migrate_access_memory(struct xe_migrate *m, struct xe_bo *bo, __fence = xe_migrate_vram(m, current_bytes, (unsigned long)buf & ~PAGE_MASK, - dma_addr + current_page, + &pagemap_addr[current_page], vram_addr, write ? XE_MIGRATE_COPY_TO_VRAM : XE_MIGRATE_COPY_TO_SRAM); @@ -1923,10 +2125,46 @@ int xe_migrate_access_memory(struct xe_migrate *m, struct xe_bo *bo, dma_fence_put(fence); out_err: - xe_migrate_dma_unmap(xe, dma_addr, len + page_offset, write); + xe_migrate_dma_unmap(xe, pagemap_addr, len + page_offset, write); return IS_ERR(fence) ? PTR_ERR(fence) : 0; } +/** + * xe_migrate_job_lock() - Lock migrate job lock + * @m: The migration context. + * @q: Queue associated with the operation which requires a lock + * + * Lock the migrate job lock if the queue is a migration queue, otherwise + * assert the VM's dma-resv is held (user queue's have own locking). + */ +void xe_migrate_job_lock(struct xe_migrate *m, struct xe_exec_queue *q) +{ + bool is_migrate = q == m->q; + + if (is_migrate) + mutex_lock(&m->job_mutex); + else + xe_vm_assert_held(q->vm); /* User queues VM's should be locked */ +} + +/** + * xe_migrate_job_unlock() - Unlock migrate job lock + * @m: The migration context. + * @q: Queue associated with the operation which requires a lock + * + * Unlock the migrate job lock if the queue is a migration queue, otherwise + * assert the VM's dma-resv is held (user queue's have own locking). + */ +void xe_migrate_job_unlock(struct xe_migrate *m, struct xe_exec_queue *q) +{ + bool is_migrate = q == m->q; + + if (is_migrate) + mutex_unlock(&m->job_mutex); + else + xe_vm_assert_held(q->vm); /* User queues VM's should be locked */ +} + #if IS_ENABLED(CONFIG_DRM_XE_KUNIT_TEST) #include "tests/xe_migrate.c" #endif diff --git a/drivers/gpu/drm/xe/xe_migrate.h b/drivers/gpu/drm/xe/xe_migrate.h index fb9839c1bae0..4fad324b6253 100644 --- a/drivers/gpu/drm/xe/xe_migrate.h +++ b/drivers/gpu/drm/xe/xe_migrate.h @@ -9,11 +9,13 @@ #include <linux/types.h> struct dma_fence; +struct drm_pagemap_addr; struct iosys_map; struct ttm_resource; struct xe_bo; struct xe_gt; +struct xe_tlb_inval_job; struct xe_exec_queue; struct xe_migrate; struct xe_migrate_pt_update; @@ -24,6 +26,8 @@ struct xe_vm; struct xe_vm_pgtable_update; struct xe_vma; +enum xe_sriov_vf_ccs_rw_ctxs; + /** * struct xe_migrate_pt_update_ops - Callbacks for the * xe_migrate_update_pgtables() function. @@ -89,21 +93,30 @@ struct xe_migrate_pt_update { struct xe_vma_ops *vops; /** @job: The job if a GPU page-table update. NULL otherwise */ struct xe_sched_job *job; + /** + * @ijob: The TLB invalidation job for primary GT. NULL otherwise + */ + struct xe_tlb_inval_job *ijob; + /** + * @mjob: The TLB invalidation job for media GT. NULL otherwise + */ + struct xe_tlb_inval_job *mjob; /** @tile_id: Tile ID of the update */ u8 tile_id; }; -struct xe_migrate *xe_migrate_init(struct xe_tile *tile); +struct xe_migrate *xe_migrate_alloc(struct xe_tile *tile); +int xe_migrate_init(struct xe_migrate *m); struct dma_fence *xe_migrate_to_vram(struct xe_migrate *m, unsigned long npages, - dma_addr_t *src_addr, + struct drm_pagemap_addr *src_addr, u64 dst_addr); struct dma_fence *xe_migrate_from_vram(struct xe_migrate *m, unsigned long npages, u64 src_addr, - dma_addr_t *dst_addr); + struct drm_pagemap_addr *dst_addr); struct dma_fence *xe_migrate_copy(struct xe_migrate *m, struct xe_bo *src_bo, @@ -112,6 +125,12 @@ struct dma_fence *xe_migrate_copy(struct xe_migrate *m, struct ttm_resource *dst, bool copy_only_ccs); +int xe_migrate_ccs_rw_copy(struct xe_tile *tile, struct xe_exec_queue *q, + struct xe_bo *src_bo, + enum xe_sriov_vf_ccs_rw_ctxs read_write); + +struct xe_lrc *xe_migrate_lrc(struct xe_migrate *migrate); +struct xe_exec_queue *xe_migrate_exec_queue(struct xe_migrate *migrate); int xe_migrate_access_memory(struct xe_migrate *m, struct xe_bo *bo, unsigned long offset, void *buf, int len, int write); @@ -133,5 +152,7 @@ xe_migrate_update_pgtables(struct xe_migrate *m, void xe_migrate_wait(struct xe_migrate *m); -struct xe_exec_queue *xe_tile_migrate_exec_queue(struct xe_tile *tile); +void xe_migrate_job_lock(struct xe_migrate *m, struct xe_exec_queue *q); +void xe_migrate_job_unlock(struct xe_migrate *m, struct xe_exec_queue *q); + #endif diff --git a/drivers/gpu/drm/xe/xe_mmio.c b/drivers/gpu/drm/xe/xe_mmio.c index e4db8d58ea2d..ef6f3ea573a2 100644 --- a/drivers/gpu/drm/xe/xe_mmio.c +++ b/drivers/gpu/drm/xe/xe_mmio.c @@ -58,7 +58,6 @@ static void tiles_fini(void *arg) static void mmio_multi_tile_setup(struct xe_device *xe, size_t tile_mmio_size) { struct xe_tile *tile; - struct xe_gt *gt; u8 id; /* @@ -68,38 +67,6 @@ static void mmio_multi_tile_setup(struct xe_device *xe, size_t tile_mmio_size) if (xe->info.tile_count == 1) return; - /* Possibly override number of tile based on configuration register */ - if (!xe->info.skip_mtcfg) { - struct xe_mmio *mmio = xe_root_tile_mmio(xe); - u8 tile_count, gt_count; - u32 mtcfg; - - /* - * Although the per-tile mmio regs are not yet initialized, this - * is fine as it's going to the root tile's mmio, that's - * guaranteed to be initialized earlier in xe_mmio_probe_early() - */ - mtcfg = xe_mmio_read32(mmio, XEHP_MTCFG_ADDR); - tile_count = REG_FIELD_GET(TILE_COUNT, mtcfg) + 1; - - if (tile_count < xe->info.tile_count) { - drm_info(&xe->drm, "tile_count: %d, reduced_tile_count %d\n", - xe->info.tile_count, tile_count); - xe->info.tile_count = tile_count; - - /* - * We've already setup gt_count according to the full - * tile count. Re-calculate it to only include the GTs - * that belong to the remaining tile(s). - */ - gt_count = 0; - for_each_gt(gt, xe, id) - if (gt->info.id < tile_count * xe->info.max_gt_per_tile) - gt_count++; - xe->info.gt_count = gt_count; - } - } - for_each_remote_tile(tile, xe, id) xe_mmio_init(&tile->mmio, tile, xe->mmio.regs + id * tile_mmio_size, SZ_4M); } diff --git a/drivers/gpu/drm/xe/xe_mmio_gem.c b/drivers/gpu/drm/xe/xe_mmio_gem.c new file mode 100644 index 000000000000..9a97c4387e4f --- /dev/null +++ b/drivers/gpu/drm/xe/xe_mmio_gem.c @@ -0,0 +1,226 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright © 2025 Intel Corporation + */ + +#include "xe_mmio_gem.h" + +#include <drm/drm_drv.h> +#include <drm/drm_gem.h> +#include <drm/drm_managed.h> + +#include "xe_device_types.h" + +/** + * DOC: Exposing MMIO regions to userspace + * + * In certain cases, the driver may allow userspace to mmap a portion of the hardware registers. + * + * This can be done as follows: + * 1. Call xe_mmio_gem_create() to create a GEM object with an mmap-able fake offset. + * 2. Use xe_mmio_gem_mmap_offset() on the created GEM object to retrieve the fake offset. + * 3. Provide the fake offset to userspace. + * 4. Userspace can call mmap with the fake offset. The length provided to mmap + * must match the size of the GEM object. + * 5. When the region is no longer needed, call xe_mmio_gem_destroy() to release the GEM object. + * + * NOTE: The exposed MMIO region must be page-aligned with regards to its BAR offset and size. + * + * WARNING: Exposing MMIO regions to userspace can have security and stability implications. + * Make sure not to expose any sensitive registers. + */ + +static void xe_mmio_gem_free(struct drm_gem_object *); +static int xe_mmio_gem_mmap(struct drm_gem_object *, struct vm_area_struct *); +static vm_fault_t xe_mmio_gem_vm_fault(struct vm_fault *); + +struct xe_mmio_gem { + struct drm_gem_object base; + phys_addr_t phys_addr; +}; + +static const struct vm_operations_struct vm_ops = { + .open = drm_gem_vm_open, + .close = drm_gem_vm_close, + .fault = xe_mmio_gem_vm_fault, +}; + +static const struct drm_gem_object_funcs xe_mmio_gem_funcs = { + .free = xe_mmio_gem_free, + .mmap = xe_mmio_gem_mmap, + .vm_ops = &vm_ops, +}; + +static inline struct xe_mmio_gem *to_xe_mmio_gem(struct drm_gem_object *obj) +{ + return container_of(obj, struct xe_mmio_gem, base); +} + +/** + * xe_mmio_gem_create - Expose an MMIO region to userspace + * @xe: The xe device + * @file: DRM file descriptor + * @phys_addr: Start of the exposed MMIO region + * @size: The size of the exposed MMIO region + * + * This function creates a GEM object that exposes an MMIO region with an mmap-able + * fake offset. + * + * See: "Exposing MMIO regions to userspace" + */ +struct xe_mmio_gem *xe_mmio_gem_create(struct xe_device *xe, struct drm_file *file, + phys_addr_t phys_addr, size_t size) +{ + struct xe_mmio_gem *obj; + struct drm_gem_object *base; + int err; + + if ((phys_addr % PAGE_SIZE != 0) || (size % PAGE_SIZE != 0)) + return ERR_PTR(-EINVAL); + + obj = kzalloc(sizeof(*obj), GFP_KERNEL); + if (!obj) + return ERR_PTR(-ENOMEM); + + base = &obj->base; + base->funcs = &xe_mmio_gem_funcs; + obj->phys_addr = phys_addr; + + drm_gem_private_object_init(&xe->drm, base, size); + + err = drm_gem_create_mmap_offset(base); + if (err) + goto free_gem; + + err = drm_vma_node_allow(&base->vma_node, file); + if (err) + goto free_gem; + + return obj; + +free_gem: + xe_mmio_gem_free(base); + return ERR_PTR(err); +} + +/** + * xe_mmio_gem_mmap_offset - Return the mmap-able fake offset + * @gem: the GEM object created with xe_mmio_gem_create() + * + * This function returns the mmap-able fake offset allocated during + * xe_mmio_gem_create(). + * + * See: "Exposing MMIO regions to userspace" + */ +u64 xe_mmio_gem_mmap_offset(struct xe_mmio_gem *gem) +{ + return drm_vma_node_offset_addr(&gem->base.vma_node); +} + +static void xe_mmio_gem_free(struct drm_gem_object *base) +{ + struct xe_mmio_gem *obj = to_xe_mmio_gem(base); + + drm_gem_object_release(base); + kfree(obj); +} + +/** + * xe_mmio_gem_destroy - Destroy the GEM object that exposes an MMIO region + * @gem: the GEM object to destroy + * + * This function releases resources associated with the GEM object created by + * xe_mmio_gem_create(). + * + * See: "Exposing MMIO regions to userspace" + */ +void xe_mmio_gem_destroy(struct xe_mmio_gem *gem) +{ + xe_mmio_gem_free(&gem->base); +} + +static int xe_mmio_gem_mmap(struct drm_gem_object *base, struct vm_area_struct *vma) +{ + if (vma->vm_end - vma->vm_start != base->size) + return -EINVAL; + + if ((vma->vm_flags & VM_SHARED) == 0) + return -EINVAL; + + /* Set vm_pgoff (used as a fake buffer offset by DRM) to 0 */ + vma->vm_pgoff = 0; + vma->vm_page_prot = pgprot_noncached(vm_get_page_prot(vma->vm_flags)); + vm_flags_set(vma, VM_IO | VM_PFNMAP | VM_DONTEXPAND | VM_DONTDUMP | + VM_DONTCOPY | VM_NORESERVE); + + /* Defer actual mapping to the fault handler. */ + return 0; +} + +static void xe_mmio_gem_release_dummy_page(struct drm_device *dev, void *res) +{ + __free_page((struct page *)res); +} + +static vm_fault_t xe_mmio_gem_vm_fault_dummy_page(struct vm_area_struct *vma) +{ + struct drm_gem_object *base = vma->vm_private_data; + struct drm_device *dev = base->dev; + vm_fault_t ret = VM_FAULT_NOPAGE; + struct page *page; + unsigned long pfn; + unsigned long i; + + page = alloc_page(GFP_KERNEL | __GFP_ZERO); + if (!page) + return VM_FAULT_OOM; + + if (drmm_add_action_or_reset(dev, xe_mmio_gem_release_dummy_page, page)) + return VM_FAULT_OOM; + + pfn = page_to_pfn(page); + + /* Map the entire VMA to the same dummy page */ + for (i = 0; i < base->size; i += PAGE_SIZE) { + unsigned long addr = vma->vm_start + i; + + ret = vmf_insert_pfn(vma, addr, pfn); + if (ret & VM_FAULT_ERROR) + break; + } + + return ret; +} + +static vm_fault_t xe_mmio_gem_vm_fault(struct vm_fault *vmf) +{ + struct vm_area_struct *vma = vmf->vma; + struct drm_gem_object *base = vma->vm_private_data; + struct xe_mmio_gem *obj = to_xe_mmio_gem(base); + struct drm_device *dev = base->dev; + vm_fault_t ret = VM_FAULT_NOPAGE; + unsigned long i; + int idx; + + if (!drm_dev_enter(dev, &idx)) { + /* + * Provide a dummy page to avoid SIGBUS for events such as hot-unplug. + * This gives the userspace the option to recover instead of crashing. + * It is assumed the userspace will receive the notification via some + * other channel (e.g. drm uevent). + */ + return xe_mmio_gem_vm_fault_dummy_page(vma); + } + + for (i = 0; i < base->size; i += PAGE_SIZE) { + unsigned long addr = vma->vm_start + i; + unsigned long phys_addr = obj->phys_addr + i; + + ret = vmf_insert_pfn(vma, addr, PHYS_PFN(phys_addr)); + if (ret & VM_FAULT_ERROR) + break; + } + + drm_dev_exit(idx); + return ret; +} diff --git a/drivers/gpu/drm/xe/xe_mmio_gem.h b/drivers/gpu/drm/xe/xe_mmio_gem.h new file mode 100644 index 000000000000..4b76d5586ebb --- /dev/null +++ b/drivers/gpu/drm/xe/xe_mmio_gem.h @@ -0,0 +1,20 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2025 Intel Corporation + */ + +#ifndef _XE_MMIO_GEM_H_ +#define _XE_MMIO_GEM_H_ + +#include <linux/types.h> + +struct drm_file; +struct xe_device; +struct xe_mmio_gem; + +struct xe_mmio_gem *xe_mmio_gem_create(struct xe_device *xe, struct drm_file *file, + phys_addr_t phys_addr, size_t size); +u64 xe_mmio_gem_mmap_offset(struct xe_mmio_gem *gem); +void xe_mmio_gem_destroy(struct xe_mmio_gem *gem); + +#endif /* _XE_MMIO_GEM_H_ */ diff --git a/drivers/gpu/drm/xe/xe_module.c b/drivers/gpu/drm/xe/xe_module.c index d9391bd08194..d08338fc3bc1 100644 --- a/drivers/gpu/drm/xe/xe_module.c +++ b/drivers/gpu/drm/xe/xe_module.c @@ -135,24 +135,17 @@ static const struct init_funcs init_funcs[] = { }, }; -static int __init xe_call_init_func(unsigned int i) +static int __init xe_call_init_func(const struct init_funcs *func) { - if (WARN_ON(i >= ARRAY_SIZE(init_funcs))) - return 0; - if (!init_funcs[i].init) - return 0; - - return init_funcs[i].init(); + if (func->init) + return func->init(); + return 0; } -static void xe_call_exit_func(unsigned int i) +static void xe_call_exit_func(const struct init_funcs *func) { - if (WARN_ON(i >= ARRAY_SIZE(init_funcs))) - return; - if (!init_funcs[i].exit) - return; - - init_funcs[i].exit(); + if (func->exit) + func->exit(); } static int __init xe_init(void) @@ -160,10 +153,12 @@ static int __init xe_init(void) int err, i; for (i = 0; i < ARRAY_SIZE(init_funcs); i++) { - err = xe_call_init_func(i); + err = xe_call_init_func(init_funcs + i); if (err) { + pr_info("%s: module_init aborted at %ps %pe\n", + DRIVER_NAME, init_funcs[i].init, ERR_PTR(err)); while (i--) - xe_call_exit_func(i); + xe_call_exit_func(init_funcs + i); return err; } } @@ -176,7 +171,7 @@ static void __exit xe_exit(void) int i; for (i = ARRAY_SIZE(init_funcs) - 1; i >= 0; i--) - xe_call_exit_func(i); + xe_call_exit_func(init_funcs + i); } module_init(xe_init); diff --git a/drivers/gpu/drm/xe/xe_nvm.c b/drivers/gpu/drm/xe/xe_nvm.c index 61b0a1531a53..9ca4a5ae1693 100644 --- a/drivers/gpu/drm/xe/xe_nvm.c +++ b/drivers/gpu/drm/xe/xe_nvm.c @@ -39,17 +39,17 @@ static void xe_nvm_release_dev(struct device *dev) static bool xe_nvm_non_posted_erase(struct xe_device *xe) { - struct xe_gt *gt = xe_root_mmio_gt(xe); + struct xe_mmio *mmio = xe_root_tile_mmio(xe); if (xe->info.platform != XE_BATTLEMAGE) return false; - return !(xe_mmio_read32(>->mmio, XE_REG(GEN12_CNTL_PROTECTED_NVM_REG)) & + return !(xe_mmio_read32(mmio, XE_REG(GEN12_CNTL_PROTECTED_NVM_REG)) & NVM_NON_POSTED_ERASE_CHICKEN_BIT); } static bool xe_nvm_writable_override(struct xe_device *xe) { - struct xe_gt *gt = xe_root_mmio_gt(xe); + struct xe_mmio *mmio = xe_root_tile_mmio(xe); bool writable_override; resource_size_t base; @@ -72,7 +72,7 @@ static bool xe_nvm_writable_override(struct xe_device *xe) } writable_override = - !(xe_mmio_read32(>->mmio, HECI_FWSTS2(base)) & + !(xe_mmio_read32(mmio, HECI_FWSTS2(base)) & HECI_FW_STATUS_2_NVM_ACCESS_MODE); if (writable_override) drm_info(&xe->drm, "NVM access overridden by jumper\n"); diff --git a/drivers/gpu/drm/xe/xe_oa.c b/drivers/gpu/drm/xe/xe_oa.c index 5729e7d3e335..a188bad172ad 100644 --- a/drivers/gpu/drm/xe/xe_oa.c +++ b/drivers/gpu/drm/xe/xe_oa.c @@ -822,7 +822,7 @@ static void xe_oa_disable_metric_set(struct xe_oa_stream *stream) u32 sqcnt1; /* Enable thread stall DOP gating and EU DOP gating. */ - if (XE_WA(stream->gt, 1508761755)) { + if (XE_GT_WA(stream->gt, 1508761755)) { xe_gt_mcr_multicast_write(stream->gt, ROW_CHICKEN, _MASKED_BIT_DISABLE(STALL_DOP_GATING_DISABLE)); xe_gt_mcr_multicast_write(stream->gt, ROW_CHICKEN2, @@ -1079,7 +1079,7 @@ static int xe_oa_enable_metric_set(struct xe_oa_stream *stream) * EU NOA signals behave incorrectly if EU clock gating is enabled. * Disable thread stall DOP gating and EU DOP gating. */ - if (XE_WA(stream->gt, 1508761755)) { + if (XE_GT_WA(stream->gt, 1508761755)) { xe_gt_mcr_multicast_write(stream->gt, ROW_CHICKEN, _MASKED_BIT_ENABLE(STALL_DOP_GATING_DISABLE)); xe_gt_mcr_multicast_write(stream->gt, ROW_CHICKEN2, @@ -1754,7 +1754,7 @@ static int xe_oa_stream_init(struct xe_oa_stream *stream, * GuC reset of engines causes OA to lose configuration * state. Prevent this by overriding GUCRC mode. */ - if (XE_WA(stream->gt, 1509372804)) { + if (XE_GT_WA(stream->gt, 1509372804)) { ret = xe_guc_pc_override_gucrc_mode(>->uc.guc.pc, SLPC_GUCRC_MODE_GUCRC_NO_RC6); if (ret) @@ -1886,7 +1886,7 @@ u32 xe_oa_timestamp_frequency(struct xe_gt *gt) { u32 reg, shift; - if (XE_WA(gt, 18013179988) || XE_WA(gt, 14015568240)) { + if (XE_GT_WA(gt, 18013179988) || XE_GT_WA(gt, 14015568240)) { xe_pm_runtime_get(gt_to_xe(gt)); reg = xe_mmio_read32(>->mmio, RPM_CONFIG0); xe_pm_runtime_put(gt_to_xe(gt)); diff --git a/drivers/gpu/drm/xe/xe_pci.c b/drivers/gpu/drm/xe/xe_pci.c index 3c40ef426f0c..046d330bad34 100644 --- a/drivers/gpu/drm/xe/xe_pci.c +++ b/drivers/gpu/drm/xe/xe_pci.c @@ -17,6 +17,8 @@ #include "display/xe_display.h" #include "regs/xe_gt_regs.h" +#include "regs/xe_regs.h" +#include "xe_configfs.h" #include "xe_device.h" #include "xe_drv.h" #include "xe_gt.h" @@ -55,7 +57,7 @@ static const struct xe_graphics_desc graphics_xelp = { }; #define XE_HP_FEATURES \ - .has_range_tlb_invalidation = true, \ + .has_range_tlb_inval = true, \ .va_bits = 48, \ .vm_max_level = 3 @@ -103,7 +105,7 @@ static const struct xe_graphics_desc graphics_xelpg = { .has_asid = 1, \ .has_atomic_enable_pte_bit = 1, \ .has_flat_ccs = 1, \ - .has_range_tlb_invalidation = 1, \ + .has_range_tlb_inval = 1, \ .has_usm = 1, \ .has_64bit_timestamp = 1, \ .va_bits = 48, \ @@ -169,6 +171,7 @@ static const struct xe_device_desc tgl_desc = { .dma_mask_size = 39, .has_display = true, .has_llc = true, + .has_sriov = true, .max_gt_per_tile = 1, .require_force_probe = true, }; @@ -193,6 +196,7 @@ static const struct xe_device_desc adl_s_desc = { .dma_mask_size = 39, .has_display = true, .has_llc = true, + .has_sriov = true, .max_gt_per_tile = 1, .require_force_probe = true, .subplatforms = (const struct xe_subplatform_desc[]) { @@ -210,6 +214,7 @@ static const struct xe_device_desc adl_p_desc = { .dma_mask_size = 39, .has_display = true, .has_llc = true, + .has_sriov = true, .max_gt_per_tile = 1, .require_force_probe = true, .subplatforms = (const struct xe_subplatform_desc[]) { @@ -225,6 +230,7 @@ static const struct xe_device_desc adl_n_desc = { .dma_mask_size = 39, .has_display = true, .has_llc = true, + .has_sriov = true, .max_gt_per_tile = 1, .require_force_probe = true, }; @@ -270,6 +276,7 @@ static const struct xe_device_desc ats_m_desc = { DG2_FEATURES, .has_display = false, + .has_sriov = true, }; static const struct xe_device_desc dg2_desc = { @@ -599,6 +606,44 @@ static int xe_info_init_early(struct xe_device *xe, } /* + * Possibly override number of tile based on configuration register. + */ +static void xe_info_probe_tile_count(struct xe_device *xe) +{ + struct xe_mmio *mmio; + u8 tile_count; + u32 mtcfg; + + KUNIT_STATIC_STUB_REDIRECT(xe_info_probe_tile_count, xe); + + /* + * Probe for tile count only for platforms that support multiple + * tiles. + */ + if (xe->info.tile_count == 1) + return; + + if (xe->info.skip_mtcfg) + return; + + mmio = xe_root_tile_mmio(xe); + + /* + * Although the per-tile mmio regs are not yet initialized, this + * is fine as it's going to the root tile's mmio, that's + * guaranteed to be initialized earlier in xe_mmio_probe_early() + */ + mtcfg = xe_mmio_read32(mmio, XEHP_MTCFG_ADDR); + tile_count = REG_FIELD_GET(TILE_COUNT, mtcfg) + 1; + + if (tile_count < xe->info.tile_count) { + drm_info(&xe->drm, "tile_count: %d, reduced_tile_count %d\n", + xe->info.tile_count, tile_count); + xe->info.tile_count = tile_count; + } +} + +/* * Initialize device info content that does require knowledge about * graphics / media IP version. * Make sure that GT / tile structures allocated by the driver match the data @@ -668,10 +713,12 @@ static int xe_info_init(struct xe_device *xe, /* Runtime detection may change this later */ xe->info.has_flat_ccs = graphics_desc->has_flat_ccs; - xe->info.has_range_tlb_invalidation = graphics_desc->has_range_tlb_invalidation; + xe->info.has_range_tlb_inval = graphics_desc->has_range_tlb_inval; xe->info.has_usm = graphics_desc->has_usm; xe->info.has_64bit_timestamp = graphics_desc->has_64bit_timestamp; + xe_info_probe_tile_count(xe); + for_each_remote_tile(tile, xe, id) { int err; @@ -687,12 +734,17 @@ static int xe_info_init(struct xe_device *xe, * All of these together determine the overall GT count. */ for_each_tile(tile, xe, id) { + int err; + gt = tile->primary_gt; gt->info.type = XE_GT_TYPE_MAIN; gt->info.id = tile->id * xe->info.max_gt_per_tile; gt->info.has_indirect_ring_state = graphics_desc->has_indirect_ring_state; gt->info.engine_mask = graphics_desc->hw_engine_mask; - xe->info.gt_count++; + + err = xe_tile_alloc_vram(tile); + if (err) + return err; if (MEDIA_VER(xe) < 13 && media_desc) gt->info.engine_mask |= media_desc->hw_engine_mask; @@ -713,9 +765,15 @@ static int xe_info_init(struct xe_device *xe, gt->info.id = tile->id * xe->info.max_gt_per_tile + 1; gt->info.has_indirect_ring_state = media_desc->has_indirect_ring_state; gt->info.engine_mask = media_desc->hw_engine_mask; - xe->info.gt_count++; } + /* + * Now that we have tiles and GTs defined, let's loop over valid GTs + * in order to define gt_count. + */ + for_each_gt(gt, xe, id) + xe->info.gt_count++; + return 0; } @@ -726,7 +784,7 @@ static void xe_pci_remove(struct pci_dev *pdev) if (IS_SRIOV_PF(xe)) xe_pci_sriov_configure(pdev, 0); - if (xe_survivability_mode_is_enabled(xe)) + if (xe_survivability_mode_is_boot_enabled(xe)) return; xe_device_remove(xe); @@ -759,6 +817,8 @@ static int xe_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent) struct xe_device *xe; int err; + xe_configfs_check_device(pdev); + if (desc->require_force_probe && !id_forced(pdev->device)) { dev_info(&pdev->dev, "Your graphics device %04x is not officially supported\n" @@ -806,7 +866,7 @@ static int xe_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent) * flashed through mei. Return success, if survivability mode * is enabled due to pcode failure or configfs being set */ - if (xe_survivability_mode_is_enabled(xe)) + if (xe_survivability_mode_is_boot_enabled(xe)) return 0; if (err) @@ -900,7 +960,7 @@ static int xe_pci_suspend(struct device *dev) struct xe_device *xe = pdev_to_xe_device(pdev); int err; - if (xe_survivability_mode_is_enabled(xe)) + if (xe_survivability_mode_is_boot_enabled(xe)) return -EBUSY; err = xe_pm_suspend(xe); diff --git a/drivers/gpu/drm/xe/xe_pci_types.h b/drivers/gpu/drm/xe/xe_pci_types.h index 4de6f69ed975..b63002fc0f67 100644 --- a/drivers/gpu/drm/xe/xe_pci_types.h +++ b/drivers/gpu/drm/xe/xe_pci_types.h @@ -60,7 +60,7 @@ struct xe_graphics_desc { u8 has_atomic_enable_pte_bit:1; u8 has_flat_ccs:1; u8 has_indirect_ring_state:1; - u8 has_range_tlb_invalidation:1; + u8 has_range_tlb_inval:1; u8 has_usm:1; u8 has_64bit_timestamp:1; }; diff --git a/drivers/gpu/drm/xe/xe_pm.c b/drivers/gpu/drm/xe/xe_pm.c index e279b47ba03b..d4b54cd34cff 100644 --- a/drivers/gpu/drm/xe/xe_pm.c +++ b/drivers/gpu/drm/xe/xe_pm.c @@ -18,11 +18,12 @@ #include "xe_device.h" #include "xe_ggtt.h" #include "xe_gt.h" -#include "xe_guc.h" +#include "xe_gt_idle.h" #include "xe_i2c.h" #include "xe_irq.h" #include "xe_pcode.h" #include "xe_pxp.h" +#include "xe_sriov_vf_ccs.h" #include "xe_trace.h" #include "xe_wa.h" @@ -176,6 +177,9 @@ int xe_pm_resume(struct xe_device *xe) drm_dbg(&xe->drm, "Resuming device\n"); trace_xe_pm_resume(xe, __builtin_return_address(0)); + for_each_gt(gt, xe, id) + xe_gt_idle_disable_c6(gt); + for_each_tile(tile, xe, id) xe_wa_apply_tile_workarounds(tile); @@ -208,6 +212,9 @@ int xe_pm_resume(struct xe_device *xe) xe_pxp_pm_resume(xe->pxp); + if (IS_SRIOV_VF(xe)) + xe_sriov_vf_ccs_register_context(xe); + drm_dbg(&xe->drm, "Device resumed\n"); return 0; err: @@ -243,6 +250,10 @@ static void xe_pm_runtime_init(struct xe_device *xe) { struct device *dev = xe->drm.dev; + /* Our current VFs do not support RPM. so, disable it */ + if (IS_SRIOV_VF(xe)) + return; + /* * Disable the system suspend direct complete optimization. * We need to ensure that the regular device suspend/resume functions @@ -367,6 +378,10 @@ static void xe_pm_runtime_fini(struct xe_device *xe) { struct device *dev = xe->drm.dev; + /* Our current VFs do not support RPM. so, disable it */ + if (IS_SRIOV_VF(xe)) + return; + pm_runtime_get_sync(dev); pm_runtime_forbid(dev); } @@ -525,6 +540,9 @@ int xe_pm_runtime_resume(struct xe_device *xe) xe_rpm_lockmap_acquire(xe); + for_each_gt(gt, xe, id) + xe_gt_idle_disable_c6(gt); + if (xe->d3cold.allowed) { err = xe_pcode_ready(xe, true); if (err) @@ -558,6 +576,9 @@ int xe_pm_runtime_resume(struct xe_device *xe) xe_pxp_pm_resume(xe->pxp); + if (IS_SRIOV_VF(xe)) + xe_sriov_vf_ccs_register_context(xe); + out: xe_rpm_lockmap_release(xe); xe_pm_write_callback_task(xe, NULL); diff --git a/drivers/gpu/drm/xe/xe_psmi.c b/drivers/gpu/drm/xe/xe_psmi.c new file mode 100644 index 000000000000..a2c9ff5bfd59 --- /dev/null +++ b/drivers/gpu/drm/xe/xe_psmi.c @@ -0,0 +1,306 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright © 2025 Intel Corporation + */ + +#include <linux/debugfs.h> + +#include "xe_bo.h" +#include "xe_device.h" +#include "xe_configfs.h" +#include "xe_psmi.h" + +/* + * PSMI capture support + * + * Requirement for PSMI capture is to have a physically contiguous buffer. The + * PSMI tool owns doing all necessary configuration (MMIO register writes are + * done from user-space). However, KMD needs to provide the PSMI tool with the + * required physical address of the base of PSMI buffer in case of VRAM. + * + * VRAM backed PSMI buffer: + * Buffer is allocated as GEM object and with XE_BO_CREATE_PINNED_BIT flag which + * creates a contiguous allocation. The physical address is returned from + * psmi_debugfs_capture_addr_show(). PSMI tool can mmap the buffer via the + * PCIBAR through sysfs. + * + * SYSTEM memory backed PSMI buffer: + * Interface here does not support allocating from SYSTEM memory region. The + * PSMI tool needs to allocate memory themselves using hugetlbfs. In order to + * get the physical address, user-space can query /proc/[pid]/pagemap. As an + * alternative, CMA debugfs could also be used to allocate reserved CMA memory. + */ + +static bool psmi_enabled(struct xe_device *xe) +{ + return xe_configfs_get_psmi_enabled(to_pci_dev(xe->drm.dev)); +} + +static void psmi_free_object(struct xe_bo *bo) +{ + xe_bo_lock(bo, NULL); + xe_bo_unpin(bo); + xe_bo_unlock(bo); + xe_bo_put(bo); +} + +/* + * Free PSMI capture buffer objects. + */ +static void psmi_cleanup(struct xe_device *xe) +{ + unsigned long id, region_mask = xe->psmi.region_mask; + struct xe_bo *bo; + + for_each_set_bit(id, ®ion_mask, + ARRAY_SIZE(xe->psmi.capture_obj)) { + /* smem should never be set */ + xe_assert(xe, id); + + bo = xe->psmi.capture_obj[id]; + if (bo) { + psmi_free_object(bo); + xe->psmi.capture_obj[id] = NULL; + } + } +} + +static struct xe_bo *psmi_alloc_object(struct xe_device *xe, + unsigned int id, size_t bo_size) +{ + struct xe_bo *bo = NULL; + struct xe_tile *tile; + int err; + + if (!id || !bo_size) + return NULL; + + tile = &xe->tiles[id - 1]; + + /* VRAM: Allocate GEM object for the capture buffer */ + bo = xe_bo_create_locked(xe, tile, NULL, bo_size, + ttm_bo_type_kernel, + XE_BO_FLAG_VRAM_IF_DGFX(tile) | + XE_BO_FLAG_PINNED | + XE_BO_FLAG_PINNED_LATE_RESTORE | + XE_BO_FLAG_NEEDS_CPU_ACCESS); + + if (!IS_ERR(bo)) { + /* Buffer written by HW, ensure stays resident */ + err = xe_bo_pin(bo); + if (err) + bo = ERR_PTR(err); + xe_bo_unlock(bo); + } + + return bo; +} + +/* + * Allocate PSMI capture buffer objects (via debugfs set function), based on + * which regions the user has selected in region_mask. @size: size in bytes + * (should be power of 2) + * + * Always release/free the current buffer objects before attempting to allocate + * new ones. Size == 0 will free all current buffers. + * + * Note, we don't write any registers as the capture tool is already configuring + * all PSMI registers itself via mmio space. + */ +static int psmi_resize_object(struct xe_device *xe, size_t size) +{ + unsigned long id, region_mask = xe->psmi.region_mask; + struct xe_bo *bo = NULL; + int err = 0; + + /* if resizing, free currently allocated buffers first */ + psmi_cleanup(xe); + + /* can set size to 0, in which case, now done */ + if (!size) + return 0; + + for_each_set_bit(id, ®ion_mask, + ARRAY_SIZE(xe->psmi.capture_obj)) { + /* smem should never be set */ + xe_assert(xe, id); + + bo = psmi_alloc_object(xe, id, size); + if (IS_ERR(bo)) { + err = PTR_ERR(bo); + break; + } + xe->psmi.capture_obj[id] = bo; + + drm_info(&xe->drm, + "PSMI capture size requested: %zu bytes, allocated: %lu:%zu\n", + size, id, bo ? xe_bo_size(bo) : 0); + } + + /* on error, reverse what was allocated */ + if (err) + psmi_cleanup(xe); + + return err; +} + +/* + * Returns an address for the capture tool to use to find start of capture + * buffer. Capture tool requires the capability to have a buffer allocated per + * each tile (VRAM region), thus we return an address for each region. + */ +static int psmi_debugfs_capture_addr_show(struct seq_file *m, void *data) +{ + struct xe_device *xe = m->private; + unsigned long id, region_mask; + struct xe_bo *bo; + u64 val; + + region_mask = xe->psmi.region_mask; + for_each_set_bit(id, ®ion_mask, + ARRAY_SIZE(xe->psmi.capture_obj)) { + /* smem should never be set */ + xe_assert(xe, id); + + /* VRAM region */ + bo = xe->psmi.capture_obj[id]; + if (!bo) + continue; + + /* pinned, so don't need bo_lock */ + val = __xe_bo_addr(bo, 0, PAGE_SIZE); + seq_printf(m, "%ld: 0x%llx\n", id, val); + } + + return 0; +} + +/* + * Return capture buffer size, using the size from first allocated object that + * is found. This works because all objects must be of the same size. + */ +static int psmi_debugfs_capture_size_get(void *data, u64 *val) +{ + unsigned long id, region_mask; + struct xe_device *xe = data; + struct xe_bo *bo; + + region_mask = xe->psmi.region_mask; + for_each_set_bit(id, ®ion_mask, + ARRAY_SIZE(xe->psmi.capture_obj)) { + /* smem should never be set */ + xe_assert(xe, id); + + bo = xe->psmi.capture_obj[id]; + if (bo) { + *val = xe_bo_size(bo); + return 0; + } + } + + /* no capture objects are allocated */ + *val = 0; + + return 0; +} + +/* + * Set size of PSMI capture buffer. This triggers the allocation of capture + * buffer in each memory region as specified with prior write to + * psmi_capture_region_mask. + */ +static int psmi_debugfs_capture_size_set(void *data, u64 val) +{ + struct xe_device *xe = data; + + /* user must have specified at least one region */ + if (!xe->psmi.region_mask) + return -EINVAL; + + return psmi_resize_object(xe, val); +} + +static int psmi_debugfs_capture_region_mask_get(void *data, u64 *val) +{ + struct xe_device *xe = data; + + *val = xe->psmi.region_mask; + + return 0; +} + +/* + * Select VRAM regions for multi-tile devices, only allowed when buffer is not + * currently allocated. + */ +static int psmi_debugfs_capture_region_mask_set(void *data, u64 region_mask) +{ + struct xe_device *xe = data; + u64 size = 0; + + /* SMEM is not supported (see comments at top of file) */ + if (region_mask & 0x1) + return -EOPNOTSUPP; + + /* input bitmask should contain only valid TTM regions */ + if (!region_mask || region_mask & ~xe->info.mem_region_mask) + return -EINVAL; + + /* only allow setting mask if buffer is not yet allocated */ + psmi_debugfs_capture_size_get(xe, &size); + if (size) + return -EBUSY; + + xe->psmi.region_mask = region_mask; + + return 0; +} + +DEFINE_SHOW_ATTRIBUTE(psmi_debugfs_capture_addr); + +DEFINE_DEBUGFS_ATTRIBUTE(psmi_debugfs_capture_region_mask_fops, + psmi_debugfs_capture_region_mask_get, + psmi_debugfs_capture_region_mask_set, + "0x%llx\n"); + +DEFINE_DEBUGFS_ATTRIBUTE(psmi_debugfs_capture_size_fops, + psmi_debugfs_capture_size_get, + psmi_debugfs_capture_size_set, + "%lld\n"); + +void xe_psmi_debugfs_register(struct xe_device *xe) +{ + struct drm_minor *minor; + + if (!psmi_enabled(xe)) + return; + + minor = xe->drm.primary; + if (!minor->debugfs_root) + return; + + debugfs_create_file("psmi_capture_addr", + 0400, minor->debugfs_root, xe, + &psmi_debugfs_capture_addr_fops); + + debugfs_create_file("psmi_capture_region_mask", + 0600, minor->debugfs_root, xe, + &psmi_debugfs_capture_region_mask_fops); + + debugfs_create_file("psmi_capture_size", + 0600, minor->debugfs_root, xe, + &psmi_debugfs_capture_size_fops); +} + +static void psmi_fini(void *arg) +{ + psmi_cleanup(arg); +} + +int xe_psmi_init(struct xe_device *xe) +{ + if (!psmi_enabled(xe)) + return 0; + + return devm_add_action(xe->drm.dev, psmi_fini, xe); +} diff --git a/drivers/gpu/drm/xe/xe_psmi.h b/drivers/gpu/drm/xe/xe_psmi.h new file mode 100644 index 000000000000..b1dfba80d893 --- /dev/null +++ b/drivers/gpu/drm/xe/xe_psmi.h @@ -0,0 +1,14 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2025 Intel Corporation + */ + +#ifndef _XE_PSMI_H_ +#define _XE_PSMI_H_ + +struct xe_device; + +int xe_psmi_init(struct xe_device *xe); +void xe_psmi_debugfs_register(struct xe_device *xe); + +#endif diff --git a/drivers/gpu/drm/xe/xe_pt.c b/drivers/gpu/drm/xe/xe_pt.c index c8e63bd23300..c129048a9a09 100644 --- a/drivers/gpu/drm/xe/xe_pt.c +++ b/drivers/gpu/drm/xe/xe_pt.c @@ -13,7 +13,7 @@ #include "xe_drm_client.h" #include "xe_exec_queue.h" #include "xe_gt.h" -#include "xe_gt_tlb_invalidation.h" +#include "xe_tlb_inval_job.h" #include "xe_migrate.h" #include "xe_pt_types.h" #include "xe_pt_walk.h" @@ -21,6 +21,7 @@ #include "xe_sched_job.h" #include "xe_sync.h" #include "xe_svm.h" +#include "xe_tlb_inval_job.h" #include "xe_trace.h" #include "xe_ttm_stolen_mgr.h" #include "xe_vm.h" @@ -69,7 +70,7 @@ static u64 __xe_pt_empty_pte(struct xe_tile *tile, struct xe_vm *vm, if (level > MAX_HUGEPTE_LEVEL) return vm->pt_ops->pde_encode_bo(vm->scratch_pt[id][level - 1]->bo, - 0, pat_index); + 0); return vm->pt_ops->pte_encode_addr(xe, 0, pat_index, level, IS_DGFX(xe), 0) | XE_PTE_NULL; @@ -518,7 +519,7 @@ xe_pt_stage_bind_entry(struct xe_ptw *parent, pgoff_t offset, { struct xe_pt_stage_bind_walk *xe_walk = container_of(walk, typeof(*xe_walk), base); - u16 pat_index = xe_walk->vma->pat_index; + u16 pat_index = xe_walk->vma->attr.pat_index; struct xe_pt *xe_parent = container_of(parent, typeof(*xe_parent), base); struct xe_vm *vm = xe_walk->vm; struct xe_pt *xe_child; @@ -616,7 +617,7 @@ xe_pt_stage_bind_entry(struct xe_ptw *parent, pgoff_t offset, xe_child->is_compact = true; } - pte = vm->pt_ops->pde_encode_bo(xe_child->bo, 0, pat_index) | flags; + pte = vm->pt_ops->pde_encode_bo(xe_child->bo, 0) | flags; ret = xe_pt_insert_entry(xe_walk, xe_parent, offset, xe_child, pte); } @@ -640,28 +641,31 @@ static const struct xe_pt_walk_ops xe_pt_stage_bind_ops = { * - In all other cases device atomics will be disabled with AE=0 until an application * request differently using a ioctl like madvise. */ -static bool xe_atomic_for_vram(struct xe_vm *vm) +static bool xe_atomic_for_vram(struct xe_vm *vm, struct xe_vma *vma) { + if (vma->attr.atomic_access == DRM_XE_ATOMIC_CPU) + return false; + return true; } -static bool xe_atomic_for_system(struct xe_vm *vm, struct xe_bo *bo) +static bool xe_atomic_for_system(struct xe_vm *vm, struct xe_vma *vma) { struct xe_device *xe = vm->xe; + struct xe_bo *bo = xe_vma_bo(vma); - if (!xe->info.has_device_atomics_on_smem) + if (!xe->info.has_device_atomics_on_smem || + vma->attr.atomic_access == DRM_XE_ATOMIC_CPU) return false; + if (vma->attr.atomic_access == DRM_XE_ATOMIC_DEVICE) + return true; + /* * If a SMEM+LMEM allocation is backed by SMEM, a device * atomics will cause a gpu page fault and which then * gets migrated to LMEM, bind such allocations with * device atomics enabled. - * - * TODO: Revisit this. Perhaps add something like a - * fault_on_atomics_in_system UAPI flag. - * Note that this also prohibits GPU atomics in LR mode for - * userptr and system memory on DGFX. */ return (!IS_DGFX(xe) || (!xe_vm_in_lr_mode(vm) || (bo && xe_bo_has_single_placement(bo)))); @@ -744,8 +748,8 @@ xe_pt_stage_bind(struct xe_tile *tile, struct xe_vma *vma, goto walk_pt; if (vma->gpuva.flags & XE_VMA_ATOMIC_PTE_BIT) { - xe_walk.default_vram_pte = xe_atomic_for_vram(vm) ? XE_USM_PPGTT_PTE_AE : 0; - xe_walk.default_system_pte = xe_atomic_for_system(vm, bo) ? + xe_walk.default_vram_pte = xe_atomic_for_vram(vm, vma) ? XE_USM_PPGTT_PTE_AE : 0; + xe_walk.default_system_pte = xe_atomic_for_system(vm, vma) ? XE_USM_PPGTT_PTE_AE : 0; } @@ -950,7 +954,19 @@ bool xe_pt_zap_ptes_range(struct xe_tile *tile, struct xe_vm *vm, struct xe_pt *pt = vm->pt_root[tile->id]; u8 pt_mask = (range->tile_present & ~range->tile_invalidated); - xe_svm_assert_in_notifier(vm); + /* + * Locking rules: + * + * - notifier_lock (write): full protection against page table changes + * and MMU notifier invalidations. + * + * - notifier_lock (read) + vm_lock (write): combined protection against + * invalidations and concurrent page table modifications. (e.g., madvise) + * + */ + lockdep_assert(lockdep_is_held_type(&vm->svm.gpusvm.notifier_lock, 0) || + (lockdep_is_held_type(&vm->svm.gpusvm.notifier_lock, 1) && + lockdep_is_held_type(&vm->lock, 0))); if (!(pt_mask & BIT(tile->id))) return false; @@ -1261,6 +1277,8 @@ static int op_add_deps(struct xe_vm *vm, struct xe_vma_op *op, } static int xe_pt_vm_dependencies(struct xe_sched_job *job, + struct xe_tlb_inval_job *ijob, + struct xe_tlb_inval_job *mjob, struct xe_vm *vm, struct xe_vma_ops *vops, struct xe_vm_pgtable_update_ops *pt_update_ops, @@ -1328,6 +1346,20 @@ static int xe_pt_vm_dependencies(struct xe_sched_job *job, for (i = 0; job && !err && i < vops->num_syncs; i++) err = xe_sync_entry_add_deps(&vops->syncs[i], job); + if (job) { + if (ijob) { + err = xe_tlb_inval_job_alloc_dep(ijob); + if (err) + return err; + } + + if (mjob) { + err = xe_tlb_inval_job_alloc_dep(mjob); + if (err) + return err; + } + } + return err; } @@ -1339,7 +1371,8 @@ static int xe_pt_pre_commit(struct xe_migrate_pt_update *pt_update) struct xe_vm_pgtable_update_ops *pt_update_ops = &vops->pt_update_ops[pt_update->tile_id]; - return xe_pt_vm_dependencies(pt_update->job, vm, pt_update->vops, + return xe_pt_vm_dependencies(pt_update->job, pt_update->ijob, + pt_update->mjob, vm, pt_update->vops, pt_update_ops, rftree); } @@ -1509,75 +1542,6 @@ static int xe_pt_svm_pre_commit(struct xe_migrate_pt_update *pt_update) } #endif -struct invalidation_fence { - struct xe_gt_tlb_invalidation_fence base; - struct xe_gt *gt; - struct dma_fence *fence; - struct dma_fence_cb cb; - struct work_struct work; - u64 start; - u64 end; - u32 asid; -}; - -static void invalidation_fence_cb(struct dma_fence *fence, - struct dma_fence_cb *cb) -{ - struct invalidation_fence *ifence = - container_of(cb, struct invalidation_fence, cb); - struct xe_device *xe = gt_to_xe(ifence->gt); - - trace_xe_gt_tlb_invalidation_fence_cb(xe, &ifence->base); - if (!ifence->fence->error) { - queue_work(system_wq, &ifence->work); - } else { - ifence->base.base.error = ifence->fence->error; - xe_gt_tlb_invalidation_fence_signal(&ifence->base); - } - dma_fence_put(ifence->fence); -} - -static void invalidation_fence_work_func(struct work_struct *w) -{ - struct invalidation_fence *ifence = - container_of(w, struct invalidation_fence, work); - struct xe_device *xe = gt_to_xe(ifence->gt); - - trace_xe_gt_tlb_invalidation_fence_work_func(xe, &ifence->base); - xe_gt_tlb_invalidation_range(ifence->gt, &ifence->base, ifence->start, - ifence->end, ifence->asid); -} - -static void invalidation_fence_init(struct xe_gt *gt, - struct invalidation_fence *ifence, - struct dma_fence *fence, - u64 start, u64 end, u32 asid) -{ - int ret; - - trace_xe_gt_tlb_invalidation_fence_create(gt_to_xe(gt), &ifence->base); - - xe_gt_tlb_invalidation_fence_init(gt, &ifence->base, false); - - ifence->fence = fence; - ifence->gt = gt; - ifence->start = start; - ifence->end = end; - ifence->asid = asid; - - INIT_WORK(&ifence->work, invalidation_fence_work_func); - ret = dma_fence_add_callback(fence, &ifence->cb, invalidation_fence_cb); - if (ret == -ENOENT) { - dma_fence_put(ifence->fence); /* Usually dropped in CB */ - invalidation_fence_work_func(&ifence->work); - } else if (ret) { - dma_fence_put(&ifence->base.base); /* Caller ref */ - dma_fence_put(&ifence->base.base); /* Creation ref */ - } - - xe_gt_assert(gt, !ret || ret == -ENOENT); -} - struct xe_pt_stage_unbind_walk { /** @base: The pagewalk base-class. */ struct xe_pt_walk base; @@ -2390,6 +2354,15 @@ static const struct xe_migrate_pt_update_ops svm_migrate_ops = { static const struct xe_migrate_pt_update_ops svm_migrate_ops; #endif +static struct xe_dep_scheduler *to_dep_scheduler(struct xe_exec_queue *q, + struct xe_gt *gt) +{ + if (xe_gt_is_media_type(gt)) + return q->tlb_inval[XE_EXEC_QUEUE_TLB_INVAL_MEDIA_GT].dep_scheduler; + + return q->tlb_inval[XE_EXEC_QUEUE_TLB_INVAL_PRIMARY_GT].dep_scheduler; +} + /** * xe_pt_update_ops_run() - Run PT update operations * @tile: Tile of PT update operations @@ -2407,8 +2380,8 @@ xe_pt_update_ops_run(struct xe_tile *tile, struct xe_vma_ops *vops) struct xe_vm *vm = vops->vm; struct xe_vm_pgtable_update_ops *pt_update_ops = &vops->pt_update_ops[tile->id]; - struct dma_fence *fence; - struct invalidation_fence *ifence = NULL, *mfence = NULL; + struct dma_fence *fence, *ifence, *mfence; + struct xe_tlb_inval_job *ijob = NULL, *mjob = NULL; struct dma_fence **fences = NULL; struct dma_fence_array *cf = NULL; struct xe_range_fence *rfence; @@ -2440,26 +2413,45 @@ xe_pt_update_ops_run(struct xe_tile *tile, struct xe_vma_ops *vops) #endif if (pt_update_ops->needs_invalidation) { - ifence = kzalloc(sizeof(*ifence), GFP_KERNEL); - if (!ifence) { - err = -ENOMEM; + struct xe_exec_queue *q = pt_update_ops->q; + struct xe_dep_scheduler *dep_scheduler = + to_dep_scheduler(q, tile->primary_gt); + + ijob = xe_tlb_inval_job_create(q, &tile->primary_gt->tlb_inval, + dep_scheduler, + pt_update_ops->start, + pt_update_ops->last, + vm->usm.asid); + if (IS_ERR(ijob)) { + err = PTR_ERR(ijob); goto kill_vm_tile1; } + update.ijob = ijob; + if (tile->media_gt) { - mfence = kzalloc(sizeof(*ifence), GFP_KERNEL); - if (!mfence) { - err = -ENOMEM; - goto free_ifence; + dep_scheduler = to_dep_scheduler(q, tile->media_gt); + + mjob = xe_tlb_inval_job_create(q, + &tile->media_gt->tlb_inval, + dep_scheduler, + pt_update_ops->start, + pt_update_ops->last, + vm->usm.asid); + if (IS_ERR(mjob)) { + err = PTR_ERR(mjob); + goto free_ijob; } + update.mjob = mjob; + fences = kmalloc_array(2, sizeof(*fences), GFP_KERNEL); if (!fences) { err = -ENOMEM; - goto free_ifence; + goto free_ijob; } cf = dma_fence_array_alloc(2); if (!cf) { err = -ENOMEM; - goto free_ifence; + goto free_ijob; } } } @@ -2467,7 +2459,7 @@ xe_pt_update_ops_run(struct xe_tile *tile, struct xe_vma_ops *vops) rfence = kzalloc(sizeof(*rfence), GFP_KERNEL); if (!rfence) { err = -ENOMEM; - goto free_ifence; + goto free_ijob; } fence = xe_migrate_update_pgtables(tile->migrate, &update); @@ -2491,30 +2483,31 @@ xe_pt_update_ops_run(struct xe_tile *tile, struct xe_vma_ops *vops) pt_update_ops->last, fence)) dma_fence_wait(fence, false); - /* tlb invalidation must be done before signaling rebind */ - if (ifence) { - if (mfence) - dma_fence_get(fence); - invalidation_fence_init(tile->primary_gt, ifence, fence, - pt_update_ops->start, - pt_update_ops->last, vm->usm.asid); - if (mfence) { - invalidation_fence_init(tile->media_gt, mfence, fence, - pt_update_ops->start, - pt_update_ops->last, vm->usm.asid); - fences[0] = &ifence->base.base; - fences[1] = &mfence->base.base; + /* tlb invalidation must be done before signaling unbind/rebind */ + if (ijob) { + struct dma_fence *__fence; + + ifence = xe_tlb_inval_job_push(ijob, tile->migrate, fence); + __fence = ifence; + + if (mjob) { + fences[0] = ifence; + mfence = xe_tlb_inval_job_push(mjob, tile->migrate, + fence); + fences[1] = mfence; + dma_fence_array_init(cf, 2, fences, vm->composite_fence_ctx, vm->composite_fence_seqno++, false); - fence = &cf->base; - } else { - fence = &ifence->base.base; + __fence = &cf->base; } + + dma_fence_put(fence); + fence = __fence; } - if (!mfence) { + if (!mjob) { dma_resv_add_fence(xe_vm_resv(vm), fence, pt_update_ops->wait_vm_bookkeep ? DMA_RESV_USAGE_KERNEL : @@ -2523,19 +2516,19 @@ xe_pt_update_ops_run(struct xe_tile *tile, struct xe_vma_ops *vops) list_for_each_entry(op, &vops->list, link) op_commit(vops->vm, tile, pt_update_ops, op, fence, NULL); } else { - dma_resv_add_fence(xe_vm_resv(vm), &ifence->base.base, + dma_resv_add_fence(xe_vm_resv(vm), ifence, pt_update_ops->wait_vm_bookkeep ? DMA_RESV_USAGE_KERNEL : DMA_RESV_USAGE_BOOKKEEP); - dma_resv_add_fence(xe_vm_resv(vm), &mfence->base.base, + dma_resv_add_fence(xe_vm_resv(vm), mfence, pt_update_ops->wait_vm_bookkeep ? DMA_RESV_USAGE_KERNEL : DMA_RESV_USAGE_BOOKKEEP); list_for_each_entry(op, &vops->list, link) - op_commit(vops->vm, tile, pt_update_ops, op, - &ifence->base.base, &mfence->base.base); + op_commit(vops->vm, tile, pt_update_ops, op, ifence, + mfence); } if (pt_update_ops->needs_svm_lock) @@ -2543,15 +2536,18 @@ xe_pt_update_ops_run(struct xe_tile *tile, struct xe_vma_ops *vops) if (pt_update_ops->needs_userptr_lock) up_read(&vm->userptr.notifier_lock); + xe_tlb_inval_job_put(mjob); + xe_tlb_inval_job_put(ijob); + return fence; free_rfence: kfree(rfence); -free_ifence: +free_ijob: kfree(cf); kfree(fences); - kfree(mfence); - kfree(ifence); + xe_tlb_inval_job_put(mjob); + xe_tlb_inval_job_put(ijob); kill_vm_tile1: if (err != -EAGAIN && err != -ENODATA && tile->id) xe_vm_kill(vops->vm, false); diff --git a/drivers/gpu/drm/xe/xe_pt_types.h b/drivers/gpu/drm/xe/xe_pt_types.h index 69eab6f37cfe..17cdd7c7e9f5 100644 --- a/drivers/gpu/drm/xe/xe_pt_types.h +++ b/drivers/gpu/drm/xe/xe_pt_types.h @@ -45,8 +45,7 @@ struct xe_pt_ops { u64 (*pte_encode_addr)(struct xe_device *xe, u64 addr, u16 pat_index, u32 pt_level, bool devmem, u64 flags); - u64 (*pde_encode_bo)(struct xe_bo *bo, u64 bo_offset, - u16 pat_index); + u64 (*pde_encode_bo)(struct xe_bo *bo, u64 bo_offset); }; struct xe_pt_entry { diff --git a/drivers/gpu/drm/xe/xe_pxp_submit.c b/drivers/gpu/drm/xe/xe_pxp_submit.c index d92ec0f515b0..ca95f2a4d4ef 100644 --- a/drivers/gpu/drm/xe/xe_pxp_submit.c +++ b/drivers/gpu/drm/xe/xe_pxp_submit.c @@ -101,7 +101,7 @@ static int allocate_gsc_client_resources(struct xe_gt *gt, xe_assert(xe, hwe); /* PXP instructions must be issued from PPGTT */ - vm = xe_vm_create(xe, XE_VM_FLAG_GSC); + vm = xe_vm_create(xe, XE_VM_FLAG_GSC, NULL); if (IS_ERR(vm)) return PTR_ERR(vm); diff --git a/drivers/gpu/drm/xe/xe_query.c b/drivers/gpu/drm/xe/xe_query.c index d517ec9ddcbf..4dbe5732cb7f 100644 --- a/drivers/gpu/drm/xe/xe_query.c +++ b/drivers/gpu/drm/xe/xe_query.c @@ -27,6 +27,7 @@ #include "xe_oa.h" #include "xe_pxp.h" #include "xe_ttm_vram_mgr.h" +#include "xe_vram_types.h" #include "xe_wa.h" static const u16 xe_to_user_engine_class[] = { @@ -337,7 +338,7 @@ static int query_config(struct xe_device *xe, struct drm_xe_device_query *query) config->num_params = num_params; config->info[DRM_XE_QUERY_CONFIG_REV_AND_DEVICE_ID] = xe->info.devid | (xe->info.revid << 16); - if (xe_device_get_root_tile(xe)->mem.vram.usable_size) + if (xe->mem.vram) config->info[DRM_XE_QUERY_CONFIG_FLAGS] |= DRM_XE_QUERY_CONFIG_FLAG_HAS_VRAM; if (xe->info.has_usm && IS_ENABLED(CONFIG_DRM_XE_GPUSVM)) @@ -410,7 +411,7 @@ static int query_gt_list(struct xe_device *xe, struct drm_xe_device_query *query gt_list->gt_list[iter].near_mem_regions = 0x1; else gt_list->gt_list[iter].near_mem_regions = - BIT(gt_to_tile(gt)->id) << 1; + BIT(gt_to_tile(gt)->mem.vram->id) << 1; gt_list->gt_list[iter].far_mem_regions = xe->info.mem_region_mask ^ gt_list->gt_list[iter].near_mem_regions; @@ -476,7 +477,7 @@ static size_t calc_topo_query_size(struct xe_device *xe) sizeof_field(struct xe_gt, fuse_topo.eu_mask_per_dss); /* L3bank mask may not be available for some GTs */ - if (!XE_WA(gt, no_media_l3)) + if (!XE_GT_WA(gt, no_media_l3)) query_size += sizeof(struct drm_xe_query_topology_mask) + sizeof_field(struct xe_gt, fuse_topo.l3_bank_mask); } @@ -539,7 +540,7 @@ static int query_gt_topology(struct xe_device *xe, * mask, then it's better to omit L3 from the query rather than * reporting bogus or zeroed information to userspace. */ - if (!XE_WA(gt, no_media_l3)) { + if (!XE_GT_WA(gt, no_media_l3)) { topo.type = DRM_XE_TOPO_L3_BANK; err = copy_mask(&query_ptr, &topo, gt->fuse_topo.l3_bank_mask, sizeof(gt->fuse_topo.l3_bank_mask)); @@ -748,10 +749,8 @@ static int query_eu_stall(struct xe_device *xe, u32 num_rates; int ret; - if (!xe_eu_stall_supported_on_platform(xe)) { - drm_dbg(&xe->drm, "EU stall monitoring is not supported on this platform\n"); + if (!xe_eu_stall_supported_on_platform(xe)) return -ENODEV; - } array_size = xe_eu_stall_get_sampling_rates(&num_rates, &rates); size = sizeof(struct drm_xe_query_eu_stall) + array_size; diff --git a/drivers/gpu/drm/xe/xe_res_cursor.h b/drivers/gpu/drm/xe/xe_res_cursor.h index d1a403cfb628..4e00008b7081 100644 --- a/drivers/gpu/drm/xe/xe_res_cursor.h +++ b/drivers/gpu/drm/xe/xe_res_cursor.h @@ -55,8 +55,8 @@ struct xe_res_cursor { u32 mem_type; /** @sgl: Scatterlist for cursor */ struct scatterlist *sgl; - /** @dma_addr: Current element in a struct drm_pagemap_device_addr array */ - const struct drm_pagemap_device_addr *dma_addr; + /** @dma_addr: Current element in a struct drm_pagemap_addr array */ + const struct drm_pagemap_addr *dma_addr; /** @mm: Buddy allocator for VRAM cursor */ struct drm_buddy *mm; /** @@ -170,7 +170,7 @@ static inline void __xe_res_sg_next(struct xe_res_cursor *cur) */ static inline void __xe_res_dma_next(struct xe_res_cursor *cur) { - const struct drm_pagemap_device_addr *addr = cur->dma_addr; + const struct drm_pagemap_addr *addr = cur->dma_addr; u64 start = cur->start; while (start >= cur->dma_seg_size) { @@ -222,14 +222,14 @@ static inline void xe_res_first_sg(const struct sg_table *sg, /** * xe_res_first_dma - initialize a xe_res_cursor with dma_addr array * - * @dma_addr: struct drm_pagemap_device_addr array to walk + * @dma_addr: struct drm_pagemap_addr array to walk * @start: Start of the range * @size: Size of the range * @cur: cursor object to initialize * * Start walking over the range of allocations between @start and @size. */ -static inline void xe_res_first_dma(const struct drm_pagemap_device_addr *dma_addr, +static inline void xe_res_first_dma(const struct drm_pagemap_addr *dma_addr, u64 start, u64 size, struct xe_res_cursor *cur) { diff --git a/drivers/gpu/drm/xe/xe_ring_ops.c b/drivers/gpu/drm/xe/xe_ring_ops.c index 7b50c7c1ee21..d71837773d6c 100644 --- a/drivers/gpu/drm/xe/xe_ring_ops.c +++ b/drivers/gpu/drm/xe/xe_ring_ops.c @@ -110,10 +110,10 @@ static int emit_bb_start(u64 batch_addr, u32 ppgtt_flag, u32 *dw, int i) return i; } -static int emit_flush_invalidate(u32 addr, u32 val, u32 *dw, int i) +static int emit_flush_invalidate(u32 addr, u32 val, u32 flush_flags, u32 *dw, int i) { - dw[i++] = MI_FLUSH_DW | MI_INVALIDATE_TLB | MI_FLUSH_DW_OP_STOREDW | - MI_FLUSH_IMM_DW; + dw[i++] = MI_FLUSH_DW | MI_FLUSH_DW_OP_STOREDW | + MI_FLUSH_IMM_DW | (flush_flags & MI_INVALIDATE_TLB) ?: 0; dw[i++] = addr | MI_FLUSH_DW_USE_GTT; dw[i++] = 0; @@ -179,7 +179,7 @@ static int emit_render_cache_flush(struct xe_sched_job *job, u32 *dw, int i) bool lacks_render = !(gt->info.engine_mask & XE_HW_ENGINE_RCS_MASK); u32 flags; - if (XE_WA(gt, 14016712196)) + if (XE_GT_WA(gt, 14016712196)) i = emit_pipe_control(dw, i, 0, PIPE_CONTROL_DEPTH_CACHE_FLUSH, LRC_PPHWSP_FLUSH_INVAL_SCRATCH_ADDR, 0); @@ -190,7 +190,7 @@ static int emit_render_cache_flush(struct xe_sched_job *job, u32 *dw, int i) PIPE_CONTROL_DC_FLUSH_ENABLE | PIPE_CONTROL_FLUSH_ENABLE); - if (XE_WA(gt, 1409600907)) + if (XE_GT_WA(gt, 1409600907)) flags |= PIPE_CONTROL_DEPTH_STALL; if (lacks_render) @@ -206,7 +206,7 @@ static int emit_pipe_control_to_ring_end(struct xe_hw_engine *hwe, u32 *dw, int if (hwe->class != XE_ENGINE_CLASS_RENDER) return i; - if (XE_WA(hwe->gt, 16020292621)) + if (XE_GT_WA(hwe->gt, 16020292621)) i = emit_pipe_control(dw, i, 0, PIPE_CONTROL_LRI_POST_SYNC, RING_NOPID(hwe->mmio_base).addr, 0); @@ -410,16 +410,14 @@ static void emit_migration_job_gen12(struct xe_sched_job *job, i = emit_bb_start(job->ptrs[0].batch_addr, BIT(8), dw, i); dw[i++] = preparser_disable(true); - i = emit_flush_invalidate(saddr, seqno, dw, i); + i = emit_flush_invalidate(saddr, seqno, job->migrate_flush_flags, dw, i); dw[i++] = preparser_disable(false); i = emit_bb_start(job->ptrs[1].batch_addr, BIT(8), dw, i); - dw[i++] = MI_FLUSH_DW | MI_INVALIDATE_TLB | job->migrate_flush_flags | - MI_FLUSH_DW_OP_STOREDW | MI_FLUSH_IMM_DW; - dw[i++] = xe_lrc_seqno_ggtt_addr(lrc) | MI_FLUSH_DW_USE_GTT; - dw[i++] = 0; - dw[i++] = seqno; /* value */ + i = emit_flush_imm_ggtt(xe_lrc_seqno_ggtt_addr(lrc), seqno, + job->migrate_flush_flags, + dw, i); i = emit_user_interrupt(dw, i); diff --git a/drivers/gpu/drm/xe/xe_rtp.c b/drivers/gpu/drm/xe/xe_rtp.c index 95571b87aa73..47ea1521dc80 100644 --- a/drivers/gpu/drm/xe/xe_rtp.c +++ b/drivers/gpu/drm/xe/xe_rtp.c @@ -9,6 +9,7 @@ #include <uapi/drm/xe_drm.h> +#include "xe_configfs.h" #include "xe_gt.h" #include "xe_gt_topology.h" #include "xe_macros.h" @@ -363,3 +364,9 @@ bool xe_rtp_match_not_sriov_vf(const struct xe_gt *gt, { return !IS_SRIOV_VF(gt_to_xe(gt)); } + +bool xe_rtp_match_psmi_enabled(const struct xe_gt *gt, + const struct xe_hw_engine *hwe) +{ + return xe_configfs_get_psmi_enabled(to_pci_dev(gt_to_xe(gt)->drm.dev)); +} diff --git a/drivers/gpu/drm/xe/xe_rtp.h b/drivers/gpu/drm/xe/xe_rtp.h index 5ed6c14b9ae3..7951fefdbe04 100644 --- a/drivers/gpu/drm/xe/xe_rtp.h +++ b/drivers/gpu/drm/xe/xe_rtp.h @@ -477,4 +477,7 @@ bool xe_rtp_match_first_render_or_compute(const struct xe_gt *gt, bool xe_rtp_match_not_sriov_vf(const struct xe_gt *gt, const struct xe_hw_engine *hwe); +bool xe_rtp_match_psmi_enabled(const struct xe_gt *gt, + const struct xe_hw_engine *hwe); + #endif diff --git a/drivers/gpu/drm/xe/xe_sa.c b/drivers/gpu/drm/xe/xe_sa.c index 1d43e183ca21..fedd017d6dd3 100644 --- a/drivers/gpu/drm/xe/xe_sa.c +++ b/drivers/gpu/drm/xe/xe_sa.c @@ -69,7 +69,6 @@ struct xe_sa_manager *__xe_sa_bo_manager_init(struct xe_tile *tile, u32 size, u3 } sa_manager->bo = bo; sa_manager->is_iomem = bo->vmap.is_iomem; - sa_manager->gpu_addr = xe_bo_ggtt_addr(bo); if (bo->vmap.is_iomem) { sa_manager->cpu_ptr = kvzalloc(managed_size, GFP_KERNEL); diff --git a/drivers/gpu/drm/xe/xe_sa.h b/drivers/gpu/drm/xe/xe_sa.h index 1170ee5a81a8..99dbf0eea540 100644 --- a/drivers/gpu/drm/xe/xe_sa.h +++ b/drivers/gpu/drm/xe/xe_sa.h @@ -7,6 +7,8 @@ #include <linux/sizes.h> #include <linux/types.h> + +#include "xe_bo.h" #include "xe_sa_types.h" struct dma_fence; @@ -43,9 +45,20 @@ to_xe_sa_manager(struct drm_suballoc_manager *mng) return container_of(mng, struct xe_sa_manager, base); } +/** + * xe_sa_manager_gpu_addr - Retrieve GPU address of a back storage BO + * within suballocator. + * @sa_manager: the &xe_sa_manager struct instance + * Return: GGTT address of the back storage BO. + */ +static inline u64 xe_sa_manager_gpu_addr(struct xe_sa_manager *sa_manager) +{ + return xe_bo_ggtt_addr(sa_manager->bo); +} + static inline u64 xe_sa_bo_gpu_addr(struct drm_suballoc *sa) { - return to_xe_sa_manager(sa->manager)->gpu_addr + + return xe_sa_manager_gpu_addr(to_xe_sa_manager(sa->manager)) + drm_suballoc_soffset(sa); } diff --git a/drivers/gpu/drm/xe/xe_sa_types.h b/drivers/gpu/drm/xe/xe_sa_types.h index 2b070ff1292e..cb7238799dcb 100644 --- a/drivers/gpu/drm/xe/xe_sa_types.h +++ b/drivers/gpu/drm/xe/xe_sa_types.h @@ -12,7 +12,6 @@ struct xe_bo; struct xe_sa_manager { struct drm_suballoc_manager base; struct xe_bo *bo; - u64 gpu_addr; void *cpu_ptr; bool is_iomem; }; diff --git a/drivers/gpu/drm/xe/xe_sriov.c b/drivers/gpu/drm/xe/xe_sriov.c index a0eab44c0e76..87911fb4eea7 100644 --- a/drivers/gpu/drm/xe/xe_sriov.c +++ b/drivers/gpu/drm/xe/xe_sriov.c @@ -15,6 +15,7 @@ #include "xe_sriov.h" #include "xe_sriov_pf.h" #include "xe_sriov_vf.h" +#include "xe_sriov_vf_ccs.h" /** * xe_sriov_mode_to_string - Convert enum value to string. @@ -157,3 +158,21 @@ const char *xe_sriov_function_name(unsigned int n, char *buf, size_t size) strscpy(buf, "PF", size); return buf; } + +/** + * xe_sriov_late_init() - SR-IOV late initialization functions. + * @xe: the &xe_device to initialize + * + * On VF this function will initialize code for CCS migration. + * + * Return: 0 on success or a negative error code on failure. + */ +int xe_sriov_late_init(struct xe_device *xe) +{ + int err = 0; + + if (IS_VF_CCS_INIT_NEEDED(xe)) + err = xe_sriov_vf_ccs_init(xe); + + return err; +} diff --git a/drivers/gpu/drm/xe/xe_sriov.h b/drivers/gpu/drm/xe/xe_sriov.h index 688fbabf08f1..0e0c1abf2d14 100644 --- a/drivers/gpu/drm/xe/xe_sriov.h +++ b/drivers/gpu/drm/xe/xe_sriov.h @@ -18,6 +18,7 @@ const char *xe_sriov_function_name(unsigned int n, char *buf, size_t len); void xe_sriov_probe_early(struct xe_device *xe); void xe_sriov_print_info(struct xe_device *xe, struct drm_printer *p); int xe_sriov_init(struct xe_device *xe); +int xe_sriov_late_init(struct xe_device *xe); static inline enum xe_sriov_mode xe_device_sriov_mode(const struct xe_device *xe) { diff --git a/drivers/gpu/drm/xe/xe_sriov_vf.c b/drivers/gpu/drm/xe/xe_sriov_vf.c index 26e243c28994..5de81f213d83 100644 --- a/drivers/gpu/drm/xe/xe_sriov_vf.c +++ b/drivers/gpu/drm/xe/xe_sriov_vf.c @@ -11,6 +11,9 @@ #include "xe_gt_sriov_printk.h" #include "xe_gt_sriov_vf.h" #include "xe_guc_ct.h" +#include "xe_guc_submit.h" +#include "xe_irq.h" +#include "xe_lrc.h" #include "xe_pm.h" #include "xe_sriov.h" #include "xe_sriov_printk.h" @@ -147,6 +150,56 @@ void xe_sriov_vf_init_early(struct xe_device *xe) xe_sriov_info(xe, "migration not supported by this module version\n"); } +/** + * vf_post_migration_shutdown - Stop the driver activities after VF migration. + * @xe: the &xe_device struct instance + * + * After this VM is migrated and assigned to a new VF, it is running on a new + * hardware, and therefore many hardware-dependent states and related structures + * require fixups. Without fixups, the hardware cannot do any work, and therefore + * all GPU pipelines are stalled. + * Stop some of kernel activities to make the fixup process faster. + */ +static void vf_post_migration_shutdown(struct xe_device *xe) +{ + struct xe_gt *gt; + unsigned int id; + int ret = 0; + + for_each_gt(gt, xe, id) { + xe_guc_submit_pause(>->uc.guc); + ret |= xe_guc_submit_reset_block(>->uc.guc); + } + + if (ret) + drm_info(&xe->drm, "migration recovery encountered ongoing reset\n"); +} + +/** + * vf_post_migration_kickstart - Re-start the driver activities under new hardware. + * @xe: the &xe_device struct instance + * + * After we have finished with all post-migration fixups, restart the driver + * activities to continue feeding the GPU with workloads. + */ +static void vf_post_migration_kickstart(struct xe_device *xe) +{ + struct xe_gt *gt; + unsigned int id; + + /* + * Make sure interrupts on the new HW are properly set. The GuC IRQ + * must be working at this point, since the recovery did started, + * but the rest was not enabled using the procedure from spec. + */ + xe_irq_resume(xe); + + for_each_gt(gt, xe, id) { + xe_guc_submit_reset_unblock(>->uc.guc); + xe_guc_submit_unpause(>->uc.guc); + } +} + static bool gt_vf_post_migration_needed(struct xe_gt *gt) { return test_bit(gt->info.id, >_to_xe(gt)->sriov.vf.migration.gt_flags); @@ -192,6 +245,11 @@ static int vf_get_next_migrated_gt_id(struct xe_device *xe) return -1; } +static size_t post_migration_scratch_size(struct xe_device *xe) +{ + return max(xe_lrc_reg_size(xe), LRC_WA_BB_SIZE); +} + /** * Perform post-migration fixups on a single GT. * @@ -208,19 +266,31 @@ static int vf_get_next_migrated_gt_id(struct xe_device *xe) static int gt_vf_post_migration_fixups(struct xe_gt *gt) { s64 shift; + void *buf; int err; + buf = kmalloc(post_migration_scratch_size(gt_to_xe(gt)), GFP_KERNEL); + if (!buf) + return -ENOMEM; + err = xe_gt_sriov_vf_query_config(gt); if (err) - return err; + goto out; shift = xe_gt_sriov_vf_ggtt_shift(gt); if (shift) { xe_tile_sriov_vf_fixup_ggtt_nodes(gt_to_tile(gt), shift); - /* FIXME: add the recovery steps */ + xe_gt_sriov_vf_default_lrcs_hwsp_rebase(gt); + err = xe_guc_contexts_hwsp_rebase(>->uc.guc, buf); + if (err) + goto out; + xe_guc_jobs_ring_rebase(>->uc.guc); xe_guc_ct_fixup_messages_with_ggtt(>->uc.guc.ct, shift); } - return 0; + +out: + kfree(buf); + return err; } static void vf_post_migration_recovery(struct xe_device *xe) @@ -230,6 +300,7 @@ static void vf_post_migration_recovery(struct xe_device *xe) drm_dbg(&xe->drm, "migration recovery in progress\n"); xe_pm_runtime_get(xe); + vf_post_migration_shutdown(xe); if (!vf_migration_supported(xe)) { xe_sriov_err(xe, "migration not supported by this module version\n"); @@ -247,6 +318,7 @@ static void vf_post_migration_recovery(struct xe_device *xe) set_bit(id, &fixed_gts); } + vf_post_migration_kickstart(xe); err = vf_post_migration_notify_resfix_done(xe, fixed_gts); if (err) goto fail; diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c new file mode 100644 index 000000000000..4872e43eb440 --- /dev/null +++ b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c @@ -0,0 +1,377 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright © 2025 Intel Corporation + */ + +#include "instructions/xe_mi_commands.h" +#include "instructions/xe_gpu_commands.h" +#include "xe_bb.h" +#include "xe_bo.h" +#include "xe_device.h" +#include "xe_exec_queue.h" +#include "xe_exec_queue_types.h" +#include "xe_guc_submit.h" +#include "xe_lrc.h" +#include "xe_migrate.h" +#include "xe_sa.h" +#include "xe_sriov_printk.h" +#include "xe_sriov_vf_ccs.h" +#include "xe_sriov_vf_ccs_types.h" + +/** + * DOC: VF save/restore of compression Meta Data + * + * VF KMD registers two special contexts/LRCAs. + * + * Save Context/LRCA: contain necessary cmds+page table to trigger Meta data / + * compression control surface (Aka CCS) save in regular System memory in VM. + * + * Restore Context/LRCA: contain necessary cmds+page table to trigger Meta data / + * compression control surface (Aka CCS) Restore from regular System memory in + * VM to corresponding CCS pool. + * + * Below diagram explain steps needed for VF save/Restore of compression Meta Data:: + * + * CCS Save CCS Restore VF KMD Guc BCS + * LRCA LRCA + * | | | | | + * | | | | | + * | Create Save LRCA | | | + * [ ]<----------------------------- [ ] | | + * | | | | | + * | | | | | + * | | | Register save LRCA | | + * | | | with Guc | | + * | | [ ]--------------------------->[ ] | + * | | | | | + * | | Create restore LRCA | | | + * | [ ]<------------------[ ] | | + * | | | | | + * | | | Register restore LRCA | | + * | | | with Guc | | + * | | [ ]--------------------------->[ ] | + * | | | | | + * | | | | | + * | | [ ]------------------------- | | + * | | [ ] Allocate main memory. | | | + * | | [ ] Allocate CCS memory. | | | + * | | [ ] Update Main memory & | | | + * [ ]<------------------------------[ ] CCS pages PPGTT + BB | | | + * | [ ]<------------------[ ] cmds to save & restore.| | | + * | | [ ]<------------------------ | | + * | | | | | + * | | | | | + * | | | | | + * : : : : : + * ---------------------------- VF Paused ------------------------------------- + * | | | | | + * | | | | | + * | | | |Schedule | + * | | | |CCS Save | + * | | | | LRCA | + * | | | [ ]------>[ ] + * | | | | | + * | | | | | + * | | | |CCS save | + * | | | |completed| + * | | | [ ]<------[ ] + * | | | | | + * : : : : : + * ---------------------------- VM Migrated ----------------------------------- + * | | | | | + * | | | | | + * : : : : : + * ---------------------------- VF Resumed ------------------------------------ + * | | | | | + * | | | | | + * | | [ ]-------------- | | + * | | [ ] Fix up GGTT | | | + * | | [ ]<------------- | | + * | | | | | + * | | | | | + * | | | Notify VF_RESFIX_DONE | | + * | | [ ]--------------------------->[ ] | + * | | | | | + * | | | |Schedule | + * | | | |CCS | + * | | | |Restore | + * | | | |LRCA | + * | | | [ ]------>[ ] + * | | | | | + * | | | | | + * | | | |CCS | + * | | | |restore | + * | | | |completed| + * | | | [ ]<------[ ] + * | | | | | + * | | | | | + * | | | VF_RESFIX_DONE complete | | + * | | | notification | | + * | | [ ]<---------------------------[ ] | + * | | | | | + * | | | | | + * : : : : : + * ------------------------- Continue VM restore ------------------------------ + */ + +static u64 get_ccs_bb_pool_size(struct xe_device *xe) +{ + u64 sys_mem_size, ccs_mem_size, ptes, bb_pool_size; + struct sysinfo si; + + si_meminfo(&si); + sys_mem_size = si.totalram * si.mem_unit; + ccs_mem_size = div64_u64(sys_mem_size, NUM_BYTES_PER_CCS_BYTE(xe)); + ptes = DIV_ROUND_UP_ULL(sys_mem_size + ccs_mem_size, XE_PAGE_SIZE); + + /** + * We need below BB size to hold PTE mappings and some DWs for copy + * command. In reality, we need space for many copy commands. So, let + * us allocate double the calculated size which is enough to holds GPU + * instructions for the whole region. + */ + bb_pool_size = ptes * sizeof(u32); + + return round_up(bb_pool_size * 2, SZ_1M); +} + +static int alloc_bb_pool(struct xe_tile *tile, struct xe_tile_vf_ccs *ctx) +{ + struct xe_device *xe = tile_to_xe(tile); + struct xe_sa_manager *sa_manager; + u64 bb_pool_size; + int offset, err; + + bb_pool_size = get_ccs_bb_pool_size(xe); + xe_sriov_info(xe, "Allocating %s CCS BB pool size = %lldMB\n", + ctx->ctx_id ? "Restore" : "Save", bb_pool_size / SZ_1M); + + sa_manager = xe_sa_bo_manager_init(tile, bb_pool_size, SZ_16); + + if (IS_ERR(sa_manager)) { + xe_sriov_err(xe, "Suballocator init failed with error: %pe\n", + sa_manager); + err = PTR_ERR(sa_manager); + return err; + } + + offset = 0; + xe_map_memset(xe, &sa_manager->bo->vmap, offset, MI_NOOP, + bb_pool_size); + + offset = bb_pool_size - sizeof(u32); + xe_map_wr(xe, &sa_manager->bo->vmap, offset, u32, MI_BATCH_BUFFER_END); + + ctx->mem.ccs_bb_pool = sa_manager; + + return 0; +} + +static void ccs_rw_update_ring(struct xe_tile_vf_ccs *ctx) +{ + u64 addr = xe_sa_manager_gpu_addr(ctx->mem.ccs_bb_pool); + struct xe_lrc *lrc = xe_exec_queue_lrc(ctx->mig_q); + u32 dw[10], i = 0; + + dw[i++] = MI_ARB_ON_OFF | MI_ARB_ENABLE; + dw[i++] = MI_BATCH_BUFFER_START | XE_INSTR_NUM_DW(3); + dw[i++] = lower_32_bits(addr); + dw[i++] = upper_32_bits(addr); + dw[i++] = MI_NOOP; + dw[i++] = MI_NOOP; + + xe_lrc_write_ring(lrc, dw, i * sizeof(u32)); + xe_lrc_set_ring_tail(lrc, lrc->ring.tail); +} + +static int register_save_restore_context(struct xe_tile_vf_ccs *ctx) +{ + int err = -EINVAL; + int ctx_type; + + switch (ctx->ctx_id) { + case XE_SRIOV_VF_CCS_READ_CTX: + ctx_type = GUC_CONTEXT_COMPRESSION_SAVE; + break; + case XE_SRIOV_VF_CCS_WRITE_CTX: + ctx_type = GUC_CONTEXT_COMPRESSION_RESTORE; + break; + default: + return err; + } + + xe_guc_register_exec_queue(ctx->mig_q, ctx_type); + return 0; +} + +/** + * xe_sriov_vf_ccs_register_context - Register read/write contexts with guc. + * @xe: the &xe_device to register contexts on. + * + * This function registers read and write contexts with Guc. Re-registration + * is needed whenever resuming from pm runtime suspend. + * + * Return: 0 on success. Negative error code on failure. + */ +int xe_sriov_vf_ccs_register_context(struct xe_device *xe) +{ + struct xe_tile *tile = xe_device_get_root_tile(xe); + enum xe_sriov_vf_ccs_rw_ctxs ctx_id; + struct xe_tile_vf_ccs *ctx; + int err; + + if (!IS_VF_CCS_READY(xe)) + return 0; + + for_each_ccs_rw_ctx(ctx_id) { + ctx = &tile->sriov.vf.ccs[ctx_id]; + err = register_save_restore_context(ctx); + if (err) + return err; + } + + return err; +} + +static void xe_sriov_vf_ccs_fini(void *arg) +{ + struct xe_tile_vf_ccs *ctx = arg; + struct xe_lrc *lrc = xe_exec_queue_lrc(ctx->mig_q); + + /* + * Make TAIL = HEAD in the ring so that no issues are seen if Guc + * submits this context to HW on VF pause after unbinding device. + */ + xe_lrc_set_ring_tail(lrc, xe_lrc_ring_head(lrc)); + xe_exec_queue_put(ctx->mig_q); +} + +/** + * xe_sriov_vf_ccs_init - Setup LRCA for save & restore. + * @xe: the &xe_device to start recovery on + * + * This function shall be called only by VF. It initializes + * LRCA and suballocator needed for CCS save & restore. + * + * Return: 0 on success. Negative error code on failure. + */ +int xe_sriov_vf_ccs_init(struct xe_device *xe) +{ + struct xe_tile *tile = xe_device_get_root_tile(xe); + enum xe_sriov_vf_ccs_rw_ctxs ctx_id; + struct xe_tile_vf_ccs *ctx; + struct xe_exec_queue *q; + u32 flags; + int err; + + xe_assert(xe, IS_SRIOV_VF(xe)); + xe_assert(xe, !IS_DGFX(xe)); + xe_assert(xe, xe_device_has_flat_ccs(xe)); + + for_each_ccs_rw_ctx(ctx_id) { + ctx = &tile->sriov.vf.ccs[ctx_id]; + ctx->ctx_id = ctx_id; + + flags = EXEC_QUEUE_FLAG_KERNEL | + EXEC_QUEUE_FLAG_PERMANENT | + EXEC_QUEUE_FLAG_MIGRATE; + q = xe_exec_queue_create_bind(xe, tile, flags, 0); + if (IS_ERR(q)) { + err = PTR_ERR(q); + goto err_ret; + } + ctx->mig_q = q; + + err = alloc_bb_pool(tile, ctx); + if (err) + goto err_free_queue; + + ccs_rw_update_ring(ctx); + + err = register_save_restore_context(ctx); + if (err) + goto err_free_queue; + + err = devm_add_action_or_reset(xe->drm.dev, + xe_sriov_vf_ccs_fini, + ctx); + if (err) + goto err_ret; + } + + xe->sriov.vf.ccs.initialized = 1; + + return 0; + +err_free_queue: + xe_exec_queue_put(q); + +err_ret: + return err; +} + +/** + * xe_sriov_vf_ccs_attach_bo - Insert CCS read write commands in the BO. + * @bo: the &buffer object to which batch buffer commands will be added. + * + * This function shall be called only by VF. It inserts the PTEs and copy + * command instructions in the BO by calling xe_migrate_ccs_rw_copy() + * function. + * + * Returns: 0 if successful, negative error code on failure. + */ +int xe_sriov_vf_ccs_attach_bo(struct xe_bo *bo) +{ + struct xe_device *xe = xe_bo_device(bo); + enum xe_sriov_vf_ccs_rw_ctxs ctx_id; + struct xe_tile_vf_ccs *ctx; + struct xe_tile *tile; + struct xe_bb *bb; + int err = 0; + + if (!IS_VF_CCS_READY(xe)) + return 0; + + tile = xe_device_get_root_tile(xe); + + for_each_ccs_rw_ctx(ctx_id) { + bb = bo->bb_ccs[ctx_id]; + /* bb should be NULL here. Assert if not NULL */ + xe_assert(xe, !bb); + + ctx = &tile->sriov.vf.ccs[ctx_id]; + err = xe_migrate_ccs_rw_copy(tile, ctx->mig_q, bo, ctx_id); + } + return err; +} + +/** + * xe_sriov_vf_ccs_detach_bo - Remove CCS read write commands from the BO. + * @bo: the &buffer object from which batch buffer commands will be removed. + * + * This function shall be called only by VF. It removes the PTEs and copy + * command instructions from the BO. Make sure to update the BB with MI_NOOP + * before freeing. + * + * Returns: 0 if successful. + */ +int xe_sriov_vf_ccs_detach_bo(struct xe_bo *bo) +{ + struct xe_device *xe = xe_bo_device(bo); + enum xe_sriov_vf_ccs_rw_ctxs ctx_id; + struct xe_bb *bb; + + if (!IS_VF_CCS_READY(xe)) + return 0; + + for_each_ccs_rw_ctx(ctx_id) { + bb = bo->bb_ccs[ctx_id]; + if (!bb) + continue; + + memset(bb->cs, MI_NOOP, bb->len * sizeof(u32)); + xe_bb_free(bb, NULL); + bo->bb_ccs[ctx_id] = NULL; + } + return 0; +} diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.h b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.h new file mode 100644 index 000000000000..1f1baf685fec --- /dev/null +++ b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.h @@ -0,0 +1,17 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2025 Intel Corporation + */ + +#ifndef _XE_SRIOV_VF_CCS_H_ +#define _XE_SRIOV_VF_CCS_H_ + +struct xe_device; +struct xe_bo; + +int xe_sriov_vf_ccs_init(struct xe_device *xe); +int xe_sriov_vf_ccs_attach_bo(struct xe_bo *bo); +int xe_sriov_vf_ccs_detach_bo(struct xe_bo *bo); +int xe_sriov_vf_ccs_register_context(struct xe_device *xe); + +#endif diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_ccs_types.h b/drivers/gpu/drm/xe/xe_sriov_vf_ccs_types.h new file mode 100644 index 000000000000..93435a6f4cb6 --- /dev/null +++ b/drivers/gpu/drm/xe/xe_sriov_vf_ccs_types.h @@ -0,0 +1,53 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2025 Intel Corporation + */ + +#ifndef _XE_SRIOV_VF_CCS_TYPES_H_ +#define _XE_SRIOV_VF_CCS_TYPES_H_ + +#define for_each_ccs_rw_ctx(id__) \ + for ((id__) = 0; (id__) < XE_SRIOV_VF_CCS_CTX_COUNT; (id__)++) + +#define IS_VF_CCS_READY(xe) ({ \ + struct xe_device *___xe = (xe); \ + xe_assert(___xe, IS_SRIOV_VF(___xe)); \ + ___xe->sriov.vf.ccs.initialized; \ + }) + +#define IS_VF_CCS_INIT_NEEDED(xe) ({\ + struct xe_device *___xe = (xe); \ + IS_SRIOV_VF(___xe) && !IS_DGFX(___xe) && \ + xe_device_has_flat_ccs(___xe) && GRAPHICS_VER(___xe) >= 20; \ + }) + +enum xe_sriov_vf_ccs_rw_ctxs { + XE_SRIOV_VF_CCS_READ_CTX, + XE_SRIOV_VF_CCS_WRITE_CTX, + XE_SRIOV_VF_CCS_CTX_COUNT +}; + +#define IS_VF_CCS_BB_VALID(xe, bo) ({ \ + struct xe_device *___xe = (xe); \ + struct xe_bo *___bo = (bo); \ + IS_SRIOV_VF(___xe) && \ + ___bo->bb_ccs[XE_SRIOV_VF_CCS_READ_CTX] && \ + ___bo->bb_ccs[XE_SRIOV_VF_CCS_WRITE_CTX]; \ + }) + +struct xe_migrate; +struct xe_sa_manager; + +struct xe_tile_vf_ccs { + /** @id: Id to which context it belongs to */ + enum xe_sriov_vf_ccs_rw_ctxs ctx_id; + /** @mig_q: exec queues used for migration */ + struct xe_exec_queue *mig_q; + + struct { + /** @ccs_bb_pool: Pool from which batch buffers are allocated. */ + struct xe_sa_manager *ccs_bb_pool; + } mem; +}; + +#endif diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_types.h b/drivers/gpu/drm/xe/xe_sriov_vf_types.h index 8300416a6226..24a873c50c49 100644 --- a/drivers/gpu/drm/xe/xe_sriov_vf_types.h +++ b/drivers/gpu/drm/xe/xe_sriov_vf_types.h @@ -36,6 +36,12 @@ struct xe_device_vf { /** @migration.gt_flags: Per-GT request flags for VF migration recovery */ unsigned long gt_flags; } migration; + + /** @ccs: VF CCS state data */ + struct { + /** @ccs.initialized: Initilalization of VF CCS is completed or not */ + bool initialized; + } ccs; }; #endif diff --git a/drivers/gpu/drm/xe/xe_survivability_mode.c b/drivers/gpu/drm/xe/xe_survivability_mode.c index 41705f5d52e3..53c5af4b810c 100644 --- a/drivers/gpu/drm/xe/xe_survivability_mode.c +++ b/drivers/gpu/drm/xe/xe_survivability_mode.c @@ -22,15 +22,18 @@ #define MAX_SCRATCH_MMIO 8 /** - * DOC: Xe Boot Survivability + * DOC: Survivability Mode * - * Boot Survivability is a software based workflow for recovering a system in a failed boot state + * Survivability Mode is a software based workflow for recovering a system in a failed boot state * Here system recoverability is concerned with recovering the firmware responsible for boot. * - * This is implemented by loading the driver with bare minimum (no drm card) to allow the firmware - * to be flashed through mei and collect telemetry. The driver's probe flow is modified - * such that it enters survivability mode when pcode initialization is incomplete and boot status - * denotes a failure. + * Boot Survivability + * =================== + * + * Boot Survivability is implemented by loading the driver with bare minimum (no drm card) to allow + * the firmware to be flashed through mei driver and collect telemetry. The driver's probe flow is + * modified such that it enters survivability mode when pcode initialization is incomplete and boot + * status denotes a failure. * * Survivability mode can also be entered manually using the survivability mode attribute available * through configfs which is beneficial in several usecases. It can be used to address scenarios @@ -46,7 +49,7 @@ * Survivability mode is indicated by the below admin-only readable sysfs which provides additional * debug information:: * - * /sys/bus/pci/devices/<device>/surivability_mode + * /sys/bus/pci/devices/<device>/survivability_mode * * Capability Information: * Provides boot status @@ -56,6 +59,22 @@ * Provides history of previous failures * Auxiliary Information * Certain failures may have information in addition to postcode information + * + * Runtime Survivability + * ===================== + * + * Certain runtime firmware errors can cause the device to enter a wedged state + * (:ref:`xe-device-wedging`) requiring a firmware flash to restore normal operation. + * Runtime Survivability Mode indicates that a firmware flash is necessary to recover the device and + * is indicated by the presence of survivability mode sysfs:: + * + * /sys/bus/pci/devices/<device>/survivability_mode + * + * Survivability mode sysfs provides information about the type of survivability mode. + * + * When such errors occur, userspace is notified with the drm device wedged uevent and runtime + * survivability mode. User can then initiate a firmware flash using userspace tools like fwupd + * to restore device to normal operation. */ static u32 aux_history_offset(u32 reg_value) @@ -121,6 +140,14 @@ static void log_survivability_info(struct pci_dev *pdev) } } +static int check_boot_failure(struct xe_device *xe) +{ + struct xe_survivability *survivability = &xe->survivability; + + return survivability->boot_status == NON_CRITICAL_FAILURE || + survivability->boot_status == CRITICAL_FAILURE; +} + static ssize_t survivability_mode_show(struct device *dev, struct device_attribute *attr, char *buff) { @@ -130,6 +157,12 @@ static ssize_t survivability_mode_show(struct device *dev, struct xe_survivability_info *info = survivability->info; int index = 0, count = 0; + count += sysfs_emit_at(buff, count, "Survivability mode type: %s\n", + survivability->type ? "Runtime" : "Boot"); + + if (!check_boot_failure(xe)) + return count; + for (index = 0; index < MAX_SCRATCH_MMIO; index++) { if (info[index].reg) count += sysfs_emit_at(buff, count, "%s: 0x%x - 0x%x\n", info[index].name, @@ -151,12 +184,11 @@ static void xe_survivability_mode_fini(void *arg) sysfs_remove_file(&dev->kobj, &dev_attr_survivability_mode.attr); } -static int enable_survivability_mode(struct pci_dev *pdev) +static int create_survivability_sysfs(struct pci_dev *pdev) { struct device *dev = &pdev->dev; struct xe_device *xe = pdev_to_xe_device(pdev); - struct xe_survivability *survivability = &xe->survivability; - int ret = 0; + int ret; /* create survivability mode sysfs */ ret = sysfs_create_file(&dev->kobj, &dev_attr_survivability_mode.attr); @@ -170,6 +202,20 @@ static int enable_survivability_mode(struct pci_dev *pdev) if (ret) return ret; + return 0; +} + +static int enable_boot_survivability_mode(struct pci_dev *pdev) +{ + struct device *dev = &pdev->dev; + struct xe_device *xe = pdev_to_xe_device(pdev); + struct xe_survivability *survivability = &xe->survivability; + int ret = 0; + + ret = create_survivability_sysfs(pdev); + if (ret) + return ret; + /* Make sure xe_heci_gsc_init() knows about survivability mode */ survivability->mode = true; @@ -192,15 +238,36 @@ err: return ret; } +static int init_survivability_mode(struct xe_device *xe) +{ + struct xe_survivability *survivability = &xe->survivability; + struct xe_survivability_info *info; + + survivability->size = MAX_SCRATCH_MMIO; + + info = devm_kcalloc(xe->drm.dev, survivability->size, sizeof(*info), + GFP_KERNEL); + if (!info) + return -ENOMEM; + + survivability->info = info; + + populate_survivability_info(xe); + + return 0; +} + /** - * xe_survivability_mode_is_enabled - check if survivability mode is enabled + * xe_survivability_mode_is_boot_enabled- check if boot survivability mode is enabled * @xe: xe device instance * - * Returns true if in survivability mode, false otherwise + * Returns true if in boot survivability mode of type, else false */ -bool xe_survivability_mode_is_enabled(struct xe_device *xe) +bool xe_survivability_mode_is_boot_enabled(struct xe_device *xe) { - return xe->survivability.mode; + struct xe_survivability *survivability = &xe->survivability; + + return survivability->mode && survivability->type == XE_SURVIVABILITY_TYPE_BOOT; } /** @@ -241,44 +308,78 @@ bool xe_survivability_mode_is_requested(struct xe_device *xe) data = xe_mmio_read32(mmio, PCODE_SCRATCH(0)); survivability->boot_status = REG_FIELD_GET(BOOT_STATUS, data); - return survivability->boot_status == NON_CRITICAL_FAILURE || - survivability->boot_status == CRITICAL_FAILURE; + return check_boot_failure(xe); } /** - * xe_survivability_mode_enable - Initialize and enable the survivability mode + * xe_survivability_mode_runtime_enable - Initialize and enable runtime survivability mode * @xe: xe device instance * - * Initialize survivability information and enable survivability mode + * Initialize survivability information and enable runtime survivability mode. + * Runtime survivability mode is enabled when certain errors cause the device to be + * in non-recoverable state. The device is declared wedged with the appropriate + * recovery method and survivability mode sysfs exposed to userspace * - * Return: 0 if survivability mode is enabled or not requested; negative error - * code otherwise. + * Return: 0 if runtime survivability mode is enabled, negative error code otherwise. */ -int xe_survivability_mode_enable(struct xe_device *xe) +int xe_survivability_mode_runtime_enable(struct xe_device *xe) { struct xe_survivability *survivability = &xe->survivability; - struct xe_survivability_info *info; struct pci_dev *pdev = to_pci_dev(xe->drm.dev); + int ret; - if (!xe_survivability_mode_is_requested(xe)) - return 0; + if (!IS_DGFX(xe) || IS_SRIOV_VF(xe) || xe->info.platform < XE_BATTLEMAGE) { + dev_err(&pdev->dev, "Runtime Survivability Mode not supported\n"); + return -EINVAL; + } - survivability->size = MAX_SCRATCH_MMIO; + ret = init_survivability_mode(xe); + if (ret) + return ret; - info = devm_kcalloc(xe->drm.dev, survivability->size, sizeof(*info), - GFP_KERNEL); - if (!info) - return -ENOMEM; + ret = create_survivability_sysfs(pdev); + if (ret) + dev_err(&pdev->dev, "Failed to create survivability mode sysfs\n"); - survivability->info = info; + survivability->type = XE_SURVIVABILITY_TYPE_RUNTIME; + dev_err(&pdev->dev, "Runtime Survivability mode enabled\n"); - populate_survivability_info(xe); + xe_device_set_wedged_method(xe, DRM_WEDGE_RECOVERY_VENDOR); + xe_device_declare_wedged(xe); + dev_err(&pdev->dev, "Firmware flash required, Please refer to the userspace documentation for more details!\n"); + + return 0; +} - /* Only log debug information and exit if it is a critical failure */ +/** + * xe_survivability_mode_boot_enable - Initialize and enable boot survivability mode + * @xe: xe device instance + * + * Initialize survivability information and enable boot survivability mode + * + * Return: 0 if boot survivability mode is enabled or not requested, negative error + * code otherwise. + */ +int xe_survivability_mode_boot_enable(struct xe_device *xe) +{ + struct xe_survivability *survivability = &xe->survivability; + struct pci_dev *pdev = to_pci_dev(xe->drm.dev); + int ret; + + if (!xe_survivability_mode_is_requested(xe)) + return 0; + + ret = init_survivability_mode(xe); + if (ret) + return ret; + + /* Log breadcrumbs but do not enter survivability mode for Critical boot errors */ if (survivability->boot_status == CRITICAL_FAILURE) { log_survivability_info(pdev); return -ENXIO; } - return enable_survivability_mode(pdev); + survivability->type = XE_SURVIVABILITY_TYPE_BOOT; + + return enable_boot_survivability_mode(pdev); } diff --git a/drivers/gpu/drm/xe/xe_survivability_mode.h b/drivers/gpu/drm/xe/xe_survivability_mode.h index 02231c2bf008..1cc94226aa82 100644 --- a/drivers/gpu/drm/xe/xe_survivability_mode.h +++ b/drivers/gpu/drm/xe/xe_survivability_mode.h @@ -10,8 +10,9 @@ struct xe_device; -int xe_survivability_mode_enable(struct xe_device *xe); -bool xe_survivability_mode_is_enabled(struct xe_device *xe); +int xe_survivability_mode_boot_enable(struct xe_device *xe); +int xe_survivability_mode_runtime_enable(struct xe_device *xe); +bool xe_survivability_mode_is_boot_enabled(struct xe_device *xe); bool xe_survivability_mode_is_requested(struct xe_device *xe); #endif /* _XE_SURVIVABILITY_MODE_H_ */ diff --git a/drivers/gpu/drm/xe/xe_survivability_mode_types.h b/drivers/gpu/drm/xe/xe_survivability_mode_types.h index 19d433e253df..cd65a5d167c9 100644 --- a/drivers/gpu/drm/xe/xe_survivability_mode_types.h +++ b/drivers/gpu/drm/xe/xe_survivability_mode_types.h @@ -9,6 +9,11 @@ #include <linux/limits.h> #include <linux/types.h> +enum xe_survivability_type { + XE_SURVIVABILITY_TYPE_BOOT, + XE_SURVIVABILITY_TYPE_RUNTIME, +}; + struct xe_survivability_info { char name[NAME_MAX]; u32 reg; @@ -30,6 +35,9 @@ struct xe_survivability { /** @mode: boolean to indicate survivability mode */ bool mode; + + /** @type: survivability type */ + enum xe_survivability_type type; }; #endif /* _XE_SURVIVABILITY_MODE_TYPES_H_ */ diff --git a/drivers/gpu/drm/xe/xe_svm.c b/drivers/gpu/drm/xe/xe_svm.c index a7ff5975873f..76c6d74c1208 100644 --- a/drivers/gpu/drm/xe/xe_svm.c +++ b/drivers/gpu/drm/xe/xe_svm.c @@ -7,7 +7,6 @@ #include "xe_bo.h" #include "xe_gt_stats.h" -#include "xe_gt_tlb_invalidation.h" #include "xe_migrate.h" #include "xe_module.h" #include "xe_pm.h" @@ -17,6 +16,7 @@ #include "xe_ttm_vram_mgr.h" #include "xe_vm.h" #include "xe_vm_types.h" +#include "xe_vram_types.h" static bool xe_svm_range_in_vram(struct xe_svm_range *range) { @@ -224,7 +224,7 @@ static void xe_svm_invalidate(struct drm_gpusvm *gpusvm, xe_device_wmb(xe); - err = xe_vm_range_tilemask_tlb_invalidation(vm, adj_start, adj_end, tile_mask); + err = xe_vm_range_tilemask_tlb_inval(vm, adj_start, adj_end, tile_mask); WARN_ON_ONCE(err); range_notifier_event_end: @@ -252,10 +252,56 @@ static int __xe_svm_garbage_collector(struct xe_vm *vm, return 0; } +static int xe_svm_range_set_default_attr(struct xe_vm *vm, u64 range_start, u64 range_end) +{ + struct xe_vma *vma; + struct xe_vma_mem_attr default_attr = { + .preferred_loc = { + .devmem_fd = DRM_XE_PREFERRED_LOC_DEFAULT_DEVICE, + .migration_policy = DRM_XE_MIGRATE_ALL_PAGES, + }, + .atomic_access = DRM_XE_ATOMIC_UNDEFINED, + }; + int err = 0; + + vma = xe_vm_find_vma_by_addr(vm, range_start); + if (!vma) + return -EINVAL; + + if (xe_vma_has_default_mem_attrs(vma)) + return 0; + + vm_dbg(&vm->xe->drm, "Existing VMA start=0x%016llx, vma_end=0x%016llx", + xe_vma_start(vma), xe_vma_end(vma)); + + if (xe_vma_start(vma) == range_start && xe_vma_end(vma) == range_end) { + default_attr.pat_index = vma->attr.default_pat_index; + default_attr.default_pat_index = vma->attr.default_pat_index; + vma->attr = default_attr; + } else { + vm_dbg(&vm->xe->drm, "Split VMA start=0x%016llx, vma_end=0x%016llx", + range_start, range_end); + err = xe_vm_alloc_cpu_addr_mirror_vma(vm, range_start, range_end - range_start); + if (err) { + drm_warn(&vm->xe->drm, "VMA SPLIT failed: %pe\n", ERR_PTR(err)); + xe_vm_kill(vm, true); + return err; + } + } + + /* + * On call from xe_svm_handle_pagefault original VMA might be changed + * signal this to lookup for VMA again. + */ + return -EAGAIN; +} + static int xe_svm_garbage_collector(struct xe_vm *vm) { struct xe_svm_range *range; - int err; + u64 range_start; + u64 range_end; + int err, ret = 0; lockdep_assert_held_write(&vm->lock); @@ -270,6 +316,9 @@ static int xe_svm_garbage_collector(struct xe_vm *vm) if (!range) break; + range_start = xe_svm_range_start(range); + range_end = xe_svm_range_end(range); + list_del(&range->garbage_collector_link); spin_unlock(&vm->svm.garbage_collector.lock); @@ -282,11 +331,19 @@ static int xe_svm_garbage_collector(struct xe_vm *vm) return err; } + err = xe_svm_range_set_default_attr(vm, range_start, range_end); + if (err) { + if (err == -EAGAIN) + ret = -EAGAIN; + else + return err; + } + spin_lock(&vm->svm.garbage_collector.lock); } spin_unlock(&vm->svm.garbage_collector.lock); - return 0; + return ret; } static void xe_svm_garbage_collector_work_func(struct work_struct *w) @@ -306,21 +363,15 @@ static struct xe_vram_region *page_to_vr(struct page *page) return container_of(page_pgmap(page), struct xe_vram_region, pagemap); } -static struct xe_tile *vr_to_tile(struct xe_vram_region *vr) -{ - return container_of(vr, struct xe_tile, mem.vram); -} - static u64 xe_vram_region_page_to_dpa(struct xe_vram_region *vr, struct page *page) { u64 dpa; - struct xe_tile *tile = vr_to_tile(vr); u64 pfn = page_to_pfn(page); u64 offset; - xe_tile_assert(tile, is_device_private_page(page)); - xe_tile_assert(tile, (pfn << PAGE_SHIFT) >= vr->hpa_base); + xe_assert(vr->xe, is_device_private_page(page)); + xe_assert(vr->xe, (pfn << PAGE_SHIFT) >= vr->hpa_base); offset = (pfn << PAGE_SHIFT) - vr->hpa_base; dpa = vr->dpa_base + offset; @@ -333,11 +384,12 @@ enum xe_svm_copy_dir { XE_SVM_COPY_TO_SRAM, }; -static int xe_svm_copy(struct page **pages, dma_addr_t *dma_addr, +static int xe_svm_copy(struct page **pages, + struct drm_pagemap_addr *pagemap_addr, unsigned long npages, const enum xe_svm_copy_dir dir) { struct xe_vram_region *vr = NULL; - struct xe_tile *tile; + struct xe_device *xe; struct dma_fence *fence = NULL; unsigned long i; #define XE_VRAM_ADDR_INVALID ~0x0ull @@ -365,12 +417,12 @@ static int xe_svm_copy(struct page **pages, dma_addr_t *dma_addr, last = (i + 1) == npages; /* No CPU page and no device pages queue'd to copy */ - if (!dma_addr[i] && vram_addr == XE_VRAM_ADDR_INVALID) + if (!pagemap_addr[i].addr && vram_addr == XE_VRAM_ADDR_INVALID) continue; if (!vr && spage) { vr = page_to_vr(spage); - tile = vr_to_tile(vr); + xe = vr->xe; } XE_WARN_ON(spage && page_to_vr(spage) != vr); @@ -379,7 +431,7 @@ static int xe_svm_copy(struct page **pages, dma_addr_t *dma_addr, * first device page, check if physical contiguous on subsequent * device pages. */ - if (dma_addr[i] && spage) { + if (pagemap_addr[i].addr && spage) { __vram_addr = xe_vram_region_page_to_dpa(vr, spage); if (vram_addr == XE_VRAM_ADDR_INVALID) { vram_addr = __vram_addr; @@ -387,6 +439,14 @@ static int xe_svm_copy(struct page **pages, dma_addr_t *dma_addr, } match = vram_addr + PAGE_SIZE * (i - pos) == __vram_addr; + /* Expected with contiguous memory */ + xe_assert(vr->xe, match); + + if (pagemap_addr[i].order) { + i += NR_PAGES(pagemap_addr[i].order) - 1; + chunk = (i - pos) == (XE_MIGRATE_CHUNK_SIZE / PAGE_SIZE); + last = (i + 1) == npages; + } } /* @@ -402,20 +462,22 @@ static int xe_svm_copy(struct page **pages, dma_addr_t *dma_addr, if (vram_addr != XE_VRAM_ADDR_INVALID) { if (sram) { - vm_dbg(&tile->xe->drm, + vm_dbg(&xe->drm, "COPY TO SRAM - 0x%016llx -> 0x%016llx, NPAGES=%ld", - vram_addr, (u64)dma_addr[pos], i - pos + incr); - __fence = xe_migrate_from_vram(tile->migrate, + vram_addr, + (u64)pagemap_addr[pos].addr, i - pos + incr); + __fence = xe_migrate_from_vram(vr->migrate, i - pos + incr, vram_addr, - dma_addr + pos); + &pagemap_addr[pos]); } else { - vm_dbg(&tile->xe->drm, + vm_dbg(&xe->drm, "COPY TO VRAM - 0x%016llx -> 0x%016llx, NPAGES=%ld", - (u64)dma_addr[pos], vram_addr, i - pos + incr); - __fence = xe_migrate_to_vram(tile->migrate, + (u64)pagemap_addr[pos].addr, vram_addr, + i - pos + incr); + __fence = xe_migrate_to_vram(vr->migrate, i - pos + incr, - dma_addr + pos, + &pagemap_addr[pos], vram_addr); } if (IS_ERR(__fence)) { @@ -428,7 +490,7 @@ static int xe_svm_copy(struct page **pages, dma_addr_t *dma_addr, } /* Setup physical address of next device page */ - if (dma_addr[i] && spage) { + if (pagemap_addr[i].addr && spage) { vram_addr = __vram_addr; pos = i; } else { @@ -438,18 +500,18 @@ static int xe_svm_copy(struct page **pages, dma_addr_t *dma_addr, /* Extra mismatched device page, copy it */ if (!match && last && vram_addr != XE_VRAM_ADDR_INVALID) { if (sram) { - vm_dbg(&tile->xe->drm, + vm_dbg(&xe->drm, "COPY TO SRAM - 0x%016llx -> 0x%016llx, NPAGES=%d", - vram_addr, (u64)dma_addr[pos], 1); - __fence = xe_migrate_from_vram(tile->migrate, 1, + vram_addr, (u64)pagemap_addr[pos].addr, 1); + __fence = xe_migrate_from_vram(vr->migrate, 1, vram_addr, - dma_addr + pos); + &pagemap_addr[pos]); } else { - vm_dbg(&tile->xe->drm, + vm_dbg(&xe->drm, "COPY TO VRAM - 0x%016llx -> 0x%016llx, NPAGES=%d", - (u64)dma_addr[pos], vram_addr, 1); - __fence = xe_migrate_to_vram(tile->migrate, 1, - dma_addr + pos, + (u64)pagemap_addr[pos].addr, vram_addr, 1); + __fence = xe_migrate_to_vram(vr->migrate, 1, + &pagemap_addr[pos], vram_addr); } if (IS_ERR(__fence)) { @@ -475,16 +537,18 @@ err_out: #undef XE_VRAM_ADDR_INVALID } -static int xe_svm_copy_to_devmem(struct page **pages, dma_addr_t *dma_addr, +static int xe_svm_copy_to_devmem(struct page **pages, + struct drm_pagemap_addr *pagemap_addr, unsigned long npages) { - return xe_svm_copy(pages, dma_addr, npages, XE_SVM_COPY_TO_VRAM); + return xe_svm_copy(pages, pagemap_addr, npages, XE_SVM_COPY_TO_VRAM); } -static int xe_svm_copy_to_ram(struct page **pages, dma_addr_t *dma_addr, +static int xe_svm_copy_to_ram(struct page **pages, + struct drm_pagemap_addr *pagemap_addr, unsigned long npages) { - return xe_svm_copy(pages, dma_addr, npages, XE_SVM_COPY_TO_SRAM); + return xe_svm_copy(pages, pagemap_addr, npages, XE_SVM_COPY_TO_SRAM); } static struct xe_bo *to_xe_bo(struct drm_pagemap_devmem *devmem_allocation) @@ -506,9 +570,9 @@ static u64 block_offset_to_pfn(struct xe_vram_region *vr, u64 offset) return PHYS_PFN(offset + vr->hpa_base); } -static struct drm_buddy *tile_to_buddy(struct xe_tile *tile) +static struct drm_buddy *vram_to_buddy(struct xe_vram_region *vram) { - return &tile->mem.vram.ttm.mm; + return &vram->ttm.mm; } static int xe_svm_populate_devmem_pfn(struct drm_pagemap_devmem *devmem_allocation, @@ -522,8 +586,7 @@ static int xe_svm_populate_devmem_pfn(struct drm_pagemap_devmem *devmem_allocati list_for_each_entry(block, blocks, link) { struct xe_vram_region *vr = block->private; - struct xe_tile *tile = vr_to_tile(vr); - struct drm_buddy *buddy = tile_to_buddy(tile); + struct drm_buddy *buddy = vram_to_buddy(vr); u64 block_pfn = block_offset_to_pfn(vr, drm_buddy_block_offset(block)); int i; @@ -683,20 +746,14 @@ u64 xe_svm_find_vma_start(struct xe_vm *vm, u64 start, u64 end, struct xe_vma *v } #if IS_ENABLED(CONFIG_DRM_XE_PAGEMAP) -static struct xe_vram_region *tile_to_vr(struct xe_tile *tile) -{ - return &tile->mem.vram; -} - static int xe_drm_pagemap_populate_mm(struct drm_pagemap *dpagemap, unsigned long start, unsigned long end, struct mm_struct *mm, unsigned long timeslice_ms) { - struct xe_tile *tile = container_of(dpagemap, typeof(*tile), mem.vram.dpagemap); - struct xe_device *xe = tile_to_xe(tile); + struct xe_vram_region *vr = container_of(dpagemap, typeof(*vr), dpagemap); + struct xe_device *xe = vr->xe; struct device *dev = xe->drm.dev; - struct xe_vram_region *vr = tile_to_vr(tile); struct drm_buddy_block *block; struct list_head *blocks; struct xe_bo *bo; @@ -709,9 +766,9 @@ static int xe_drm_pagemap_populate_mm(struct drm_pagemap *dpagemap, xe_pm_runtime_get(xe); retry: - bo = xe_bo_create_locked(tile_to_xe(tile), NULL, NULL, end - start, + bo = xe_bo_create_locked(vr->xe, NULL, NULL, end - start, ttm_bo_type_device, - XE_BO_FLAG_VRAM_IF_DGFX(tile) | + (IS_DGFX(xe) ? XE_BO_FLAG_VRAM(vr) : XE_BO_FLAG_SYSTEM) | XE_BO_FLAG_CPU_ADDR_MIRROR); if (IS_ERR(bo)) { err = PTR_ERR(bo); @@ -721,9 +778,7 @@ static int xe_drm_pagemap_populate_mm(struct drm_pagemap *dpagemap, } drm_pagemap_devmem_init(&bo->devmem_allocation, dev, mm, - &dpagemap_devmem_ops, - &tile->mem.vram.dpagemap, - end - start); + &dpagemap_devmem_ops, dpagemap, end - start); blocks = &to_xe_ttm_vram_mgr_resource(bo->ttm.resource)->blocks; list_for_each_entry(block, blocks, link) @@ -790,22 +845,9 @@ bool xe_svm_range_needs_migrate_to_vram(struct xe_svm_range *range, struct xe_vm return true; } -/** - * xe_svm_handle_pagefault() - SVM handle page fault - * @vm: The VM. - * @vma: The CPU address mirror VMA. - * @gt: The gt upon the fault occurred. - * @fault_addr: The GPU fault address. - * @atomic: The fault atomic access bit. - * - * Create GPU bindings for a SVM page fault. Optionally migrate to device - * memory. - * - * Return: 0 on success, negative error code on error. - */ -int xe_svm_handle_pagefault(struct xe_vm *vm, struct xe_vma *vma, - struct xe_gt *gt, u64 fault_addr, - bool atomic) +static int __xe_svm_handle_pagefault(struct xe_vm *vm, struct xe_vma *vma, + struct xe_gt *gt, u64 fault_addr, + bool need_vram) { struct drm_gpusvm_ctx ctx = { .read_only = xe_vma_read_only(vma), @@ -813,14 +855,14 @@ int xe_svm_handle_pagefault(struct xe_vm *vm, struct xe_vma *vma, IS_ENABLED(CONFIG_DRM_XE_PAGEMAP), .check_pages_threshold = IS_DGFX(vm->xe) && IS_ENABLED(CONFIG_DRM_XE_PAGEMAP) ? SZ_64K : 0, - .devmem_only = atomic && IS_DGFX(vm->xe) && - IS_ENABLED(CONFIG_DRM_XE_PAGEMAP), - .timeslice_ms = atomic && IS_DGFX(vm->xe) && + .devmem_only = need_vram && IS_ENABLED(CONFIG_DRM_XE_PAGEMAP), + .timeslice_ms = need_vram && IS_DGFX(vm->xe) && IS_ENABLED(CONFIG_DRM_XE_PAGEMAP) ? vm->xe->atomic_svm_timeslice_ms : 0, }; struct xe_svm_range *range; struct dma_fence *fence; + struct drm_pagemap *dpagemap; struct xe_tile *tile = gt_to_tile(gt); int migrate_try_count = ctx.devmem_only ? 3 : 1; ktime_t end = 0; @@ -850,8 +892,14 @@ retry: range_debug(range, "PAGE FAULT"); + dpagemap = xe_vma_resolve_pagemap(vma, tile); if (--migrate_try_count >= 0 && - xe_svm_range_needs_migrate_to_vram(range, vma, IS_DGFX(vm->xe))) { + xe_svm_range_needs_migrate_to_vram(range, vma, !!dpagemap || ctx.devmem_only)) { + /* TODO : For multi-device dpagemap will be used to find the + * remote tile and remote device. Will need to modify + * xe_svm_alloc_vram to use dpagemap for future multi-device + * support. + */ err = xe_svm_alloc_vram(tile, range, &ctx); ctx.timeslice_ms <<= 1; /* Double timeslice if we have to retry */ if (err) { @@ -919,6 +967,45 @@ err_out: } /** + * xe_svm_handle_pagefault() - SVM handle page fault + * @vm: The VM. + * @vma: The CPU address mirror VMA. + * @gt: The gt upon the fault occurred. + * @fault_addr: The GPU fault address. + * @atomic: The fault atomic access bit. + * + * Create GPU bindings for a SVM page fault. Optionally migrate to device + * memory. + * + * Return: 0 on success, negative error code on error. + */ +int xe_svm_handle_pagefault(struct xe_vm *vm, struct xe_vma *vma, + struct xe_gt *gt, u64 fault_addr, + bool atomic) +{ + int need_vram, ret; +retry: + need_vram = xe_vma_need_vram_for_atomic(vm->xe, vma, atomic); + if (need_vram < 0) + return need_vram; + + ret = __xe_svm_handle_pagefault(vm, vma, gt, fault_addr, + need_vram ? true : false); + if (ret == -EAGAIN) { + /* + * Retry once on -EAGAIN to re-lookup the VMA, as the original VMA + * may have been split by xe_svm_range_set_default_attr. + */ + vma = xe_vm_find_vma_by_addr(vm, fault_addr); + if (!vma) + return -EINVAL; + + goto retry; + } + return ret; +} + +/** * xe_svm_has_mapping() - SVM has mappings * @vm: The VM. * @start: Start address. @@ -934,6 +1021,41 @@ bool xe_svm_has_mapping(struct xe_vm *vm, u64 start, u64 end) } /** + * xe_svm_unmap_address_range - UNMAP SVM mappings and ranges + * @vm: The VM + * @start: start addr + * @end: end addr + * + * This function UNMAPS svm ranges if start or end address are inside them. + */ +void xe_svm_unmap_address_range(struct xe_vm *vm, u64 start, u64 end) +{ + struct drm_gpusvm_notifier *notifier, *next; + + lockdep_assert_held_write(&vm->lock); + + drm_gpusvm_for_each_notifier_safe(notifier, next, &vm->svm.gpusvm, start, end) { + struct drm_gpusvm_range *range, *__next; + + drm_gpusvm_for_each_range_safe(range, __next, notifier, start, end) { + if (start > drm_gpusvm_range_start(range) || + end < drm_gpusvm_range_end(range)) { + if (IS_DGFX(vm->xe) && xe_svm_range_in_vram(to_xe_range(range))) + drm_gpusvm_range_evict(&vm->svm.gpusvm, range); + drm_gpusvm_range_get(range); + __xe_svm_garbage_collector(vm, to_xe_range(range)); + if (!list_empty(&to_xe_range(range)->garbage_collector_link)) { + spin_lock(&vm->svm.garbage_collector.lock); + list_del(&to_xe_range(range)->garbage_collector_link); + spin_unlock(&vm->svm.garbage_collector.lock); + } + drm_gpusvm_range_put(range); + } + } + } +} + +/** * xe_svm_bo_evict() - SVM evict BO to system memory * @bo: BO to evict * @@ -997,8 +1119,94 @@ int xe_svm_range_get_pages(struct xe_vm *vm, struct xe_svm_range *range, return err; } +/** + * xe_svm_ranges_zap_ptes_in_range - clear ptes of svm ranges in input range + * @vm: Pointer to the xe_vm structure + * @start: Start of the input range + * @end: End of the input range + * + * This function removes the page table entries (PTEs) associated + * with the svm ranges within the given input start and end + * + * Return: tile_mask for which gt's need to be tlb invalidated. + */ +u8 xe_svm_ranges_zap_ptes_in_range(struct xe_vm *vm, u64 start, u64 end) +{ + struct drm_gpusvm_notifier *notifier; + struct xe_svm_range *range; + u64 adj_start, adj_end; + struct xe_tile *tile; + u8 tile_mask = 0; + u8 id; + + lockdep_assert(lockdep_is_held_type(&vm->svm.gpusvm.notifier_lock, 1) && + lockdep_is_held_type(&vm->lock, 0)); + + drm_gpusvm_for_each_notifier(notifier, &vm->svm.gpusvm, start, end) { + struct drm_gpusvm_range *r = NULL; + + adj_start = max(start, drm_gpusvm_notifier_start(notifier)); + adj_end = min(end, drm_gpusvm_notifier_end(notifier)); + drm_gpusvm_for_each_range(r, notifier, adj_start, adj_end) { + range = to_xe_range(r); + for_each_tile(tile, vm->xe, id) { + if (xe_pt_zap_ptes_range(tile, vm, range)) { + tile_mask |= BIT(id); + /* + * WRITE_ONCE pairs with READ_ONCE in + * xe_vm_has_valid_gpu_mapping(). + * Must not fail after setting + * tile_invalidated and before + * TLB invalidation. + */ + WRITE_ONCE(range->tile_invalidated, + range->tile_invalidated | BIT(id)); + } + } + } + } + + return tile_mask; +} + #if IS_ENABLED(CONFIG_DRM_XE_PAGEMAP) +static struct drm_pagemap *tile_local_pagemap(struct xe_tile *tile) +{ + return &tile->mem.vram->dpagemap; +} + +/** + * xe_vma_resolve_pagemap - Resolve the appropriate DRM pagemap for a VMA + * @vma: Pointer to the xe_vma structure containing memory attributes + * @tile: Pointer to the xe_tile structure used as fallback for VRAM mapping + * + * This function determines the correct DRM pagemap to use for a given VMA. + * It first checks if a valid devmem_fd is provided in the VMA's preferred + * location. If the devmem_fd is negative, it returns NULL, indicating no + * pagemap is available and smem to be used as preferred location. + * If the devmem_fd is equal to the default faulting + * GT identifier, it returns the VRAM pagemap associated with the tile. + * + * Future support for multi-device configurations may use drm_pagemap_from_fd() + * to resolve pagemaps from arbitrary file descriptors. + * + * Return: A pointer to the resolved drm_pagemap, or NULL if none is applicable. + */ +struct drm_pagemap *xe_vma_resolve_pagemap(struct xe_vma *vma, struct xe_tile *tile) +{ + s32 fd = (s32)vma->attr.preferred_loc.devmem_fd; + + if (fd == DRM_XE_PREFERRED_LOC_DEFAULT_SYSTEM) + return NULL; + + if (fd == DRM_XE_PREFERRED_LOC_DEFAULT_DEVICE) + return IS_DGFX(tile_to_xe(tile)) ? tile_local_pagemap(tile) : NULL; + + /* TODO: Support multi-device with drm_pagemap_from_fd(fd) */ + return NULL; +} + /** * xe_svm_alloc_vram()- Allocate device memory pages for range, * migrating existing data. @@ -1016,14 +1224,14 @@ int xe_svm_alloc_vram(struct xe_tile *tile, struct xe_svm_range *range, xe_assert(tile_to_xe(tile), range->base.flags.migrate_devmem); range_debug(range, "ALLOCATE VRAM"); - dpagemap = xe_tile_local_pagemap(tile); + dpagemap = tile_local_pagemap(tile); return drm_pagemap_populate_mm(dpagemap, xe_svm_range_start(range), xe_svm_range_end(range), range->base.gpusvm->mm, ctx->timeslice_ms); } -static struct drm_pagemap_device_addr +static struct drm_pagemap_addr xe_drm_pagemap_device_map(struct drm_pagemap *dpagemap, struct device *dev, struct page *page, @@ -1042,7 +1250,7 @@ xe_drm_pagemap_device_map(struct drm_pagemap *dpagemap, prot = 0; } - return drm_pagemap_device_addr_encode(addr, prot, order, dir); + return drm_pagemap_addr_encode(addr, prot, order, dir); } static const struct drm_pagemap_ops xe_drm_pagemap_ops = { @@ -1111,6 +1319,11 @@ int xe_devm_add(struct xe_tile *tile, struct xe_vram_region *vr) { return 0; } + +struct drm_pagemap *xe_vma_resolve_pagemap(struct xe_vma *vma, struct xe_tile *tile) +{ + return NULL; +} #endif /** diff --git a/drivers/gpu/drm/xe/xe_svm.h b/drivers/gpu/drm/xe/xe_svm.h index da9a69ea0bb1..9d6a8840a8b7 100644 --- a/drivers/gpu/drm/xe/xe_svm.h +++ b/drivers/gpu/drm/xe/xe_svm.h @@ -90,6 +90,12 @@ bool xe_svm_range_validate(struct xe_vm *vm, u64 xe_svm_find_vma_start(struct xe_vm *vm, u64 addr, u64 end, struct xe_vma *vma); +void xe_svm_unmap_address_range(struct xe_vm *vm, u64 start, u64 end); + +u8 xe_svm_ranges_zap_ptes_in_range(struct xe_vm *vm, u64 start, u64 end); + +struct drm_pagemap *xe_vma_resolve_pagemap(struct xe_vma *vma, struct xe_tile *tile); + /** * xe_svm_range_has_dma_mapping() - SVM range has DMA mapping * @range: SVM range @@ -163,7 +169,7 @@ void xe_svm_flush(struct xe_vm *vm); #else #include <linux/interval_tree.h> -struct drm_pagemap_device_addr; +struct drm_pagemap_addr; struct drm_gpusvm_ctx; struct drm_gpusvm_range; struct xe_bo; @@ -178,7 +184,7 @@ struct xe_vram_region; struct xe_svm_range { struct { struct interval_tree_node itree; - const struct drm_pagemap_device_addr *dma_addr; + const struct drm_pagemap_addr *dma_addr; } base; u32 tile_present; u32 tile_invalidated; @@ -303,6 +309,23 @@ u64 xe_svm_find_vma_start(struct xe_vm *vm, u64 addr, u64 end, struct xe_vma *vm return ULONG_MAX; } +static inline +void xe_svm_unmap_address_range(struct xe_vm *vm, u64 start, u64 end) +{ +} + +static inline +u8 xe_svm_ranges_zap_ptes_in_range(struct xe_vm *vm, u64 start, u64 end) +{ + return 0; +} + +static inline +struct drm_pagemap *xe_vma_resolve_pagemap(struct xe_vma *vma, struct xe_tile *tile) +{ + return NULL; +} + #define xe_svm_assert_in_notifier(...) do {} while (0) #define xe_svm_range_has_dma_mapping(...) false diff --git a/drivers/gpu/drm/xe/xe_sync.c b/drivers/gpu/drm/xe/xe_sync.c index f87276df18f2..82872a51f098 100644 --- a/drivers/gpu/drm/xe/xe_sync.c +++ b/drivers/gpu/drm/xe/xe_sync.c @@ -77,6 +77,7 @@ static void user_fence_worker(struct work_struct *w) { struct xe_user_fence *ufence = container_of(w, struct xe_user_fence, worker); + WRITE_ONCE(ufence->signalled, 1); if (mmget_not_zero(ufence->mm)) { kthread_use_mm(ufence->mm); if (copy_to_user(ufence->addr, &ufence->value, sizeof(ufence->value))) @@ -91,7 +92,6 @@ static void user_fence_worker(struct work_struct *w) * Wake up waiters only after updating the ufence state, allowing the UMD * to safely reuse the same ufence without encountering -EBUSY errors. */ - WRITE_ONCE(ufence->signalled, 1); wake_up_all(&ufence->xe->ufence_wq); user_fence_put(ufence); } diff --git a/drivers/gpu/drm/xe/xe_tile.c b/drivers/gpu/drm/xe/xe_tile.c index 86e9811e60ba..d49ba3401963 100644 --- a/drivers/gpu/drm/xe/xe_tile.c +++ b/drivers/gpu/drm/xe/xe_tile.c @@ -7,6 +7,7 @@ #include <drm/drm_managed.h> +#include "xe_bo.h" #include "xe_device.h" #include "xe_ggtt.h" #include "xe_gt.h" @@ -19,6 +20,8 @@ #include "xe_tile_sysfs.h" #include "xe_ttm_vram_mgr.h" #include "xe_wa.h" +#include "xe_vram.h" +#include "xe_vram_types.h" /** * DOC: Multi-tile Design @@ -92,6 +95,35 @@ static int xe_tile_alloc(struct xe_tile *tile) if (!tile->mem.ggtt) return -ENOMEM; + tile->migrate = xe_migrate_alloc(tile); + if (!tile->migrate) + return -ENOMEM; + + return 0; +} + +/** + * xe_tile_alloc_vram - Perform per-tile VRAM structs allocation + * @tile: Tile to perform allocations for + * + * Allocates VRAM per-tile data structures using DRM-managed allocations. + * Does not touch the hardware. + * + * Returns -ENOMEM if allocations fail, otherwise 0. + */ +int xe_tile_alloc_vram(struct xe_tile *tile) +{ + struct xe_device *xe = tile_to_xe(tile); + struct xe_vram_region *vram; + + if (!IS_DGFX(xe)) + return 0; + + vram = xe_vram_region_alloc(xe, tile->id, XE_PL_VRAM0 + tile->id); + if (!vram) + return -ENOMEM; + tile->mem.vram = vram; + return 0; } @@ -127,21 +159,6 @@ int xe_tile_init_early(struct xe_tile *tile, struct xe_device *xe, u8 id) } ALLOW_ERROR_INJECTION(xe_tile_init_early, ERRNO); /* See xe_pci_probe() */ -static int tile_ttm_mgr_init(struct xe_tile *tile) -{ - struct xe_device *xe = tile_to_xe(tile); - int err; - - if (tile->mem.vram.usable_size) { - err = xe_ttm_vram_mgr_init(tile, &tile->mem.vram.ttm); - if (err) - return err; - xe->info.mem_region_mask |= BIT(tile->id) << 1; - } - - return 0; -} - /** * xe_tile_init_noalloc - Init tile up to the point where allocations can happen. * @tile: The tile to initialize. @@ -159,16 +176,19 @@ static int tile_ttm_mgr_init(struct xe_tile *tile) int xe_tile_init_noalloc(struct xe_tile *tile) { struct xe_device *xe = tile_to_xe(tile); - int err; - - err = tile_ttm_mgr_init(tile); - if (err) - return err; xe_wa_apply_tile_workarounds(tile); if (xe->info.has_usm && IS_DGFX(xe)) - xe_devm_add(tile, &tile->mem.vram); + xe_devm_add(tile, tile->mem.vram); + + if (IS_DGFX(xe) && !ttm_resource_manager_used(&tile->mem.vram->ttm.manager)) { + int err = xe_ttm_vram_mgr_init(xe, tile->mem.vram); + + if (err) + return err; + xe->info.mem_region_mask |= BIT(tile->mem.vram->id) << 1; + } return xe_tile_sysfs_init(tile); } diff --git a/drivers/gpu/drm/xe/xe_tile.h b/drivers/gpu/drm/xe/xe_tile.h index cc33e8733983..dceb6297aa01 100644 --- a/drivers/gpu/drm/xe/xe_tile.h +++ b/drivers/gpu/drm/xe/xe_tile.h @@ -14,19 +14,9 @@ int xe_tile_init_early(struct xe_tile *tile, struct xe_device *xe, u8 id); int xe_tile_init_noalloc(struct xe_tile *tile); int xe_tile_init(struct xe_tile *tile); -void xe_tile_migrate_wait(struct xe_tile *tile); +int xe_tile_alloc_vram(struct xe_tile *tile); -#if IS_ENABLED(CONFIG_DRM_XE_PAGEMAP) -static inline struct drm_pagemap *xe_tile_local_pagemap(struct xe_tile *tile) -{ - return &tile->mem.vram.dpagemap; -} -#else -static inline struct drm_pagemap *xe_tile_local_pagemap(struct xe_tile *tile) -{ - return NULL; -} -#endif +void xe_tile_migrate_wait(struct xe_tile *tile); static inline bool xe_tile_is_root(struct xe_tile *tile) { diff --git a/drivers/gpu/drm/xe/xe_tlb_inval.c b/drivers/gpu/drm/xe/xe_tlb_inval.c new file mode 100644 index 000000000000..e6e97b5a7b5c --- /dev/null +++ b/drivers/gpu/drm/xe/xe_tlb_inval.c @@ -0,0 +1,434 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright © 2023 Intel Corporation + */ + +#include <drm/drm_managed.h> + +#include "abi/guc_actions_abi.h" +#include "xe_device.h" +#include "xe_force_wake.h" +#include "xe_gt.h" +#include "xe_gt_printk.h" +#include "xe_guc.h" +#include "xe_guc_ct.h" +#include "xe_guc_tlb_inval.h" +#include "xe_gt_stats.h" +#include "xe_tlb_inval.h" +#include "xe_mmio.h" +#include "xe_pm.h" +#include "xe_tlb_inval.h" +#include "xe_trace.h" + +/** + * DOC: Xe TLB invalidation + * + * Xe TLB invalidation is implemented in two layers. The first is the frontend + * API, which provides an interface for TLB invalidations to the driver code. + * The frontend handles seqno assignment, synchronization (fences), and the + * timeout mechanism. The frontend is implemented via an embedded structure + * xe_tlb_inval that includes a set of ops hooking into the backend. The backend + * interacts with the hardware (or firmware) to perform the actual invalidation. + */ + +#define FENCE_STACK_BIT DMA_FENCE_FLAG_USER_BITS + +static void xe_tlb_inval_fence_fini(struct xe_tlb_inval_fence *fence) +{ + if (WARN_ON_ONCE(!fence->tlb_inval)) + return; + + xe_pm_runtime_put(fence->tlb_inval->xe); + fence->tlb_inval = NULL; /* fini() should be called once */ +} + +static void +xe_tlb_inval_fence_signal(struct xe_tlb_inval_fence *fence) +{ + bool stack = test_bit(FENCE_STACK_BIT, &fence->base.flags); + + lockdep_assert_held(&fence->tlb_inval->pending_lock); + + list_del(&fence->link); + trace_xe_tlb_inval_fence_signal(fence->tlb_inval->xe, fence); + xe_tlb_inval_fence_fini(fence); + dma_fence_signal(&fence->base); + if (!stack) + dma_fence_put(&fence->base); +} + +static void +xe_tlb_inval_fence_signal_unlocked(struct xe_tlb_inval_fence *fence) +{ + struct xe_tlb_inval *tlb_inval = fence->tlb_inval; + + spin_lock_irq(&tlb_inval->pending_lock); + xe_tlb_inval_fence_signal(fence); + spin_unlock_irq(&tlb_inval->pending_lock); +} + +static void xe_tlb_inval_fence_timeout(struct work_struct *work) +{ + struct xe_tlb_inval *tlb_inval = container_of(work, struct xe_tlb_inval, + fence_tdr.work); + struct xe_device *xe = tlb_inval->xe; + struct xe_tlb_inval_fence *fence, *next; + long timeout_delay = tlb_inval->ops->timeout_delay(tlb_inval); + + tlb_inval->ops->flush(tlb_inval); + + spin_lock_irq(&tlb_inval->pending_lock); + list_for_each_entry_safe(fence, next, + &tlb_inval->pending_fences, link) { + s64 since_inval_ms = ktime_ms_delta(ktime_get(), + fence->inval_time); + + if (msecs_to_jiffies(since_inval_ms) < timeout_delay) + break; + + trace_xe_tlb_inval_fence_timeout(xe, fence); + drm_err(&xe->drm, + "TLB invalidation fence timeout, seqno=%d recv=%d", + fence->seqno, tlb_inval->seqno_recv); + + fence->base.error = -ETIME; + xe_tlb_inval_fence_signal(fence); + } + if (!list_empty(&tlb_inval->pending_fences)) + queue_delayed_work(system_wq, &tlb_inval->fence_tdr, + timeout_delay); + spin_unlock_irq(&tlb_inval->pending_lock); +} + +/** + * tlb_inval_fini - Clean up TLB invalidation state + * @drm: @drm_device + * @arg: pointer to struct @xe_tlb_inval + * + * Cancel pending fence workers and clean up any additional + * TLB invalidation state. + */ +static void tlb_inval_fini(struct drm_device *drm, void *arg) +{ + struct xe_tlb_inval *tlb_inval = arg; + + xe_tlb_inval_reset(tlb_inval); +} + +/** + * xe_gt_tlb_inval_init - Initialize TLB invalidation state + * @gt: GT structure + * + * Initialize TLB invalidation state, purely software initialization, should + * be called once during driver load. + * + * Return: 0 on success, negative error code on error. + */ +int xe_gt_tlb_inval_init_early(struct xe_gt *gt) +{ + struct xe_device *xe = gt_to_xe(gt); + struct xe_tlb_inval *tlb_inval = >->tlb_inval; + int err; + + tlb_inval->xe = xe; + tlb_inval->seqno = 1; + INIT_LIST_HEAD(&tlb_inval->pending_fences); + spin_lock_init(&tlb_inval->pending_lock); + spin_lock_init(&tlb_inval->lock); + INIT_DELAYED_WORK(&tlb_inval->fence_tdr, xe_tlb_inval_fence_timeout); + + err = drmm_mutex_init(&xe->drm, &tlb_inval->seqno_lock); + if (err) + return err; + + tlb_inval->job_wq = drmm_alloc_ordered_workqueue(&xe->drm, + "gt-tbl-inval-job-wq", + WQ_MEM_RECLAIM); + if (IS_ERR(tlb_inval->job_wq)) + return PTR_ERR(tlb_inval->job_wq); + + /* XXX: Blindly setting up backend to GuC */ + xe_guc_tlb_inval_init_early(>->uc.guc, tlb_inval); + + return drmm_add_action_or_reset(&xe->drm, tlb_inval_fini, tlb_inval); +} + +/** + * xe_tlb_inval_reset() - TLB invalidation reset + * @tlb_inval: TLB invalidation client + * + * Signal any pending invalidation fences, should be called during a GT reset + */ +void xe_tlb_inval_reset(struct xe_tlb_inval *tlb_inval) +{ + struct xe_tlb_inval_fence *fence, *next; + int pending_seqno; + + /* + * we can get here before the backends are even initialized if we're + * wedging very early, in which case there are not going to be any + * pendind fences so we can bail immediately. + */ + if (!tlb_inval->ops->initialized(tlb_inval)) + return; + + /* + * Backend is already disabled at this point. No new TLB requests can + * appear. + */ + + mutex_lock(&tlb_inval->seqno_lock); + spin_lock_irq(&tlb_inval->pending_lock); + cancel_delayed_work(&tlb_inval->fence_tdr); + /* + * We might have various kworkers waiting for TLB flushes to complete + * which are not tracked with an explicit TLB fence, however at this + * stage that will never happen since the backend is already disabled, + * so make sure we signal them here under the assumption that we have + * completed a full GT reset. + */ + if (tlb_inval->seqno == 1) + pending_seqno = TLB_INVALIDATION_SEQNO_MAX - 1; + else + pending_seqno = tlb_inval->seqno - 1; + WRITE_ONCE(tlb_inval->seqno_recv, pending_seqno); + + list_for_each_entry_safe(fence, next, + &tlb_inval->pending_fences, link) + xe_tlb_inval_fence_signal(fence); + spin_unlock_irq(&tlb_inval->pending_lock); + mutex_unlock(&tlb_inval->seqno_lock); +} + +static bool xe_tlb_inval_seqno_past(struct xe_tlb_inval *tlb_inval, int seqno) +{ + int seqno_recv = READ_ONCE(tlb_inval->seqno_recv); + + lockdep_assert_held(&tlb_inval->pending_lock); + + if (seqno - seqno_recv < -(TLB_INVALIDATION_SEQNO_MAX / 2)) + return false; + + if (seqno - seqno_recv > (TLB_INVALIDATION_SEQNO_MAX / 2)) + return true; + + return seqno_recv >= seqno; +} + +static void xe_tlb_inval_fence_prep(struct xe_tlb_inval_fence *fence) +{ + struct xe_tlb_inval *tlb_inval = fence->tlb_inval; + + fence->seqno = tlb_inval->seqno; + trace_xe_tlb_inval_fence_send(tlb_inval->xe, fence); + + spin_lock_irq(&tlb_inval->pending_lock); + fence->inval_time = ktime_get(); + list_add_tail(&fence->link, &tlb_inval->pending_fences); + + if (list_is_singular(&tlb_inval->pending_fences)) + queue_delayed_work(system_wq, &tlb_inval->fence_tdr, + tlb_inval->ops->timeout_delay(tlb_inval)); + spin_unlock_irq(&tlb_inval->pending_lock); + + tlb_inval->seqno = (tlb_inval->seqno + 1) % + TLB_INVALIDATION_SEQNO_MAX; + if (!tlb_inval->seqno) + tlb_inval->seqno = 1; +} + +#define xe_tlb_inval_issue(__tlb_inval, __fence, op, args...) \ +({ \ + int __ret; \ + \ + xe_assert((__tlb_inval)->xe, (__tlb_inval)->ops); \ + xe_assert((__tlb_inval)->xe, (__fence)); \ + \ + mutex_lock(&(__tlb_inval)->seqno_lock); \ + xe_tlb_inval_fence_prep((__fence)); \ + __ret = op((__tlb_inval), (__fence)->seqno, ##args); \ + if (__ret < 0) \ + xe_tlb_inval_fence_signal_unlocked((__fence)); \ + mutex_unlock(&(__tlb_inval)->seqno_lock); \ + \ + __ret == -ECANCELED ? 0 : __ret; \ +}) + +/** + * xe_tlb_inval_all() - Issue a TLB invalidation for all TLBs + * @tlb_inval: TLB invalidation client + * @fence: invalidation fence which will be signal on TLB invalidation + * completion + * + * Issue a TLB invalidation for all TLBs. Completion of TLB is asynchronous and + * caller can use the invalidation fence to wait for completion. + * + * Return: 0 on success, negative error code on error + */ +int xe_tlb_inval_all(struct xe_tlb_inval *tlb_inval, + struct xe_tlb_inval_fence *fence) +{ + return xe_tlb_inval_issue(tlb_inval, fence, tlb_inval->ops->all); +} + +/** + * xe_tlb_inval_ggtt() - Issue a TLB invalidation for the GGTT + * @tlb_inval: TLB invalidation client + * + * Issue a TLB invalidation for the GGTT. Completion of TLB is asynchronous and + * caller can use the invalidation fence to wait for completion. + * + * Return: 0 on success, negative error code on error + */ +int xe_tlb_inval_ggtt(struct xe_tlb_inval *tlb_inval) +{ + struct xe_tlb_inval_fence fence, *fence_ptr = &fence; + int ret; + + xe_tlb_inval_fence_init(tlb_inval, fence_ptr, true); + ret = xe_tlb_inval_issue(tlb_inval, fence_ptr, tlb_inval->ops->ggtt); + xe_tlb_inval_fence_wait(fence_ptr); + + return ret; +} + +/** + * xe_tlb_inval_range() - Issue a TLB invalidation for an address range + * @tlb_inval: TLB invalidation client + * @fence: invalidation fence which will be signal on TLB invalidation + * completion + * @start: start address + * @end: end address + * @asid: address space id + * + * Issue a range based TLB invalidation if supported, if not fallback to a full + * TLB invalidation. Completion of TLB is asynchronous and caller can use + * the invalidation fence to wait for completion. + * + * Return: Negative error code on error, 0 on success + */ +int xe_tlb_inval_range(struct xe_tlb_inval *tlb_inval, + struct xe_tlb_inval_fence *fence, u64 start, u64 end, + u32 asid) +{ + return xe_tlb_inval_issue(tlb_inval, fence, tlb_inval->ops->ppgtt, + start, end, asid); +} + +/** + * xe_tlb_inval_vm() - Issue a TLB invalidation for a VM + * @tlb_inval: TLB invalidation client + * @vm: VM to invalidate + * + * Invalidate entire VM's address space + */ +void xe_tlb_inval_vm(struct xe_tlb_inval *tlb_inval, struct xe_vm *vm) +{ + struct xe_tlb_inval_fence fence; + u64 range = 1ull << vm->xe->info.va_bits; + + xe_tlb_inval_fence_init(tlb_inval, &fence, true); + xe_tlb_inval_range(tlb_inval, &fence, 0, range, vm->usm.asid); + xe_tlb_inval_fence_wait(&fence); +} + +/** + * xe_tlb_inval_done_handler() - TLB invalidation done handler + * @tlb_inval: TLB invalidation client + * @seqno: seqno of invalidation that is done + * + * Update recv seqno, signal any TLB invalidation fences, and restart TDR + */ +void xe_tlb_inval_done_handler(struct xe_tlb_inval *tlb_inval, int seqno) +{ + struct xe_device *xe = tlb_inval->xe; + struct xe_tlb_inval_fence *fence, *next; + unsigned long flags; + + /* + * This can also be run both directly from the IRQ handler and also in + * process_g2h_msg(). Only one may process any individual CT message, + * however the order they are processed here could result in skipping a + * seqno. To handle that we just process all the seqnos from the last + * seqno_recv up to and including the one in msg[0]. The delta should be + * very small so there shouldn't be much of pending_fences we actually + * need to iterate over here. + * + * From GuC POV we expect the seqnos to always appear in-order, so if we + * see something later in the timeline we can be sure that anything + * appearing earlier has already signalled, just that we have yet to + * officially process the CT message like if racing against + * process_g2h_msg(). + */ + spin_lock_irqsave(&tlb_inval->pending_lock, flags); + if (xe_tlb_inval_seqno_past(tlb_inval, seqno)) { + spin_unlock_irqrestore(&tlb_inval->pending_lock, flags); + return; + } + + WRITE_ONCE(tlb_inval->seqno_recv, seqno); + + list_for_each_entry_safe(fence, next, + &tlb_inval->pending_fences, link) { + trace_xe_tlb_inval_fence_recv(xe, fence); + + if (!xe_tlb_inval_seqno_past(tlb_inval, fence->seqno)) + break; + + xe_tlb_inval_fence_signal(fence); + } + + if (!list_empty(&tlb_inval->pending_fences)) + mod_delayed_work(system_wq, + &tlb_inval->fence_tdr, + tlb_inval->ops->timeout_delay(tlb_inval)); + else + cancel_delayed_work(&tlb_inval->fence_tdr); + + spin_unlock_irqrestore(&tlb_inval->pending_lock, flags); +} + +static const char * +xe_inval_fence_get_driver_name(struct dma_fence *dma_fence) +{ + return "xe"; +} + +static const char * +xe_inval_fence_get_timeline_name(struct dma_fence *dma_fence) +{ + return "tlb_inval_fence"; +} + +static const struct dma_fence_ops inval_fence_ops = { + .get_driver_name = xe_inval_fence_get_driver_name, + .get_timeline_name = xe_inval_fence_get_timeline_name, +}; + +/** + * xe_tlb_inval_fence_init() - Initialize TLB invalidation fence + * @tlb_inval: TLB invalidation client + * @fence: TLB invalidation fence to initialize + * @stack: fence is stack variable + * + * Initialize TLB invalidation fence for use. xe_tlb_inval_fence_fini + * will be automatically called when fence is signalled (all fences must signal), + * even on error. + */ +void xe_tlb_inval_fence_init(struct xe_tlb_inval *tlb_inval, + struct xe_tlb_inval_fence *fence, + bool stack) +{ + xe_pm_runtime_get_noresume(tlb_inval->xe); + + spin_lock_irq(&tlb_inval->lock); + dma_fence_init(&fence->base, &inval_fence_ops, &tlb_inval->lock, + dma_fence_context_alloc(1), 1); + spin_unlock_irq(&tlb_inval->lock); + INIT_LIST_HEAD(&fence->link); + if (stack) + set_bit(FENCE_STACK_BIT, &fence->base.flags); + else + dma_fence_get(&fence->base); + fence->tlb_inval = tlb_inval; +} diff --git a/drivers/gpu/drm/xe/xe_tlb_inval.h b/drivers/gpu/drm/xe/xe_tlb_inval.h new file mode 100644 index 000000000000..554634dfd4e2 --- /dev/null +++ b/drivers/gpu/drm/xe/xe_tlb_inval.h @@ -0,0 +1,46 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2025 Intel Corporation + */ + +#ifndef _XE_TLB_INVAL_H_ +#define _XE_TLB_INVAL_H_ + +#include <linux/types.h> + +#include "xe_tlb_inval_types.h" + +struct xe_gt; +struct xe_guc; +struct xe_vm; + +int xe_gt_tlb_inval_init_early(struct xe_gt *gt); + +void xe_tlb_inval_reset(struct xe_tlb_inval *tlb_inval); +int xe_tlb_inval_all(struct xe_tlb_inval *tlb_inval, + struct xe_tlb_inval_fence *fence); +int xe_tlb_inval_ggtt(struct xe_tlb_inval *tlb_inval); +void xe_tlb_inval_vm(struct xe_tlb_inval *tlb_inval, struct xe_vm *vm); +int xe_tlb_inval_range(struct xe_tlb_inval *tlb_inval, + struct xe_tlb_inval_fence *fence, + u64 start, u64 end, u32 asid); + +void xe_tlb_inval_fence_init(struct xe_tlb_inval *tlb_inval, + struct xe_tlb_inval_fence *fence, + bool stack); + +/** + * xe_tlb_inval_fence_wait() - TLB invalidiation fence wait + * @fence: TLB invalidation fence to wait on + * + * Wait on a TLB invalidiation fence until it signals, non interruptable + */ +static inline void +xe_tlb_inval_fence_wait(struct xe_tlb_inval_fence *fence) +{ + dma_fence_wait(&fence->base, false); +} + +void xe_tlb_inval_done_handler(struct xe_tlb_inval *tlb_inval, int seqno); + +#endif /* _XE_TLB_INVAL_ */ diff --git a/drivers/gpu/drm/xe/xe_tlb_inval_job.c b/drivers/gpu/drm/xe/xe_tlb_inval_job.c new file mode 100644 index 000000000000..492def04a559 --- /dev/null +++ b/drivers/gpu/drm/xe/xe_tlb_inval_job.c @@ -0,0 +1,268 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright © 2025 Intel Corporation + */ + +#include "xe_assert.h" +#include "xe_dep_job_types.h" +#include "xe_dep_scheduler.h" +#include "xe_exec_queue.h" +#include "xe_gt_types.h" +#include "xe_tlb_inval.h" +#include "xe_tlb_inval_job.h" +#include "xe_migrate.h" +#include "xe_pm.h" + +/** struct xe_tlb_inval_job - TLB invalidation job */ +struct xe_tlb_inval_job { + /** @dep: base generic dependency Xe job */ + struct xe_dep_job dep; + /** @tlb_inval: TLB invalidation client */ + struct xe_tlb_inval *tlb_inval; + /** @q: exec queue issuing the invalidate */ + struct xe_exec_queue *q; + /** @refcount: ref count of this job */ + struct kref refcount; + /** + * @fence: dma fence to indicate completion. 1 way relationship - job + * can safely reference fence, fence cannot safely reference job. + */ + struct dma_fence *fence; + /** @start: Start address to invalidate */ + u64 start; + /** @end: End address to invalidate */ + u64 end; + /** @asid: Address space ID to invalidate */ + u32 asid; + /** @fence_armed: Fence has been armed */ + bool fence_armed; +}; + +static struct dma_fence *xe_tlb_inval_job_run(struct xe_dep_job *dep_job) +{ + struct xe_tlb_inval_job *job = + container_of(dep_job, typeof(*job), dep); + struct xe_tlb_inval_fence *ifence = + container_of(job->fence, typeof(*ifence), base); + + xe_tlb_inval_range(job->tlb_inval, ifence, job->start, + job->end, job->asid); + + return job->fence; +} + +static void xe_tlb_inval_job_free(struct xe_dep_job *dep_job) +{ + struct xe_tlb_inval_job *job = + container_of(dep_job, typeof(*job), dep); + + /* Pairs with get in xe_tlb_inval_job_push */ + xe_tlb_inval_job_put(job); +} + +static const struct xe_dep_job_ops dep_job_ops = { + .run_job = xe_tlb_inval_job_run, + .free_job = xe_tlb_inval_job_free, +}; + +/** + * xe_tlb_inval_job_create() - TLB invalidation job create + * @q: exec queue issuing the invalidate + * @tlb_inval: TLB invalidation client + * @dep_scheduler: Dependency scheduler for job + * @start: Start address to invalidate + * @end: End address to invalidate + * @asid: Address space ID to invalidate + * + * Create a TLB invalidation job and initialize internal fields. The caller is + * responsible for releasing the creation reference. + * + * Return: TLB invalidation job object on success, ERR_PTR failure + */ +struct xe_tlb_inval_job * +xe_tlb_inval_job_create(struct xe_exec_queue *q, struct xe_tlb_inval *tlb_inval, + struct xe_dep_scheduler *dep_scheduler, u64 start, + u64 end, u32 asid) +{ + struct xe_tlb_inval_job *job; + struct drm_sched_entity *entity = + xe_dep_scheduler_entity(dep_scheduler); + struct xe_tlb_inval_fence *ifence; + int err; + + job = kmalloc(sizeof(*job), GFP_KERNEL); + if (!job) + return ERR_PTR(-ENOMEM); + + job->q = q; + job->tlb_inval = tlb_inval; + job->start = start; + job->end = end; + job->asid = asid; + job->fence_armed = false; + job->dep.ops = &dep_job_ops; + kref_init(&job->refcount); + xe_exec_queue_get(q); /* Pairs with put in xe_tlb_inval_job_destroy */ + + ifence = kmalloc(sizeof(*ifence), GFP_KERNEL); + if (!ifence) { + err = -ENOMEM; + goto err_job; + } + job->fence = &ifence->base; + + err = drm_sched_job_init(&job->dep.drm, entity, 1, NULL, + q->xef ? q->xef->drm->client_id : 0); + if (err) + goto err_fence; + + /* Pairs with put in xe_tlb_inval_job_destroy */ + xe_pm_runtime_get_noresume(gt_to_xe(q->gt)); + + return job; + +err_fence: + kfree(ifence); +err_job: + xe_exec_queue_put(q); + kfree(job); + + return ERR_PTR(err); +} + +static void xe_tlb_inval_job_destroy(struct kref *ref) +{ + struct xe_tlb_inval_job *job = container_of(ref, typeof(*job), + refcount); + struct xe_tlb_inval_fence *ifence = + container_of(job->fence, typeof(*ifence), base); + struct xe_exec_queue *q = job->q; + struct xe_device *xe = gt_to_xe(q->gt); + + if (!job->fence_armed) + kfree(ifence); + else + /* Ref from xe_tlb_inval_fence_init */ + dma_fence_put(job->fence); + + drm_sched_job_cleanup(&job->dep.drm); + kfree(job); + xe_exec_queue_put(q); /* Pairs with get from xe_tlb_inval_job_create */ + xe_pm_runtime_put(xe); /* Pairs with get from xe_tlb_inval_job_create */ +} + +/** + * xe_tlb_inval_alloc_dep() - TLB invalidation job alloc dependency + * @job: TLB invalidation job to alloc dependency for + * + * Allocate storage for a dependency in the TLB invalidation fence. This + * function should be called at most once per job and must be paired with + * xe_tlb_inval_job_push being called with a real fence. + * + * Return: 0 on success, -errno on failure + */ +int xe_tlb_inval_job_alloc_dep(struct xe_tlb_inval_job *job) +{ + xe_assert(gt_to_xe(job->q->gt), !xa_load(&job->dep.drm.dependencies, 0)); + might_alloc(GFP_KERNEL); + + return drm_sched_job_add_dependency(&job->dep.drm, + dma_fence_get_stub()); +} + +/** + * xe_tlb_inval_job_push() - TLB invalidation job push + * @job: TLB invalidation job to push + * @m: The migration object being used + * @fence: Dependency for TLB invalidation job + * + * Pushes a TLB invalidation job for execution, using @fence as a dependency. + * Storage for @fence must be preallocated with xe_tlb_inval_job_alloc_dep + * prior to this call if @fence is not signaled. Takes a reference to the job’s + * finished fence, which the caller is responsible for releasing, and return it + * to the caller. This function is safe to be called in the path of reclaim. + * + * Return: Job's finished fence on success, cannot fail + */ +struct dma_fence *xe_tlb_inval_job_push(struct xe_tlb_inval_job *job, + struct xe_migrate *m, + struct dma_fence *fence) +{ + struct xe_tlb_inval_fence *ifence = + container_of(job->fence, typeof(*ifence), base); + + if (!dma_fence_is_signaled(fence)) { + void *ptr; + + /* + * Can be in path of reclaim, hence the preallocation of fence + * storage in xe_tlb_inval_job_alloc_dep. Verify caller did + * this correctly. + */ + xe_assert(gt_to_xe(job->q->gt), + xa_load(&job->dep.drm.dependencies, 0) == + dma_fence_get_stub()); + + dma_fence_get(fence); /* ref released once dependency processed by scheduler */ + ptr = xa_store(&job->dep.drm.dependencies, 0, fence, + GFP_ATOMIC); + xe_assert(gt_to_xe(job->q->gt), !xa_is_err(ptr)); + } + + xe_tlb_inval_job_get(job); /* Pairs with put in free_job */ + job->fence_armed = true; + + /* + * We need the migration lock to protect the job's seqno and the spsc + * queue, only taken on migration queue, user queues protected dma-resv + * VM lock. + */ + xe_migrate_job_lock(m, job->q); + + /* Creation ref pairs with put in xe_tlb_inval_job_destroy */ + xe_tlb_inval_fence_init(job->tlb_inval, ifence, false); + dma_fence_get(job->fence); /* Pairs with put in DRM scheduler */ + + drm_sched_job_arm(&job->dep.drm); + /* + * caller ref, get must be done before job push as it could immediately + * signal and free. + */ + dma_fence_get(&job->dep.drm.s_fence->finished); + drm_sched_entity_push_job(&job->dep.drm); + + xe_migrate_job_unlock(m, job->q); + + /* + * Not using job->fence, as it has its own dma-fence context, which does + * not allow TLB invalidation fences on the same queue, GT tuple to + * be squashed in dma-resv/DRM scheduler. Instead, we use the DRM scheduler + * context and job's finished fence, which enables squashing. + */ + return &job->dep.drm.s_fence->finished; +} + +/** + * xe_tlb_inval_job_get() - Get a reference to TLB invalidation job + * @job: TLB invalidation job object + * + * Increment the TLB invalidation job's reference count + */ +void xe_tlb_inval_job_get(struct xe_tlb_inval_job *job) +{ + kref_get(&job->refcount); +} + +/** + * xe_tlb_inval_job_put() - Put a reference to TLB invalidation job + * @job: TLB invalidation job object + * + * Decrement the TLB invalidation job's reference count, call + * xe_tlb_inval_job_destroy when reference count == 0. Skips decrement if + * input @job is NULL or IS_ERR. + */ +void xe_tlb_inval_job_put(struct xe_tlb_inval_job *job) +{ + if (!IS_ERR_OR_NULL(job)) + kref_put(&job->refcount, xe_tlb_inval_job_destroy); +} diff --git a/drivers/gpu/drm/xe/xe_tlb_inval_job.h b/drivers/gpu/drm/xe/xe_tlb_inval_job.h new file mode 100644 index 000000000000..e63edcb26b50 --- /dev/null +++ b/drivers/gpu/drm/xe/xe_tlb_inval_job.h @@ -0,0 +1,33 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2025 Intel Corporation + */ + +#ifndef _XE_TLB_INVAL_JOB_H_ +#define _XE_TLB_INVAL_JOB_H_ + +#include <linux/types.h> + +struct dma_fence; +struct xe_dep_scheduler; +struct xe_exec_queue; +struct xe_tlb_inval; +struct xe_tlb_inval_job; +struct xe_migrate; + +struct xe_tlb_inval_job * +xe_tlb_inval_job_create(struct xe_exec_queue *q, struct xe_tlb_inval *tlb_inval, + struct xe_dep_scheduler *dep_scheduler, + u64 start, u64 end, u32 asid); + +int xe_tlb_inval_job_alloc_dep(struct xe_tlb_inval_job *job); + +struct dma_fence *xe_tlb_inval_job_push(struct xe_tlb_inval_job *job, + struct xe_migrate *m, + struct dma_fence *fence); + +void xe_tlb_inval_job_get(struct xe_tlb_inval_job *job); + +void xe_tlb_inval_job_put(struct xe_tlb_inval_job *job); + +#endif diff --git a/drivers/gpu/drm/xe/xe_tlb_inval_types.h b/drivers/gpu/drm/xe/xe_tlb_inval_types.h new file mode 100644 index 000000000000..8f8b060e9005 --- /dev/null +++ b/drivers/gpu/drm/xe/xe_tlb_inval_types.h @@ -0,0 +1,130 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2023 Intel Corporation + */ + +#ifndef _XE_TLB_INVAL_TYPES_H_ +#define _XE_TLB_INVAL_TYPES_H_ + +#include <linux/workqueue.h> +#include <linux/dma-fence.h> + +struct xe_tlb_inval; + +/** struct xe_tlb_inval_ops - TLB invalidation ops (backend) */ +struct xe_tlb_inval_ops { + /** + * @all: Invalidate all TLBs + * @tlb_inval: TLB invalidation client + * @seqno: Seqno of TLB invalidation + * + * Return 0 on success, -ECANCELED if backend is mid-reset, error on + * failure + */ + int (*all)(struct xe_tlb_inval *tlb_inval, u32 seqno); + + /** + * @ggtt: Invalidate global translation TLBs + * @tlb_inval: TLB invalidation client + * @seqno: Seqno of TLB invalidation + * + * Return 0 on success, -ECANCELED if backend is mid-reset, error on + * failure + */ + int (*ggtt)(struct xe_tlb_inval *tlb_inval, u32 seqno); + + /** + * @ppgtt: Invalidate per-process translation TLBs + * @tlb_inval: TLB invalidation client + * @seqno: Seqno of TLB invalidation + * @start: Start address + * @end: End address + * @asid: Address space ID + * + * Return 0 on success, -ECANCELED if backend is mid-reset, error on + * failure + */ + int (*ppgtt)(struct xe_tlb_inval *tlb_inval, u32 seqno, u64 start, + u64 end, u32 asid); + + /** + * @initialized: Backend is initialized + * @tlb_inval: TLB invalidation client + * + * Return: True if back is initialized, False otherwise + */ + bool (*initialized)(struct xe_tlb_inval *tlb_inval); + + /** + * @flush: Flush pending TLB invalidations + * @tlb_inval: TLB invalidation client + */ + void (*flush)(struct xe_tlb_inval *tlb_inval); + + /** + * @timeout_delay: Timeout delay for TLB invalidation + * @tlb_inval: TLB invalidation client + * + * Return: Timeout delay for TLB invalidation in jiffies + */ + long (*timeout_delay)(struct xe_tlb_inval *tlb_inval); +}; + +/** struct xe_tlb_inval - TLB invalidation client (frontend) */ +struct xe_tlb_inval { + /** @private: Backend private pointer */ + void *private; + /** @xe: Pointer to Xe device */ + struct xe_device *xe; + /** @ops: TLB invalidation ops */ + const struct xe_tlb_inval_ops *ops; + /** @tlb_inval.seqno: TLB invalidation seqno, protected by CT lock */ +#define TLB_INVALIDATION_SEQNO_MAX 0x100000 + int seqno; + /** @tlb_invalidation.seqno_lock: protects @tlb_invalidation.seqno */ + struct mutex seqno_lock; + /** + * @seqno_recv: last received TLB invalidation seqno, protected by + * CT lock + */ + int seqno_recv; + /** + * @pending_fences: list of pending fences waiting TLB invaliations, + * protected CT lock + */ + struct list_head pending_fences; + /** + * @pending_lock: protects @pending_fences and updating @seqno_recv. + */ + spinlock_t pending_lock; + /** + * @fence_tdr: schedules a delayed call to xe_tlb_fence_timeout after + * the timeout interval is over. + */ + struct delayed_work fence_tdr; + /** @job_wq: schedules TLB invalidation jobs */ + struct workqueue_struct *job_wq; + /** @tlb_inval.lock: protects TLB invalidation fences */ + spinlock_t lock; +}; + +/** + * struct xe_tlb_inval_fence - TLB invalidation fence + * + * Optionally passed to xe_tlb_inval* functions and will be signaled upon TLB + * invalidation completion. + */ +struct xe_tlb_inval_fence { + /** @base: dma fence base */ + struct dma_fence base; + /** @tlb_inval: TLB invalidation client which fence belong to */ + struct xe_tlb_inval *tlb_inval; + /** @link: link into list of pending tlb fences */ + struct list_head link; + /** @seqno: seqno of TLB invalidation to signal fence one */ + int seqno; + /** @inval_time: time of TLB invalidation */ + ktime_t inval_time; +}; + +#endif diff --git a/drivers/gpu/drm/xe/xe_trace.h b/drivers/gpu/drm/xe/xe_trace.h index b4a3577df70c..314f42fcbcbd 100644 --- a/drivers/gpu/drm/xe/xe_trace.h +++ b/drivers/gpu/drm/xe/xe_trace.h @@ -14,10 +14,10 @@ #include "xe_exec_queue_types.h" #include "xe_gpu_scheduler_types.h" -#include "xe_gt_tlb_invalidation_types.h" #include "xe_gt_types.h" #include "xe_guc_exec_queue_types.h" #include "xe_sched_job.h" +#include "xe_tlb_inval_types.h" #include "xe_vm.h" #define __dev_name_xe(xe) dev_name((xe)->drm.dev) @@ -25,13 +25,13 @@ #define __dev_name_gt(gt) __dev_name_xe(gt_to_xe((gt))) #define __dev_name_eq(q) __dev_name_gt((q)->gt) -DECLARE_EVENT_CLASS(xe_gt_tlb_invalidation_fence, - TP_PROTO(struct xe_device *xe, struct xe_gt_tlb_invalidation_fence *fence), +DECLARE_EVENT_CLASS(xe_tlb_inval_fence, + TP_PROTO(struct xe_device *xe, struct xe_tlb_inval_fence *fence), TP_ARGS(xe, fence), TP_STRUCT__entry( __string(dev, __dev_name_xe(xe)) - __field(struct xe_gt_tlb_invalidation_fence *, fence) + __field(struct xe_tlb_inval_fence *, fence) __field(int, seqno) ), @@ -45,39 +45,23 @@ DECLARE_EVENT_CLASS(xe_gt_tlb_invalidation_fence, __get_str(dev), __entry->fence, __entry->seqno) ); -DEFINE_EVENT(xe_gt_tlb_invalidation_fence, xe_gt_tlb_invalidation_fence_create, - TP_PROTO(struct xe_device *xe, struct xe_gt_tlb_invalidation_fence *fence), +DEFINE_EVENT(xe_tlb_inval_fence, xe_tlb_inval_fence_send, + TP_PROTO(struct xe_device *xe, struct xe_tlb_inval_fence *fence), TP_ARGS(xe, fence) ); -DEFINE_EVENT(xe_gt_tlb_invalidation_fence, - xe_gt_tlb_invalidation_fence_work_func, - TP_PROTO(struct xe_device *xe, struct xe_gt_tlb_invalidation_fence *fence), +DEFINE_EVENT(xe_tlb_inval_fence, xe_tlb_inval_fence_recv, + TP_PROTO(struct xe_device *xe, struct xe_tlb_inval_fence *fence), TP_ARGS(xe, fence) ); -DEFINE_EVENT(xe_gt_tlb_invalidation_fence, xe_gt_tlb_invalidation_fence_cb, - TP_PROTO(struct xe_device *xe, struct xe_gt_tlb_invalidation_fence *fence), +DEFINE_EVENT(xe_tlb_inval_fence, xe_tlb_inval_fence_signal, + TP_PROTO(struct xe_device *xe, struct xe_tlb_inval_fence *fence), TP_ARGS(xe, fence) ); -DEFINE_EVENT(xe_gt_tlb_invalidation_fence, xe_gt_tlb_invalidation_fence_send, - TP_PROTO(struct xe_device *xe, struct xe_gt_tlb_invalidation_fence *fence), - TP_ARGS(xe, fence) -); - -DEFINE_EVENT(xe_gt_tlb_invalidation_fence, xe_gt_tlb_invalidation_fence_recv, - TP_PROTO(struct xe_device *xe, struct xe_gt_tlb_invalidation_fence *fence), - TP_ARGS(xe, fence) -); - -DEFINE_EVENT(xe_gt_tlb_invalidation_fence, xe_gt_tlb_invalidation_fence_signal, - TP_PROTO(struct xe_device *xe, struct xe_gt_tlb_invalidation_fence *fence), - TP_ARGS(xe, fence) -); - -DEFINE_EVENT(xe_gt_tlb_invalidation_fence, xe_gt_tlb_invalidation_fence_timeout, - TP_PROTO(struct xe_device *xe, struct xe_gt_tlb_invalidation_fence *fence), +DEFINE_EVENT(xe_tlb_inval_fence, xe_tlb_inval_fence_timeout, + TP_PROTO(struct xe_device *xe, struct xe_tlb_inval_fence *fence), TP_ARGS(xe, fence) ); diff --git a/drivers/gpu/drm/xe/xe_ttm_stolen_mgr.c b/drivers/gpu/drm/xe/xe_ttm_stolen_mgr.c index d9c9d2547aad..dc588255674d 100644 --- a/drivers/gpu/drm/xe/xe_ttm_stolen_mgr.c +++ b/drivers/gpu/drm/xe/xe_ttm_stolen_mgr.c @@ -25,6 +25,7 @@ #include "xe_ttm_stolen_mgr.h" #include "xe_ttm_vram_mgr.h" #include "xe_wa.h" +#include "xe_vram.h" struct xe_ttm_stolen_mgr { struct xe_ttm_vram_mgr base; @@ -82,15 +83,16 @@ static u32 get_wopcm_size(struct xe_device *xe) static s64 detect_bar2_dgfx(struct xe_device *xe, struct xe_ttm_stolen_mgr *mgr) { - struct xe_tile *tile = xe_device_get_root_tile(xe); + struct xe_vram_region *tile_vram = xe_device_get_root_tile(xe)->mem.vram; + resource_size_t tile_io_start = xe_vram_region_io_start(tile_vram); struct xe_mmio *mmio = xe_root_tile_mmio(xe); struct pci_dev *pdev = to_pci_dev(xe->drm.dev); u64 stolen_size, wopcm_size; u64 tile_offset; u64 tile_size; - tile_offset = tile->mem.vram.io_start - xe->mem.vram.io_start; - tile_size = tile->mem.vram.actual_physical_size; + tile_offset = tile_io_start - xe_vram_region_io_start(xe->mem.vram); + tile_size = xe_vram_region_actual_physical_size(tile_vram); /* Use DSM base address instead for stolen memory */ mgr->stolen_base = (xe_mmio_read64_2x32(mmio, DSMBASE) & BDSM_MASK) - tile_offset; @@ -107,7 +109,7 @@ static s64 detect_bar2_dgfx(struct xe_device *xe, struct xe_ttm_stolen_mgr *mgr) /* Verify usage fits in the actual resource available */ if (mgr->stolen_base + stolen_size <= pci_resource_len(pdev, LMEM_BAR)) - mgr->io_base = tile->mem.vram.io_start + mgr->stolen_base; + mgr->io_base = tile_io_start + mgr->stolen_base; /* * There may be few KB of platform dependent reserved memory at the end @@ -164,7 +166,7 @@ static u32 detect_bar2_integrated(struct xe_device *xe, struct xe_ttm_stolen_mgr stolen_size -= wopcm_size; - if (media_gt && XE_WA(media_gt, 14019821291)) { + if (media_gt && XE_GT_WA(media_gt, 14019821291)) { u64 gscpsmi_base = xe_mmio_read64_2x32(&media_gt->mmio, GSCPSMI_BASE) & ~GENMASK_ULL(5, 0); diff --git a/drivers/gpu/drm/xe/xe_ttm_vram_mgr.c b/drivers/gpu/drm/xe/xe_ttm_vram_mgr.c index 9e375a40aee9..9175b4a2214b 100644 --- a/drivers/gpu/drm/xe/xe_ttm_vram_mgr.c +++ b/drivers/gpu/drm/xe/xe_ttm_vram_mgr.c @@ -15,6 +15,7 @@ #include "xe_gt.h" #include "xe_res_cursor.h" #include "xe_ttm_vram_mgr.h" +#include "xe_vram_types.h" static inline struct drm_buddy_block * xe_ttm_vram_mgr_first_block(struct list_head *list) @@ -337,13 +338,20 @@ int __xe_ttm_vram_mgr_init(struct xe_device *xe, struct xe_ttm_vram_mgr *mgr, return drmm_add_action_or_reset(&xe->drm, ttm_vram_mgr_fini, mgr); } -int xe_ttm_vram_mgr_init(struct xe_tile *tile, struct xe_ttm_vram_mgr *mgr) +/** + * xe_ttm_vram_mgr_init - initialize TTM VRAM region + * @xe: pointer to Xe device + * @vram: pointer to xe_vram_region that contains the memory region attributes + * + * Initialize the Xe TTM for given @vram region using the given parameters. + * + * Returns 0 for success, negative error code otherwise. + */ +int xe_ttm_vram_mgr_init(struct xe_device *xe, struct xe_vram_region *vram) { - struct xe_device *xe = tile_to_xe(tile); - struct xe_vram_region *vram = &tile->mem.vram; - - return __xe_ttm_vram_mgr_init(xe, mgr, XE_PL_VRAM0 + tile->id, - vram->usable_size, vram->io_size, + return __xe_ttm_vram_mgr_init(xe, &vram->ttm, vram->placement, + xe_vram_region_usable_size(vram), + xe_vram_region_io_size(vram), PAGE_SIZE); } @@ -392,7 +400,7 @@ int xe_ttm_vram_mgr_alloc_sgt(struct xe_device *xe, */ xe_res_first(res, offset, length, &cursor); for_each_sgtable_sg((*sgt), sg, i) { - phys_addr_t phys = cursor.start + tile->mem.vram.io_start; + phys_addr_t phys = cursor.start + xe_vram_region_io_start(tile->mem.vram); size_t size = min_t(u64, cursor.size, SZ_2G); dma_addr_t addr; diff --git a/drivers/gpu/drm/xe/xe_ttm_vram_mgr.h b/drivers/gpu/drm/xe/xe_ttm_vram_mgr.h index cc76050e376d..87b7fae5edba 100644 --- a/drivers/gpu/drm/xe/xe_ttm_vram_mgr.h +++ b/drivers/gpu/drm/xe/xe_ttm_vram_mgr.h @@ -11,11 +11,12 @@ enum dma_data_direction; struct xe_device; struct xe_tile; +struct xe_vram_region; int __xe_ttm_vram_mgr_init(struct xe_device *xe, struct xe_ttm_vram_mgr *mgr, u32 mem_type, u64 size, u64 io_size, u64 default_page_size); -int xe_ttm_vram_mgr_init(struct xe_tile *tile, struct xe_ttm_vram_mgr *mgr); +int xe_ttm_vram_mgr_init(struct xe_device *xe, struct xe_vram_region *vram); int xe_ttm_vram_mgr_alloc_sgt(struct xe_device *xe, struct ttm_resource *res, u64 offset, u64 length, diff --git a/drivers/gpu/drm/xe/xe_tuning.c b/drivers/gpu/drm/xe/xe_tuning.c index 828b45b24c23..a524170a04d0 100644 --- a/drivers/gpu/drm/xe/xe_tuning.c +++ b/drivers/gpu/drm/xe/xe_tuning.c @@ -99,7 +99,7 @@ static const struct xe_rtp_entry_sr engine_tunings[] = { XE_RTP_ACTIONS(SET(SAMPLER_MODE, INDIRECT_STATE_BASE_ADDR_OVERRIDE)) }, { XE_RTP_NAME("Tuning: Disable NULL query for Anyhit Shader"), - XE_RTP_RULES(GRAPHICS_VERSION_RANGE(3000, XE_RTP_END_VERSION_UNDEFINED), + XE_RTP_RULES(GRAPHICS_VERSION_RANGE(2000, XE_RTP_END_VERSION_UNDEFINED), FUNC(xe_rtp_match_first_render_or_compute)), XE_RTP_ACTIONS(SET(RT_CTRL, DIS_NULL_QUERY)) }, diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c index 86842247e9d8..3ff3c67aa79d 100644 --- a/drivers/gpu/drm/xe/xe_vm.c +++ b/drivers/gpu/drm/xe/xe_vm.c @@ -28,7 +28,6 @@ #include "xe_drm_client.h" #include "xe_exec_queue.h" #include "xe_gt_pagefault.h" -#include "xe_gt_tlb_invalidation.h" #include "xe_migrate.h" #include "xe_pat.h" #include "xe_pm.h" @@ -38,6 +37,8 @@ #include "xe_res_cursor.h" #include "xe_svm.h" #include "xe_sync.h" +#include "xe_tile.h" +#include "xe_tlb_inval.h" #include "xe_trace_bo.h" #include "xe_wa.h" #include "xe_hmm.h" @@ -953,7 +954,7 @@ struct dma_fence *xe_vma_rebind(struct xe_vm *vm, struct xe_vma *vma, u8 tile_ma for_each_tile(tile, vm->xe, id) { vops.pt_update_ops[id].wait_vm_bookkeep = true; vops.pt_update_ops[tile->id].q = - xe_tile_migrate_exec_queue(tile); + xe_migrate_exec_queue(tile->migrate); } err = xe_vm_ops_add_rebind(&vops, vma, tile_mask); @@ -1043,7 +1044,7 @@ struct dma_fence *xe_vm_range_rebind(struct xe_vm *vm, for_each_tile(tile, vm->xe, id) { vops.pt_update_ops[id].wait_vm_bookkeep = true; vops.pt_update_ops[tile->id].q = - xe_tile_migrate_exec_queue(tile); + xe_migrate_exec_queue(tile->migrate); } err = xe_vm_ops_add_range_rebind(&vops, vma, range, tile_mask); @@ -1126,7 +1127,7 @@ struct dma_fence *xe_vm_range_unbind(struct xe_vm *vm, for_each_tile(tile, vm->xe, id) { vops.pt_update_ops[id].wait_vm_bookkeep = true; vops.pt_update_ops[tile->id].q = - xe_tile_migrate_exec_queue(tile); + xe_migrate_exec_queue(tile->migrate); } err = xe_vm_ops_add_range_unbind(&vops, range); @@ -1168,7 +1169,8 @@ static struct xe_vma *xe_vma_create(struct xe_vm *vm, struct xe_bo *bo, u64 bo_offset_or_userptr, u64 start, u64 end, - u16 pat_index, unsigned int flags) + struct xe_vma_mem_attr *attr, + unsigned int flags) { struct xe_vma *vma; struct xe_tile *tile; @@ -1223,7 +1225,7 @@ static struct xe_vma *xe_vma_create(struct xe_vm *vm, if (vm->xe->info.has_atomic_enable_pte_bit) vma->gpuva.flags |= XE_VMA_ATOMIC_PTE_BIT; - vma->pat_index = pat_index; + vma->attr = *attr; if (bo) { struct drm_gpuvm_bo *vm_bo; @@ -1512,14 +1514,39 @@ static u64 pte_encode_ps(u32 pt_level) return 0; } -static u64 xelp_pde_encode_bo(struct xe_bo *bo, u64 bo_offset, - const u16 pat_index) +static u16 pde_pat_index(struct xe_bo *bo) +{ + struct xe_device *xe = xe_bo_device(bo); + u16 pat_index; + + /* + * We only have two bits to encode the PAT index in non-leaf nodes, but + * these only point to other paging structures so we only need a minimal + * selection of options. The user PAT index is only for encoding leaf + * nodes, where we have use of more bits to do the encoding. The + * non-leaf nodes are instead under driver control so the chosen index + * here should be distict from the user PAT index. Also the + * corresponding coherency of the PAT index should be tied to the + * allocation type of the page table (or at least we should pick + * something which is always safe). + */ + if (!xe_bo_is_vram(bo) && bo->ttm.ttm->caching == ttm_cached) + pat_index = xe->pat.idx[XE_CACHE_WB]; + else + pat_index = xe->pat.idx[XE_CACHE_NONE]; + + xe_assert(xe, pat_index <= 3); + + return pat_index; +} + +static u64 xelp_pde_encode_bo(struct xe_bo *bo, u64 bo_offset) { u64 pde; pde = xe_bo_addr(bo, bo_offset, XE_PAGE_SIZE); pde |= XE_PAGE_PRESENT | XE_PAGE_RW; - pde |= pde_encode_pat_index(pat_index); + pde |= pde_encode_pat_index(pde_pat_index(bo)); return pde; } @@ -1610,8 +1637,12 @@ static int xe_vm_create_scratch(struct xe_device *xe, struct xe_tile *tile, for (i = MAX_HUGEPTE_LEVEL; i < vm->pt_root[id]->level; i++) { vm->scratch_pt[id][i] = xe_pt_create(vm, tile, i); - if (IS_ERR(vm->scratch_pt[id][i])) - return PTR_ERR(vm->scratch_pt[id][i]); + if (IS_ERR(vm->scratch_pt[id][i])) { + int err = PTR_ERR(vm->scratch_pt[id][i]); + + vm->scratch_pt[id][i] = NULL; + return err; + } xe_pt_populate_empty(tile, vm, vm->scratch_pt[id][i]); } @@ -1640,7 +1671,7 @@ static void xe_vm_free_scratch(struct xe_vm *vm) } } -struct xe_vm *xe_vm_create(struct xe_device *xe, u32 flags) +struct xe_vm *xe_vm_create(struct xe_device *xe, u32 flags, struct xe_file *xef) { struct drm_gem_object *vm_resv_obj; struct xe_vm *vm; @@ -1661,9 +1692,10 @@ struct xe_vm *xe_vm_create(struct xe_device *xe, u32 flags) vm->xe = xe; vm->size = 1ull << xe->info.va_bits; - vm->flags = flags; + if (xef) + vm->xef = xe_file_get(xef); /** * GSC VMs are kernel-owned, only used for PXP ops and can sometimes be * manipulated under the PXP mutex. However, the PXP mutex can be taken @@ -1794,6 +1826,20 @@ struct xe_vm *xe_vm_create(struct xe_device *xe, u32 flags) if (number_tiles > 1) vm->composite_fence_ctx = dma_fence_context_alloc(1); + if (xef && xe->info.has_asid) { + u32 asid; + + down_write(&xe->usm.lock); + err = xa_alloc_cyclic(&xe->usm.asid_to_vm, &asid, vm, + XA_LIMIT(1, XE_MAX_ASID - 1), + &xe->usm.next_asid, GFP_KERNEL); + up_write(&xe->usm.lock); + if (err < 0) + goto err_unlock_close; + + vm->usm.asid = asid; + } + trace_xe_vm_create(vm); return vm; @@ -1814,6 +1860,8 @@ err_no_resv: for_each_tile(tile, xe, id) xe_range_fence_tree_fini(&vm->rftree[id]); ttm_lru_bulk_move_fini(&xe->ttm, &vm->lru_bulk_move); + if (vm->xef) + xe_file_put(vm->xef); kfree(vm); if (flags & XE_VM_FLAG_LR_MODE) xe_pm_runtime_put(xe); @@ -1850,7 +1898,7 @@ static void xe_vm_close(struct xe_vm *vm) xe_pt_clear(xe, vm->pt_root[id]); for_each_gt(gt, xe, id) - xe_gt_tlb_invalidation_vm(gt, vm); + xe_tlb_inval_vm(>->tlb_inval, vm); } } @@ -2024,8 +2072,7 @@ struct xe_vm *xe_vm_lookup(struct xe_file *xef, u32 id) u64 xe_vm_pdp4_descriptor(struct xe_vm *vm, struct xe_tile *tile) { - return vm->pt_ops->pde_encode_bo(vm->pt_root[tile->id]->bo, 0, - tile_to_xe(tile)->pat.idx[XE_CACHE_WB]); + return vm->pt_ops->pde_encode_bo(vm->pt_root[tile->id]->bo, 0); } static struct xe_exec_queue * @@ -2059,16 +2106,15 @@ int xe_vm_create_ioctl(struct drm_device *dev, void *data, struct xe_device *xe = to_xe_device(dev); struct xe_file *xef = to_xe_file(file); struct drm_xe_vm_create *args = data; - struct xe_tile *tile; struct xe_vm *vm; - u32 id, asid; + u32 id; int err; u32 flags = 0; if (XE_IOCTL_DBG(xe, args->extensions)) return -EINVAL; - if (XE_WA(xe_root_mmio_gt(xe), 14016763929)) + if (XE_GT_WA(xe_root_mmio_gt(xe), 14016763929)) args->flags |= DRM_XE_VM_CREATE_FLAG_SCRATCH_PAGE; if (XE_IOCTL_DBG(xe, args->flags & DRM_XE_VM_CREATE_FLAG_FAULT_MODE && @@ -2097,29 +2143,10 @@ int xe_vm_create_ioctl(struct drm_device *dev, void *data, if (args->flags & DRM_XE_VM_CREATE_FLAG_FAULT_MODE) flags |= XE_VM_FLAG_FAULT_MODE; - vm = xe_vm_create(xe, flags); + vm = xe_vm_create(xe, flags, xef); if (IS_ERR(vm)) return PTR_ERR(vm); - if (xe->info.has_asid) { - down_write(&xe->usm.lock); - err = xa_alloc_cyclic(&xe->usm.asid_to_vm, &asid, vm, - XA_LIMIT(1, XE_MAX_ASID - 1), - &xe->usm.next_asid, GFP_KERNEL); - up_write(&xe->usm.lock); - if (err < 0) - goto err_close_and_put; - - vm->usm.asid = asid; - } - - vm->xef = xe_file_get(xef); - - /* Record BO memory for VM pagetable created against client */ - for_each_tile(tile, xe, id) - if (vm->pt_root[id]) - xe_drm_client_add_bo(vm->xef->client, vm->pt_root[id]->bo); - #if IS_ENABLED(CONFIG_DRM_XE_DEBUG_MEM) /* Warning: Security issue - never enable by default */ args->reserved[0] = xe_bo_main_addr(vm->pt_root[0]->bo, XE_PAGE_SIZE); @@ -2169,6 +2196,108 @@ int xe_vm_destroy_ioctl(struct drm_device *dev, void *data, return err; } +static int xe_vm_query_vmas(struct xe_vm *vm, u64 start, u64 end) +{ + struct drm_gpuva *gpuva; + u32 num_vmas = 0; + + lockdep_assert_held(&vm->lock); + drm_gpuvm_for_each_va_range(gpuva, &vm->gpuvm, start, end) + num_vmas++; + + return num_vmas; +} + +static int get_mem_attrs(struct xe_vm *vm, u32 *num_vmas, u64 start, + u64 end, struct drm_xe_mem_range_attr *attrs) +{ + struct drm_gpuva *gpuva; + int i = 0; + + lockdep_assert_held(&vm->lock); + + drm_gpuvm_for_each_va_range(gpuva, &vm->gpuvm, start, end) { + struct xe_vma *vma = gpuva_to_vma(gpuva); + + if (i == *num_vmas) + return -ENOSPC; + + attrs[i].start = xe_vma_start(vma); + attrs[i].end = xe_vma_end(vma); + attrs[i].atomic.val = vma->attr.atomic_access; + attrs[i].pat_index.val = vma->attr.pat_index; + attrs[i].preferred_mem_loc.devmem_fd = vma->attr.preferred_loc.devmem_fd; + attrs[i].preferred_mem_loc.migration_policy = + vma->attr.preferred_loc.migration_policy; + + i++; + } + + *num_vmas = i; + return 0; +} + +int xe_vm_query_vmas_attrs_ioctl(struct drm_device *dev, void *data, struct drm_file *file) +{ + struct xe_device *xe = to_xe_device(dev); + struct xe_file *xef = to_xe_file(file); + struct drm_xe_mem_range_attr *mem_attrs; + struct drm_xe_vm_query_mem_range_attr *args = data; + u64 __user *attrs_user = u64_to_user_ptr(args->vector_of_mem_attr); + struct xe_vm *vm; + int err = 0; + + if (XE_IOCTL_DBG(xe, + ((args->num_mem_ranges == 0 && + (attrs_user || args->sizeof_mem_range_attr != 0)) || + (args->num_mem_ranges > 0 && + (!attrs_user || + args->sizeof_mem_range_attr != + sizeof(struct drm_xe_mem_range_attr)))))) + return -EINVAL; + + vm = xe_vm_lookup(xef, args->vm_id); + if (XE_IOCTL_DBG(xe, !vm)) + return -EINVAL; + + err = down_read_interruptible(&vm->lock); + if (err) + goto put_vm; + + attrs_user = u64_to_user_ptr(args->vector_of_mem_attr); + + if (args->num_mem_ranges == 0 && !attrs_user) { + args->num_mem_ranges = xe_vm_query_vmas(vm, args->start, args->start + args->range); + args->sizeof_mem_range_attr = sizeof(struct drm_xe_mem_range_attr); + goto unlock_vm; + } + + mem_attrs = kvmalloc_array(args->num_mem_ranges, args->sizeof_mem_range_attr, + GFP_KERNEL | __GFP_ACCOUNT | + __GFP_RETRY_MAYFAIL | __GFP_NOWARN); + if (!mem_attrs) { + err = args->num_mem_ranges > 1 ? -ENOBUFS : -ENOMEM; + goto unlock_vm; + } + + memset(mem_attrs, 0, args->num_mem_ranges * args->sizeof_mem_range_attr); + err = get_mem_attrs(vm, &args->num_mem_ranges, args->start, + args->start + args->range, mem_attrs); + if (err) + goto free_mem_attrs; + + err = copy_to_user(attrs_user, mem_attrs, + args->sizeof_mem_range_attr * args->num_mem_ranges); + +free_mem_attrs: + kvfree(mem_attrs); +unlock_vm: + up_read(&vm->lock); +put_vm: + xe_vm_put(vm); + return err; +} + static bool vma_matches(struct xe_vma *vma, u64 page_addr) { if (page_addr > xe_vma_end(vma) - 1 || @@ -2374,9 +2503,10 @@ vm_bind_ioctl_ops_create(struct xe_vm *vm, struct xe_vma_ops *vops, __xe_vm_needs_clear_scratch_pages(vm, flags); } else if (__op->op == DRM_GPUVA_OP_PREFETCH) { struct xe_vma *vma = gpuva_to_vma(op->base.prefetch.va); + struct xe_tile *tile; struct xe_svm_range *svm_range; struct drm_gpusvm_ctx ctx = {}; - struct xe_tile *tile; + struct drm_pagemap *dpagemap; u8 id, tile_mask = 0; u32 i; @@ -2393,8 +2523,24 @@ vm_bind_ioctl_ops_create(struct xe_vm *vm, struct xe_vma_ops *vops, tile_mask |= 0x1 << id; xa_init_flags(&op->prefetch_range.range, XA_FLAGS_ALLOC); - op->prefetch_range.region = prefetch_region; op->prefetch_range.ranges_count = 0; + tile = NULL; + + if (prefetch_region == DRM_XE_CONSULT_MEM_ADVISE_PREF_LOC) { + dpagemap = xe_vma_resolve_pagemap(vma, + xe_device_get_root_tile(vm->xe)); + /* + * TODO: Once multigpu support is enabled will need + * something to dereference tile from dpagemap. + */ + if (dpagemap) + tile = xe_device_get_root_tile(vm->xe); + } else if (prefetch_region) { + tile = &vm->xe->tiles[region_to_mem_type[prefetch_region] - + XE_PL_VRAM0]; + } + + op->prefetch_range.tile = tile; alloc_next_range: svm_range = xe_svm_range_find_or_insert(vm, addr, vma, &ctx); @@ -2413,7 +2559,7 @@ alloc_next_range: goto unwind_prefetch_ops; } - if (xe_svm_range_validate(vm, svm_range, tile_mask, !!prefetch_region)) { + if (xe_svm_range_validate(vm, svm_range, tile_mask, !!tile)) { xe_svm_range_debug(svm_range, "PREFETCH - RANGE IS VALID"); goto check_next_range; } @@ -2450,7 +2596,7 @@ unwind_prefetch_ops: ALLOW_ERROR_INJECTION(vm_bind_ioctl_ops_create, ERRNO); static struct xe_vma *new_vma(struct xe_vm *vm, struct drm_gpuva_op_map *op, - u16 pat_index, unsigned int flags) + struct xe_vma_mem_attr *attr, unsigned int flags) { struct xe_bo *bo = op->gem.obj ? gem_to_xe_bo(op->gem.obj) : NULL; struct drm_exec exec; @@ -2479,7 +2625,7 @@ static struct xe_vma *new_vma(struct xe_vm *vm, struct drm_gpuva_op_map *op, } vma = xe_vma_create(vm, bo, op->gem.offset, op->va.addr, op->va.addr + - op->va.range - 1, pat_index, flags); + op->va.range - 1, attr, flags); if (IS_ERR(vma)) goto err_unlock; @@ -2596,6 +2742,29 @@ static int xe_vma_op_commit(struct xe_vm *vm, struct xe_vma_op *op) return err; } +/** + * xe_vma_has_default_mem_attrs - Check if a VMA has default memory attributes + * @vma: Pointer to the xe_vma structure to check + * + * This function determines whether the given VMA (Virtual Memory Area) + * has its memory attributes set to their default values. Specifically, + * it checks the following conditions: + * + * - `atomic_access` is `DRM_XE_VMA_ATOMIC_UNDEFINED` + * - `pat_index` is equal to `default_pat_index` + * - `preferred_loc.devmem_fd` is `DRM_XE_PREFERRED_LOC_DEFAULT_DEVICE` + * - `preferred_loc.migration_policy` is `DRM_XE_MIGRATE_ALL_PAGES` + * + * Return: true if all attributes are at their default values, false otherwise. + */ +bool xe_vma_has_default_mem_attrs(struct xe_vma *vma) +{ + return (vma->attr.atomic_access == DRM_XE_ATOMIC_UNDEFINED && + vma->attr.pat_index == vma->attr.default_pat_index && + vma->attr.preferred_loc.devmem_fd == DRM_XE_PREFERRED_LOC_DEFAULT_DEVICE && + vma->attr.preferred_loc.migration_policy == DRM_XE_MIGRATE_ALL_PAGES); +} + static int vm_bind_ioctl_ops_parse(struct xe_vm *vm, struct drm_gpuva_ops *ops, struct xe_vma_ops *vops) { @@ -2622,6 +2791,16 @@ static int vm_bind_ioctl_ops_parse(struct xe_vm *vm, struct drm_gpuva_ops *ops, switch (op->base.op) { case DRM_GPUVA_OP_MAP: { + struct xe_vma_mem_attr default_attr = { + .preferred_loc = { + .devmem_fd = DRM_XE_PREFERRED_LOC_DEFAULT_DEVICE, + .migration_policy = DRM_XE_MIGRATE_ALL_PAGES, + }, + .atomic_access = DRM_XE_ATOMIC_UNDEFINED, + .default_pat_index = op->map.pat_index, + .pat_index = op->map.pat_index, + }; + flags |= op->map.read_only ? VMA_CREATE_FLAG_READ_ONLY : 0; flags |= op->map.is_null ? @@ -2631,7 +2810,7 @@ static int vm_bind_ioctl_ops_parse(struct xe_vm *vm, struct drm_gpuva_ops *ops, flags |= op->map.is_cpu_addr_mirror ? VMA_CREATE_FLAG_IS_SYSTEM_ALLOCATOR : 0; - vma = new_vma(vm, &op->base.map, op->map.pat_index, + vma = new_vma(vm, &op->base.map, &default_attr, flags); if (IS_ERR(vma)) return PTR_ERR(vma); @@ -2659,8 +2838,12 @@ static int vm_bind_ioctl_ops_parse(struct xe_vm *vm, struct drm_gpuva_ops *ops, end = op->base.remap.next->va.addr; if (xe_vma_is_cpu_addr_mirror(old) && - xe_svm_has_mapping(vm, start, end)) - return -EBUSY; + xe_svm_has_mapping(vm, start, end)) { + if (vops->flags & XE_VMA_OPS_FLAG_MADVISE) + xe_svm_unmap_address_range(vm, start, end); + else + return -EBUSY; + } op->remap.start = xe_vma_start(old); op->remap.range = xe_vma_size(old); @@ -2679,7 +2862,7 @@ static int vm_bind_ioctl_ops_parse(struct xe_vm *vm, struct drm_gpuva_ops *ops, if (op->base.remap.prev) { vma = new_vma(vm, op->base.remap.prev, - old->pat_index, flags); + &old->attr, flags); if (IS_ERR(vma)) return PTR_ERR(vma); @@ -2709,7 +2892,7 @@ static int vm_bind_ioctl_ops_parse(struct xe_vm *vm, struct drm_gpuva_ops *ops, if (op->base.remap.next) { vma = new_vma(vm, op->base.remap.next, - old->pat_index, flags); + &old->attr, flags); if (IS_ERR(vma)) return PTR_ERR(vma); @@ -2896,30 +3079,26 @@ static int prefetch_ranges(struct xe_vm *vm, struct xe_vma_op *op) { bool devmem_possible = IS_DGFX(vm->xe) && IS_ENABLED(CONFIG_DRM_XE_PAGEMAP); struct xe_vma *vma = gpuva_to_vma(op->base.prefetch.va); + struct xe_tile *tile = op->prefetch_range.tile; int err = 0; struct xe_svm_range *svm_range; struct drm_gpusvm_ctx ctx = {}; - struct xe_tile *tile; unsigned long i; - u32 region; if (!xe_vma_is_cpu_addr_mirror(vma)) return 0; - region = op->prefetch_range.region; - ctx.read_only = xe_vma_read_only(vma); ctx.devmem_possible = devmem_possible; ctx.check_pages_threshold = devmem_possible ? SZ_64K : 0; /* TODO: Threading the migration */ xa_for_each(&op->prefetch_range.range, i, svm_range) { - if (!region) + if (!tile) xe_svm_range_migrate_to_smem(vm, svm_range); - if (xe_svm_range_needs_migrate_to_vram(svm_range, vma, region)) { - tile = &vm->xe->tiles[region_to_mem_type[region] - XE_PL_VRAM0]; + if (xe_svm_range_needs_migrate_to_vram(svm_range, vma, !!tile)) { err = xe_svm_alloc_vram(tile, svm_range, &ctx); if (err) { drm_dbg(&vm->xe->drm, "VRAM allocation failed, retry from userspace, asid=%u, gpusvm=%p, errno=%pe\n", @@ -2982,12 +3161,11 @@ static int op_lock_and_prep(struct drm_exec *exec, struct xe_vm *vm, struct xe_vma *vma = gpuva_to_vma(op->base.prefetch.va); u32 region; - if (xe_vma_is_cpu_addr_mirror(vma)) - region = op->prefetch_range.region; - else + if (!xe_vma_is_cpu_addr_mirror(vma)) { region = op->prefetch.region; - - xe_assert(vm->xe, region <= ARRAY_SIZE(region_to_mem_type)); + xe_assert(vm->xe, region == DRM_XE_CONSULT_MEM_ADVISE_PREF_LOC || + region <= ARRAY_SIZE(region_to_mem_type)); + } err = vma_lock_and_validate(exec, gpuva_to_vma(op->base.prefetch.va), @@ -3405,8 +3583,8 @@ static int vm_bind_ioctl_check_args(struct xe_device *xe, struct xe_vm *vm, op == DRM_XE_VM_BIND_OP_PREFETCH) || XE_IOCTL_DBG(xe, prefetch_region && op != DRM_XE_VM_BIND_OP_PREFETCH) || - XE_IOCTL_DBG(xe, !(BIT(prefetch_region) & - xe->info.mem_region_mask)) || + XE_IOCTL_DBG(xe, (prefetch_region != DRM_XE_CONSULT_MEM_ADVISE_PREF_LOC && + !(BIT(prefetch_region) & xe->info.mem_region_mask))) || XE_IOCTL_DBG(xe, obj && op == DRM_XE_VM_BIND_OP_UNMAP)) { err = -EINVAL; @@ -3428,6 +3606,7 @@ static int vm_bind_ioctl_check_args(struct xe_device *xe, struct xe_vm *vm, free_bind_ops: if (args->num_binds > 1) kvfree(*bind_ops); + *bind_ops = NULL; return err; } @@ -3534,7 +3713,7 @@ int xe_vm_bind_ioctl(struct drm_device *dev, void *data, struct drm_file *file) struct xe_exec_queue *q = NULL; u32 num_syncs, num_ufence = 0; struct xe_sync_entry *syncs = NULL; - struct drm_xe_vm_bind_op *bind_ops; + struct drm_xe_vm_bind_op *bind_ops = NULL; struct xe_vma_ops vops; struct dma_fence *fence; int err; @@ -3552,7 +3731,7 @@ int xe_vm_bind_ioctl(struct drm_device *dev, void *data, struct drm_file *file) q = xe_exec_queue_lookup(xef, args->exec_queue_id); if (XE_IOCTL_DBG(xe, !q)) { err = -ENOENT; - goto put_vm; + goto free_bind_ops; } if (XE_IOCTL_DBG(xe, !(q->flags & EXEC_QUEUE_FLAG_VM))) { @@ -3598,7 +3777,7 @@ int xe_vm_bind_ioctl(struct drm_device *dev, void *data, struct drm_file *file) __GFP_RETRY_MAYFAIL | __GFP_NOWARN); if (!ops) { err = -ENOMEM; - goto release_vm_lock; + goto free_bos; } } @@ -3732,17 +3911,20 @@ free_syncs: put_obj: for (i = 0; i < args->num_binds; ++i) xe_bo_put(bos[i]); + + kvfree(ops); +free_bos: + kvfree(bos); release_vm_lock: up_write(&vm->lock); put_exec_queue: if (q) xe_exec_queue_put(q); -put_vm: - xe_vm_put(vm); - kvfree(bos); - kvfree(ops); +free_bind_ops: if (args->num_binds > 1) kvfree(bind_ops); +put_vm: + xe_vm_put(vm); return err; } @@ -3850,7 +4032,7 @@ void xe_vm_unlock(struct xe_vm *vm) } /** - * xe_vm_range_tilemask_tlb_invalidation - Issue a TLB invalidation on this tilemask for an + * xe_vm_range_tilemask_tlb_inval - Issue a TLB invalidation on this tilemask for an * address range * @vm: The VM * @start: start address @@ -3861,10 +4043,11 @@ void xe_vm_unlock(struct xe_vm *vm) * * Returns 0 for success, negative error code otherwise. */ -int xe_vm_range_tilemask_tlb_invalidation(struct xe_vm *vm, u64 start, - u64 end, u8 tile_mask) +int xe_vm_range_tilemask_tlb_inval(struct xe_vm *vm, u64 start, + u64 end, u8 tile_mask) { - struct xe_gt_tlb_invalidation_fence fence[XE_MAX_TILES_PER_DEVICE * XE_MAX_GT_PER_TILE]; + struct xe_tlb_inval_fence + fence[XE_MAX_TILES_PER_DEVICE * XE_MAX_GT_PER_TILE]; struct xe_tile *tile; u32 fence_id = 0; u8 id; @@ -3874,39 +4057,36 @@ int xe_vm_range_tilemask_tlb_invalidation(struct xe_vm *vm, u64 start, return 0; for_each_tile(tile, vm->xe, id) { - if (tile_mask & BIT(id)) { - xe_gt_tlb_invalidation_fence_init(tile->primary_gt, - &fence[fence_id], true); - - err = xe_gt_tlb_invalidation_range(tile->primary_gt, - &fence[fence_id], - start, - end, - vm->usm.asid); - if (err) - goto wait; - ++fence_id; + if (!(tile_mask & BIT(id))) + continue; - if (!tile->media_gt) - continue; + xe_tlb_inval_fence_init(&tile->primary_gt->tlb_inval, + &fence[fence_id], true); - xe_gt_tlb_invalidation_fence_init(tile->media_gt, - &fence[fence_id], true); + err = xe_tlb_inval_range(&tile->primary_gt->tlb_inval, + &fence[fence_id], start, end, + vm->usm.asid); + if (err) + goto wait; + ++fence_id; - err = xe_gt_tlb_invalidation_range(tile->media_gt, - &fence[fence_id], - start, - end, - vm->usm.asid); - if (err) - goto wait; - ++fence_id; - } + if (!tile->media_gt) + continue; + + xe_tlb_inval_fence_init(&tile->media_gt->tlb_inval, + &fence[fence_id], true); + + err = xe_tlb_inval_range(&tile->media_gt->tlb_inval, + &fence[fence_id], start, end, + vm->usm.asid); + if (err) + goto wait; + ++fence_id; } wait: for (id = 0; id < fence_id; ++id) - xe_gt_tlb_invalidation_fence_wait(&fence[id]); + xe_tlb_inval_fence_wait(&fence[id]); return err; } @@ -3965,8 +4145,8 @@ int xe_vm_invalidate_vma(struct xe_vma *vma) xe_device_wmb(xe); - ret = xe_vm_range_tilemask_tlb_invalidation(xe_vma_vm(vma), xe_vma_start(vma), - xe_vma_end(vma), tile_mask); + ret = xe_vm_range_tilemask_tlb_inval(xe_vma_vm(vma), xe_vma_start(vma), + xe_vma_end(vma), tile_mask); /* WRITE_ONCE pairs with READ_ONCE in xe_vm_has_valid_gpu_mapping() */ WRITE_ONCE(vma->tile_invalidated, vma->tile_mask); @@ -4168,3 +4348,223 @@ void xe_vm_snapshot_free(struct xe_vm_snapshot *snap) } kvfree(snap); } + +/** + * xe_vma_need_vram_for_atomic - Check if VMA needs VRAM migration for atomic operations + * @xe: Pointer to the XE device structure + * @vma: Pointer to the virtual memory area (VMA) structure + * @is_atomic: In pagefault path and atomic operation + * + * This function determines whether the given VMA needs to be migrated to + * VRAM in order to do atomic GPU operation. + * + * Return: + * 1 - Migration to VRAM is required + * 0 - Migration is not required + * -EACCES - Invalid access for atomic memory attr + * + */ +int xe_vma_need_vram_for_atomic(struct xe_device *xe, struct xe_vma *vma, bool is_atomic) +{ + u32 atomic_access = xe_vma_bo(vma) ? xe_vma_bo(vma)->attr.atomic_access : + vma->attr.atomic_access; + + if (!IS_DGFX(xe) || !is_atomic) + return false; + + /* + * NOTE: The checks implemented here are platform-specific. For + * instance, on a device supporting CXL atomics, these would ideally + * work universally without additional handling. + */ + switch (atomic_access) { + case DRM_XE_ATOMIC_DEVICE: + return !xe->info.has_device_atomics_on_smem; + + case DRM_XE_ATOMIC_CPU: + return -EACCES; + + case DRM_XE_ATOMIC_UNDEFINED: + case DRM_XE_ATOMIC_GLOBAL: + default: + return 1; + } +} + +static int xe_vm_alloc_vma(struct xe_vm *vm, + struct drm_gpuvm_map_req *map_req, + bool is_madvise) +{ + struct xe_vma_ops vops; + struct drm_gpuva_ops *ops = NULL; + struct drm_gpuva_op *__op; + bool is_cpu_addr_mirror = false; + bool remap_op = false; + struct xe_vma_mem_attr tmp_attr; + u16 default_pat; + int err; + + lockdep_assert_held_write(&vm->lock); + + if (is_madvise) + ops = drm_gpuvm_madvise_ops_create(&vm->gpuvm, map_req); + else + ops = drm_gpuvm_sm_map_ops_create(&vm->gpuvm, map_req); + + if (IS_ERR(ops)) + return PTR_ERR(ops); + + if (list_empty(&ops->list)) { + err = 0; + goto free_ops; + } + + drm_gpuva_for_each_op(__op, ops) { + struct xe_vma_op *op = gpuva_op_to_vma_op(__op); + struct xe_vma *vma = NULL; + + if (!is_madvise) { + if (__op->op == DRM_GPUVA_OP_UNMAP) { + vma = gpuva_to_vma(op->base.unmap.va); + XE_WARN_ON(!xe_vma_has_default_mem_attrs(vma)); + default_pat = vma->attr.default_pat_index; + } + + if (__op->op == DRM_GPUVA_OP_REMAP) { + vma = gpuva_to_vma(op->base.remap.unmap->va); + default_pat = vma->attr.default_pat_index; + } + + if (__op->op == DRM_GPUVA_OP_MAP) { + op->map.is_cpu_addr_mirror = true; + op->map.pat_index = default_pat; + } + } else { + if (__op->op == DRM_GPUVA_OP_REMAP) { + vma = gpuva_to_vma(op->base.remap.unmap->va); + xe_assert(vm->xe, !remap_op); + xe_assert(vm->xe, xe_vma_has_no_bo(vma)); + remap_op = true; + + if (xe_vma_is_cpu_addr_mirror(vma)) + is_cpu_addr_mirror = true; + else + is_cpu_addr_mirror = false; + } + + if (__op->op == DRM_GPUVA_OP_MAP) { + xe_assert(vm->xe, remap_op); + remap_op = false; + /* + * In case of madvise ops DRM_GPUVA_OP_MAP is + * always after DRM_GPUVA_OP_REMAP, so ensure + * we assign op->map.is_cpu_addr_mirror true + * if REMAP is for xe_vma_is_cpu_addr_mirror vma + */ + op->map.is_cpu_addr_mirror = is_cpu_addr_mirror; + } + } + print_op(vm->xe, __op); + } + + xe_vma_ops_init(&vops, vm, NULL, NULL, 0); + + if (is_madvise) + vops.flags |= XE_VMA_OPS_FLAG_MADVISE; + + err = vm_bind_ioctl_ops_parse(vm, ops, &vops); + if (err) + goto unwind_ops; + + xe_vm_lock(vm, false); + + drm_gpuva_for_each_op(__op, ops) { + struct xe_vma_op *op = gpuva_op_to_vma_op(__op); + struct xe_vma *vma; + + if (__op->op == DRM_GPUVA_OP_UNMAP) { + vma = gpuva_to_vma(op->base.unmap.va); + /* There should be no unmap for madvise */ + if (is_madvise) + XE_WARN_ON("UNEXPECTED UNMAP"); + + xe_vma_destroy(vma, NULL); + } else if (__op->op == DRM_GPUVA_OP_REMAP) { + vma = gpuva_to_vma(op->base.remap.unmap->va); + /* In case of madvise ops Store attributes for REMAP UNMAPPED + * VMA, so they can be assigned to newly MAP created vma. + */ + if (is_madvise) + tmp_attr = vma->attr; + + xe_vma_destroy(gpuva_to_vma(op->base.remap.unmap->va), NULL); + } else if (__op->op == DRM_GPUVA_OP_MAP) { + vma = op->map.vma; + /* In case of madvise call, MAP will always be follwed by REMAP. + * Therefore temp_attr will always have sane values, making it safe to + * copy them to new vma. + */ + if (is_madvise) + vma->attr = tmp_attr; + } + } + + xe_vm_unlock(vm); + drm_gpuva_ops_free(&vm->gpuvm, ops); + return 0; + +unwind_ops: + vm_bind_ioctl_ops_unwind(vm, &ops, 1); +free_ops: + drm_gpuva_ops_free(&vm->gpuvm, ops); + return err; +} + +/** + * xe_vm_alloc_madvise_vma - Allocate VMA's with madvise ops + * @vm: Pointer to the xe_vm structure + * @start: Starting input address + * @range: Size of the input range + * + * This function splits existing vma to create new vma for user provided input range + * + * Return: 0 if success + */ +int xe_vm_alloc_madvise_vma(struct xe_vm *vm, uint64_t start, uint64_t range) +{ + struct drm_gpuvm_map_req map_req = { + .map.va.addr = start, + .map.va.range = range, + }; + + lockdep_assert_held_write(&vm->lock); + + vm_dbg(&vm->xe->drm, "MADVISE_OPS_CREATE: addr=0x%016llx, size=0x%016llx", start, range); + + return xe_vm_alloc_vma(vm, &map_req, true); +} + +/** + * xe_vm_alloc_cpu_addr_mirror_vma - Allocate CPU addr mirror vma + * @vm: Pointer to the xe_vm structure + * @start: Starting input address + * @range: Size of the input range + * + * This function splits/merges existing vma to create new vma for user provided input range + * + * Return: 0 if success + */ +int xe_vm_alloc_cpu_addr_mirror_vma(struct xe_vm *vm, uint64_t start, uint64_t range) +{ + struct drm_gpuvm_map_req map_req = { + .map.va.addr = start, + .map.va.range = range, + }; + + lockdep_assert_held_write(&vm->lock); + + vm_dbg(&vm->xe->drm, "CPU_ADDR_MIRROR_VMA_OPS_CREATE: addr=0x%016llx, size=0x%016llx", + start, range); + + return xe_vm_alloc_vma(vm, &map_req, false); +} diff --git a/drivers/gpu/drm/xe/xe_vm.h b/drivers/gpu/drm/xe/xe_vm.h index 3475a118f666..b3e5bec0fa58 100644 --- a/drivers/gpu/drm/xe/xe_vm.h +++ b/drivers/gpu/drm/xe/xe_vm.h @@ -26,7 +26,7 @@ struct xe_sync_entry; struct xe_svm_range; struct drm_exec; -struct xe_vm *xe_vm_create(struct xe_device *xe, u32 flags); +struct xe_vm *xe_vm_create(struct xe_device *xe, u32 flags, struct xe_file *xef); struct xe_vm *xe_vm_lookup(struct xe_file *xef, u32 id); int xe_vma_cmp_vma_cb(const void *key, const struct rb_node *node); @@ -66,6 +66,8 @@ static inline bool xe_vm_is_closed_or_banned(struct xe_vm *vm) struct xe_vma * xe_vm_find_overlapping_vma(struct xe_vm *vm, u64 start, u64 range); +bool xe_vma_has_default_mem_attrs(struct xe_vma *vma); + /** * xe_vm_has_scratch() - Whether the vm is configured for scratch PTEs * @vm: The vm @@ -171,6 +173,12 @@ static inline bool xe_vma_is_userptr(struct xe_vma *vma) struct xe_vma *xe_vm_find_vma_by_addr(struct xe_vm *vm, u64 page_addr); +int xe_vma_need_vram_for_atomic(struct xe_device *xe, struct xe_vma *vma, bool is_atomic); + +int xe_vm_alloc_madvise_vma(struct xe_vm *vm, uint64_t addr, uint64_t size); + +int xe_vm_alloc_cpu_addr_mirror_vma(struct xe_vm *vm, uint64_t addr, uint64_t size); + /** * to_userptr_vma() - Return a pointer to an embedding userptr vma * @vma: Pointer to the embedded struct xe_vma @@ -191,7 +199,7 @@ int xe_vm_destroy_ioctl(struct drm_device *dev, void *data, struct drm_file *file); int xe_vm_bind_ioctl(struct drm_device *dev, void *data, struct drm_file *file); - +int xe_vm_query_vmas_attrs_ioctl(struct drm_device *dev, void *data, struct drm_file *file); void xe_vm_close_and_put(struct xe_vm *vm); static inline bool xe_vm_in_fault_mode(struct xe_vm *vm) @@ -228,8 +236,8 @@ struct dma_fence *xe_vm_range_rebind(struct xe_vm *vm, struct dma_fence *xe_vm_range_unbind(struct xe_vm *vm, struct xe_svm_range *range); -int xe_vm_range_tilemask_tlb_invalidation(struct xe_vm *vm, u64 start, - u64 end, u8 tile_mask); +int xe_vm_range_tilemask_tlb_inval(struct xe_vm *vm, u64 start, + u64 end, u8 tile_mask); int xe_vm_invalidate_vma(struct xe_vma *vma); @@ -315,22 +323,14 @@ void xe_vm_snapshot_free(struct xe_vm_snapshot *snap); * Register this task as currently making bos resident for the vm. Intended * to avoid eviction by the same task of shared bos bound to the vm. * Call with the vm's resv lock held. - * - * Return: A pin cookie that should be used for xe_vm_clear_validating(). */ -static inline struct pin_cookie xe_vm_set_validating(struct xe_vm *vm, - bool allow_res_evict) +static inline void xe_vm_set_validating(struct xe_vm *vm, bool allow_res_evict) { - struct pin_cookie cookie = {}; - if (vm && !allow_res_evict) { xe_vm_assert_held(vm); - cookie = lockdep_pin_lock(&xe_vm_resv(vm)->lock.base); /* Pairs with READ_ONCE in xe_vm_is_validating() */ WRITE_ONCE(vm->validating, current); } - - return cookie; } /** @@ -338,17 +338,14 @@ static inline struct pin_cookie xe_vm_set_validating(struct xe_vm *vm, * @vm: Pointer to the vm or NULL * @allow_res_evict: Eviction from @vm was allowed. Must be set to the same * value as for xe_vm_set_validation(). - * @cookie: Cookie obtained from xe_vm_set_validating(). * * Register this task as currently making bos resident for the vm. Intended * to avoid eviction by the same task of shared bos bound to the vm. * Call with the vm's resv lock held. */ -static inline void xe_vm_clear_validating(struct xe_vm *vm, bool allow_res_evict, - struct pin_cookie cookie) +static inline void xe_vm_clear_validating(struct xe_vm *vm, bool allow_res_evict) { if (vm && !allow_res_evict) { - lockdep_unpin_lock(&xe_vm_resv(vm)->lock.base, cookie); /* Pairs with READ_ONCE in xe_vm_is_validating() */ WRITE_ONCE(vm->validating, NULL); } diff --git a/drivers/gpu/drm/xe/xe_vm_madvise.c b/drivers/gpu/drm/xe/xe_vm_madvise.c new file mode 100644 index 000000000000..09c5783ee523 --- /dev/null +++ b/drivers/gpu/drm/xe/xe_vm_madvise.c @@ -0,0 +1,445 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright © 2025 Intel Corporation + */ + +#include "xe_vm_madvise.h" + +#include <linux/nospec.h> +#include <drm/xe_drm.h> + +#include "xe_bo.h" +#include "xe_pat.h" +#include "xe_pt.h" +#include "xe_svm.h" + +struct xe_vmas_in_madvise_range { + u64 addr; + u64 range; + struct xe_vma **vmas; + int num_vmas; + bool has_svm_vmas; + bool has_bo_vmas; + bool has_userptr_vmas; +}; + +static int get_vmas(struct xe_vm *vm, struct xe_vmas_in_madvise_range *madvise_range) +{ + u64 addr = madvise_range->addr; + u64 range = madvise_range->range; + + struct xe_vma **__vmas; + struct drm_gpuva *gpuva; + int max_vmas = 8; + + lockdep_assert_held(&vm->lock); + + madvise_range->num_vmas = 0; + madvise_range->vmas = kmalloc_array(max_vmas, sizeof(*madvise_range->vmas), GFP_KERNEL); + if (!madvise_range->vmas) + return -ENOMEM; + + vm_dbg(&vm->xe->drm, "VMA's in range: start=0x%016llx, end=0x%016llx", addr, addr + range); + + drm_gpuvm_for_each_va_range(gpuva, &vm->gpuvm, addr, addr + range) { + struct xe_vma *vma = gpuva_to_vma(gpuva); + + if (xe_vma_bo(vma)) + madvise_range->has_bo_vmas = true; + else if (xe_vma_is_cpu_addr_mirror(vma)) + madvise_range->has_svm_vmas = true; + else if (xe_vma_is_userptr(vma)) + madvise_range->has_userptr_vmas = true; + + if (madvise_range->num_vmas == max_vmas) { + max_vmas <<= 1; + __vmas = krealloc(madvise_range->vmas, + max_vmas * sizeof(*madvise_range->vmas), + GFP_KERNEL); + if (!__vmas) { + kfree(madvise_range->vmas); + return -ENOMEM; + } + madvise_range->vmas = __vmas; + } + + madvise_range->vmas[madvise_range->num_vmas] = vma; + (madvise_range->num_vmas)++; + } + + if (!madvise_range->num_vmas) + kfree(madvise_range->vmas); + + vm_dbg(&vm->xe->drm, "madvise_range-num_vmas = %d\n", madvise_range->num_vmas); + + return 0; +} + +static void madvise_preferred_mem_loc(struct xe_device *xe, struct xe_vm *vm, + struct xe_vma **vmas, int num_vmas, + struct drm_xe_madvise *op) +{ + int i; + + xe_assert(vm->xe, op->type == DRM_XE_MEM_RANGE_ATTR_PREFERRED_LOC); + + for (i = 0; i < num_vmas; i++) { + /*TODO: Extend attributes to bo based vmas */ + if ((vmas[i]->attr.preferred_loc.devmem_fd == op->preferred_mem_loc.devmem_fd && + vmas[i]->attr.preferred_loc.migration_policy == + op->preferred_mem_loc.migration_policy) || + !xe_vma_is_cpu_addr_mirror(vmas[i])) { + vmas[i]->skip_invalidation = true; + } else { + vmas[i]->skip_invalidation = false; + vmas[i]->attr.preferred_loc.devmem_fd = op->preferred_mem_loc.devmem_fd; + /* Till multi-device support is not added migration_policy + * is of no use and can be ignored. + */ + vmas[i]->attr.preferred_loc.migration_policy = + op->preferred_mem_loc.migration_policy; + } + } +} + +static void madvise_atomic(struct xe_device *xe, struct xe_vm *vm, + struct xe_vma **vmas, int num_vmas, + struct drm_xe_madvise *op) +{ + struct xe_bo *bo; + int i; + + xe_assert(vm->xe, op->type == DRM_XE_MEM_RANGE_ATTR_ATOMIC); + xe_assert(vm->xe, op->atomic.val <= DRM_XE_ATOMIC_CPU); + + for (i = 0; i < num_vmas; i++) { + if (xe_vma_is_userptr(vmas[i]) && + !(op->atomic.val == DRM_XE_ATOMIC_DEVICE && + xe->info.has_device_atomics_on_smem)) { + vmas[i]->skip_invalidation = true; + continue; + } + + if (vmas[i]->attr.atomic_access == op->atomic.val) { + vmas[i]->skip_invalidation = true; + } else { + vmas[i]->skip_invalidation = false; + vmas[i]->attr.atomic_access = op->atomic.val; + } + + vmas[i]->attr.atomic_access = op->atomic.val; + + bo = xe_vma_bo(vmas[i]); + if (!bo || bo->attr.atomic_access == op->atomic.val) + continue; + + vmas[i]->skip_invalidation = false; + xe_bo_assert_held(bo); + bo->attr.atomic_access = op->atomic.val; + + /* Invalidate cpu page table, so bo can migrate to smem in next access */ + if (xe_bo_is_vram(bo) && + (bo->attr.atomic_access == DRM_XE_ATOMIC_CPU || + bo->attr.atomic_access == DRM_XE_ATOMIC_GLOBAL)) + ttm_bo_unmap_virtual(&bo->ttm); + } +} + +static void madvise_pat_index(struct xe_device *xe, struct xe_vm *vm, + struct xe_vma **vmas, int num_vmas, + struct drm_xe_madvise *op) +{ + int i; + + xe_assert(vm->xe, op->type == DRM_XE_MEM_RANGE_ATTR_PAT); + + for (i = 0; i < num_vmas; i++) { + if (vmas[i]->attr.pat_index == op->pat_index.val) { + vmas[i]->skip_invalidation = true; + } else { + vmas[i]->skip_invalidation = false; + vmas[i]->attr.pat_index = op->pat_index.val; + } + } +} + +typedef void (*madvise_func)(struct xe_device *xe, struct xe_vm *vm, + struct xe_vma **vmas, int num_vmas, + struct drm_xe_madvise *op); + +static const madvise_func madvise_funcs[] = { + [DRM_XE_MEM_RANGE_ATTR_PREFERRED_LOC] = madvise_preferred_mem_loc, + [DRM_XE_MEM_RANGE_ATTR_ATOMIC] = madvise_atomic, + [DRM_XE_MEM_RANGE_ATTR_PAT] = madvise_pat_index, +}; + +static u8 xe_zap_ptes_in_madvise_range(struct xe_vm *vm, u64 start, u64 end) +{ + struct drm_gpuva *gpuva; + struct xe_tile *tile; + u8 id, tile_mask = 0; + + lockdep_assert_held_write(&vm->lock); + + /* Wait for pending binds */ + if (dma_resv_wait_timeout(xe_vm_resv(vm), DMA_RESV_USAGE_BOOKKEEP, + false, MAX_SCHEDULE_TIMEOUT) <= 0) + XE_WARN_ON(1); + + drm_gpuvm_for_each_va_range(gpuva, &vm->gpuvm, start, end) { + struct xe_vma *vma = gpuva_to_vma(gpuva); + + if (vma->skip_invalidation || xe_vma_is_null(vma)) + continue; + + if (xe_vma_is_cpu_addr_mirror(vma)) { + tile_mask |= xe_svm_ranges_zap_ptes_in_range(vm, + xe_vma_start(vma), + xe_vma_end(vma)); + } else { + for_each_tile(tile, vm->xe, id) { + if (xe_pt_zap_ptes(tile, vma)) { + tile_mask |= BIT(id); + + /* + * WRITE_ONCE pairs with READ_ONCE + * in xe_vm_has_valid_gpu_mapping() + */ + WRITE_ONCE(vma->tile_invalidated, + vma->tile_invalidated | BIT(id)); + } + } + } + } + + return tile_mask; +} + +static int xe_vm_invalidate_madvise_range(struct xe_vm *vm, u64 start, u64 end) +{ + u8 tile_mask = xe_zap_ptes_in_madvise_range(vm, start, end); + + if (!tile_mask) + return 0; + + xe_device_wmb(vm->xe); + + return xe_vm_range_tilemask_tlb_inval(vm, start, end, tile_mask); +} + +static bool madvise_args_are_sane(struct xe_device *xe, const struct drm_xe_madvise *args) +{ + if (XE_IOCTL_DBG(xe, !args)) + return false; + + if (XE_IOCTL_DBG(xe, !IS_ALIGNED(args->start, SZ_4K))) + return false; + + if (XE_IOCTL_DBG(xe, !IS_ALIGNED(args->range, SZ_4K))) + return false; + + if (XE_IOCTL_DBG(xe, args->range < SZ_4K)) + return false; + + switch (args->type) { + case DRM_XE_MEM_RANGE_ATTR_PREFERRED_LOC: + { + s32 fd = (s32)args->preferred_mem_loc.devmem_fd; + + if (XE_IOCTL_DBG(xe, fd < DRM_XE_PREFERRED_LOC_DEFAULT_SYSTEM)) + return false; + + if (XE_IOCTL_DBG(xe, args->preferred_mem_loc.migration_policy > + DRM_XE_MIGRATE_ONLY_SYSTEM_PAGES)) + return false; + + if (XE_IOCTL_DBG(xe, args->preferred_mem_loc.pad)) + return false; + + if (XE_IOCTL_DBG(xe, args->atomic.reserved)) + return false; + break; + } + case DRM_XE_MEM_RANGE_ATTR_ATOMIC: + if (XE_IOCTL_DBG(xe, args->atomic.val > DRM_XE_ATOMIC_CPU)) + return false; + + if (XE_IOCTL_DBG(xe, args->atomic.pad)) + return false; + + if (XE_IOCTL_DBG(xe, args->atomic.reserved)) + return false; + + break; + case DRM_XE_MEM_RANGE_ATTR_PAT: + { + u16 coh_mode = xe_pat_index_get_coh_mode(xe, args->pat_index.val); + + if (XE_IOCTL_DBG(xe, !coh_mode)) + return false; + + if (XE_WARN_ON(coh_mode > XE_COH_AT_LEAST_1WAY)) + return false; + + if (XE_IOCTL_DBG(xe, args->pat_index.pad)) + return false; + + if (XE_IOCTL_DBG(xe, args->pat_index.reserved)) + return false; + break; + } + default: + if (XE_IOCTL_DBG(xe, 1)) + return false; + } + + if (XE_IOCTL_DBG(xe, args->reserved[0] || args->reserved[1])) + return false; + + return true; +} + +static bool check_bo_args_are_sane(struct xe_vm *vm, struct xe_vma **vmas, + int num_vmas, u32 atomic_val) +{ + struct xe_device *xe = vm->xe; + struct xe_bo *bo; + int i; + + for (i = 0; i < num_vmas; i++) { + bo = xe_vma_bo(vmas[i]); + if (!bo) + continue; + /* + * NOTE: The following atomic checks are platform-specific. For example, + * if a device supports CXL atomics, these may not be necessary or + * may behave differently. + */ + if (XE_IOCTL_DBG(xe, atomic_val == DRM_XE_ATOMIC_CPU && + !(bo->flags & XE_BO_FLAG_SYSTEM))) + return false; + + if (XE_IOCTL_DBG(xe, atomic_val == DRM_XE_ATOMIC_DEVICE && + !(bo->flags & XE_BO_FLAG_VRAM0) && + !(bo->flags & XE_BO_FLAG_VRAM1) && + !(bo->flags & XE_BO_FLAG_SYSTEM && + xe->info.has_device_atomics_on_smem))) + return false; + + if (XE_IOCTL_DBG(xe, atomic_val == DRM_XE_ATOMIC_GLOBAL && + (!(bo->flags & XE_BO_FLAG_SYSTEM) || + (!(bo->flags & XE_BO_FLAG_VRAM0) && + !(bo->flags & XE_BO_FLAG_VRAM1))))) + return false; + } + return true; +} +/** + * xe_vm_madvise_ioctl - Handle MADVise ioctl for a VM + * @dev: DRM device pointer + * @data: Pointer to ioctl data (drm_xe_madvise*) + * @file: DRM file pointer + * + * Handles the MADVISE ioctl to provide memory advice for vma's within + * input range. + * + * Return: 0 on success or a negative error code on failure. + */ +int xe_vm_madvise_ioctl(struct drm_device *dev, void *data, struct drm_file *file) +{ + struct xe_device *xe = to_xe_device(dev); + struct xe_file *xef = to_xe_file(file); + struct drm_xe_madvise *args = data; + struct xe_vmas_in_madvise_range madvise_range = {.addr = args->start, + .range = args->range, }; + struct xe_vm *vm; + struct drm_exec exec; + int err, attr_type; + + vm = xe_vm_lookup(xef, args->vm_id); + if (XE_IOCTL_DBG(xe, !vm)) + return -EINVAL; + + if (!madvise_args_are_sane(vm->xe, args)) { + err = -EINVAL; + goto put_vm; + } + + xe_svm_flush(vm); + + err = down_write_killable(&vm->lock); + if (err) + goto put_vm; + + if (XE_IOCTL_DBG(xe, xe_vm_is_closed_or_banned(vm))) { + err = -ENOENT; + goto unlock_vm; + } + + err = xe_vm_alloc_madvise_vma(vm, args->start, args->range); + if (err) + goto unlock_vm; + + err = get_vmas(vm, &madvise_range); + if (err || !madvise_range.num_vmas) + goto unlock_vm; + + if (madvise_range.has_bo_vmas) { + if (args->type == DRM_XE_MEM_RANGE_ATTR_ATOMIC) { + if (!check_bo_args_are_sane(vm, madvise_range.vmas, + madvise_range.num_vmas, + args->atomic.val)) { + err = -EINVAL; + goto unlock_vm; + } + } + + drm_exec_init(&exec, DRM_EXEC_IGNORE_DUPLICATES | DRM_EXEC_INTERRUPTIBLE_WAIT, 0); + drm_exec_until_all_locked(&exec) { + for (int i = 0; i < madvise_range.num_vmas; i++) { + struct xe_bo *bo = xe_vma_bo(madvise_range.vmas[i]); + + if (!bo) + continue; + err = drm_exec_lock_obj(&exec, &bo->ttm.base); + drm_exec_retry_on_contention(&exec); + if (err) + goto err_fini; + } + } + } + + if (madvise_range.has_userptr_vmas) { + err = down_read_interruptible(&vm->userptr.notifier_lock); + if (err) + goto err_fini; + } + + if (madvise_range.has_svm_vmas) { + err = down_read_interruptible(&vm->svm.gpusvm.notifier_lock); + if (err) + goto unlock_userptr; + } + + attr_type = array_index_nospec(args->type, ARRAY_SIZE(madvise_funcs)); + madvise_funcs[attr_type](xe, vm, madvise_range.vmas, madvise_range.num_vmas, args); + + err = xe_vm_invalidate_madvise_range(vm, args->start, args->start + args->range); + + if (madvise_range.has_svm_vmas) + xe_svm_notifier_unlock(vm); + +unlock_userptr: + if (madvise_range.has_userptr_vmas) + up_read(&vm->userptr.notifier_lock); +err_fini: + if (madvise_range.has_bo_vmas) + drm_exec_fini(&exec); + kfree(madvise_range.vmas); + madvise_range.vmas = NULL; +unlock_vm: + up_write(&vm->lock); +put_vm: + xe_vm_put(vm); + return err; +} diff --git a/drivers/gpu/drm/xe/xe_vm_madvise.h b/drivers/gpu/drm/xe/xe_vm_madvise.h new file mode 100644 index 000000000000..b0e1fc445f23 --- /dev/null +++ b/drivers/gpu/drm/xe/xe_vm_madvise.h @@ -0,0 +1,15 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2025 Intel Corporation + */ + +#ifndef _XE_VM_MADVISE_H_ +#define _XE_VM_MADVISE_H_ + +struct drm_device; +struct drm_file; + +int xe_vm_madvise_ioctl(struct drm_device *dev, void *data, + struct drm_file *file); + +#endif diff --git a/drivers/gpu/drm/xe/xe_vm_types.h b/drivers/gpu/drm/xe/xe_vm_types.h index 8a07feef503b..b5108d010786 100644 --- a/drivers/gpu/drm/xe/xe_vm_types.h +++ b/drivers/gpu/drm/xe/xe_vm_types.h @@ -77,6 +77,44 @@ struct xe_userptr { #endif }; +/** + * struct xe_vma_mem_attr - memory attributes associated with vma + */ +struct xe_vma_mem_attr { + /** @preferred_loc: perferred memory_location */ + struct { + /** @preferred_loc.migration_policy: Pages migration policy */ + u32 migration_policy; + + /** + * @preferred_loc.devmem_fd: used for determining pagemap_fd + * requested by user DRM_XE_PREFERRED_LOC_DEFAULT_SYSTEM and + * DRM_XE_PREFERRED_LOC_DEFAULT_DEVICE mean system memory or + * closest device memory respectively. + */ + u32 devmem_fd; + } preferred_loc; + + /** + * @atomic_access: The atomic access type for the vma + * See %DRM_XE_VMA_ATOMIC_UNDEFINED, %DRM_XE_VMA_ATOMIC_DEVICE, + * %DRM_XE_VMA_ATOMIC_GLOBAL, and %DRM_XE_VMA_ATOMIC_CPU for possible + * values. These are defined in uapi/drm/xe_drm.h. + */ + u32 atomic_access; + + /** + * @default_pat_index: The pat index for VMA set during first bind by user. + */ + u16 default_pat_index; + + /** + * @pat_index: The pat index to use when encoding the PTEs for this vma. + * same as default_pat_index unless overwritten by madvise. + */ + u16 pat_index; +}; + struct xe_vma { /** @gpuva: Base GPUVA object */ struct drm_gpuva gpuva; @@ -126,15 +164,22 @@ struct xe_vma { u8 tile_staged; /** - * @pat_index: The pat index to use when encoding the PTEs for this vma. + * @skip_invalidation: Used in madvise to avoid invalidation + * if mem attributes doesn't change */ - u16 pat_index; + bool skip_invalidation; /** * @ufence: The user fence that was provided with MAP. * Needs to be signalled before UNMAP can be processed. */ struct xe_user_fence *ufence; + + /** + * @attr: The attributes of vma which determines the migration policy + * and encoding of the PTEs for this vma. + */ + struct xe_vma_mem_attr attr; }; /** @@ -395,8 +440,11 @@ struct xe_vma_op_prefetch_range { struct xarray range; /** @ranges_count: number of svm ranges to map */ u32 ranges_count; - /** @region: memory region to prefetch to */ - u32 region; + /** + * @tile: Pointer to the tile structure containing memory to prefetch. + * NULL if prefetch requested region is smem + */ + struct xe_tile *tile; }; /** enum xe_vma_op_flags - flags for VMA operation */ @@ -462,6 +510,7 @@ struct xe_vma_ops { struct xe_vm_pgtable_update_ops pt_update_ops[XE_MAX_TILES_PER_DEVICE]; /** @flag: signify the properties within xe_vma_ops*/ #define XE_VMA_OPS_FLAG_HAS_SVM_PREFETCH BIT(0) +#define XE_VMA_OPS_FLAG_MADVISE BIT(1) u32 flags; #ifdef TEST_VM_OPS_ERROR /** @inject_error: inject error to test error handling */ diff --git a/drivers/gpu/drm/xe/xe_vram.c b/drivers/gpu/drm/xe/xe_vram.c index e421a74fb87c..b44ebf50fedb 100644 --- a/drivers/gpu/drm/xe/xe_vram.c +++ b/drivers/gpu/drm/xe/xe_vram.c @@ -3,6 +3,7 @@ * Copyright © 2021-2024 Intel Corporation */ +#include <kunit/visibility.h> #include <linux/pci.h> #include <drm/drm_managed.h> @@ -19,7 +20,9 @@ #include "xe_mmio.h" #include "xe_module.h" #include "xe_sriov.h" +#include "xe_ttm_vram_mgr.h" #include "xe_vram.h" +#include "xe_vram_types.h" #define BAR_SIZE_SHIFT 20 @@ -136,7 +139,7 @@ static bool resource_is_valid(struct pci_dev *pdev, int bar) return true; } -static int determine_lmem_bar_size(struct xe_device *xe) +static int determine_lmem_bar_size(struct xe_device *xe, struct xe_vram_region *lmem_bar) { struct pci_dev *pdev = to_pci_dev(xe->drm.dev); @@ -147,16 +150,16 @@ static int determine_lmem_bar_size(struct xe_device *xe) resize_vram_bar(xe); - xe->mem.vram.io_start = pci_resource_start(pdev, LMEM_BAR); - xe->mem.vram.io_size = pci_resource_len(pdev, LMEM_BAR); - if (!xe->mem.vram.io_size) + lmem_bar->io_start = pci_resource_start(pdev, LMEM_BAR); + lmem_bar->io_size = pci_resource_len(pdev, LMEM_BAR); + if (!lmem_bar->io_size) return -EIO; /* XXX: Need to change when xe link code is ready */ - xe->mem.vram.dpa_base = 0; + lmem_bar->dpa_base = 0; /* set up a map to the total memory area. */ - xe->mem.vram.mapping = ioremap_wc(xe->mem.vram.io_start, xe->mem.vram.io_size); + lmem_bar->mapping = devm_ioremap_wc(&pdev->dev, lmem_bar->io_start, lmem_bar->io_size); return 0; } @@ -278,13 +281,71 @@ static void vram_fini(void *arg) struct xe_tile *tile; int id; - if (xe->mem.vram.mapping) - iounmap(xe->mem.vram.mapping); - - xe->mem.vram.mapping = NULL; + xe->mem.vram->mapping = NULL; for_each_tile(tile, xe, id) - tile->mem.vram.mapping = NULL; + tile->mem.vram->mapping = NULL; +} + +struct xe_vram_region *xe_vram_region_alloc(struct xe_device *xe, u8 id, u32 placement) +{ + struct xe_vram_region *vram; + struct drm_device *drm = &xe->drm; + + xe_assert(xe, id < xe->info.tile_count); + + vram = drmm_kzalloc(drm, sizeof(*vram), GFP_KERNEL); + if (!vram) + return NULL; + + vram->xe = xe; + vram->id = id; + vram->placement = placement; +#if defined(CONFIG_DRM_XE_PAGEMAP) + vram->migrate = xe->tiles[id].migrate; +#endif + return vram; +} + +static void print_vram_region_info(struct xe_device *xe, struct xe_vram_region *vram) +{ + struct drm_device *drm = &xe->drm; + + if (vram->io_size < vram->usable_size) + drm_info(drm, "Small BAR device\n"); + + drm_info(drm, + "VRAM[%u]: Actual physical size %pa, usable size exclude stolen %pa, CPU accessible size %pa\n", + vram->id, &vram->actual_physical_size, &vram->usable_size, &vram->io_size); + drm_info(drm, "VRAM[%u]: DPA range: [%pa-%llx], io range: [%pa-%llx]\n", + vram->id, &vram->dpa_base, vram->dpa_base + (u64)vram->actual_physical_size, + &vram->io_start, vram->io_start + (u64)vram->io_size); +} + +static int vram_region_init(struct xe_device *xe, struct xe_vram_region *vram, + struct xe_vram_region *lmem_bar, u64 offset, u64 usable_size, + u64 region_size, resource_size_t remain_io_size) +{ + /* Check if VRAM region is already initialized */ + if (vram->mapping) + return 0; + + vram->actual_physical_size = region_size; + vram->io_start = lmem_bar->io_start + offset; + vram->io_size = min_t(u64, usable_size, remain_io_size); + + if (!vram->io_size) { + drm_err(&xe->drm, "Tile without any CPU visible VRAM. Aborting.\n"); + return -ENODEV; + } + + vram->dpa_base = lmem_bar->dpa_base + offset; + vram->mapping = lmem_bar->mapping + offset; + vram->usable_size = usable_size; + + print_vram_region_info(xe, vram); + + return 0; } /** @@ -298,78 +359,108 @@ static void vram_fini(void *arg) int xe_vram_probe(struct xe_device *xe) { struct xe_tile *tile; - resource_size_t io_size; + struct xe_vram_region lmem_bar; + resource_size_t remain_io_size; u64 available_size = 0; u64 total_size = 0; - u64 tile_offset; - u64 tile_size; - u64 vram_size; int err; u8 id; if (!IS_DGFX(xe)) return 0; - /* Get the size of the root tile's vram for later accessibility comparison */ - tile = xe_device_get_root_tile(xe); - err = tile_vram_size(tile, &vram_size, &tile_size, &tile_offset); + err = determine_lmem_bar_size(xe, &lmem_bar); if (err) return err; + drm_info(&xe->drm, "VISIBLE VRAM: %pa, %pa\n", &lmem_bar.io_start, &lmem_bar.io_size); - err = determine_lmem_bar_size(xe); - if (err) - return err; - - drm_info(&xe->drm, "VISIBLE VRAM: %pa, %pa\n", &xe->mem.vram.io_start, - &xe->mem.vram.io_size); - - io_size = xe->mem.vram.io_size; + remain_io_size = lmem_bar.io_size; - /* tile specific ranges */ for_each_tile(tile, xe, id) { - err = tile_vram_size(tile, &vram_size, &tile_size, &tile_offset); + u64 region_size; + u64 usable_size; + u64 tile_offset; + + err = tile_vram_size(tile, &usable_size, ®ion_size, &tile_offset); if (err) return err; - tile->mem.vram.actual_physical_size = tile_size; - tile->mem.vram.io_start = xe->mem.vram.io_start + tile_offset; - tile->mem.vram.io_size = min_t(u64, vram_size, io_size); + total_size += region_size; + available_size += usable_size; + + err = vram_region_init(xe, tile->mem.vram, &lmem_bar, tile_offset, usable_size, + region_size, remain_io_size); + if (err) + return err; - if (!tile->mem.vram.io_size) { - drm_err(&xe->drm, "Tile without any CPU visible VRAM. Aborting.\n"); - return -ENODEV; + if (total_size > lmem_bar.io_size) { + drm_info(&xe->drm, "VRAM: %pa is larger than resource %pa\n", + &total_size, &lmem_bar.io_size); } - tile->mem.vram.dpa_base = xe->mem.vram.dpa_base + tile_offset; - tile->mem.vram.usable_size = vram_size; - tile->mem.vram.mapping = xe->mem.vram.mapping + tile_offset; + remain_io_size -= min_t(u64, tile->mem.vram->actual_physical_size, remain_io_size); + } - if (tile->mem.vram.io_size < tile->mem.vram.usable_size) - drm_info(&xe->drm, "Small BAR device\n"); - drm_info(&xe->drm, "VRAM[%u, %u]: Actual physical size %pa, usable size exclude stolen %pa, CPU accessible size %pa\n", id, - tile->id, &tile->mem.vram.actual_physical_size, &tile->mem.vram.usable_size, &tile->mem.vram.io_size); - drm_info(&xe->drm, "VRAM[%u, %u]: DPA range: [%pa-%llx], io range: [%pa-%llx]\n", id, tile->id, - &tile->mem.vram.dpa_base, tile->mem.vram.dpa_base + (u64)tile->mem.vram.actual_physical_size, - &tile->mem.vram.io_start, tile->mem.vram.io_start + (u64)tile->mem.vram.io_size); + err = vram_region_init(xe, xe->mem.vram, &lmem_bar, 0, available_size, total_size, + lmem_bar.io_size); + if (err) + return err; - /* calculate total size using tile size to get the correct HW sizing */ - total_size += tile_size; - available_size += vram_size; + return devm_add_action_or_reset(xe->drm.dev, vram_fini, xe); +} - if (total_size > xe->mem.vram.io_size) { - drm_info(&xe->drm, "VRAM: %pa is larger than resource %pa\n", - &total_size, &xe->mem.vram.io_size); - } +/** + * xe_vram_region_io_start - Get the IO start of a VRAM region + * @vram: the VRAM region + * + * Return: the IO start of the VRAM region, or 0 if not valid + */ +resource_size_t xe_vram_region_io_start(const struct xe_vram_region *vram) +{ + return vram ? vram->io_start : 0; +} - io_size -= min_t(u64, tile_size, io_size); - } +/** + * xe_vram_region_io_size - Get the IO size of a VRAM region + * @vram: the VRAM region + * + * Return: the IO size of the VRAM region, or 0 if not valid + */ +resource_size_t xe_vram_region_io_size(const struct xe_vram_region *vram) +{ + return vram ? vram->io_size : 0; +} - xe->mem.vram.actual_physical_size = total_size; +/** + * xe_vram_region_dpa_base - Get the DPA base of a VRAM region + * @vram: the VRAM region + * + * Return: the DPA base of the VRAM region, or 0 if not valid + */ +resource_size_t xe_vram_region_dpa_base(const struct xe_vram_region *vram) +{ + return vram ? vram->dpa_base : 0; +} - drm_info(&xe->drm, "Total VRAM: %pa, %pa\n", &xe->mem.vram.io_start, - &xe->mem.vram.actual_physical_size); - drm_info(&xe->drm, "Available VRAM: %pa, %pa\n", &xe->mem.vram.io_start, - &available_size); +/** + * xe_vram_region_usable_size - Get the usable size of a VRAM region + * @vram: the VRAM region + * + * Return: the usable size of the VRAM region, or 0 if not valid + */ +resource_size_t xe_vram_region_usable_size(const struct xe_vram_region *vram) +{ + return vram ? vram->usable_size : 0; +} - return devm_add_action_or_reset(xe->drm.dev, vram_fini, xe); +/** + * xe_vram_region_actual_physical_size - Get the actual physical size of a VRAM region + * @vram: the VRAM region + * + * Return: the actual physical size of the VRAM region, or 0 if not valid + */ +resource_size_t xe_vram_region_actual_physical_size(const struct xe_vram_region *vram) +{ + return vram ? vram->actual_physical_size : 0; } +EXPORT_SYMBOL_IF_KUNIT(xe_vram_region_actual_physical_size); diff --git a/drivers/gpu/drm/xe/xe_vram.h b/drivers/gpu/drm/xe/xe_vram.h index e31cc04ec0db..72860f714fc6 100644 --- a/drivers/gpu/drm/xe/xe_vram.h +++ b/drivers/gpu/drm/xe/xe_vram.h @@ -6,8 +6,19 @@ #ifndef _XE_VRAM_H_ #define _XE_VRAM_H_ +#include <linux/types.h> + struct xe_device; +struct xe_vram_region; int xe_vram_probe(struct xe_device *xe); +struct xe_vram_region *xe_vram_region_alloc(struct xe_device *xe, u8 id, u32 placement); + +resource_size_t xe_vram_region_io_start(const struct xe_vram_region *vram); +resource_size_t xe_vram_region_io_size(const struct xe_vram_region *vram); +resource_size_t xe_vram_region_dpa_base(const struct xe_vram_region *vram); +resource_size_t xe_vram_region_usable_size(const struct xe_vram_region *vram); +resource_size_t xe_vram_region_actual_physical_size(const struct xe_vram_region *vram); + #endif diff --git a/drivers/gpu/drm/xe/xe_vram_freq.c b/drivers/gpu/drm/xe/xe_vram_freq.c index b26e26d73dae..17bc84da4cdc 100644 --- a/drivers/gpu/drm/xe/xe_vram_freq.c +++ b/drivers/gpu/drm/xe/xe_vram_freq.c @@ -34,7 +34,7 @@ static ssize_t max_freq_show(struct device *dev, struct device_attribute *attr, char *buf) { struct xe_tile *tile = dev_to_tile(dev); - u32 val, mbox; + u32 val = 0, mbox; int err; mbox = REG_FIELD_PREP(PCODE_MB_COMMAND, PCODE_FREQUENCY_CONFIG) @@ -56,7 +56,7 @@ static ssize_t min_freq_show(struct device *dev, struct device_attribute *attr, char *buf) { struct xe_tile *tile = dev_to_tile(dev); - u32 val, mbox; + u32 val = 0, mbox; int err; mbox = REG_FIELD_PREP(PCODE_MB_COMMAND, PCODE_FREQUENCY_CONFIG) diff --git a/drivers/gpu/drm/xe/xe_vram_types.h b/drivers/gpu/drm/xe/xe_vram_types.h new file mode 100644 index 000000000000..83772dcbf1af --- /dev/null +++ b/drivers/gpu/drm/xe/xe_vram_types.h @@ -0,0 +1,85 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2025 Intel Corporation + */ + +#ifndef _XE_VRAM_TYPES_H_ +#define _XE_VRAM_TYPES_H_ + +#if IS_ENABLED(CONFIG_DRM_XE_PAGEMAP) +#include <drm/drm_pagemap.h> +#endif + +#include "xe_ttm_vram_mgr_types.h" + +struct xe_device; +struct xe_migrate; + +/** + * struct xe_vram_region - memory region structure + * This is used to describe a memory region in xe + * device, such as HBM memory or CXL extension memory. + */ +struct xe_vram_region { + /** @xe: Back pointer to xe device */ + struct xe_device *xe; + /** + * @id: VRAM region instance id + * + * The value should be unique for VRAM region. + */ + u8 id; + /** @io_start: IO start address of this VRAM instance */ + resource_size_t io_start; + /** + * @io_size: IO size of this VRAM instance + * + * This represents how much of this VRAM we can access + * via the CPU through the VRAM BAR. This can be smaller + * than @usable_size, in which case only part of VRAM is CPU + * accessible (typically the first 256M). This + * configuration is known as small-bar. + */ + resource_size_t io_size; + /** @dpa_base: This memory regions's DPA (device physical address) base */ + resource_size_t dpa_base; + /** + * @usable_size: usable size of VRAM + * + * Usable size of VRAM excluding reserved portions + * (e.g stolen mem) + */ + resource_size_t usable_size; + /** + * @actual_physical_size: Actual VRAM size + * + * Actual VRAM size including reserved portions + * (e.g stolen mem) + */ + resource_size_t actual_physical_size; + /** @mapping: pointer to VRAM mappable space */ + void __iomem *mapping; + /** @ttm: VRAM TTM manager */ + struct xe_ttm_vram_mgr ttm; + /** @placement: TTM placement dedicated for this region */ + u32 placement; +#if IS_ENABLED(CONFIG_DRM_XE_PAGEMAP) + /** @migrate: Back pointer to migrate */ + struct xe_migrate *migrate; + /** @pagemap: Used to remap device memory as ZONE_DEVICE */ + struct dev_pagemap pagemap; + /** + * @dpagemap: The struct drm_pagemap of the ZONE_DEVICE memory + * pages of this tile. + */ + struct drm_pagemap dpagemap; + /** + * @hpa_base: base host physical address + * + * This is generated when remap device memory as ZONE_DEVICE + */ + resource_size_t hpa_base; +#endif +}; + +#endif diff --git a/drivers/gpu/drm/xe/xe_wa.c b/drivers/gpu/drm/xe/xe_wa.c index 22a98600fd8f..52c7df4c3afd 100644 --- a/drivers/gpu/drm/xe/xe_wa.c +++ b/drivers/gpu/drm/xe/xe_wa.c @@ -538,6 +538,11 @@ static const struct xe_rtp_entry_sr engine_was[] = { XE_RTP_RULES(GRAPHICS_VERSION(2004), ENGINE_CLASS(RENDER)), XE_RTP_ACTIONS(SET(HALF_SLICE_CHICKEN7, CLEAR_OPTIMIZATION_DISABLE)) }, + { XE_RTP_NAME("13012615864"), + XE_RTP_RULES(GRAPHICS_VERSION(2004), + FUNC(xe_rtp_match_first_render_or_compute)), + XE_RTP_ACTIONS(SET(TDL_TSL_CHICKEN, RES_CHK_SPR_DIS)) + }, /* Xe2_HPG */ @@ -602,6 +607,11 @@ static const struct xe_rtp_entry_sr engine_was[] = { FUNC(xe_rtp_match_first_render_or_compute)), XE_RTP_ACTIONS(SET(TDL_TSL_CHICKEN, STK_ID_RESTRICT)) }, + { XE_RTP_NAME("13012615864"), + XE_RTP_RULES(GRAPHICS_VERSION_RANGE(2001, 2002), + FUNC(xe_rtp_match_first_render_or_compute)), + XE_RTP_ACTIONS(SET(TDL_TSL_CHICKEN, RES_CHK_SPR_DIS)) + }, /* Xe2_LPM */ @@ -647,7 +657,8 @@ static const struct xe_rtp_entry_sr engine_was[] = { XE_RTP_ACTIONS(SET(TDL_CHICKEN, QID_WAIT_FOR_THREAD_NOT_RUN_DISABLE)) }, { XE_RTP_NAME("13012615864"), - XE_RTP_RULES(GRAPHICS_VERSION_RANGE(3000, 3001), + XE_RTP_RULES(GRAPHICS_VERSION_RANGE(3000, 3001), OR, + GRAPHICS_VERSION(3003), FUNC(xe_rtp_match_first_render_or_compute)), XE_RTP_ACTIONS(SET(TDL_TSL_CHICKEN, RES_CHK_SPR_DIS)) }, @@ -905,13 +916,13 @@ void xe_wa_process_device_oob(struct xe_device *xe) } /** - * xe_wa_process_oob - process OOB workaround table + * xe_wa_process_gt_oob - process GT OOB workaround table * @gt: GT instance to process workarounds for * * Process OOB workaround table for this platform, marking in @gt the * workarounds that are active. */ -void xe_wa_process_oob(struct xe_gt *gt) +void xe_wa_process_gt_oob(struct xe_gt *gt) { struct xe_rtp_process_ctx ctx = XE_RTP_PROCESS_CTX_INITIALIZER(gt); @@ -995,12 +1006,12 @@ int xe_wa_device_init(struct xe_device *xe) } /** - * xe_wa_init - initialize gt with workaround bookkeeping + * xe_wa_gt_init - initialize gt with workaround bookkeeping * @gt: GT instance to initialize * * Returns 0 for success, negative error code otherwise. */ -int xe_wa_init(struct xe_gt *gt) +int xe_wa_gt_init(struct xe_gt *gt) { struct xe_device *xe = gt_to_xe(gt); size_t n_oob, n_lrc, n_engine, n_gt, total; @@ -1026,7 +1037,7 @@ int xe_wa_init(struct xe_gt *gt) return 0; } -ALLOW_ERROR_INJECTION(xe_wa_init, ERRNO); /* See xe_pci_probe() */ +ALLOW_ERROR_INJECTION(xe_wa_gt_init, ERRNO); /* See xe_pci_probe() */ void xe_wa_device_dump(struct xe_device *xe, struct drm_printer *p) { @@ -1079,6 +1090,6 @@ void xe_wa_apply_tile_workarounds(struct xe_tile *tile) if (IS_SRIOV_VF(tile->xe)) return; - if (XE_WA(tile->primary_gt, 22010954014)) + if (XE_GT_WA(tile->primary_gt, 22010954014)) xe_mmio_rmw32(mmio, XEHP_CLOCK_GATE_DIS, 0, SGSI_SIDECLK_DIS); } diff --git a/drivers/gpu/drm/xe/xe_wa.h b/drivers/gpu/drm/xe/xe_wa.h index f3880c65cb8d..6a869b2de643 100644 --- a/drivers/gpu/drm/xe/xe_wa.h +++ b/drivers/gpu/drm/xe/xe_wa.h @@ -14,9 +14,9 @@ struct xe_hw_engine; struct xe_tile; int xe_wa_device_init(struct xe_device *xe); -int xe_wa_init(struct xe_gt *gt); +int xe_wa_gt_init(struct xe_gt *gt); void xe_wa_process_device_oob(struct xe_device *xe); -void xe_wa_process_oob(struct xe_gt *gt); +void xe_wa_process_gt_oob(struct xe_gt *gt); void xe_wa_process_gt(struct xe_gt *gt); void xe_wa_process_engine(struct xe_hw_engine *hwe); void xe_wa_process_lrc(struct xe_hw_engine *hwe); @@ -25,11 +25,11 @@ void xe_wa_device_dump(struct xe_device *xe, struct drm_printer *p); void xe_wa_dump(struct xe_gt *gt, struct drm_printer *p); /** - * XE_WA - Out-of-band workarounds, to be queried and called as needed. + * XE_GT_WA - Out-of-band GT workarounds, to be queried and called as needed. * @gt__: gt instance * @id__: XE_OOB_<id__>, as generated by build system in generated/xe_wa_oob.h */ -#define XE_WA(gt__, id__) ({ \ +#define XE_GT_WA(gt__, id__) ({ \ xe_gt_assert(gt__, (gt__)->wa_active.oob_initialized); \ test_bit(XE_WA_OOB_ ## id__, (gt__)->wa_active.oob); \ }) diff --git a/drivers/gpu/drm/xe/xe_wa_oob.rules b/drivers/gpu/drm/xe/xe_wa_oob.rules index e990f20eccfe..0cf2b3c03532 100644 --- a/drivers/gpu/drm/xe/xe_wa_oob.rules +++ b/drivers/gpu/drm/xe/xe_wa_oob.rules @@ -1,4 +1,6 @@ 1607983814 GRAPHICS_VERSION_RANGE(1200, 1210) +16010904313 GRAPHICS_VERSION_RANGE(1200, 1210) +18022495364 GRAPHICS_VERSION_RANGE(1200, 1210) 22012773006 GRAPHICS_VERSION_RANGE(1200, 1250) 14014475959 GRAPHICS_VERSION_RANGE(1270, 1271), GRAPHICS_STEP(A0, B0) PLATFORM(DG2) @@ -46,7 +48,7 @@ 16023588340 GRAPHICS_VERSION(2001), FUNC(xe_rtp_match_not_sriov_vf) 14019789679 GRAPHICS_VERSION(1255) GRAPHICS_VERSION_RANGE(1270, 2004) -no_media_l3 MEDIA_VERSION(3000) +no_media_l3 MEDIA_VERSION_RANGE(3000, 3002) 14022866841 GRAPHICS_VERSION(3000), GRAPHICS_STEP(A0, B0) MEDIA_VERSION(3000), MEDIA_STEP(A0, B0) 16021333562 GRAPHICS_VERSION_RANGE(1200, 1274) @@ -66,9 +68,16 @@ no_media_l3 MEDIA_VERSION(3000) MEDIA_VERSION_RANGE(1300, 3000) MEDIA_VERSION(3002) GRAPHICS_VERSION(3003) +14020001231 GRAPHICS_VERSION_RANGE(2001,2004), FUNC(xe_rtp_match_psmi_enabled) + MEDIA_VERSION(2000), FUNC(xe_rtp_match_psmi_enabled) + MEDIA_VERSION(3000), FUNC(xe_rtp_match_psmi_enabled) + MEDIA_VERSION(3002), FUNC(xe_rtp_match_psmi_enabled) +16023683509 MEDIA_VERSION(2000), FUNC(xe_rtp_match_psmi_enabled) + MEDIA_VERSION(3000), MEDIA_STEP(A0, B0), FUNC(xe_rtp_match_psmi_enabled) # SoC workaround - currently applies to all platforms with the following # primary GT GMDID 14022085890 GRAPHICS_VERSION(2001) 15015404425_disable PLATFORM(PANTHERLAKE), MEDIA_STEP(B0, FOREVER) +16026007364 MEDIA_VERSION(3000) diff --git a/include/drm/drm_device.h b/include/drm/drm_device.h index a33aedd5e9ec..59fd3f4d5995 100644 --- a/include/drm/drm_device.h +++ b/include/drm/drm_device.h @@ -26,10 +26,14 @@ struct pci_controller; * Recovery methods for wedged device in order of less to more side-effects. * To be used with drm_dev_wedged_event() as recovery @method. Callers can * use any one, multiple (or'd) or none depending on their needs. + * + * Refer to "Device Wedging" chapter in Documentation/gpu/drm-uapi.rst for more + * details. */ #define DRM_WEDGE_RECOVERY_NONE BIT(0) /* optional telemetry collection */ #define DRM_WEDGE_RECOVERY_REBIND BIT(1) /* unbind + bind driver */ #define DRM_WEDGE_RECOVERY_BUS_RESET BIT(2) /* unbind + reset bus device + bind */ +#define DRM_WEDGE_RECOVERY_VENDOR BIT(3) /* vendor specific recovery method */ /** * struct drm_wedge_task_info - information about the guilty task of a wedge dev diff --git a/include/drm/drm_gpusvm.h b/include/drm/drm_gpusvm.h index 142fc2af1716..0e336148309d 100644 --- a/include/drm/drm_gpusvm.h +++ b/include/drm/drm_gpusvm.h @@ -17,7 +17,7 @@ struct drm_gpusvm_notifier; struct drm_gpusvm_ops; struct drm_gpusvm_range; struct drm_pagemap; -struct drm_pagemap_device_addr; +struct drm_pagemap_addr; /** * struct drm_gpusvm_ops - Operations structure for GPU SVM @@ -154,7 +154,7 @@ struct drm_gpusvm_range { struct interval_tree_node itree; struct list_head entry; unsigned long notifier_seq; - struct drm_pagemap_device_addr *dma_addr; + struct drm_pagemap_addr *dma_addr; struct drm_pagemap *dpagemap; struct drm_gpusvm_range_flags flags; }; diff --git a/include/drm/drm_pagemap.h b/include/drm/drm_pagemap.h index e5f20a1235be..f6e7e234c089 100644 --- a/include/drm/drm_pagemap.h +++ b/include/drm/drm_pagemap.h @@ -6,6 +6,8 @@ #include <linux/hmm.h> #include <linux/types.h> +#define NR_PAGES(order) (1U << (order)) + struct drm_pagemap; struct drm_pagemap_zdd; struct device; @@ -23,7 +25,7 @@ enum drm_interconnect_protocol { }; /** - * struct drm_pagemap_device_addr - Device address representation. + * struct drm_pagemap_addr - Address representation. * @addr: The dma address or driver-defined address for driver private interconnects. * @proto: The interconnect protocol. * @order: The page order of the device mapping. (Size is PAGE_SIZE << order). @@ -32,7 +34,7 @@ enum drm_interconnect_protocol { * Note: There is room for improvement here. We should be able to pack into * 64 bits. */ -struct drm_pagemap_device_addr { +struct drm_pagemap_addr { dma_addr_t addr; u64 proto : 54; u64 order : 8; @@ -40,21 +42,21 @@ struct drm_pagemap_device_addr { }; /** - * drm_pagemap_device_addr_encode() - Encode a dma address with metadata + * drm_pagemap_addr_encode() - Encode a dma address with metadata * @addr: The dma address or driver-defined address for driver private interconnects. * @proto: The interconnect protocol. * @order: The page order of the dma mapping. (Size is PAGE_SIZE << order). * @dir: The DMA direction. * - * Return: A struct drm_pagemap_device_addr encoding the above information. + * Return: A struct drm_pagemap_addr encoding the above information. */ -static inline struct drm_pagemap_device_addr -drm_pagemap_device_addr_encode(dma_addr_t addr, - enum drm_interconnect_protocol proto, - unsigned int order, - enum dma_data_direction dir) +static inline struct drm_pagemap_addr +drm_pagemap_addr_encode(dma_addr_t addr, + enum drm_interconnect_protocol proto, + unsigned int order, + enum dma_data_direction dir) { - return (struct drm_pagemap_device_addr) { + return (struct drm_pagemap_addr) { .addr = addr, .proto = proto, .order = order, @@ -75,11 +77,11 @@ struct drm_pagemap_ops { * @order: The page order of the device mapping. (Size is PAGE_SIZE << order). * @dir: The transfer direction. */ - struct drm_pagemap_device_addr (*device_map)(struct drm_pagemap *dpagemap, - struct device *dev, - struct page *page, - unsigned int order, - enum dma_data_direction dir); + struct drm_pagemap_addr (*device_map)(struct drm_pagemap *dpagemap, + struct device *dev, + struct page *page, + unsigned int order, + enum dma_data_direction dir); /** * @device_unmap: Unmap a device address previously obtained using @device_map. @@ -90,7 +92,7 @@ struct drm_pagemap_ops { */ void (*device_unmap)(struct drm_pagemap *dpagemap, struct device *dev, - struct drm_pagemap_device_addr addr); + struct drm_pagemap_addr addr); /** * @populate_mm: Populate part of the mm with @dpagemap memory, @@ -170,29 +172,33 @@ struct drm_pagemap_devmem_ops { /** * @copy_to_devmem: Copy to device memory (required for migration) * @pages: Pointer to array of device memory pages (destination) - * @dma_addr: Pointer to array of DMA addresses (source) + * @pagemap_addr: Pointer to array of DMA information (source) * @npages: Number of pages to copy * - * Copy pages to device memory. + * Copy pages to device memory. If the order of a @pagemap_addr entry + * is greater than 0, the entry is populated but subsequent entries + * within the range of that order are not populated. * * Return: 0 on success, a negative error code on failure. */ int (*copy_to_devmem)(struct page **pages, - dma_addr_t *dma_addr, + struct drm_pagemap_addr *pagemap_addr, unsigned long npages); /** * @copy_to_ram: Copy to system RAM (required for migration) * @pages: Pointer to array of device memory pages (source) - * @dma_addr: Pointer to array of DMA addresses (destination) + * @pagemap_addr: Pointer to array of DMA information (destination) * @npages: Number of pages to copy * - * Copy pages to system RAM. + * Copy pages to system RAM. If the order of a @pagemap_addr entry + * is greater than 0, the entry is populated but subsequent entries + * within the range of that order are not populated. * * Return: 0 on success, a negative error code on failure. */ int (*copy_to_ram)(struct page **pages, - dma_addr_t *dma_addr, + struct drm_pagemap_addr *pagemap_addr, unsigned long npages); }; diff --git a/include/drm/intel/pciids.h b/include/drm/intel/pciids.h index 76f8d26f9cc9..da6301a6fcea 100644 --- a/include/drm/intel/pciids.h +++ b/include/drm/intel/pciids.h @@ -26,6 +26,11 @@ #define __PCIIDS_H__ #ifdef __KERNEL__ +#define INTEL_PCI_DEVICE(_id, _info) { \ + PCI_DEVICE(PCI_VENDOR_ID_INTEL, (_id)), \ + .driver_data = (kernel_ulong_t)(_info), \ +} + #define INTEL_VGA_DEVICE(_id, _info) { \ PCI_DEVICE(PCI_VENDOR_ID_INTEL, (_id)), \ .class = PCI_BASE_CLASS_DISPLAY << 16, .class_mask = 0xff << 16, \ diff --git a/include/linux/iopoll.h b/include/linux/iopoll.h index 91324c331a4b..bdd2e0652bc3 100644 --- a/include/linux/iopoll.h +++ b/include/linux/iopoll.h @@ -14,62 +14,64 @@ #include <linux/io.h> /** - * read_poll_timeout - Periodically poll an address until a condition is - * met or a timeout occurs - * @op: accessor function (takes @args as its arguments) - * @val: Variable to read the value into - * @cond: Break condition (usually involving @val) - * @sleep_us: Maximum time to sleep between reads in us (0 tight-loops). Please - * read usleep_range() function description for details and + * poll_timeout_us - Periodically poll and perform an operation until + * a condition is met or a timeout occurs + * + * @op: Operation + * @cond: Break condition + * @sleep_us: Maximum time to sleep between operations in us (0 tight-loops). + * Please read usleep_range() function description for details and * limitations. * @timeout_us: Timeout in us, 0 means never timeout - * @sleep_before_read: if it is true, sleep @sleep_us before read. - * @args: arguments for @op poll + * @sleep_before_op: if it is true, sleep @sleep_us before operation. * * When available, you'll probably want to use one of the specialized * macros defined below rather than this macro directly. * - * Returns: 0 on success and -ETIMEDOUT upon a timeout. In either - * case, the last read value at @args is stored in @val. Must not + * Returns: 0 on success and -ETIMEDOUT upon a timeout. Must not * be called from atomic context if sleep_us or timeout_us are used. */ -#define read_poll_timeout(op, val, cond, sleep_us, timeout_us, \ - sleep_before_read, args...) \ +#define poll_timeout_us(op, cond, sleep_us, timeout_us, sleep_before_op) \ ({ \ u64 __timeout_us = (timeout_us); \ unsigned long __sleep_us = (sleep_us); \ ktime_t __timeout = ktime_add_us(ktime_get(), __timeout_us); \ + int ___ret; \ might_sleep_if((__sleep_us) != 0); \ - if (sleep_before_read && __sleep_us) \ + if ((sleep_before_op) && __sleep_us) \ usleep_range((__sleep_us >> 2) + 1, __sleep_us); \ for (;;) { \ - (val) = op(args); \ - if (cond) \ + bool __expired = __timeout_us && \ + ktime_compare(ktime_get(), __timeout) > 0; \ + /* guarantee 'op' and 'cond' are evaluated after timeout expired */ \ + barrier(); \ + op; \ + if (cond) { \ + ___ret = 0; \ break; \ - if (__timeout_us && \ - ktime_compare(ktime_get(), __timeout) > 0) { \ - (val) = op(args); \ + } \ + if (__expired) { \ + ___ret = -ETIMEDOUT; \ break; \ } \ if (__sleep_us) \ usleep_range((__sleep_us >> 2) + 1, __sleep_us); \ cpu_relax(); \ } \ - (cond) ? 0 : -ETIMEDOUT; \ + ___ret; \ }) /** - * read_poll_timeout_atomic - Periodically poll an address until a condition is - * met or a timeout occurs - * @op: accessor function (takes @args as its arguments) - * @val: Variable to read the value into - * @cond: Break condition (usually involving @val) - * @delay_us: Time to udelay between reads in us (0 tight-loops). Please - * read udelay() function description for details and + * poll_timeout_us_atomic - Periodically poll and perform an operation until + * a condition is met or a timeout occurs + * + * @op: Operation + * @cond: Break condition + * @delay_us: Time to udelay between operations in us (0 tight-loops). + * Please read udelay() function description for details and * limitations. * @timeout_us: Timeout in us, 0 means never timeout - * @delay_before_read: if it is true, delay @delay_us before read. - * @args: arguments for @op poll + * @delay_before_op: if it is true, delay @delay_us before operation. * * This macro does not rely on timekeeping. Hence it is safe to call even when * timekeeping is suspended, at the expense of an underestimation of wall clock @@ -78,27 +80,32 @@ * When available, you'll probably want to use one of the specialized * macros defined below rather than this macro directly. * - * Returns: 0 on success and -ETIMEDOUT upon a timeout. In either - * case, the last read value at @args is stored in @val. + * Returns: 0 on success and -ETIMEDOUT upon a timeout. */ -#define read_poll_timeout_atomic(op, val, cond, delay_us, timeout_us, \ - delay_before_read, args...) \ +#define poll_timeout_us_atomic(op, cond, delay_us, timeout_us, \ + delay_before_op) \ ({ \ u64 __timeout_us = (timeout_us); \ s64 __left_ns = __timeout_us * NSEC_PER_USEC; \ unsigned long __delay_us = (delay_us); \ u64 __delay_ns = __delay_us * NSEC_PER_USEC; \ - if (delay_before_read && __delay_us) { \ + int ___ret; \ + if ((delay_before_op) && __delay_us) { \ udelay(__delay_us); \ if (__timeout_us) \ __left_ns -= __delay_ns; \ } \ for (;;) { \ - (val) = op(args); \ - if (cond) \ + bool __expired = __timeout_us && __left_ns < 0; \ + /* guarantee 'op' and 'cond' are evaluated after timeout expired */ \ + barrier(); \ + op; \ + if (cond) { \ + ___ret = 0; \ break; \ - if (__timeout_us && __left_ns < 0) { \ - (val) = op(args); \ + } \ + if (__expired) { \ + ___ret = -ETIMEDOUT; \ break; \ } \ if (__delay_us) { \ @@ -110,10 +117,61 @@ if (__timeout_us) \ __left_ns--; \ } \ - (cond) ? 0 : -ETIMEDOUT; \ + ___ret; \ }) /** + * read_poll_timeout - Periodically poll an address until a condition is + * met or a timeout occurs + * @op: accessor function (takes @args as its arguments) + * @val: Variable to read the value into + * @cond: Break condition (usually involving @val) + * @sleep_us: Maximum time to sleep between reads in us (0 tight-loops). Please + * read usleep_range() function description for details and + * limitations. + * @timeout_us: Timeout in us, 0 means never timeout + * @sleep_before_read: if it is true, sleep @sleep_us before read. + * @args: arguments for @op poll + * + * When available, you'll probably want to use one of the specialized + * macros defined below rather than this macro directly. + * + * Returns: 0 on success and -ETIMEDOUT upon a timeout. In either + * case, the last read value at @args is stored in @val. Must not + * be called from atomic context if sleep_us or timeout_us are used. + */ +#define read_poll_timeout(op, val, cond, sleep_us, timeout_us, \ + sleep_before_read, args...) \ + poll_timeout_us((val) = op(args), cond, sleep_us, timeout_us, sleep_before_read) + +/** + * read_poll_timeout_atomic - Periodically poll an address until a condition is + * met or a timeout occurs + * @op: accessor function (takes @args as its arguments) + * @val: Variable to read the value into + * @cond: Break condition (usually involving @val) + * @delay_us: Time to udelay between reads in us (0 tight-loops). Please + * read udelay() function description for details and + * limitations. + * @timeout_us: Timeout in us, 0 means never timeout + * @delay_before_read: if it is true, delay @delay_us before read. + * @args: arguments for @op poll + * + * This macro does not rely on timekeeping. Hence it is safe to call even when + * timekeeping is suspended, at the expense of an underestimation of wall clock + * time, which is rather minimal with a non-zero delay_us. + * + * When available, you'll probably want to use one of the specialized + * macros defined below rather than this macro directly. + * + * Returns: 0 on success and -ETIMEDOUT upon a timeout. In either + * case, the last read value at @args is stored in @val. + */ +#define read_poll_timeout_atomic(op, val, cond, sleep_us, timeout_us, \ + sleep_before_read, args...) \ + poll_timeout_us_atomic((val) = op(args), cond, sleep_us, timeout_us, sleep_before_read) + +/** * readx_poll_timeout - Periodically poll an address until a condition is met or a timeout occurs * @op: accessor function (takes @addr as its only argument) * @addr: Address to poll diff --git a/include/uapi/drm/xe_drm.h b/include/uapi/drm/xe_drm.h index e2426413488f..40ff19f52a8d 100644 --- a/include/uapi/drm/xe_drm.h +++ b/include/uapi/drm/xe_drm.h @@ -81,6 +81,8 @@ extern "C" { * - &DRM_IOCTL_XE_EXEC * - &DRM_IOCTL_XE_WAIT_USER_FENCE * - &DRM_IOCTL_XE_OBSERVATION + * - &DRM_IOCTL_XE_MADVISE + * - &DRM_IOCTL_XE_VM_QUERY_MEM_RANGE_ATTRS */ /* @@ -102,6 +104,8 @@ extern "C" { #define DRM_XE_EXEC 0x09 #define DRM_XE_WAIT_USER_FENCE 0x0a #define DRM_XE_OBSERVATION 0x0b +#define DRM_XE_MADVISE 0x0c +#define DRM_XE_VM_QUERY_MEM_RANGE_ATTRS 0x0d /* Must be kept compact -- no holes */ @@ -117,6 +121,8 @@ extern "C" { #define DRM_IOCTL_XE_EXEC DRM_IOW(DRM_COMMAND_BASE + DRM_XE_EXEC, struct drm_xe_exec) #define DRM_IOCTL_XE_WAIT_USER_FENCE DRM_IOWR(DRM_COMMAND_BASE + DRM_XE_WAIT_USER_FENCE, struct drm_xe_wait_user_fence) #define DRM_IOCTL_XE_OBSERVATION DRM_IOW(DRM_COMMAND_BASE + DRM_XE_OBSERVATION, struct drm_xe_observation_param) +#define DRM_IOCTL_XE_MADVISE DRM_IOW(DRM_COMMAND_BASE + DRM_XE_MADVISE, struct drm_xe_madvise) +#define DRM_IOCTL_XE_VM_QUERY_MEM_RANGE_ATTRS DRM_IOWR(DRM_COMMAND_BASE + DRM_XE_VM_QUERY_MEM_RANGE_ATTRS, struct drm_xe_vm_query_mem_range_attr) /** * DOC: Xe IOCTL Extensions @@ -760,7 +766,11 @@ struct drm_xe_device_query { * gem creation * * The @flags can be: - * - %DRM_XE_GEM_CREATE_FLAG_DEFER_BACKING + * - %DRM_XE_GEM_CREATE_FLAG_DEFER_BACKING - Modify the GEM object + * allocation strategy by deferring physical memory allocation + * until the object is either bound to a virtual memory region via + * VM_BIND or accessed by the CPU. As a result, no backing memory is + * reserved at the time of GEM object creation. * - %DRM_XE_GEM_CREATE_FLAG_SCANOUT * - %DRM_XE_GEM_CREATE_FLAG_NEEDS_VISIBLE_VRAM - When using VRAM as a * possible placement, ensure that the corresponding VRAM allocation @@ -1003,6 +1013,10 @@ struct drm_xe_vm_destroy { * valid on VMs with DRM_XE_VM_CREATE_FLAG_FAULT_MODE set. The CPU address * mirror flag are only valid for DRM_XE_VM_BIND_OP_MAP operations, the BO * handle MBZ, and the BO offset MBZ. + * + * The @prefetch_mem_region_instance for %DRM_XE_VM_BIND_OP_PREFETCH can also be: + * - %DRM_XE_CONSULT_MEM_ADVISE_PREF_LOC, which ensures prefetching occurs in + * the memory region advised by madvise. */ struct drm_xe_vm_bind_op { /** @extensions: Pointer to the first extension struct, if any */ @@ -1108,6 +1122,7 @@ struct drm_xe_vm_bind_op { /** @flags: Bind flags */ __u32 flags; +#define DRM_XE_CONSULT_MEM_ADVISE_PREF_LOC -1 /** * @prefetch_mem_region_instance: Memory region to prefetch VMA to. * It is a region instance, not a mask. @@ -1974,6 +1989,271 @@ struct drm_xe_query_eu_stall { __u64 sampling_rates[]; }; +/** + * struct drm_xe_madvise - Input of &DRM_IOCTL_XE_MADVISE + * + * This structure is used to set memory attributes for a virtual address range + * in a VM. The type of attribute is specified by @type, and the corresponding + * union member is used to provide additional parameters for @type. + * + * Supported attribute types: + * - DRM_XE_MEM_RANGE_ATTR_PREFERRED_LOC: Set preferred memory location. + * - DRM_XE_MEM_RANGE_ATTR_ATOMIC: Set atomic access policy. + * - DRM_XE_MEM_RANGE_ATTR_PAT: Set page attribute table index. + * + * Example: + * + * .. code-block:: C + * + * struct drm_xe_madvise madvise = { + * .vm_id = vm_id, + * .start = 0x100000, + * .range = 0x2000, + * .type = DRM_XE_MEM_RANGE_ATTR_ATOMIC, + * .atomic_val = DRM_XE_ATOMIC_DEVICE, + * }; + * + * ioctl(fd, DRM_IOCTL_XE_MADVISE, &madvise); + * + */ +struct drm_xe_madvise { + /** @extensions: Pointer to the first extension struct, if any */ + __u64 extensions; + + /** @start: start of the virtual address range */ + __u64 start; + + /** @range: size of the virtual address range */ + __u64 range; + + /** @vm_id: vm_id of the virtual range */ + __u32 vm_id; + +#define DRM_XE_MEM_RANGE_ATTR_PREFERRED_LOC 0 +#define DRM_XE_MEM_RANGE_ATTR_ATOMIC 1 +#define DRM_XE_MEM_RANGE_ATTR_PAT 2 + /** @type: type of attribute */ + __u32 type; + + union { + /** + * @preferred_mem_loc: preferred memory location + * + * Used when @type == DRM_XE_MEM_RANGE_ATTR_PREFERRED_LOC + * + * Supported values for @preferred_mem_loc.devmem_fd: + * - DRM_XE_PREFERRED_LOC_DEFAULT_DEVICE: set vram of fault tile as preferred loc + * - DRM_XE_PREFERRED_LOC_DEFAULT_SYSTEM: set smem as preferred loc + * + * Supported values for @preferred_mem_loc.migration_policy: + * - DRM_XE_MIGRATE_ALL_PAGES + * - DRM_XE_MIGRATE_ONLY_SYSTEM_PAGES + */ + struct { +#define DRM_XE_PREFERRED_LOC_DEFAULT_DEVICE 0 +#define DRM_XE_PREFERRED_LOC_DEFAULT_SYSTEM -1 + /** @preferred_mem_loc.devmem_fd: fd for preferred loc */ + __u32 devmem_fd; + +#define DRM_XE_MIGRATE_ALL_PAGES 0 +#define DRM_XE_MIGRATE_ONLY_SYSTEM_PAGES 1 + /** @preferred_mem_loc.migration_policy: Page migration policy */ + __u16 migration_policy; + + /** @preferred_mem_loc.pad : MBZ */ + __u16 pad; + + /** @preferred_mem_loc.reserved : Reserved */ + __u64 reserved; + } preferred_mem_loc; + + /** + * @atomic: Atomic access policy + * + * Used when @type == DRM_XE_MEM_RANGE_ATTR_ATOMIC. + * + * Supported values for @atomic.val: + * - DRM_XE_ATOMIC_UNDEFINED: Undefined or default behaviour. + * Support both GPU and CPU atomic operations for system allocator. + * Support GPU atomic operations for normal(bo) allocator. + * - DRM_XE_ATOMIC_DEVICE: Support GPU atomic operations. + * - DRM_XE_ATOMIC_GLOBAL: Support both GPU and CPU atomic operations. + * - DRM_XE_ATOMIC_CPU: Support CPU atomic only, no GPU atomics supported. + */ + struct { +#define DRM_XE_ATOMIC_UNDEFINED 0 +#define DRM_XE_ATOMIC_DEVICE 1 +#define DRM_XE_ATOMIC_GLOBAL 2 +#define DRM_XE_ATOMIC_CPU 3 + /** @atomic.val: value of atomic operation */ + __u32 val; + + /** @atomic.pad: MBZ */ + __u32 pad; + + /** @atomic.reserved: Reserved */ + __u64 reserved; + } atomic; + + /** + * @pat_index: Page attribute table index + * + * Used when @type == DRM_XE_MEM_RANGE_ATTR_PAT. + */ + struct { + /** @pat_index.val: PAT index value */ + __u32 val; + + /** @pat_index.pad: MBZ */ + __u32 pad; + + /** @pat_index.reserved: Reserved */ + __u64 reserved; + } pat_index; + }; + + /** @reserved: Reserved */ + __u64 reserved[2]; +}; + +/** + * struct drm_xe_mem_range_attr - Output of &DRM_IOCTL_XE_VM_QUERY_MEM_RANGES_ATTRS + * + * This structure is provided by userspace and filled by KMD in response to the + * DRM_IOCTL_XE_VM_QUERY_MEM_RANGES_ATTRS ioctl. It describes memory attributes of + * a memory ranges within a user specified address range in a VM. + * + * The structure includes information such as atomic access policy, + * page attribute table (PAT) index, and preferred memory location. + * Userspace allocates an array of these structures and passes a pointer to the + * ioctl to retrieve attributes for each memory ranges + * + * @extensions: Pointer to the first extension struct, if any + * @start: Start address of the memory range + * @end: End address of the virtual memory range + * + */ +struct drm_xe_mem_range_attr { + /** @extensions: Pointer to the first extension struct, if any */ + __u64 extensions; + + /** @start: start of the memory range */ + __u64 start; + + /** @end: end of the memory range */ + __u64 end; + + /** @preferred_mem_loc: preferred memory location */ + struct { + /** @preferred_mem_loc.devmem_fd: fd for preferred loc */ + __u32 devmem_fd; + + /** @preferred_mem_loc.migration_policy: Page migration policy */ + __u32 migration_policy; + } preferred_mem_loc; + + /** @atomic: Atomic access policy */ + struct { + /** @atomic.val: atomic attribute */ + __u32 val; + + /** @atomic.reserved: Reserved */ + __u32 reserved; + } atomic; + + /** @pat_index: Page attribute table index */ + struct { + /** @pat_index.val: PAT index */ + __u32 val; + + /** @pat_index.reserved: Reserved */ + __u32 reserved; + } pat_index; + + /** @reserved: Reserved */ + __u64 reserved[2]; +}; + +/** + * struct drm_xe_vm_query_mem_range_attr - Input of &DRM_IOCTL_XE_VM_QUERY_MEM_ATTRIBUTES + * + * This structure is used to query memory attributes of memory regions + * within a user specified address range in a VM. It provides detailed + * information about each memory range, including atomic access policy, + * page attribute table (PAT) index, and preferred memory location. + * + * Userspace first calls the ioctl with @num_mem_ranges = 0, + * @sizeof_mem_ranges_attr = 0 and @vector_of_vma_mem_attr = NULL to retrieve + * the number of memory regions and size of each memory range attribute. + * Then, it allocates a buffer of that size and calls the ioctl again to fill + * the buffer with memory range attributes. + * + * If second call fails with -ENOSPC, it means memory ranges changed between + * first call and now, retry IOCTL again with @num_mem_ranges = 0, + * @sizeof_mem_ranges_attr = 0 and @vector_of_vma_mem_attr = NULL followed by + * Second ioctl call. + * + * Example: + * + * .. code-block:: C + * + * struct drm_xe_vm_query_mem_range_attr query = { + * .vm_id = vm_id, + * .start = 0x100000, + * .range = 0x2000, + * }; + * + * // First ioctl call to get num of mem regions and sizeof each attribute + * ioctl(fd, DRM_IOCTL_XE_VM_QUERY_MEM_RANGE_ATTRS, &query); + * + * // Allocate buffer for the memory region attributes + * void *ptr = malloc(query.num_mem_ranges * query.sizeof_mem_range_attr); + * void *ptr_start = ptr; + * + * query.vector_of_mem_attr = (uintptr_t)ptr; + * + * // Second ioctl call to actually fill the memory attributes + * ioctl(fd, DRM_IOCTL_XE_VM_QUERY_MEM_RANGE_ATTRS, &query); + * + * // Iterate over the returned memory region attributes + * for (unsigned int i = 0; i < query.num_mem_ranges; ++i) { + * struct drm_xe_mem_range_attr *attr = (struct drm_xe_mem_range_attr *)ptr; + * + * // Do something with attr + * + * // Move pointer by one entry + * ptr += query.sizeof_mem_range_attr; + * } + * + * free(ptr_start); + */ +struct drm_xe_vm_query_mem_range_attr { + /** @extensions: Pointer to the first extension struct, if any */ + __u64 extensions; + + /** @vm_id: vm_id of the virtual range */ + __u32 vm_id; + + /** @num_mem_ranges: number of mem_ranges in range */ + __u32 num_mem_ranges; + + /** @start: start of the virtual address range */ + __u64 start; + + /** @range: size of the virtual address range */ + __u64 range; + + /** @sizeof_mem_range_attr: size of struct drm_xe_mem_range_attr */ + __u64 sizeof_mem_range_attr; + + /** @vector_of_mem_attr: userptr to array of struct drm_xe_mem_range_attr */ + __u64 vector_of_mem_attr; + + /** @reserved: Reserved */ + __u64 reserved[2]; + +}; + #if defined(__cplusplus) } #endif |