Skip to contentSkip to conversation

drm+msm: Shrinker and LRU rework

PATCH New Rob Clark 1 participant 16 messages Aug 2, 2022

Subsystems (5)
lore
Share & tools
lore

Rob Clark

author
#0 lore raw

v4 cover drm+msm: Shrinker and LRU rework

From: Rob Clark <[email protected]>

Unchanged other than small update in 09/15

original description below:

This is mostly motivated by getting drm/msm to pass an i-g-t shrinker test that I've been working on. In particular the test creates and cycles between more GEM buffers than what will fit in RAM to force eviction and re-pin. (There are sub-tests that cover this case both single threaded and with many child processes in parallel.)

Getting this test to pass necessitated a few improvements:

1. Re-ordering submit path to get rid of __GFP_NORETRY (in the common case, doing this for syncobjs is still TODO)
2. Decoupling locks needed in the retire path from locks that could be held while hitting reclaim in the submit path
3. If necessary, allow stalling on active BOs for reclaim.

The latter point is because we pin objects in the synchronous part of the submit path (before queuing on the drm gpu-scheduler), which means in the parallel variant of the i-g-t test, we need to be able to block in the reclaim path until some queued work has completed/retired.

In the process of re-working how drm/msm tracks buffer state in it's various LRU lists, I refactored out a drm_gem_lru helper which, in theory, should be usable by other drivers and drm shmem helpers for implementing LRU tracking and shrinker.

v2: rebase + small fix in 13/13
v3: use lockdep_assert_held in GEM LRU helper, and add a couple patches
at the end to convert MSM from WARN_ON(!is_locked()) to lockdep asserts
v4: keep drm_gem_move_tail_locked() static until there is a user

Rob Clark (15):
drm/msm: Reorder lock vs submit alloc drm/msm: Small submit cleanup
drm/msm: Split out idr_lock
drm/msm/gem: Check for active in shrinker path drm/msm/gem: Rename update_inactive drm/msm/gem: Rename to pin/unpin_pages drm/msm/gem: Consolidate pin/unpin paths drm/msm/gem: Remove active refcnt
drm/gem: Add LRU/shrinker helper drm/msm/gem: Convert to using drm_gem_lru drm/msm/gem: Unpin buffers earlier drm/msm/gem: Consolidate shrinker trace drm/msm/gem: Evict active GEM objects when necessary drm/msm/gem: Add msm_gem_assert_locked() drm/msm/gem: Convert to lockdep assert

 drivers/gpu/drm/drm_gem.c              | 170 +++++++++++++++++++++++
 drivers/gpu/drm/msm/msm_drv.c          |  18 ++-
 drivers/gpu/drm/msm/msm_drv.h          |  70 +++++++---
 drivers/gpu/drm/msm/msm_gem.c          | 179 +++++++++----------------
 drivers/gpu/drm/msm/msm_gem.h          | 123 ++---------------
 drivers/gpu/drm/msm/msm_gem_prime.c    |   4 +-
 drivers/gpu/drm/msm/msm_gem_shrinker.c | 164 +++++++++++-----------
 drivers/gpu/drm/msm/msm_gem_submit.c   |  78 ++++-------
 drivers/gpu/drm/msm/msm_gpu.c          |   3 -
 drivers/gpu/drm/msm/msm_gpu.h          |  10 +-
 drivers/gpu/drm/msm/msm_gpu_trace.h    |  36 +++--
 drivers/gpu/drm/msm/msm_submitqueue.c  |   1 +
 include/drm/drm_gem.h                  |  55 ++++++++
 13 files changed, 491 insertions(+), 420 deletions(-)
Signature · 1 line
2.36.1
author
#1 lore raw

PATCH v4 1/15 drm/msm: Reorder lock vs submit alloc

From: Rob Clark <[email protected]>

This lets us drop the NORETRY.

---
 drivers/gpu/drm/msm/msm_gem_submit.c | 24 ++++++++++--------------
 1 file changed, 10 insertions(+), 14 deletions(-)
drivers/gpu/drm/msm/msm_gem_submit.c +10 -14
index c9e4aeb14f4a..b7c61a99d274 100644
@@ -36,7 +36,7 @@ static struct msm_gem_submit *submit_create(struct drm_device *dev,
 	if (sz > SIZE_MAX)
 		return ERR_PTR(-ENOMEM);
 
-	submit = kzalloc(sz, GFP_KERNEL | __GFP_NOWARN | __GFP_NORETRY);
+	submit = kzalloc(sz, GFP_KERNEL);
 	if (!submit)
 		return ERR_PTR(-ENOMEM);
 
@@ -771,25 +771,21 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data,
 	trace_msm_gpu_submit(pid_nr(pid), ring->id, submitid,
 		args->nr_bos, args->nr_cmds);
 
-	ret = mutex_lock_interruptible(&queue->lock);
-	if (ret)
-		goto out_post_unlock;
-
 	if (args->flags & MSM_SUBMIT_FENCE_FD_OUT) {
 		out_fence_fd = get_unused_fd_flags(O_CLOEXEC);
 		if (out_fence_fd < 0) {
 			ret = out_fence_fd;
-			goto out_unlock;
+			return ret;
 		}
 	}
 
-	submit = submit_create(dev, gpu, queue, args->nr_bos,
-		args->nr_cmds);
-	if (IS_ERR(submit)) {
-		ret = PTR_ERR(submit);
-		submit = NULL;
-		goto out_unlock;
-	}
+	submit = submit_create(dev, gpu, queue, args->nr_bos, args->nr_cmds);
+	if (IS_ERR(submit))
+		return PTR_ERR(submit);
+
+	ret = mutex_lock_interruptible(&queue->lock);
+	if (ret)
+		goto out_post_unlock;
 
 	submit->pid = pid;
 	submit->ident = submitid;
@@ -965,9 +961,9 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data,
 	if (ret && (out_fence_fd >= 0))
 		put_unused_fd(out_fence_fd);
 	mutex_unlock(&queue->lock);
+out_post_unlock:
 	if (submit)
 		msm_gem_submit_put(submit);
-out_post_unlock:
 	if (!IS_ERR_OR_NULL(post_deps)) {
 		for (i = 0; i < args->nr_out_syncobjs; ++i) {
 			kfree(post_deps[i].chain);
Signature · 1 line
2.36.1
author
#2 lore raw

PATCH v4 2/15 drm/msm: Small submit cleanup

In reply to Rob Clark

From: Rob Clark <[email protected]>

Move more initialization into submit_create().

---
 drivers/gpu/drm/msm/msm_gem_submit.c | 20 +++++++++-----------
 1 file changed, 9 insertions(+), 11 deletions(-)
drivers/gpu/drm/msm/msm_gem_submit.c +9 -11
index b7c61a99d274..c7819781879c 100644
@@ -26,6 +26,7 @@ static struct msm_gem_submit *submit_create(struct drm_device *dev,
 		struct msm_gpu_submitqueue *queue, uint32_t nr_bos,
 		uint32_t nr_cmds)
 {
+	static atomic_t ident = ATOMIC_INIT(0);
 	struct msm_gem_submit *submit;
 	uint64_t sz;
 	int ret;
@@ -52,9 +53,13 @@ static struct msm_gem_submit *submit_create(struct drm_device *dev,
 	submit->gpu = gpu;
 	submit->cmd = (void *)&submit->bos[nr_bos];
 	submit->queue = queue;
+	submit->pid = get_pid(task_pid(current));
 	submit->ring = gpu->rb[queue->ring_nr];
 	submit->fault_dumped = false;
 
+	/* Get a unique identifier for the submission for logging purposes */
+	submit->ident = atomic_inc_return(&ident) - 1;
+
 	INIT_LIST_HEAD(&submit->node);
 
 	return submit;
@@ -718,7 +723,6 @@ static void msm_process_post_deps(struct msm_submit_post_dep *post_deps,
 int msm_ioctl_gem_submit(struct drm_device *dev, void *data,
 		struct drm_file *file)
 {
-	static atomic_t ident = ATOMIC_INIT(0);
 	struct msm_drm_private *priv = dev->dev_private;
 	struct drm_msm_gem_submit *args = data;
 	struct msm_file_private *ctx = file->driver_priv;
@@ -729,10 +733,9 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data,
 	struct msm_submit_post_dep *post_deps = NULL;
 	struct drm_syncobj **syncobjs_to_reset = NULL;
 	int out_fence_fd = -1;
-	struct pid *pid = get_pid(task_pid(current));
 	bool has_ww_ticket = false;
 	unsigned i;
-	int ret, submitid;
+	int ret;
 
 	if (!gpu)
 		return -ENXIO;
@@ -764,12 +767,7 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data,
 	if (!queue)
 		return -ENOENT;
 
-	/* Get a unique identifier for the submission for logging purposes */
-	submitid = atomic_inc_return(&ident) - 1;
-
 	ring = gpu->rb[queue->ring_nr];
-	trace_msm_gpu_submit(pid_nr(pid), ring->id, submitid,
-		args->nr_bos, args->nr_cmds);
 
 	if (args->flags & MSM_SUBMIT_FENCE_FD_OUT) {
 		out_fence_fd = get_unused_fd_flags(O_CLOEXEC);
@@ -783,13 +781,13 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data,
 	if (IS_ERR(submit))
 		return PTR_ERR(submit);
 
+	trace_msm_gpu_submit(pid_nr(submit->pid), ring->id, submit->ident,
+		args->nr_bos, args->nr_cmds);
+
 	ret = mutex_lock_interruptible(&queue->lock);
 	if (ret)
 		goto out_post_unlock;
 
-	submit->pid = pid;
-	submit->ident = submitid;
-
 	if (args->flags & MSM_SUBMIT_SUDO)
 		submit->in_rb = true;
 
Signature · 1 line
2.36.1
author
#3 lore raw

PATCH v4 3/15 drm/msm: Split out idr_lock

In reply to Rob Clark

From: Rob Clark <[email protected]>

Otherwise if we hit reclaim pinning objects in the submit path, we'll be blocking retire_worker trying to free a submit.

---
 drivers/gpu/drm/msm/msm_drv.c         |  4 ++--
 drivers/gpu/drm/msm/msm_gem_submit.c  | 10 ++++++++--
 drivers/gpu/drm/msm/msm_gpu.h         |  4 +++-
 drivers/gpu/drm/msm/msm_submitqueue.c |  1 +
 4 files changed, 14 insertions(+), 5 deletions(-)
drivers/gpu/drm/msm/msm_drv.c +2 -2
index 1ed4cd09dbf8..d7ca025457b6 100644
@@ -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;
drivers/gpu/drm/msm/msm_gem_submit.c +8 -2
index c7819781879c..16c662808522 100644
@@ -72,9 +72,9 @@ void __msm_gem_submit_destroy(struct kref *kref)
 	unsigned i;
 
 	if (submit->fence_id) {
-		mutex_lock(&submit->queue->lock);
+		mutex_lock(&submit->queue->idr_lock);
 		idr_remove(&submit->queue->fence_idr, submit->fence_id);
-		mutex_unlock(&submit->queue->lock);
+		mutex_unlock(&submit->queue->idr_lock);
 	}
 
 	dma_fence_put(submit->user_fence);
@@ -881,6 +881,8 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data,
 
 	submit->nr_cmds = i;
 
+	mutex_lock(&queue->idr_lock);
+
 	/*
 	 * If using userspace provided seqno fence, validate that the id
 	 * is available before arming sched job.  Since access to fence_idr
@@ -889,6 +891,7 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data,
 	 */
 	if ((args->flags & MSM_SUBMIT_FENCE_SN_IN) &&
 			idr_find(&queue->fence_idr, args->fence)) {
+		mutex_unlock(&queue->idr_lock);
 		ret = -EINVAL;
 		goto out;
 	}
@@ -921,6 +924,9 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data,
 						    submit->user_fence, 1,
 						    INT_MAX, GFP_KERNEL);
 	}
+
+	mutex_unlock(&queue->idr_lock);
+
 	if (submit->fence_id < 0) {
 		ret = submit->fence_id;
 		submit->fence_id = 0;
drivers/gpu/drm/msm/msm_gpu.h +3 -1
index 4d935fedd2ac..962d2070bcdf 100644
@@ -466,7 +466,8 @@ static inline int msm_gpu_convert_priority(struct msm_gpu *gpu, int prio,
  * @node:      node in the context's list of submitqueues
  * @fence_idr: maps fence-id to dma_fence for userspace visible fence
  *             seqno, protected by submitqueue lock
- * @lock:      submitqueue lock
+ * @idr_lock:  for serializing access to fence_idr
+ * @lock:      submitqueue lock for serializing submits on a queue
  * @ref:       reference count
  * @entity:    the submit job-queue
  */
@@ -479,6 +480,7 @@ struct msm_gpu_submitqueue {
 	struct msm_file_private *ctx;
 	struct list_head node;
 	struct idr fence_idr;
+	struct mutex idr_lock;
 	struct mutex lock;
 	struct kref ref;
 	struct drm_sched_entity *entity;
drivers/gpu/drm/msm/msm_submitqueue.c +1 -0
index f486a3cd4e55..c6929e205b51 100644
@@ -200,6 +200,7 @@ int msm_submitqueue_create(struct drm_device *drm, struct msm_file_private *ctx,
 		*id = queue->id;
 
 	idr_init(&queue->fence_idr);
+	mutex_init(&queue->idr_lock);
 	mutex_init(&queue->lock);
 
 	list_add_tail(&queue->node, &ctx->submitqueues);
Signature · 1 line
2.36.1
author
#4 lore raw

PATCH v4 4/15 drm/msm/gem: Check for active in shrinker path

In reply to Rob Clark

From: Rob Clark <[email protected]>

Currently in our shrinker path we shouldn't be encountering anything that is active, but this will change in subsequent patches. So check if there are unsignaled fences.

---
 drivers/gpu/drm/msm/msm_gem.c          | 10 ++++++++++
 drivers/gpu/drm/msm/msm_gem.h          |  1 +
 drivers/gpu/drm/msm/msm_gem_shrinker.c |  6 ++++++
 3 files changed, 17 insertions(+)
drivers/gpu/drm/msm/msm_gem.c +10 -0
index 8ddbd2e001d4..b55d252aef17 100644
@@ -870,6 +870,16 @@ static void update_inactive(struct msm_gem_object *msm_obj)
 	mutex_unlock(&priv->mm_lock);
 }
 
+bool msm_gem_active(struct drm_gem_object *obj)
+{
+	GEM_WARN_ON(!msm_gem_is_locked(obj));
+
+	if (to_msm_bo(obj)->pin_count)
+		return true;
+
+	return !dma_resv_test_signaled(obj->resv, dma_resv_usage_rw(true));
+}
+
 int msm_gem_cpu_prep(struct drm_gem_object *obj, uint32_t op, ktime_t *timeout)
 {
 	bool write = !!(op & MSM_PREP_WRITE);
drivers/gpu/drm/msm/msm_gem.h +1 -0
index 432032ad4aed..0ab0dc4f8c25 100644
@@ -173,6 +173,7 @@ void msm_gem_put_vaddr(struct drm_gem_object *obj);
 int msm_gem_madvise(struct drm_gem_object *obj, unsigned madv);
 void msm_gem_active_get(struct drm_gem_object *obj, struct msm_gpu *gpu);
 void msm_gem_active_put(struct drm_gem_object *obj);
+bool msm_gem_active(struct drm_gem_object *obj);
 int msm_gem_cpu_prep(struct drm_gem_object *obj, uint32_t op, ktime_t *timeout);
 int msm_gem_cpu_fini(struct drm_gem_object *obj);
 int msm_gem_new_handle(struct drm_device *dev, struct drm_file *file,
drivers/gpu/drm/msm/msm_gem_shrinker.c +6 -0
index 6e39d959b9f0..ea8ed74982c1 100644
@@ -43,6 +43,9 @@ purge(struct msm_gem_object *msm_obj)
 	if (!is_purgeable(msm_obj))
 		return false;
 
+	if (msm_gem_active(&msm_obj->base))
+		return false;
+
 	/*
 	 * This will move the obj out of still_in_list to
 	 * the purged list
@@ -58,6 +61,9 @@ evict(struct msm_gem_object *msm_obj)
 	if (is_unevictable(msm_obj))
 		return false;
 
+	if (msm_gem_active(&msm_obj->base))
+		return false;
+
 	msm_gem_evict(&msm_obj->base);
 
 	return true;
Signature · 1 line
2.36.1
author
#5 lore raw

PATCH v4 5/15 drm/msm/gem: Rename update_inactive

In reply to Rob Clark

From: Rob Clark <[email protected]>

Really what this is doing is updating various LRU lists.

---
 drivers/gpu/drm/msm/msm_gem.c | 25 +++++++++++++------------
 1 file changed, 13 insertions(+), 12 deletions(-)
drivers/gpu/drm/msm/msm_gem.c +13 -12
index b55d252aef17..97467364dc0a 100644
@@ -19,7 +19,7 @@
 #include "msm_gpu.h"
 #include "msm_mmu.h"
 
-static void update_inactive(struct msm_gem_object *msm_obj);
+static void update_lru(struct drm_gem_object *obj);
 
 static dma_addr_t physaddr(struct drm_gem_object *obj)
 {
@@ -132,7 +132,7 @@ static struct page **get_pages(struct drm_gem_object *obj)
 		if (msm_obj->flags & MSM_BO_WC)
 			sync_for_device(msm_obj);
 
-		update_inactive(msm_obj);
+		update_lru(obj);
 	}
 
 	return msm_obj->pages;
@@ -193,7 +193,7 @@ struct page **msm_gem_get_pages(struct drm_gem_object *obj)
 
 	if (!IS_ERR(p)) {
 		msm_obj->pin_count++;
-		update_inactive(msm_obj);
+		update_lru(obj);
 	}
 
 	msm_gem_unlock(obj);
@@ -207,7 +207,7 @@ void msm_gem_put_pages(struct drm_gem_object *obj)
 	msm_gem_lock(obj);
 	msm_obj->pin_count--;
 	GEM_WARN_ON(msm_obj->pin_count < 0);
-	update_inactive(msm_obj);
+	update_lru(obj);
 	msm_gem_unlock(obj);
 }
 
@@ -449,7 +449,7 @@ void msm_gem_unpin_locked(struct drm_gem_object *obj)
 	msm_obj->pin_count--;
 	GEM_WARN_ON(msm_obj->pin_count < 0);
 
-	update_inactive(msm_obj);
+	update_lru(obj);
 }
 
 struct msm_gem_vma *msm_gem_get_vma_locked(struct drm_gem_object *obj,
@@ -658,7 +658,7 @@ static void *get_vaddr(struct drm_gem_object *obj, unsigned madv)
 			goto fail;
 		}
 
-		update_inactive(msm_obj);
+		update_lru(obj);
 	}
 
 	return msm_obj->vaddr;
@@ -730,7 +730,7 @@ int msm_gem_madvise(struct drm_gem_object *obj, unsigned madv)
 	 * between inactive lists
 	 */
 	if (msm_obj->active_count == 0)
-		update_inactive(msm_obj);
+		update_lru(obj);
 
 	msm_gem_unlock(obj);
 
@@ -757,7 +757,7 @@ void msm_gem_purge(struct drm_gem_object *obj)
 	put_iova_vmas(obj);
 
 	msm_obj->madv = __MSM_MADV_PURGED;
-	update_inactive(msm_obj);
+	update_lru(obj);
 
 	drm_gem_free_mmap_offset(obj);
 
@@ -792,7 +792,7 @@ void msm_gem_evict(struct drm_gem_object *obj)
 
 	put_pages(obj);
 
-	update_inactive(msm_obj);
+	update_lru(obj);
 }
 
 void msm_gem_vunmap(struct drm_gem_object *obj)
@@ -835,13 +835,14 @@ void msm_gem_active_put(struct drm_gem_object *obj)
 	GEM_WARN_ON(!msm_gem_is_locked(obj));
 
 	if (--msm_obj->active_count == 0) {
-		update_inactive(msm_obj);
+		update_lru(obj);
 	}
 }
 
-static void update_inactive(struct msm_gem_object *msm_obj)
+static void update_lru(struct drm_gem_object *obj)
 {
-	struct msm_drm_private *priv = msm_obj->base.dev->dev_private;
+	struct msm_drm_private *priv = obj->dev->dev_private;
+	struct msm_gem_object *msm_obj = to_msm_bo(obj);
 
 	GEM_WARN_ON(!msm_gem_is_locked(&msm_obj->base));
 
Signature · 1 line
2.36.1
author
#6 lore raw

PATCH v4 6/15 drm/msm/gem: Rename to pin/unpin_pages

In reply to Rob Clark

From: Rob Clark <[email protected]>

Since that is what these fxns actually do.. they are getting *pinned* pages (as opposed to cases where we need pages, but don't need them pinned, like CPU mappings).

---
 drivers/gpu/drm/msm/msm_gem.c       | 18 +++++++++++++-----
 drivers/gpu/drm/msm/msm_gem.h       |  4 ++--
 drivers/gpu/drm/msm/msm_gem_prime.c |  4 ++--
 3 files changed, 17 insertions(+), 9 deletions(-)
drivers/gpu/drm/msm/msm_gem.c +13 -5
index 97467364dc0a..3da64c7f65a2 100644
@@ -177,30 +177,38 @@ static void put_pages(struct drm_gem_object *obj)
 	}
 }
 
-struct page **msm_gem_get_pages(struct drm_gem_object *obj)
+static struct page **msm_gem_pin_pages_locked(struct drm_gem_object *obj)
 {
 	struct msm_gem_object *msm_obj = to_msm_bo(obj);
 	struct page **p;
 
-	msm_gem_lock(obj);
+	GEM_WARN_ON(!msm_gem_is_locked(obj));
 
 	if (GEM_WARN_ON(msm_obj->madv != MSM_MADV_WILLNEED)) {
-		msm_gem_unlock(obj);
 		return ERR_PTR(-EBUSY);
 	}
 
 	p = get_pages(obj);
-
 	if (!IS_ERR(p)) {
 		msm_obj->pin_count++;
 		update_lru(obj);
 	}
 
+	return p;
+}
+
+struct page **msm_gem_pin_pages(struct drm_gem_object *obj)
+{
+	struct page **p;
+
+	msm_gem_lock(obj);
+	p = msm_gem_pin_pages_locked(obj);
 	msm_gem_unlock(obj);
+
 	return p;
 }
 
-void msm_gem_put_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);
 
drivers/gpu/drm/msm/msm_gem.h +2 -2
index 0ab0dc4f8c25..6fe521ccda45 100644
@@ -159,8 +159,8 @@ int msm_gem_get_and_pin_iova(struct drm_gem_object *obj,
 		struct msm_gem_address_space *aspace, uint64_t *iova);
 void msm_gem_unpin_iova(struct drm_gem_object *obj,
 		struct msm_gem_address_space *aspace);
-struct page **msm_gem_get_pages(struct drm_gem_object *obj);
-void msm_gem_put_pages(struct drm_gem_object *obj);
+struct page **msm_gem_pin_pages(struct drm_gem_object *obj);
+void msm_gem_unpin_pages(struct drm_gem_object *obj);
 int msm_gem_dumb_create(struct drm_file *file, struct drm_device *dev,
 		struct drm_mode_create_dumb *args);
 int msm_gem_dumb_map_offset(struct drm_file *file, struct drm_device *dev,
drivers/gpu/drm/msm/msm_gem_prime.c +2 -2
index dcc8a573bc76..c1d91863df05 100644
@@ -63,12 +63,12 @@ struct drm_gem_object *msm_gem_prime_import_sg_table(struct drm_device *dev,
 int msm_gem_prime_pin(struct drm_gem_object *obj)
 {
 	if (!obj->import_attach)
-		msm_gem_get_pages(obj);
+		msm_gem_pin_pages(obj);
 	return 0;
 }
 
 void msm_gem_prime_unpin(struct drm_gem_object *obj)
 {
 	if (!obj->import_attach)
-		msm_gem_put_pages(obj);
+		msm_gem_unpin_pages(obj);
 }
Signature · 1 line
2.36.1
author
#7 lore raw

PATCH v4 7/15 drm/msm/gem: Consolidate pin/unpin paths

In reply to Rob Clark

From: Rob Clark <[email protected]>

Avoid having multiple spots where we increment/decrement pin_count (and associated LRU updating)

---
 drivers/gpu/drm/msm/msm_gem.c | 13 +++++--------
 1 file changed, 5 insertions(+), 8 deletions(-)
drivers/gpu/drm/msm/msm_gem.c +5 -8
index 3da64c7f65a2..407b18a24dc4 100644
@@ -190,7 +190,7 @@ static struct page **msm_gem_pin_pages_locked(struct drm_gem_object *obj)
 
 	p = get_pages(obj);
 	if (!IS_ERR(p)) {
-		msm_obj->pin_count++;
+		to_msm_bo(obj)->pin_count++;
 		update_lru(obj);
 	}
 
@@ -213,9 +213,7 @@ 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_obj->pin_count--;
-	GEM_WARN_ON(msm_obj->pin_count < 0);
-	update_lru(obj);
+	msm_gem_unpin_locked(obj);
 	msm_gem_unlock(obj);
 }
 
@@ -436,14 +434,13 @@ int msm_gem_pin_vma_locked(struct drm_gem_object *obj, struct msm_gem_vma *vma)
 	if (GEM_WARN_ON(msm_obj->madv != MSM_MADV_WILLNEED))
 		return -EBUSY;
 
-	pages = get_pages(obj);
+	pages = msm_gem_pin_pages_locked(obj);
 	if (IS_ERR(pages))
 		return PTR_ERR(pages);
 
 	ret = msm_gem_map_vma(vma->aspace, vma, prot, msm_obj->sgt, obj->size);
-
-	if (!ret)
-		msm_obj->pin_count++;
+	if (ret)
+		msm_gem_unpin_locked(obj);
 
 	return ret;
 }
Signature · 1 line
2.36.1
author
#8 lore raw

PATCH v4 8/15 drm/msm/gem: Remove active refcnt

In reply to Rob Clark

From: Rob Clark <[email protected]>

At this point the pinned refcnt is sufficient, and the shrinker is already prepared to encounter objects which are still active according to fences attached to the resv.

---
 drivers/gpu/drm/msm/msm_gem.c        | 45 ++--------------------------
 drivers/gpu/drm/msm/msm_gem.h        | 14 ++-------
 drivers/gpu/drm/msm/msm_gem_submit.c | 22 ++------------
 3 files changed, 8 insertions(+), 73 deletions(-)
drivers/gpu/drm/msm/msm_gem.c +3 -42
index 407b18a24dc4..209438744bab 100644
@@ -734,8 +734,7 @@ int msm_gem_madvise(struct drm_gem_object *obj, unsigned madv)
 	/* If the obj is inactive, we might need to move it
 	 * between inactive lists
 	 */
-	if (msm_obj->active_count == 0)
-		update_lru(obj);
+	update_lru(obj);
 
 	msm_gem_unlock(obj);
 
@@ -788,7 +787,6 @@ 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);
-	GEM_WARN_ON(msm_obj->active_count);
 
 	/* Get rid of any iommu mapping(s): */
 	put_iova_spaces(obj, false);
@@ -813,37 +811,6 @@ void msm_gem_vunmap(struct drm_gem_object *obj)
 	msm_obj->vaddr = NULL;
 }
 
-void msm_gem_active_get(struct drm_gem_object *obj, struct msm_gpu *gpu)
-{
-	struct msm_gem_object *msm_obj = to_msm_bo(obj);
-	struct msm_drm_private *priv = obj->dev->dev_private;
-
-	might_sleep();
-	GEM_WARN_ON(!msm_gem_is_locked(obj));
-	GEM_WARN_ON(msm_obj->madv != MSM_MADV_WILLNEED);
-	GEM_WARN_ON(msm_obj->dontneed);
-
-	if (msm_obj->active_count++ == 0) {
-		mutex_lock(&priv->mm_lock);
-		if (msm_obj->evictable)
-			mark_unevictable(msm_obj);
-		list_move_tail(&msm_obj->mm_list, &gpu->active_list);
-		mutex_unlock(&priv->mm_lock);
-	}
-}
-
-void msm_gem_active_put(struct drm_gem_object *obj)
-{
-	struct msm_gem_object *msm_obj = to_msm_bo(obj);
-
-	might_sleep();
-	GEM_WARN_ON(!msm_gem_is_locked(obj));
-
-	if (--msm_obj->active_count == 0) {
-		update_lru(obj);
-	}
-}
-
 static void update_lru(struct drm_gem_object *obj)
 {
 	struct msm_drm_private *priv = obj->dev->dev_private;
@@ -851,9 +818,6 @@ static void update_lru(struct drm_gem_object *obj)
 
 	GEM_WARN_ON(!msm_gem_is_locked(&msm_obj->base));
 
-	if (msm_obj->active_count != 0)
-		return;
-
 	mutex_lock(&priv->mm_lock);
 
 	if (msm_obj->dontneed)
@@ -926,7 +890,7 @@ void msm_gem_describe(struct drm_gem_object *obj, struct seq_file *m,
 	stats->all.count++;
 	stats->all.size += obj->size;
 
-	if (is_active(msm_obj)) {
+	if (msm_gem_active(obj)) {
 		stats->active.count++;
 		stats->active.size += obj->size;
 	}
@@ -954,7 +918,7 @@ void msm_gem_describe(struct drm_gem_object *obj, struct seq_file *m,
 	}
 
 	seq_printf(m, "%08x: %c %2d (%2d) %08llx %p",
-			msm_obj->flags, is_active(msm_obj) ? 'A' : 'I',
+			msm_obj->flags, msm_gem_active(obj) ? 'A' : 'I',
 			obj->name, kref_read(&obj->refcount),
 			off, msm_obj->vaddr);
 
@@ -1037,9 +1001,6 @@ static void msm_gem_free_object(struct drm_gem_object *obj)
 	list_del(&msm_obj->mm_list);
 	mutex_unlock(&priv->mm_lock);
 
-	/* object should not be on active list: */
-	GEM_WARN_ON(is_active(msm_obj));
-
 	put_iova_spaces(obj, true);
 
 	if (obj->import_attach) {
drivers/gpu/drm/msm/msm_gem.h +2 -12
index 6fe521ccda45..420ba49bf21a 100644
@@ -138,7 +138,6 @@ struct msm_gem_object {
 
 	char name[32]; /* Identifier to print for the debugfs files */
 
-	int active_count;
 	int pin_count;
 };
 #define to_msm_bo(x) container_of(x, struct msm_gem_object, base)
@@ -171,8 +170,6 @@ void *msm_gem_get_vaddr_active(struct drm_gem_object *obj);
 void msm_gem_put_vaddr_locked(struct drm_gem_object *obj);
 void msm_gem_put_vaddr(struct drm_gem_object *obj);
 int msm_gem_madvise(struct drm_gem_object *obj, unsigned madv);
-void msm_gem_active_get(struct drm_gem_object *obj, struct msm_gpu *gpu);
-void msm_gem_active_put(struct drm_gem_object *obj);
 bool msm_gem_active(struct drm_gem_object *obj);
 int msm_gem_cpu_prep(struct drm_gem_object *obj, uint32_t op, ktime_t *timeout);
 int msm_gem_cpu_fini(struct drm_gem_object *obj);
@@ -245,12 +242,6 @@ msm_gem_is_locked(struct drm_gem_object *obj)
 	return dma_resv_is_locked(obj->resv) || (kref_read(&obj->refcount) == 0);
 }
 
-static inline bool is_active(struct msm_gem_object *msm_obj)
-{
-	GEM_WARN_ON(!msm_gem_is_locked(&msm_obj->base));
-	return msm_obj->active_count;
-}
-
 /* imported/exported objects are not purgeable: */
 static inline bool is_unpurgeable(struct msm_gem_object *msm_obj)
 {
@@ -391,9 +382,8 @@ struct msm_gem_submit {
 /* make sure these don't conflict w/ MSM_SUBMIT_BO_x */
 #define BO_VALID	0x8000	/* is current addr in cmdstream correct/valid? */
 #define BO_LOCKED	0x4000	/* obj lock is held */
-#define BO_ACTIVE	0x2000	/* active refcnt is held */
-#define BO_OBJ_PINNED	0x1000	/* obj (pages) is pinned and on active list */
-#define BO_VMA_PINNED	0x0800	/* vma (virtual address) is pinned */
+#define BO_OBJ_PINNED	0x2000	/* obj (pages) is pinned and on active list */
+#define BO_VMA_PINNED	0x1000	/* vma (virtual address) is pinned */
 		uint32_t flags;
 		union {
 			struct msm_gem_object *obj;
drivers/gpu/drm/msm/msm_gem_submit.c +3 -19
index 16c662808522..adf358fb8e9d 100644
@@ -243,17 +243,13 @@ static void submit_cleanup_bo(struct msm_gem_submit *submit, int i,
 	if (flags & BO_OBJ_PINNED)
 		msm_gem_unpin_locked(obj);
 
-	if (flags & BO_ACTIVE)
-		msm_gem_active_put(obj);
-
 	if (flags & BO_LOCKED)
 		dma_resv_unlock(obj->resv);
 }
 
 static void submit_unlock_unpin_bo(struct msm_gem_submit *submit, int i)
 {
-	unsigned cleanup_flags = BO_VMA_PINNED | BO_OBJ_PINNED |
-				 BO_ACTIVE | BO_LOCKED;
+	unsigned cleanup_flags = BO_VMA_PINNED | BO_OBJ_PINNED | BO_LOCKED;
 	submit_cleanup_bo(submit, i, cleanup_flags);
 
 	if (!(submit->bos[i].flags & BO_VALID))
@@ -358,18 +354,6 @@ static int submit_pin_objects(struct msm_gem_submit *submit)
 
 	submit->valid = true;
 
-	/*
-	 * Increment active_count first, so if under memory pressure, we
-	 * don't inadvertently evict a bo needed by the submit in order
-	 * to pin an earlier bo in the same submit.
-	 */
-	for (i = 0; i < submit->nr_bos; i++) {
-		struct drm_gem_object *obj = &submit->bos[i].obj->base;
-
-		msm_gem_active_get(obj, submit->gpu);
-		submit->bos[i].flags |= BO_ACTIVE;
-	}
-
 	for (i = 0; i < submit->nr_bos; i++) {
 		struct drm_gem_object *obj = &submit->bos[i].obj->base;
 		struct msm_gem_vma *vma;
@@ -521,7 +505,7 @@ static void submit_cleanup(struct msm_gem_submit *submit, bool error)
 	unsigned i;
 
 	if (error)
-		cleanup_flags |= BO_VMA_PINNED | BO_OBJ_PINNED | BO_ACTIVE;
+		cleanup_flags |= BO_VMA_PINNED | BO_OBJ_PINNED;
 
 	for (i = 0; i < submit->nr_bos; i++) {
 		struct msm_gem_object *msm_obj = submit->bos[i].obj;
@@ -540,7 +524,7 @@ void msm_submit_retire(struct msm_gem_submit *submit)
 
 		msm_gem_lock(obj);
 		/* Note, VMA already fence-unpinned before submit: */
-		submit_cleanup_bo(submit, i, BO_OBJ_PINNED | BO_ACTIVE);
+		submit_cleanup_bo(submit, i, BO_OBJ_PINNED);
 		msm_gem_unlock(obj);
 		drm_gem_object_put(obj);
 	}
Signature · 1 line
2.36.1
author
#9 lore raw

PATCH v4 9/15 drm/gem: Add LRU/shrinker helper

In reply to Rob Clark

From: Rob Clark <[email protected]>

Add a simple LRU helper to assist with driver's shrinker implementation. It handles tracking the number of backing pages associated with a given LRU, and provides a helper to implement shrinker_scan.

A driver can use multiple LRU instances to track objects in various states, for example a dontneed LRU for purgeable objects, a willneed LRU for evictable objects, and an unpinned LRU for objects without backing pages.

All LRUs that the object can be moved between must share a single lock.

v2: lockdep_assert_held() instead of WARN_ON(!mutex_is_locked())
v3: make drm_gem_lru_move_tail_locked() static until there is a user

Cc Daniel Vetter <[email protected]>
Cc Thomas Zimmermann <[email protected]>
Cc Dmitry Osipenko <[email protected]>
---
 drivers/gpu/drm/drm_gem.c | 170 ++++++++++++++++++++++++++++++++++++++
 include/drm/drm_gem.h     |  55 ++++++++++++
 2 files changed, 225 insertions(+)
drivers/gpu/drm/drm_gem.c +170 -0
index eb0c2d041f13..556714c10472 100644
@@ -165,6 +165,7 @@ void drm_gem_private_object_init(struct drm_device *dev,
 		obj->resv = &obj->_resv;
 
 	drm_vma_node_reset(&obj->vma_node);
+	INIT_LIST_HEAD(&obj->lru_node);
 }
 EXPORT_SYMBOL(drm_gem_private_object_init);
 
@@ -951,6 +952,7 @@ drm_gem_object_release(struct drm_gem_object *obj)
 
 	dma_resv_fini(&obj->_resv);
 	drm_gem_free_mmap_offset(obj);
+	drm_gem_lru_remove(obj);
 }
 EXPORT_SYMBOL(drm_gem_object_release);
 
@@ -1274,3 +1276,171 @@ drm_gem_unlock_reservations(struct drm_gem_object **objs, int count,
 	ww_acquire_fini(acquire_ctx);
 }
 EXPORT_SYMBOL(drm_gem_unlock_reservations);
+
+/**
+ * drm_gem_lru_init - initialize a LRU
+ *
+ * @lru: The LRU to initialize
+ * @lock: The lock protecting the LRU
+ */
+void
+drm_gem_lru_init(struct drm_gem_lru *lru, struct mutex *lock)
+{
+	lru->lock = lock;
+	lru->count = 0;
+	INIT_LIST_HEAD(&lru->list);
+}
+EXPORT_SYMBOL(drm_gem_lru_init);
+
+static void
+drm_gem_lru_remove_locked(struct drm_gem_object *obj)
+{
+	obj->lru->count -= obj->size >> PAGE_SHIFT;
+	WARN_ON(obj->lru->count < 0);
+	list_del(&obj->lru_node);
+	obj->lru = NULL;
+}
+
+/**
+ * drm_gem_lru_remove - remove object from whatever LRU it is in
+ *
+ * If the object is currently in any LRU, remove it.
+ *
+ * @obj: The GEM object to remove from current LRU
+ */
+void
+drm_gem_lru_remove(struct drm_gem_object *obj)
+{
+	struct drm_gem_lru *lru = obj->lru;
+
+	if (!lru)
+		return;
+
+	mutex_lock(lru->lock);
+	drm_gem_lru_remove_locked(obj);
+	mutex_unlock(lru->lock);
+}
+EXPORT_SYMBOL(drm_gem_lru_remove);
+
+static void
+drm_gem_lru_move_tail_locked(struct drm_gem_lru *lru, struct drm_gem_object *obj)
+{
+	lockdep_assert_held_once(lru->lock);
+
+	if (obj->lru)
+		drm_gem_lru_remove_locked(obj);
+
+	lru->count += obj->size >> PAGE_SHIFT;
+	list_add_tail(&obj->lru_node, &lru->list);
+	obj->lru = lru;
+}
+
+/**
+ * drm_gem_lru_move_tail - move the object to the tail of the LRU
+ *
+ * If the object is already in this LRU it will be moved to the
+ * tail.  Otherwise it will be removed from whichever other LRU
+ * it is in (if any) and moved into this LRU.
+ *
+ * @lru: The LRU to move the object into.
+ * @obj: The GEM object to move into this LRU
+ */
+void
+drm_gem_lru_move_tail(struct drm_gem_lru *lru, struct drm_gem_object *obj)
+{
+	mutex_lock(lru->lock);
+	drm_gem_lru_move_tail_locked(lru, obj);
+	mutex_unlock(lru->lock);
+}
+EXPORT_SYMBOL(drm_gem_lru_move_tail);
+
+/**
+ * drm_gem_lru_scan - helper to implement shrinker.scan_objects
+ *
+ * If the shrink callback succeeds, it is expected that the driver
+ * move the object out of this LRU.
+ *
+ * If the LRU possibly contain active buffers, it is the responsibility
+ * of the shrink callback to check for this (ie. dma_resv_test_signaled())
+ * or if necessary block until the buffer becomes idle.
+ *
+ * @lru: The LRU to scan
+ * @nr_to_scan: The number of pages to try to reclaim
+ * @shrink: Callback to try to shrink/reclaim the object.
+ */
+unsigned long
+drm_gem_lru_scan(struct drm_gem_lru *lru, unsigned nr_to_scan,
+		 bool (*shrink)(struct drm_gem_object *obj))
+{
+	struct drm_gem_lru still_in_lru;
+	struct drm_gem_object *obj;
+	unsigned freed = 0;
+
+	drm_gem_lru_init(&still_in_lru, lru->lock);
+
+	mutex_lock(lru->lock);
+
+	while (freed < nr_to_scan) {
+		obj = list_first_entry_or_null(&lru->list, typeof(*obj), lru_node);
+
+		if (!obj)
+			break;
+
+		drm_gem_lru_move_tail_locked(&still_in_lru, obj);
+
+		/*
+		 * If it's in the process of being freed, gem_object->free()
+		 * may be blocked on lock waiting to remove it.  So just
+		 * skip it.
+		 */
+		if (!kref_get_unless_zero(&obj->refcount))
+			continue;
+
+		/*
+		 * Now that we own a reference, we can drop the lock for the
+		 * rest of the loop body, to reduce contention with other
+		 * code paths that need the LRU lock
+		 */
+		mutex_unlock(lru->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 (!dma_resv_trylock(obj->resv))
+			goto tail;
+
+		if (shrink(obj)) {
+			freed += obj->size >> PAGE_SHIFT;
+
+			/*
+			 * If we succeeded in releasing the object's backing
+			 * pages, we expect the driver to have moved the object
+			 * out of this LRU
+			 */
+			WARN_ON(obj->lru == &still_in_lru);
+			WARN_ON(obj->lru == lru);
+		}
+
+		dma_resv_unlock(obj->resv);
+
+tail:
+		drm_gem_object_put(obj);
+		mutex_lock(lru->lock);
+	}
+
+	/*
+	 * Move objects we've skipped over out of the temporary still_in_lru
+	 * back into this LRU
+	 */
+	list_for_each_entry (obj, &still_in_lru.list, lru_node)
+		obj->lru = lru;
+	list_splice_tail(&still_in_lru.list, &lru->list);
+	lru->count += still_in_lru.count;
+
+	mutex_unlock(lru->lock);
+
+	return freed;
+}
+EXPORT_SYMBOL(drm_gem_lru_scan);
include/drm/drm_gem.h +55 -0

71 lines not on this page · raw message

Signature · 1 line
2.36.1
author
#10 lore raw

PATCH v4 10/15 drm/msm/gem: Convert to using drm_gem_lru

In reply to Rob Clark

From: Rob Clark <[email protected]>

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.

---
 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(-)
drivers/gpu/drm/msm/msm_drv.c +9 -5
index d7ca025457b6..1ca4a92ba96e 100644
@@ -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);
drivers/gpu/drm/msm/msm_drv.h +51 -19

79 lines not on this page · raw message

drivers/gpu/drm/msm/msm_gem.c +15 -43

120 lines not on this page · raw message

drivers/gpu/drm/msm/msm_gem.h +0 -93

122 lines not on this page · raw message

drivers/gpu/drm/msm/msm_gem_shrinker.c +29 -88

179 lines not on this page · raw message

drivers/gpu/drm/msm/msm_gpu.c +0 -3
index c2bfcf3f1f40..fba85f894314 100644
@@ -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;
drivers/gpu/drm/msm/msm_gpu.h +0 -6
index 962d2070bcdf..a0885ff99e5b 100644
@@ -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:
 	 *
Signature · 1 line
2.36.1
author
#11 lore raw

PATCH v4 11/15 drm/msm/gem: Unpin buffers earlier

In reply to Rob Clark

From: Rob Clark <[email protected]>

We've already attached the fences, so obj->resv (which shrinker checks) tells us whether they are still active. So we can unpin sooner, before we drop the queue lock.

This also avoids the need to grab the obj lock in the retire path, avoiding potential for lock contention between submit and retire.

---
 drivers/gpu/drm/msm/msm_gem_submit.c | 8 ++------
 1 file changed, 2 insertions(+), 6 deletions(-)
drivers/gpu/drm/msm/msm_gem_submit.c +2 -6
index adf358fb8e9d..5599d93ec0d2 100644
@@ -501,11 +501,11 @@ static int submit_reloc(struct msm_gem_submit *submit, struct msm_gem_object *ob
  */
 static void submit_cleanup(struct msm_gem_submit *submit, bool error)
 {
-	unsigned cleanup_flags = BO_LOCKED;
+	unsigned cleanup_flags = BO_LOCKED | BO_OBJ_PINNED;
 	unsigned i;
 
 	if (error)
-		cleanup_flags |= BO_VMA_PINNED | BO_OBJ_PINNED;
+		cleanup_flags |= BO_VMA_PINNED;
 
 	for (i = 0; i < submit->nr_bos; i++) {
 		struct msm_gem_object *msm_obj = submit->bos[i].obj;
@@ -522,10 +522,6 @@ void msm_submit_retire(struct msm_gem_submit *submit)
 	for (i = 0; i < submit->nr_bos; i++) {
 		struct drm_gem_object *obj = &submit->bos[i].obj->base;
 
-		msm_gem_lock(obj);
-		/* Note, VMA already fence-unpinned before submit: */
-		submit_cleanup_bo(submit, i, BO_OBJ_PINNED);
-		msm_gem_unlock(obj);
 		drm_gem_object_put(obj);
 	}
 }
Signature · 1 line
2.36.1
author
#12 lore raw

PATCH v4 12/15 drm/msm/gem: Consolidate shrinker trace

In reply to Rob Clark

From: Rob Clark <[email protected]>

Combine separate trace events for purge vs evict into one. When we add support for purging/evicting active buffers we'll just add more info into this one trace event, rather than adding a bunch more events.

---
 drivers/gpu/drm/msm/msm_gem_shrinker.c | 19 ++++++---------
 drivers/gpu/drm/msm/msm_gpu_trace.h    | 32 +++++++++++---------------
 2 files changed, 20 insertions(+), 31 deletions(-)
drivers/gpu/drm/msm/msm_gem_shrinker.c +7 -12
index 530b1102b46d..5cc05d669a08 100644
@@ -71,25 +71,20 @@ 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;
+	unsigned long freed, purged, evicted = 0;
 
-	freed = drm_gem_lru_scan(&priv->lru.dontneed, nr, purge);
-	nr -= freed;
-
-	if (freed > 0)
-		trace_msm_gem_purge(freed << PAGE_SHIFT);
+	purged = drm_gem_lru_scan(&priv->lru.dontneed, nr, purge);
+	nr -= purged;
 
 	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);
+	freed = purged + evicted;
 
-		freed += evicted;
-	}
+	if (freed)
+		trace_msm_gem_shrink(sc->nr_to_scan, purged, evicted);
 
 	return (freed > 0) ? freed : SHRINK_STOP;
 }
drivers/gpu/drm/msm/msm_gpu_trace.h +13 -19
index ca0b08d7875b..8867fa0a0306 100644
@@ -115,29 +115,23 @@ TRACE_EVENT(msm_gmu_freq_change,
 );
 
 
-TRACE_EVENT(msm_gem_purge,
-		TP_PROTO(u32 bytes),
-		TP_ARGS(bytes),
+TRACE_EVENT(msm_gem_shrink,
+		TP_PROTO(u32 nr_to_scan, u32 purged, u32 evicted),
+		TP_ARGS(nr_to_scan, purged, evicted),
 		TP_STRUCT__entry(
-			__field(u32, bytes)
+			__field(u32, nr_to_scan)
+			__field(u32, purged)
+			__field(u32, evicted)
 			),
 		TP_fast_assign(
-			__entry->bytes = bytes;
+			__entry->nr_to_scan = nr_to_scan;
+			__entry->purged = purged;
+			__entry->evicted = evicted;
 			),
-		TP_printk("Purging %u bytes", __entry->bytes)
-);
-
-
-TRACE_EVENT(msm_gem_evict,
-		TP_PROTO(u32 bytes),
-		TP_ARGS(bytes),
-		TP_STRUCT__entry(
-			__field(u32, bytes)
-			),
-		TP_fast_assign(
-			__entry->bytes = bytes;
-			),
-		TP_printk("Evicting %u bytes", __entry->bytes)
+		TP_printk("nr_to_scan=%u pages, purged=%u pages, evicted=%u pages",
+			  __entry->nr_to_scan,
+			  __entry->purged,
+			  __entry->evicted)
 );
 
 
Signature · 1 line
2.36.1
author
#13 lore raw

PATCH v4 13/15 drm/msm/gem: Evict active GEM objects when necessary

In reply to Rob Clark

From: Rob Clark <[email protected]>

If we are under enough memory pressure, we should stall waiting for active buffers to become idle in order to evict.

v2: Check for __GFP_ATOMIC before blocking

---
 drivers/gpu/drm/msm/msm_gem_shrinker.c | 70 +++++++++++++++++++++-----
 drivers/gpu/drm/msm/msm_gpu_trace.h    | 16 +++---
 2 files changed, 68 insertions(+), 18 deletions(-)
drivers/gpu/drm/msm/msm_gem_shrinker.c +58 -12
index 5cc05d669a08..f31054d25314 100644
@@ -24,6 +24,13 @@ static bool can_swap(void)
 	return enable_eviction && get_nr_swap_pages() > 0;
 }
 
+static bool can_block(struct shrink_control *sc)
+{
+	if (sc->gfp_mask & __GFP_ATOMIC)
+		return false;
+	return current_is_kswapd() || (sc->gfp_mask & __GFP_RECLAIM);
+}
+
 static unsigned long
 msm_gem_shrinker_count(struct shrinker *shrinker, struct shrink_control *sc)
 {
@@ -65,26 +72,65 @@ evict(struct drm_gem_object *obj)
 	return true;
 }
 
+static bool
+wait_for_idle(struct drm_gem_object *obj)
+{
+	enum dma_resv_usage usage = dma_resv_usage_rw(true);
+	return dma_resv_wait_timeout(obj->resv, usage, false, 1000) > 0;
+}
+
+static bool
+active_purge(struct drm_gem_object *obj)
+{
+	if (!wait_for_idle(obj))
+		return false;
+
+	return purge(obj);
+}
+
+static bool
+active_evict(struct drm_gem_object *obj)
+{
+	if (!wait_for_idle(obj))
+		return false;
+
+	return evict(obj);
+}
+
 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);
+	struct {
+		struct drm_gem_lru *lru;
+		bool (*shrink)(struct drm_gem_object *obj);
+		bool cond;
+		unsigned long freed;
+	} stages[] = {
+		/* Stages of progressively more aggressive/expensive reclaim: */
+		{ &priv->lru.dontneed, purge,        true },
+		{ &priv->lru.willneed, evict,        can_swap() },
+		{ &priv->lru.dontneed, active_purge, can_block(sc) },
+		{ &priv->lru.willneed, active_evict, can_swap() && can_block(sc) },
+	};
 	long nr = sc->nr_to_scan;
-	unsigned long freed, purged, evicted = 0;
-
-	purged = drm_gem_lru_scan(&priv->lru.dontneed, nr, purge);
-	nr -= purged;
-
-	if (can_swap() && nr > 0) {
-		evicted = drm_gem_lru_scan(&priv->lru.willneed, nr, evict);
-		nr -= evicted;
+	unsigned long freed = 0;
+
+	for (unsigned i = 0; (nr > 0) && (i < ARRAY_SIZE(stages)); i++) {
+		if (!stages[i].cond)
+			continue;
+		stages[i].freed =
+			drm_gem_lru_scan(stages[i].lru, nr, stages[i].shrink);
+		nr -= stages[i].freed;
+		freed += stages[i].freed;
 	}
 
-	freed = purged + evicted;
-
-	if (freed)
-		trace_msm_gem_shrink(sc->nr_to_scan, purged, evicted);
+	if (freed) {
+		trace_msm_gem_shrink(sc->nr_to_scan, stages[0].freed,
+				     stages[1].freed, stages[2].freed,
+				     stages[3].freed);
+	}
 
 	return (freed > 0) ? freed : SHRINK_STOP;
 }
drivers/gpu/drm/msm/msm_gpu_trace.h +10 -6
index 8867fa0a0306..ac40d857bc45 100644
@@ -116,22 +116,26 @@ TRACE_EVENT(msm_gmu_freq_change,
 
 
 TRACE_EVENT(msm_gem_shrink,
-		TP_PROTO(u32 nr_to_scan, u32 purged, u32 evicted),
-		TP_ARGS(nr_to_scan, purged, evicted),
+		TP_PROTO(u32 nr_to_scan, u32 purged, u32 evicted,
+			 u32 active_purged, u32 active_evicted),
+		TP_ARGS(nr_to_scan, purged, evicted, active_purged, active_evicted),
 		TP_STRUCT__entry(
 			__field(u32, nr_to_scan)
 			__field(u32, purged)
 			__field(u32, evicted)
+			__field(u32, active_purged)
+			__field(u32, active_evicted)
 			),
 		TP_fast_assign(
 			__entry->nr_to_scan = nr_to_scan;
 			__entry->purged = purged;
 			__entry->evicted = evicted;
+			__entry->active_purged = active_purged;
+			__entry->active_evicted = active_evicted;
 			),
-		TP_printk("nr_to_scan=%u pages, purged=%u pages, evicted=%u pages",
-			  __entry->nr_to_scan,
-			  __entry->purged,
-			  __entry->evicted)
+		TP_printk("nr_to_scan=%u pg, purged=%u pg, evicted=%u pg, active_purged=%u pg, active_evicted=%u pg",
+			  __entry->nr_to_scan, __entry->purged, __entry->evicted,
+			  __entry->active_purged, __entry->active_evicted)
 );
 
 
Signature · 1 line
2.36.1
author
#14 lore raw

PATCH v4 14/15 drm/msm/gem: Add msm_gem_assert_locked()

In reply to Rob Clark

From: Rob Clark <[email protected]>

All use of msm_gem_is_locked() is just for WARN_ON()s, so extract out into an msm_gem_assert_locked() patch.

---
 drivers/gpu/drm/msm/msm_gem.c | 36 +++++++++++++++++------------------
 drivers/gpu/drm/msm/msm_gem.h |  8 +++++++-
 2 files changed, 25 insertions(+), 19 deletions(-)
drivers/gpu/drm/msm/msm_gem.c +18 -18
index d4e8af46f4ef..1dee0d18abbb 100644
@@ -97,7 +97,7 @@ static struct page **get_pages(struct drm_gem_object *obj)
 {
 	struct msm_gem_object *msm_obj = to_msm_bo(obj);
 
-	GEM_WARN_ON(!msm_gem_is_locked(obj));
+	msm_gem_assert_locked(obj);
 
 	if (!msm_obj->pages) {
 		struct drm_device *dev = obj->dev;
@@ -183,7 +183,7 @@ static struct page **msm_gem_pin_pages_locked(struct drm_gem_object *obj)
 	struct msm_gem_object *msm_obj = to_msm_bo(obj);
 	struct page **p;
 
-	GEM_WARN_ON(!msm_gem_is_locked(obj));
+	msm_gem_assert_locked(obj);
 
 	if (GEM_WARN_ON(msm_obj->madv != MSM_MADV_WILLNEED)) {
 		return ERR_PTR(-EBUSY);
@@ -278,7 +278,7 @@ static uint64_t mmap_offset(struct drm_gem_object *obj)
 	struct drm_device *dev = obj->dev;
 	int ret;
 
-	GEM_WARN_ON(!msm_gem_is_locked(obj));
+	msm_gem_assert_locked(obj);
 
 	/* Make it mmapable */
 	ret = drm_gem_create_mmap_offset(obj);
@@ -307,7 +307,7 @@ static struct msm_gem_vma *add_vma(struct drm_gem_object *obj,
 	struct msm_gem_object *msm_obj = to_msm_bo(obj);
 	struct msm_gem_vma *vma;
 
-	GEM_WARN_ON(!msm_gem_is_locked(obj));
+	msm_gem_assert_locked(obj);
 
 	vma = kzalloc(sizeof(*vma), GFP_KERNEL);
 	if (!vma)
@@ -326,7 +326,7 @@ static struct msm_gem_vma *lookup_vma(struct drm_gem_object *obj,
 	struct msm_gem_object *msm_obj = to_msm_bo(obj);
 	struct msm_gem_vma *vma;
 
-	GEM_WARN_ON(!msm_gem_is_locked(obj));
+	msm_gem_assert_locked(obj);
 
 	list_for_each_entry(vma, &msm_obj->vmas, list) {
 		if (vma->aspace == aspace)
@@ -357,7 +357,7 @@ put_iova_spaces(struct drm_gem_object *obj, bool close)
 	struct msm_gem_object *msm_obj = to_msm_bo(obj);
 	struct msm_gem_vma *vma;
 
-	GEM_WARN_ON(!msm_gem_is_locked(obj));
+	msm_gem_assert_locked(obj);
 
 	list_for_each_entry(vma, &msm_obj->vmas, list) {
 		if (vma->aspace) {
@@ -375,7 +375,7 @@ put_iova_vmas(struct drm_gem_object *obj)
 	struct msm_gem_object *msm_obj = to_msm_bo(obj);
 	struct msm_gem_vma *vma, *tmp;
 
-	GEM_WARN_ON(!msm_gem_is_locked(obj));
+	msm_gem_assert_locked(obj);
 
 	list_for_each_entry_safe(vma, tmp, &msm_obj->vmas, list) {
 		del_vma(vma);
@@ -388,7 +388,7 @@ static struct msm_gem_vma *get_vma_locked(struct drm_gem_object *obj,
 {
 	struct msm_gem_vma *vma;
 
-	GEM_WARN_ON(!msm_gem_is_locked(obj));
+	msm_gem_assert_locked(obj);
 
 	vma = lookup_vma(obj, aspace);
 
@@ -428,7 +428,7 @@ int msm_gem_pin_vma_locked(struct drm_gem_object *obj, struct msm_gem_vma *vma)
 	if (msm_obj->flags & MSM_BO_CACHED_COHERENT)
 		prot |= IOMMU_CACHE;
 
-	GEM_WARN_ON(!msm_gem_is_locked(obj));
+	msm_gem_assert_locked(obj);
 
 	if (GEM_WARN_ON(msm_obj->madv != MSM_MADV_WILLNEED))
 		return -EBUSY;
@@ -448,7 +448,7 @@ void msm_gem_unpin_locked(struct drm_gem_object *obj)
 {
 	struct msm_gem_object *msm_obj = to_msm_bo(obj);
 
-	GEM_WARN_ON(!msm_gem_is_locked(obj));
+	msm_gem_assert_locked(obj);
 
 	msm_obj->pin_count--;
 	GEM_WARN_ON(msm_obj->pin_count < 0);
@@ -469,7 +469,7 @@ static int get_and_pin_iova_range_locked(struct drm_gem_object *obj,
 	struct msm_gem_vma *vma;
 	int ret;
 
-	GEM_WARN_ON(!msm_gem_is_locked(obj));
+	msm_gem_assert_locked(obj);
 
 	vma = get_vma_locked(obj, aspace, range_start, range_end);
 	if (IS_ERR(vma))
@@ -630,7 +630,7 @@ static void *get_vaddr(struct drm_gem_object *obj, unsigned madv)
 	struct msm_gem_object *msm_obj = to_msm_bo(obj);
 	int ret = 0;
 
-	GEM_WARN_ON(!msm_gem_is_locked(obj));
+	msm_gem_assert_locked(obj);
 
 	if (obj->import_attach)
 		return ERR_PTR(-ENODEV);
@@ -703,7 +703,7 @@ void msm_gem_put_vaddr_locked(struct drm_gem_object *obj)
 {
 	struct msm_gem_object *msm_obj = to_msm_bo(obj);
 
-	GEM_WARN_ON(!msm_gem_is_locked(obj));
+	msm_gem_assert_locked(obj);
 	GEM_WARN_ON(msm_obj->vmap_count < 1);
 
 	msm_obj->vmap_count--;
@@ -745,7 +745,7 @@ void msm_gem_purge(struct drm_gem_object *obj)
 	struct drm_device *dev = obj->dev;
 	struct msm_gem_object *msm_obj = to_msm_bo(obj);
 
-	GEM_WARN_ON(!msm_gem_is_locked(obj));
+	msm_gem_assert_locked(obj);
 	GEM_WARN_ON(!is_purgeable(msm_obj));
 
 	/* Get rid of any iommu mapping(s): */
@@ -782,7 +782,7 @@ void msm_gem_evict(struct drm_gem_object *obj)
 	struct drm_device *dev = obj->dev;
 	struct msm_gem_object *msm_obj = to_msm_bo(obj);
 
-	GEM_WARN_ON(!msm_gem_is_locked(obj));
+	msm_gem_assert_locked(obj);
 	GEM_WARN_ON(is_unevictable(msm_obj));
 
 	/* Get rid of any iommu mapping(s): */
@@ -797,7 +797,7 @@ void msm_gem_vunmap(struct drm_gem_object *obj)
 {
 	struct msm_gem_object *msm_obj = to_msm_bo(obj);
 
-	GEM_WARN_ON(!msm_gem_is_locked(obj));
+	msm_gem_assert_locked(obj);
 
 	if (!msm_obj->vaddr || GEM_WARN_ON(!is_vunmapable(msm_obj)))
 		return;
@@ -811,7 +811,7 @@ static void update_lru(struct drm_gem_object *obj)
 	struct msm_drm_private *priv = obj->dev->dev_private;
 	struct msm_gem_object *msm_obj = to_msm_bo(obj);
 
-	GEM_WARN_ON(!msm_gem_is_locked(&msm_obj->base));
+	msm_gem_assert_locked(&msm_obj->base);
 
 	if (!msm_obj->pages) {
 		GEM_WARN_ON(msm_obj->pin_count);
@@ -831,7 +831,7 @@ static void update_lru(struct drm_gem_object *obj)
 
 bool msm_gem_active(struct drm_gem_object *obj)
 {
-	GEM_WARN_ON(!msm_gem_is_locked(obj));
+	msm_gem_assert_locked(obj);
 
 	if (to_msm_bo(obj)->pin_count)
 		return true;
drivers/gpu/drm/msm/msm_gem.h +7 -1
index 0403b27ff779..3c6add51d13b 100644
@@ -215,6 +215,12 @@ msm_gem_is_locked(struct drm_gem_object *obj)
 	return dma_resv_is_locked(obj->resv) || (kref_read(&obj->refcount) == 0);
 }
 
+static inline void
+msm_gem_assert_locked(struct drm_gem_object *obj)
+{
+	GEM_WARN_ON(!msm_gem_is_locked(obj));
+}
+
 /* imported/exported objects are not purgeable: */
 static inline bool is_unpurgeable(struct msm_gem_object *msm_obj)
 {
@@ -229,7 +235,7 @@ static inline bool is_purgeable(struct msm_gem_object *msm_obj)
 
 static inline bool is_vunmapable(struct msm_gem_object *msm_obj)
 {
-	GEM_WARN_ON(!msm_gem_is_locked(&msm_obj->base));
+	msm_gem_assert_locked(&msm_obj->base);
 	return (msm_obj->vmap_count == 0) && msm_obj->vaddr;
 }
 
Signature · 1 line
2.36.1
author
#15 lore raw

PATCH v4 15/15 drm/msm/gem: Convert to lockdep assert

In reply to Rob Clark

From: Rob Clark <[email protected]>

---
 drivers/gpu/drm/msm/msm_gem.h | 15 ++++++---------
 1 file changed, 6 insertions(+), 9 deletions(-)
drivers/gpu/drm/msm/msm_gem.h +6 -9
index 3c6add51d13b..c4844cf3a585 100644
@@ -197,8 +197,8 @@ msm_gem_unlock(struct drm_gem_object *obj)
 	dma_resv_unlock(obj->resv);
 }
 
-static inline bool
-msm_gem_is_locked(struct drm_gem_object *obj)
+static inline void
+msm_gem_assert_locked(struct drm_gem_object *obj)
 {
 	/*
 	 * Destroying the object is a special case.. msm_gem_free_object()
@@ -212,13 +212,10 @@ msm_gem_is_locked(struct drm_gem_object *obj)
 	 * Unfortunately lockdep is not aware of this detail.  So when the
 	 * refcount drops to zero, we pretend it is already locked.
 	 */
-	return dma_resv_is_locked(obj->resv) || (kref_read(&obj->refcount) == 0);
-}
-
-static inline void
-msm_gem_assert_locked(struct drm_gem_object *obj)
-{
-	GEM_WARN_ON(!msm_gem_is_locked(obj));
+	lockdep_assert_once(
+		(kref_read(&obj->refcount) == 0) ||
+		(lockdep_is_held(&obj->resv->lock.base) != LOCK_STATE_NOT_HELD)
+	);
 }
 
 /* imported/exported objects are not purgeable: */
Signature · 1 line
2.36.1