From 05ba44b3704e2115251c340fc72db609a0b97f3b Mon Sep 17 00:00:00 2001 From: Rob Clark Date: Tue, 2 Aug 2022 08:51:36 -0700 Subject: drm/msm: Split out idr_lock Otherwise if we hit reclaim pinning objects in the submit path, we'll be blocking retire_worker trying to free a submit. Signed-off-by: Rob Clark Patchwork: https://patchwork.freedesktop.org/patch/496116/ Link: https://lore.kernel.org/r/20220802155152.1727594-4-robdclark@gmail.com --- drivers/gpu/drm/msm/msm_drv.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'drivers/gpu/drm/msm/msm_drv.c') diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c index 1ed4cd09dbf8..d7ca025457b6 100644 --- a/drivers/gpu/drm/msm/msm_drv.c +++ b/drivers/gpu/drm/msm/msm_drv.c @@ -883,13 +883,13 @@ static int wait_fence(struct msm_gpu_submitqueue *queue, uint32_t fence_id, * retired, so if the fence is not found it means there is nothing * to wait for */ - ret = mutex_lock_interruptible(&queue->lock); + ret = mutex_lock_interruptible(&queue->idr_lock); if (ret) return ret; fence = idr_find(&queue->fence_idr, fence_id); if (fence) fence = dma_fence_get_rcu(fence); - mutex_unlock(&queue->lock); + mutex_unlock(&queue->idr_lock); if (!fence) return 0; -- cgit v1.2.3 From b352ba54a82072614f44c5ea352b5ea6e7f7b4c0 Mon Sep 17 00:00:00 2001 From: Rob Clark Date: Tue, 2 Aug 2022 08:51:43 -0700 Subject: drm/msm/gem: Convert to using drm_gem_lru This converts over to use the shared GEM LRU/shrinker helpers. Note that it means we are no longer tracking purgeable or willneed buffers that are active separately. But the most recently pinned buffers should be at the tail of the various LRUs, and the shrinker is already prepared to encounter objects which are still active. Signed-off-by: Rob Clark Patchwork: https://patchwork.freedesktop.org/patch/496131/ Link: https://lore.kernel.org/r/20220802155152.1727594-11-robdclark@gmail.com --- drivers/gpu/drm/msm/msm_drv.c | 14 ++-- drivers/gpu/drm/msm/msm_drv.h | 70 ++++++++++++++------ drivers/gpu/drm/msm/msm_gem.c | 58 +++++----------- drivers/gpu/drm/msm/msm_gem.h | 93 -------------------------- drivers/gpu/drm/msm/msm_gem_shrinker.c | 117 ++++++++------------------------- drivers/gpu/drm/msm/msm_gpu.c | 3 - drivers/gpu/drm/msm/msm_gpu.h | 6 -- 7 files changed, 104 insertions(+), 257 deletions(-) (limited to 'drivers/gpu/drm/msm/msm_drv.c') diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c index d7ca025457b6..1ca4a92ba96e 100644 --- a/drivers/gpu/drm/msm/msm_drv.c +++ b/drivers/gpu/drm/msm/msm_drv.c @@ -418,14 +418,18 @@ static int msm_drm_init(struct device *dev, const struct drm_driver *drv) INIT_LIST_HEAD(&priv->objects); mutex_init(&priv->obj_lock); - INIT_LIST_HEAD(&priv->inactive_willneed); - INIT_LIST_HEAD(&priv->inactive_dontneed); - INIT_LIST_HEAD(&priv->inactive_unpinned); - mutex_init(&priv->mm_lock); + /* + * Initialize the LRUs: + */ + mutex_init(&priv->lru.lock); + drm_gem_lru_init(&priv->lru.unbacked, &priv->lru.lock); + drm_gem_lru_init(&priv->lru.pinned, &priv->lru.lock); + drm_gem_lru_init(&priv->lru.willneed, &priv->lru.lock); + drm_gem_lru_init(&priv->lru.dontneed, &priv->lru.lock); /* Teach lockdep about lock ordering wrt. shrinker: */ fs_reclaim_acquire(GFP_KERNEL); - might_lock(&priv->mm_lock); + might_lock(&priv->lru.lock); fs_reclaim_release(GFP_KERNEL); drm_mode_config_init(ddev); diff --git a/drivers/gpu/drm/msm/msm_drv.h b/drivers/gpu/drm/msm/msm_drv.h index b3689a2d27d7..208ae5bc5e6b 100644 --- a/drivers/gpu/drm/msm/msm_drv.h +++ b/drivers/gpu/drm/msm/msm_drv.h @@ -142,28 +142,60 @@ struct msm_drm_private { struct mutex obj_lock; /** - * LRUs of inactive GEM objects. Every bo is either in one of the - * inactive lists (depending on whether or not it is shrinkable) or - * gpu->active_list (for the gpu it is active on[1]), or transiently - * on a temporary list as the shrinker is running. + * lru: * - * Note that inactive_willneed also contains pinned and vmap'd bos, - * but the number of pinned-but-not-active objects is small (scanout - * buffers, ringbuffer, etc). + * The various LRU's that a GEM object is in at various stages of + * it's lifetime. Objects start out in the unbacked LRU. When + * pinned (for scannout or permanently mapped GPU buffers, like + * ringbuffer, memptr, fw, etc) it moves to the pinned LRU. When + * unpinned, it moves into willneed or dontneed LRU depending on + * madvise state. When backing pages are evicted (willneed) or + * purged (dontneed) it moves back into the unbacked LRU. * - * These lists are protected by mm_lock (which should be acquired - * before per GEM object lock). One should *not* hold mm_lock in - * get_pages()/vmap()/etc paths, as they can trigger the shrinker. - * - * [1] if someone ever added support for the old 2d cores, there could be - * more than one gpu object + * The dontneed LRU is considered by the shrinker for objects + * that are candidate for purging, and the willneed LRU is + * considered for objects that could be evicted. */ - struct list_head inactive_willneed; /* inactive + potentially unpin/evictable */ - struct list_head inactive_dontneed; /* inactive + shrinkable */ - struct list_head inactive_unpinned; /* inactive + purged or unpinned */ - long shrinkable_count; /* write access under mm_lock */ - long evictable_count; /* write access under mm_lock */ - struct mutex mm_lock; + struct { + /** + * unbacked: + * + * The LRU for GEM objects without backing pages allocated. + * This mostly exists so that objects are always is one + * LRU. + */ + struct drm_gem_lru unbacked; + + /** + * pinned: + * + * The LRU for pinned GEM objects + */ + struct drm_gem_lru pinned; + + /** + * willneed: + * + * The LRU for unpinned GEM objects which are in madvise + * WILLNEED state (ie. can be evicted) + */ + struct drm_gem_lru willneed; + + /** + * dontneed: + * + * The LRU for unpinned GEM objects which are in madvise + * DONTNEED state (ie. can be purged) + */ + struct drm_gem_lru dontneed; + + /** + * lock: + * + * Protects manipulation of all of the LRUs. + */ + struct mutex lock; + } lru; struct workqueue_struct *wq; diff --git a/drivers/gpu/drm/msm/msm_gem.c b/drivers/gpu/drm/msm/msm_gem.c index 209438744bab..d4e8af46f4ef 100644 --- a/drivers/gpu/drm/msm/msm_gem.c +++ b/drivers/gpu/drm/msm/msm_gem.c @@ -174,6 +174,7 @@ static void put_pages(struct drm_gem_object *obj) put_pages_vram(obj); msm_obj->pages = NULL; + update_lru(obj); } } @@ -210,8 +211,6 @@ struct page **msm_gem_pin_pages(struct drm_gem_object *obj) void msm_gem_unpin_pages(struct drm_gem_object *obj) { - struct msm_gem_object *msm_obj = to_msm_bo(obj); - msm_gem_lock(obj); msm_gem_unpin_locked(obj); msm_gem_unlock(obj); @@ -761,7 +760,6 @@ void msm_gem_purge(struct drm_gem_object *obj) put_iova_vmas(obj); msm_obj->madv = __MSM_MADV_PURGED; - update_lru(obj); drm_gem_free_mmap_offset(obj); @@ -786,7 +784,6 @@ void msm_gem_evict(struct drm_gem_object *obj) GEM_WARN_ON(!msm_gem_is_locked(obj)); GEM_WARN_ON(is_unevictable(msm_obj)); - GEM_WARN_ON(!msm_obj->evictable); /* Get rid of any iommu mapping(s): */ put_iova_spaces(obj, false); @@ -794,8 +791,6 @@ void msm_gem_evict(struct drm_gem_object *obj) drm_vma_node_unmap(&obj->vma_node, dev->anon_inode->i_mapping); put_pages(obj); - - update_lru(obj); } void msm_gem_vunmap(struct drm_gem_object *obj) @@ -818,26 +813,20 @@ static void update_lru(struct drm_gem_object *obj) GEM_WARN_ON(!msm_gem_is_locked(&msm_obj->base)); - mutex_lock(&priv->mm_lock); - - if (msm_obj->dontneed) - mark_unpurgeable(msm_obj); - if (msm_obj->evictable) - mark_unevictable(msm_obj); - - list_del(&msm_obj->mm_list); - if ((msm_obj->madv == MSM_MADV_WILLNEED) && msm_obj->sgt) { - list_add_tail(&msm_obj->mm_list, &priv->inactive_willneed); - mark_evictable(msm_obj); - } else if (msm_obj->madv == MSM_MADV_DONTNEED) { - list_add_tail(&msm_obj->mm_list, &priv->inactive_dontneed); - mark_purgeable(msm_obj); + if (!msm_obj->pages) { + GEM_WARN_ON(msm_obj->pin_count); + GEM_WARN_ON(msm_obj->vmap_count); + + drm_gem_lru_move_tail(&priv->lru.unbacked, obj); + } else if (msm_obj->pin_count || msm_obj->vmap_count) { + drm_gem_lru_move_tail(&priv->lru.pinned, obj); + } else if (msm_obj->madv == MSM_MADV_WILLNEED) { + drm_gem_lru_move_tail(&priv->lru.willneed, obj); } else { - GEM_WARN_ON((msm_obj->madv != __MSM_MADV_PURGED) && msm_obj->sgt); - list_add_tail(&msm_obj->mm_list, &priv->inactive_unpinned); - } + GEM_WARN_ON(msm_obj->madv != MSM_MADV_DONTNEED); - mutex_unlock(&priv->mm_lock); + drm_gem_lru_move_tail(&priv->lru.dontneed, obj); + } } bool msm_gem_active(struct drm_gem_object *obj) @@ -995,12 +984,6 @@ static void msm_gem_free_object(struct drm_gem_object *obj) list_del(&msm_obj->node); mutex_unlock(&priv->obj_lock); - mutex_lock(&priv->mm_lock); - if (msm_obj->dontneed) - mark_unpurgeable(msm_obj); - list_del(&msm_obj->mm_list); - mutex_unlock(&priv->mm_lock); - put_iova_spaces(obj, true); if (obj->import_attach) { @@ -1160,13 +1143,6 @@ struct drm_gem_object *msm_gem_new(struct drm_device *dev, uint32_t size, uint32 to_msm_bo(obj)->vram_node = &vma->node; - /* Call chain get_pages() -> update_inactive() tries to - * access msm_obj->mm_list, but it is not initialized yet. - * To avoid NULL pointer dereference error, initialize - * mm_list to be empty. - */ - INIT_LIST_HEAD(&msm_obj->mm_list); - msm_gem_lock(obj); pages = get_pages(obj); msm_gem_unlock(obj); @@ -1189,9 +1165,7 @@ struct drm_gem_object *msm_gem_new(struct drm_device *dev, uint32_t size, uint32 mapping_set_gfp_mask(obj->filp->f_mapping, GFP_HIGHUSER); } - mutex_lock(&priv->mm_lock); - list_add_tail(&msm_obj->mm_list, &priv->inactive_unpinned); - mutex_unlock(&priv->mm_lock); + drm_gem_lru_move_tail(&priv->lru.unbacked, obj); mutex_lock(&priv->obj_lock); list_add_tail(&msm_obj->node, &priv->objects); @@ -1247,9 +1221,7 @@ struct drm_gem_object *msm_gem_import(struct drm_device *dev, msm_gem_unlock(obj); - mutex_lock(&priv->mm_lock); - list_add_tail(&msm_obj->mm_list, &priv->inactive_unpinned); - mutex_unlock(&priv->mm_lock); + drm_gem_lru_move_tail(&priv->lru.pinned, obj); mutex_lock(&priv->obj_lock); list_add_tail(&msm_obj->node, &priv->objects); diff --git a/drivers/gpu/drm/msm/msm_gem.h b/drivers/gpu/drm/msm/msm_gem.h index 420ba49bf21a..0403b27ff779 100644 --- a/drivers/gpu/drm/msm/msm_gem.h +++ b/drivers/gpu/drm/msm/msm_gem.h @@ -93,16 +93,6 @@ struct msm_gem_object { */ uint8_t madv; - /** - * Is object on inactive_dontneed list (ie. counted in priv->shrinkable_count)? - */ - bool dontneed : 1; - - /** - * Is object evictable (ie. counted in priv->evictable_count)? - */ - bool evictable : 1; - /** * count of active vmap'ing */ @@ -114,17 +104,6 @@ struct msm_gem_object { */ struct list_head node; - /** - * An object is either: - * inactive - on priv->inactive_dontneed or priv->inactive_willneed - * (depending on purgeability status) - * active - on one one of the gpu's active_list.. well, at - * least for now we don't have (I don't think) hw sync between - * 2d and 3d one devices which have both, meaning we need to - * block on submit if a bo is already on other ring - */ - struct list_head mm_list; - struct page **pages; struct sg_table *sgt; void *vaddr; @@ -206,12 +185,6 @@ msm_gem_lock(struct drm_gem_object *obj) dma_resv_lock(obj->resv, NULL); } -static inline bool __must_check -msm_gem_trylock(struct drm_gem_object *obj) -{ - return dma_resv_trylock(obj->resv); -} - static inline int msm_gem_lock_interruptible(struct drm_gem_object *obj) { @@ -260,77 +233,11 @@ static inline bool is_vunmapable(struct msm_gem_object *msm_obj) return (msm_obj->vmap_count == 0) && msm_obj->vaddr; } -static inline void mark_purgeable(struct msm_gem_object *msm_obj) -{ - struct msm_drm_private *priv = msm_obj->base.dev->dev_private; - - GEM_WARN_ON(!mutex_is_locked(&priv->mm_lock)); - - if (is_unpurgeable(msm_obj)) - return; - - if (GEM_WARN_ON(msm_obj->dontneed)) - return; - - priv->shrinkable_count += msm_obj->base.size >> PAGE_SHIFT; - msm_obj->dontneed = true; -} - -static inline void mark_unpurgeable(struct msm_gem_object *msm_obj) -{ - struct msm_drm_private *priv = msm_obj->base.dev->dev_private; - - GEM_WARN_ON(!mutex_is_locked(&priv->mm_lock)); - - if (is_unpurgeable(msm_obj)) - return; - - if (GEM_WARN_ON(!msm_obj->dontneed)) - return; - - priv->shrinkable_count -= msm_obj->base.size >> PAGE_SHIFT; - GEM_WARN_ON(priv->shrinkable_count < 0); - msm_obj->dontneed = false; -} - static inline bool is_unevictable(struct msm_gem_object *msm_obj) { return is_unpurgeable(msm_obj) || msm_obj->vaddr; } -static inline void mark_evictable(struct msm_gem_object *msm_obj) -{ - struct msm_drm_private *priv = msm_obj->base.dev->dev_private; - - WARN_ON(!mutex_is_locked(&priv->mm_lock)); - - if (is_unevictable(msm_obj)) - return; - - if (WARN_ON(msm_obj->evictable)) - return; - - priv->evictable_count += msm_obj->base.size >> PAGE_SHIFT; - msm_obj->evictable = true; -} - -static inline void mark_unevictable(struct msm_gem_object *msm_obj) -{ - struct msm_drm_private *priv = msm_obj->base.dev->dev_private; - - WARN_ON(!mutex_is_locked(&priv->mm_lock)); - - if (is_unevictable(msm_obj)) - return; - - if (WARN_ON(!msm_obj->evictable)) - return; - - priv->evictable_count -= msm_obj->base.size >> PAGE_SHIFT; - WARN_ON(priv->evictable_count < 0); - msm_obj->evictable = false; -} - void msm_gem_purge(struct drm_gem_object *obj); void msm_gem_evict(struct drm_gem_object *obj); void msm_gem_vunmap(struct drm_gem_object *obj); diff --git a/drivers/gpu/drm/msm/msm_gem_shrinker.c b/drivers/gpu/drm/msm/msm_gem_shrinker.c index 0ab058734777..f1e52410f521 100644 --- a/drivers/gpu/drm/msm/msm_gem_shrinker.c +++ b/drivers/gpu/drm/msm/msm_gem_shrinker.c @@ -29,121 +29,61 @@ msm_gem_shrinker_count(struct shrinker *shrinker, struct shrink_control *sc) { struct msm_drm_private *priv = container_of(shrinker, struct msm_drm_private, shrinker); - unsigned count = priv->shrinkable_count; + unsigned count = priv->lru.dontneed.count; if (can_swap()) - count += priv->evictable_count; + count += priv->lru.willneed.count; return count; } static bool -purge(struct msm_gem_object *msm_obj) +purge(struct drm_gem_object *obj) { - if (!is_purgeable(msm_obj)) + if (!is_purgeable(to_msm_bo(obj))) return false; - if (msm_gem_active(&msm_obj->base)) + if (msm_gem_active(obj)) return false; - /* - * This will move the obj out of still_in_list to - * the purged list - */ - msm_gem_purge(&msm_obj->base); + msm_gem_purge(obj); return true; } static bool -evict(struct msm_gem_object *msm_obj) +evict(struct drm_gem_object *obj) { - if (is_unevictable(msm_obj)) + if (is_unevictable(to_msm_bo(obj))) return false; - if (msm_gem_active(&msm_obj->base)) + if (msm_gem_active(obj)) return false; - msm_gem_evict(&msm_obj->base); + msm_gem_evict(obj); return true; } -static unsigned long -scan(struct msm_drm_private *priv, unsigned nr_to_scan, struct list_head *list, - bool (*shrink)(struct msm_gem_object *msm_obj)) -{ - unsigned freed = 0; - struct list_head still_in_list; - - INIT_LIST_HEAD(&still_in_list); - - mutex_lock(&priv->mm_lock); - - while (freed < nr_to_scan) { - struct msm_gem_object *msm_obj = list_first_entry_or_null( - list, typeof(*msm_obj), mm_list); - - if (!msm_obj) - break; - - list_move_tail(&msm_obj->mm_list, &still_in_list); - - /* - * If it is in the process of being freed, msm_gem_free_object - * can be blocked on mm_lock waiting to remove it. So just - * skip it. - */ - if (!kref_get_unless_zero(&msm_obj->base.refcount)) - continue; - - /* - * Now that we own a reference, we can drop mm_lock for the - * rest of the loop body, to reduce contention with the - * retire_submit path (which could make more objects purgeable) - */ - - mutex_unlock(&priv->mm_lock); - - /* - * Note that this still needs to be trylock, since we can - * hit shrinker in response to trying to get backing pages - * for this obj (ie. while it's lock is already held) - */ - if (!msm_gem_trylock(&msm_obj->base)) - goto tail; - - if (shrink(msm_obj)) - freed += msm_obj->base.size >> PAGE_SHIFT; - - msm_gem_unlock(&msm_obj->base); - -tail: - drm_gem_object_put(&msm_obj->base); - mutex_lock(&priv->mm_lock); - } - - list_splice_tail(&still_in_list, list); - mutex_unlock(&priv->mm_lock); - - return freed; -} - static unsigned long msm_gem_shrinker_scan(struct shrinker *shrinker, struct shrink_control *sc) { struct msm_drm_private *priv = container_of(shrinker, struct msm_drm_private, shrinker); + long nr = sc->nr_to_scan; unsigned long freed; - freed = scan(priv, sc->nr_to_scan, &priv->inactive_dontneed, purge); + freed = drm_gem_lru_scan(&priv->lru.dontneed, nr, purge); + nr -= freed; if (freed > 0) trace_msm_gem_purge(freed << PAGE_SHIFT); - if (can_swap() && freed < sc->nr_to_scan) { - int evicted = scan(priv, sc->nr_to_scan - freed, - &priv->inactive_willneed, evict); + if (can_swap() && nr > 0) { + unsigned long evicted; + + evicted = drm_gem_lru_scan(&priv->lru.willneed, nr, evict); + nr -= evicted; if (evicted > 0) trace_msm_gem_evict(evicted << PAGE_SHIFT); @@ -179,12 +119,12 @@ msm_gem_shrinker_shrink(struct drm_device *dev, unsigned long nr_to_scan) static const int vmap_shrink_limit = 15; static bool -vmap_shrink(struct msm_gem_object *msm_obj) +vmap_shrink(struct drm_gem_object *obj) { - if (!is_vunmapable(msm_obj)) + if (!is_vunmapable(to_msm_bo(obj))) return false; - msm_gem_vunmap(&msm_obj->base); + msm_gem_vunmap(obj); return true; } @@ -194,17 +134,18 @@ msm_gem_shrinker_vmap(struct notifier_block *nb, unsigned long event, void *ptr) { struct msm_drm_private *priv = container_of(nb, struct msm_drm_private, vmap_notifier); - struct list_head *mm_lists[] = { - &priv->inactive_dontneed, - &priv->inactive_willneed, - priv->gpu ? &priv->gpu->active_list : NULL, + struct drm_gem_lru *lrus[] = { + &priv->lru.dontneed, + &priv->lru.willneed, + &priv->lru.pinned, NULL, }; unsigned idx, unmapped = 0; - for (idx = 0; mm_lists[idx] && unmapped < vmap_shrink_limit; idx++) { - unmapped += scan(priv, vmap_shrink_limit - unmapped, - mm_lists[idx], vmap_shrink); + for (idx = 0; lrus[idx] && unmapped < vmap_shrink_limit; idx++) { + unmapped += drm_gem_lru_scan(lrus[idx], + vmap_shrink_limit - unmapped, + vmap_shrink); } *(unsigned long *)ptr += unmapped; diff --git a/drivers/gpu/drm/msm/msm_gpu.c b/drivers/gpu/drm/msm/msm_gpu.c index c2bfcf3f1f40..fba85f894314 100644 --- a/drivers/gpu/drm/msm/msm_gpu.c +++ b/drivers/gpu/drm/msm/msm_gpu.c @@ -846,7 +846,6 @@ int msm_gpu_init(struct drm_device *drm, struct platform_device *pdev, sched_set_fifo_low(gpu->worker->task); - INIT_LIST_HEAD(&gpu->active_list); mutex_init(&gpu->active_lock); mutex_init(&gpu->lock); init_waitqueue_head(&gpu->retire_event); @@ -974,8 +973,6 @@ void msm_gpu_cleanup(struct msm_gpu *gpu) DBG("%s", gpu->name); - WARN_ON(!list_empty(&gpu->active_list)); - for (i = 0; i < ARRAY_SIZE(gpu->rb); i++) { msm_ringbuffer_destroy(gpu->rb[i]); gpu->rb[i] = NULL; diff --git a/drivers/gpu/drm/msm/msm_gpu.h b/drivers/gpu/drm/msm/msm_gpu.h index 962d2070bcdf..a0885ff99e5b 100644 --- a/drivers/gpu/drm/msm/msm_gpu.h +++ b/drivers/gpu/drm/msm/msm_gpu.h @@ -187,12 +187,6 @@ struct msm_gpu { */ int cur_ctx_seqno; - /* - * List of GEM active objects on this gpu. Protected by - * msm_drm_private::mm_lock - */ - struct list_head active_list; - /** * lock: * -- cgit v1.2.3 From 6d29709de8028cfd5a827d02f93a5076ba632563 Mon Sep 17 00:00:00 2001 From: Rob Clark Date: Sun, 7 Aug 2022 10:28:48 -0700 Subject: drm/msm: Add fault-injection support Intended as a way to trigger error paths in mesa. Signed-off-by: Rob Clark Reviewed-by: Dmitry Baryshkov Patchwork: https://patchwork.freedesktop.org/patch/496710/ Link: https://lore.kernel.org/r/20220807172848.2432845-1-robdclark@gmail.com --- drivers/gpu/drm/msm/msm_debugfs.c | 8 ++++++++ drivers/gpu/drm/msm/msm_drv.c | 15 +++++++++++++++ drivers/gpu/drm/msm/msm_drv.h | 7 +++++++ 3 files changed, 30 insertions(+) (limited to 'drivers/gpu/drm/msm/msm_drv.c') diff --git a/drivers/gpu/drm/msm/msm_debugfs.c b/drivers/gpu/drm/msm/msm_debugfs.c index 7d2dab260f86..95f4374ae21c 100644 --- a/drivers/gpu/drm/msm/msm_debugfs.c +++ b/drivers/gpu/drm/msm/msm_debugfs.c @@ -7,6 +7,7 @@ #ifdef CONFIG_DEBUG_FS #include +#include #include #include @@ -326,6 +327,13 @@ void msm_debugfs_init(struct drm_minor *minor) if (priv->kms && priv->kms->funcs->debugfs_init) priv->kms->funcs->debugfs_init(priv->kms, minor); + +#ifdef CONFIG_FAULT_INJECTION + fault_create_debugfs_attr("fail_gem_alloc", minor->debugfs_root, + &fail_gem_alloc); + fault_create_debugfs_attr("fail_gem_iova", minor->debugfs_root, + &fail_gem_iova); +#endif } #endif diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c index 1ca4a92ba96e..b564ff685f67 100644 --- a/drivers/gpu/drm/msm/msm_drv.c +++ b/drivers/gpu/drm/msm/msm_drv.c @@ -6,6 +6,7 @@ */ #include +#include #include #include #include @@ -78,6 +79,11 @@ static bool modeset = true; MODULE_PARM_DESC(modeset, "Use kernel modesetting [KMS] (1=on (default), 0=disable)"); module_param(modeset, bool, 0600); +#ifdef CONFIG_FAULT_INJECTION +DECLARE_FAULT_ATTR(fail_gem_alloc); +DECLARE_FAULT_ATTR(fail_gem_iova); +#endif + static irqreturn_t msm_irq(int irq, void *arg) { struct drm_device *dev = arg; @@ -701,6 +707,9 @@ static int msm_ioctl_gem_new(struct drm_device *dev, void *data, flags |= MSM_BO_WC; } + if (should_fail(&fail_gem_alloc, args->size)) + return -ENOMEM; + return msm_gem_new_handle(dev, file, args->size, args->flags, &args->handle, NULL); } @@ -762,6 +771,9 @@ static int msm_ioctl_gem_info_iova(struct drm_device *dev, if (!priv->gpu) return -EINVAL; + if (should_fail(&fail_gem_iova, obj->size)) + return -ENOMEM; + /* * Don't pin the memory here - just get an address so that userspace can * be productive @@ -783,6 +795,9 @@ static int msm_ioctl_gem_info_set_iova(struct drm_device *dev, if (priv->gpu->aspace == ctx->aspace) return -EOPNOTSUPP; + if (should_fail(&fail_gem_iova, obj->size)) + return -ENOMEM; + return msm_gem_set_iova(obj, ctx->aspace, iova); } diff --git a/drivers/gpu/drm/msm/msm_drv.h b/drivers/gpu/drm/msm/msm_drv.h index 208ae5bc5e6b..09c4585a8064 100644 --- a/drivers/gpu/drm/msm/msm_drv.h +++ b/drivers/gpu/drm/msm/msm_drv.h @@ -34,6 +34,13 @@ #include #include +#ifdef CONFIG_FAULT_INJECTION +extern struct fault_attr fail_gem_alloc; +extern struct fault_attr fail_gem_iova; +#else +# define should_fail(attr, size) 0 +#endif + struct msm_kms; struct msm_gpu; struct msm_mmu; -- cgit v1.2.3 From c51720a6d0b402ed418c904500f78ce1897a93bc Mon Sep 17 00:00:00 2001 From: Kuogee Hsieh Date: Wed, 6 Jul 2022 12:32:08 -0700 Subject: drm/msm/dp: make eDP panel as the first connected connector Some userspace presumes that the first connected connector is the main display, where it's supposed to display e.g. the login screen. For laptops, this should be the main panel. This patch call drm_helper_move_panel_connectors_to_head() after drm_bridge_connector_init() to make sure eDP stay at head of connected connector list. This fixes unexpected corruption happen at eDP panel if eDP is not placed at head of connected connector list. Changes in v2: -- move drm_helper_move_panel_connectors_to_head() to dpu_kms_drm_obj_init() Changes in v4: -- move drm_helper_move_panel_connectors_to_head() to msm_drm_init() Signed-off-by: Kuogee Hsieh Reviewed-by: Abhinav Kumar Reviewed-by: Douglas Anderson Tested-by: Douglas Anderson Reviewed-by: Dmitry Baryshkov Reviewed-by: Stephen Boyd Fixes: ef7837ff091c ("drm/msm/dp: Add DP controllers for sc7280") Patchwork: https://patchwork.freedesktop.org/patch/492581/ Link: https://lore.kernel.org/r/1657135928-31195-1-git-send-email-quic_khsieh@quicinc.com Signed-off-by: Abhinav Kumar Signed-off-by: Rob Clark --- drivers/gpu/drm/msm/msm_drv.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'drivers/gpu/drm/msm/msm_drv.c') diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c index b564ff685f67..3403d7da6cae 100644 --- a/drivers/gpu/drm/msm/msm_drv.c +++ b/drivers/gpu/drm/msm/msm_drv.c @@ -479,6 +479,8 @@ static int msm_drm_init(struct device *dev, const struct drm_driver *drv) } } + drm_helper_move_panel_connectors_to_head(ddev); + ddev->mode_config.funcs = &mode_config_funcs; ddev->mode_config.helper_private = &mode_config_helper_funcs; -- cgit v1.2.3 From 89d529e393199502a951819a51454ddd64e62681 Mon Sep 17 00:00:00 2001 From: Javier Martinez Canillas Date: Tue, 16 Aug 2022 15:46:12 +0200 Subject: drm/msm: Make .remove and .shutdown HW shutdown consistent Drivers' .remove and .shutdown callbacks are executed on different code paths. The former is called when a device is removed from the bus, while the latter is called at system shutdown time to quiesce the device. This means that some overlap exists between the two, because both have to take care of properly shutting down the hardware. But currently the logic used in these two callbacks isn't consistent in msm drivers, which could lead to kernel panic. For example, on .remove the component is deleted and its .unbind callback leads to the hardware being shutdown but only if the DRM device has been marked as registered. That check doesn't exist in the .shutdown logic and this can lead to the driver calling drm_atomic_helper_shutdown() for a DRM device that hasn't been properly initialized. A situation like this can happen if drivers for expected sub-devices fail to probe, since the .bind callback will never be executed. If that is the case, drm_atomic_helper_shutdown() will attempt to take mutexes that are only initialized if drm_mode_config_init() is called during a device bind. This bug was attempted to be fixed in commit 623f279c7781 ("drm/msm: fix shutdown hook in case GPU components failed to bind"), but unfortunately it still happens in some cases as the one mentioned above, i.e: systemd-shutdown[1]: Powering off. kvm: exiting hardware virtualization platform wifi-firmware.0: Removing from iommu group 12 platform video-firmware.0: Removing from iommu group 10 ------------[ cut here ]------------ WARNING: CPU: 6 PID: 1 at drivers/gpu/drm/drm_modeset_lock.c:317 drm_modeset_lock_all_ctx+0x3c4/0x3d0 ... Hardware name: Google CoachZ (rev3+) (DT) pstate: a0400009 (NzCv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--) pc : drm_modeset_lock_all_ctx+0x3c4/0x3d0 lr : drm_modeset_lock_all_ctx+0x48/0x3d0 sp : ffff80000805bb80 x29: ffff80000805bb80 x28: ffff327c00128000 x27: 0000000000000000 x26: 0000000000000000 x25: 0000000000000001 x24: ffffc95d820ec030 x23: ffff327c00bbd090 x22: ffffc95d8215eca0 x21: ffff327c039c5800 x20: ffff327c039c5988 x19: ffff80000805bbe8 x18: 0000000000000034 x17: 000000040044ffff x16: ffffc95d80cac920 x15: 0000000000000000 x14: 0000000000000315 x13: 0000000000000315 x12: 0000000000000000 x11: 0000000000000000 x10: 0000000000000000 x9 : 0000000000000000 x8 : ffff80000805bc28 x7 : 0000000000000000 x6 : 0000000000000000 x5 : 0000000000000000 x4 : 0000000000000000 x3 : 0000000000000000 x2 : ffff327c00128000 x1 : 0000000000000000 x0 : ffff327c039c59b0 Call trace: drm_modeset_lock_all_ctx+0x3c4/0x3d0 drm_atomic_helper_shutdown+0x70/0x134 msm_drv_shutdown+0x30/0x40 platform_shutdown+0x28/0x40 device_shutdown+0x148/0x350 kernel_power_off+0x38/0x80 __do_sys_reboot+0x288/0x2c0 __arm64_sys_reboot+0x28/0x34 invoke_syscall+0x48/0x114 el0_svc_common.constprop.0+0x44/0xec do_el0_svc+0x2c/0xc0 el0_svc+0x2c/0x84 el0t_64_sync_handler+0x11c/0x150 el0t_64_sync+0x18c/0x190 ---[ end trace 0000000000000000 ]--- Unable to handle kernel NULL pointer dereference at virtual address 0000000000000018 Mem abort info: ESR = 0x0000000096000004 EC = 0x25: DABT (current EL), IL = 32 bits SET = 0, FnV = 0 EA = 0, S1PTW = 0 FSC = 0x04: level 0 translation fault Data abort info: ISV = 0, ISS = 0x00000004 CM = 0, WnR = 0 user pgtable: 4k pages, 48-bit VAs, pgdp=000000010eab1000 [0000000000000018] pgd=0000000000000000, p4d=0000000000000000 Internal error: Oops: 96000004 [#1] PREEMPT SMP ... Hardware name: Google CoachZ (rev3+) (DT) pstate: a0400009 (NzCv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--) pc : ww_mutex_lock+0x28/0x32c lr : drm_modeset_lock_all_ctx+0x1b0/0x3d0 sp : ffff80000805bb50 x29: ffff80000805bb50 x28: ffff327c00128000 x27: 0000000000000000 x26: 0000000000000000 x25: 0000000000000001 x24: 0000000000000018 x23: ffff80000805bc10 x22: ffff327c039c5ad8 x21: ffff327c039c5800 x20: ffff80000805bbe8 x19: 0000000000000018 x18: 0000000000000034 x17: 000000040044ffff x16: ffffc95d80cac920 x15: 0000000000000000 x14: 0000000000000315 x13: 0000000000000315 x12: 0000000000000000 x11: 0000000000000000 x10: 0000000000000000 x9 : 0000000000000000 x8 : ffff80000805bc28 x7 : 0000000000000000 x6 : 0000000000000000 x5 : 0000000000000000 x4 : 0000000000000000 x3 : 0000000000000000 x2 : ffff327c00128000 x1 : 0000000000000000 x0 : 0000000000000018 Call trace: ww_mutex_lock+0x28/0x32c drm_modeset_lock_all_ctx+0x1b0/0x3d0 drm_atomic_helper_shutdown+0x70/0x134 msm_drv_shutdown+0x30/0x40 platform_shutdown+0x28/0x40 device_shutdown+0x148/0x350 kernel_power_off+0x38/0x80 __do_sys_reboot+0x288/0x2c0 __arm64_sys_reboot+0x28/0x34 invoke_syscall+0x48/0x114 el0_svc_common.constprop.0+0x44/0xec do_el0_svc+0x2c/0xc0 el0_svc+0x2c/0x84 el0t_64_sync_handler+0x11c/0x150 el0t_64_sync+0x18c/0x190 Code: aa0103f4 d503201f d2800001 aa0103e3 (c8e37c02) ---[ end trace 0000000000000000 ]--- Kernel panic - not syncing: Attempted to kill init! exitcode=0x0000000b Kernel Offset: 0x495d77c00000 from 0xffff800008000000 PHYS_OFFSET: 0xffffcd8500000000 CPU features: 0x800,00c2a015,19801c82 Memory Limit: none ---[ end Kernel panic - not syncing: Attempted to kill init! exitcode=0x0000000b ]--- Fixes: 9d5cbf5fe46e ("drm/msm: add shutdown support for display platform_driver") Signed-off-by: Javier Martinez Canillas Reviewed-by: Abhinav Kumar Patchwork: https://patchwork.freedesktop.org/patch/497842/ Link: https://lore.kernel.org/r/20220816134612.916527-1-javierm@redhat.com Signed-off-by: Dmitry Baryshkov Signed-off-by: Rob Clark --- drivers/gpu/drm/msm/msm_drv.c | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-) (limited to 'drivers/gpu/drm/msm/msm_drv.c') diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c index 3403d7da6cae..07f66412533b 100644 --- a/drivers/gpu/drm/msm/msm_drv.c +++ b/drivers/gpu/drm/msm/msm_drv.c @@ -1263,10 +1263,15 @@ void msm_drv_shutdown(struct platform_device *pdev) struct msm_drm_private *priv = platform_get_drvdata(pdev); struct drm_device *drm = priv ? priv->dev : NULL; - if (!priv || !priv->kms) - return; - - drm_atomic_helper_shutdown(drm); + /* + * Shutdown the hw if we're far enough along where things might be on. + * If we run this too early, we'll end up panicking in any variety of + * places. Since we don't register the drm device until late in + * msm_drm_init, drm_dev->registered is used as an indicator that the + * shutdown will be successful. + */ + if (drm && drm->registered) + drm_atomic_helper_shutdown(drm); } static struct platform_driver msm_platform_driver = { -- cgit v1.2.3