From f679fd6057fbf5ab34aaee28d58b7f81af0cbf48 Mon Sep 17 00:00:00 2001 From: Ahmad Rehman Date: Mon, 4 Mar 2024 15:56:00 -0600 Subject: drm/amdgpu: Init zone device and drm client after mode-1 reset on reload In passthrough environment, when amdgpu is reloaded after unload, mode-1 is triggered after initializing the necessary IPs, That init does not include KFD, and KFD init waits until the reset is completed. KFD init is called in the reset handler, but in this case, the zone device and drm client is not initialized, causing app to create kernel panic. v2: Removing the init KFD condition from amdgpu_amdkfd_drm_client_create. As the previous version has the potential of creating DRM client twice. v3: v2 patch results in SDMA engine hung as DRM open causes VM clear to SDMA before SDMA init. Adding the condition to in drm client creation, on top of v1, to guard against drm client creation call multiple times. Signed-off-by: Ahmad Rehman Reviewed-by: Felix Kuehling Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c index 15b188aaf681..80b9642f2bc4 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c @@ -2479,8 +2479,11 @@ static void amdgpu_drv_delayed_reset_work_handler(struct work_struct *work) } for (i = 0; i < mgpu_info.num_dgpu; i++) { adev = mgpu_info.gpu_ins[i].adev; - if (!adev->kfd.init_complete) + if (!adev->kfd.init_complete) { + kgd2kfd_init_zone_device(adev); amdgpu_amdkfd_device_init(adev); + amdgpu_amdkfd_drm_client_create(adev); + } amdgpu_ttm_set_buffer_funcs_status(adev, true); } } -- cgit v1.2.3 From a3a4c0b12346a2493b41c8790d85141844a04e28 Mon Sep 17 00:00:00 2001 From: shaoyunl Date: Fri, 22 Mar 2024 12:25:16 -0400 Subject: drm/amdgpu : Add mes_log_enable to control mes log feature The MES log might slow down the performance for extra step of log the data, disable it by default and introduce a parameter can enable it when necessary Signed-off-by: shaoyunl Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu.h | 1 + drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 10 ++++++++++ drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c | 5 ++++- drivers/gpu/drm/amd/amdgpu/mes_v11_0.c | 7 +++++-- 4 files changed, 20 insertions(+), 3 deletions(-) (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index 9c62552bec34..b3b84647207e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -210,6 +210,7 @@ extern int amdgpu_async_gfx_ring; extern int amdgpu_mcbp; extern int amdgpu_discovery; extern int amdgpu_mes; +extern int amdgpu_mes_log_enable; extern int amdgpu_mes_kiq; extern int amdgpu_noretry; extern int amdgpu_force_asic_type; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c index 80b9642f2bc4..e4277298cf1a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c @@ -195,6 +195,7 @@ int amdgpu_async_gfx_ring = 1; int amdgpu_mcbp = -1; int amdgpu_discovery = -1; int amdgpu_mes; +int amdgpu_mes_log_enable = 0; int amdgpu_mes_kiq; int amdgpu_noretry = -1; int amdgpu_force_asic_type = -1; @@ -667,6 +668,15 @@ MODULE_PARM_DESC(mes, "Enable Micro Engine Scheduler (0 = disabled (default), 1 = enabled)"); module_param_named(mes, amdgpu_mes, int, 0444); +/** + * DOC: mes_log_enable (int) + * Enable Micro Engine Scheduler log. This is used to enable/disable MES internal log. + * (0 = disabled (default), 1 = enabled) + */ +MODULE_PARM_DESC(mes_log_enable, + "Enable Micro Engine Scheduler log (0 = disabled (default), 1 = enabled)"); +module_param_named(mes_log_enable, amdgpu_mes_log_enable, int, 0444); + /** * DOC: mes_kiq (int) * Enable Micro Engine Scheduler KIQ. This is a new engine pipe for kiq. diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c index a98e03e0a51f..cfd613ea39b3 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c @@ -102,6 +102,9 @@ static int amdgpu_mes_event_log_init(struct amdgpu_device *adev) { int r; + if (!amdgpu_mes_log_enable) + return 0; + r = amdgpu_bo_create_kernel(adev, PAGE_SIZE, PAGE_SIZE, AMDGPU_GEM_DOMAIN_GTT, &adev->mes.event_log_gpu_obj, @@ -1565,7 +1568,7 @@ void amdgpu_debugfs_mes_event_log_init(struct amdgpu_device *adev) #if defined(CONFIG_DEBUG_FS) struct drm_minor *minor = adev_to_drm(adev)->primary; struct dentry *root = minor->debugfs_root; - if (adev->enable_mes) + if (adev->enable_mes && amdgpu_mes_log_enable) debugfs_create_file("amdgpu_mes_event_log", 0444, root, adev, &amdgpu_debugfs_mes_event_log_fops); diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c index 072c478665ad..63f281a9984d 100644 --- a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c @@ -411,8 +411,11 @@ static int mes_v11_0_set_hw_resources(struct amdgpu_mes *mes) mes_set_hw_res_pkt.enable_reg_active_poll = 1; mes_set_hw_res_pkt.enable_level_process_quantum_check = 1; mes_set_hw_res_pkt.oversubscription_timer = 50; - mes_set_hw_res_pkt.enable_mes_event_int_logging = 1; - mes_set_hw_res_pkt.event_intr_history_gpu_mc_ptr = mes->event_log_gpu_addr; + if (amdgpu_mes_log_enable) { + mes_set_hw_res_pkt.enable_mes_event_int_logging = 1; + mes_set_hw_res_pkt.event_intr_history_gpu_mc_ptr = + mes->event_log_gpu_addr; + } return mes_v11_0_submit_pkt_and_poll_completion(mes, &mes_set_hw_res_pkt, sizeof(mes_set_hw_res_pkt), -- cgit v1.2.3