Intel-XE Archive on lore.kernel.org
 help / color / mirror / Atom feed
* [PATCH v4] drm/xe/ufence: Signal ufence faster when possible
@ 2024-10-18 15:29 Nirmoy Das
  2024-10-18 16:16 ` ✓ CI.Patch_applied: success for " Patchwork
                   ` (17 more replies)
  0 siblings, 18 replies; 20+ messages in thread
From: Nirmoy Das @ 2024-10-18 15:29 UTC (permalink / raw)
  To: intel-xe; +Cc: Nirmoy Das, Matthew Auld, Matthew Brost

When the backing fence is already signaled, the ufence can be
immediately signaled without queuing in the ordered work queue.
This should also reduce load from the xe ordered_wq and won't
block signaling a ufence which doesn't require any serialization.

v2: fix system_wq typo
v3: signal immediately instead of queuing in system_wq (Matt B)
v4: revert back to v2 of using workqueue because of locking issue
    and remote viewing a different mm struct.
    Use Xe's unordered_wq which should be less congested than global
    one.

Link: https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/1630
Cc: Matthew Auld <matthew.auld@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Nirmoy Das <nirmoy.das@intel.com>
---
 drivers/gpu/drm/xe/xe_sync.c | 24 +++++++++++++++++++++---
 1 file changed, 21 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_sync.c b/drivers/gpu/drm/xe/xe_sync.c
index a90480c6aecf..7a1558c7ce09 100644
--- a/drivers/gpu/drm/xe/xe_sync.c
+++ b/drivers/gpu/drm/xe/xe_sync.c
@@ -92,18 +92,27 @@ static void user_fence_worker(struct work_struct *w)
 	user_fence_put(ufence);
 }
 
-static void kick_ufence(struct xe_user_fence *ufence, struct dma_fence *fence)
+static void kick_ufence_ordered(struct xe_user_fence *ufence,
+				struct dma_fence *fence)
 {
 	INIT_WORK(&ufence->worker, user_fence_worker);
 	queue_work(ufence->xe->ordered_wq, &ufence->worker);
 	dma_fence_put(fence);
 }
 
+static void kick_ufence_unordered(struct xe_user_fence *ufence,
+				  struct dma_fence *fence)
+{
+	INIT_WORK(&ufence->worker, user_fence_worker);
+	queue_work(ufence->xe->unordered_wq, &ufence->worker);
+	dma_fence_put(fence);
+}
+
 static void user_fence_cb(struct dma_fence *fence, struct dma_fence_cb *cb)
 {
 	struct xe_user_fence *ufence = container_of(cb, struct xe_user_fence, cb);
 
-	kick_ufence(ufence, fence);
+	kick_ufence_ordered(ufence, fence);
 }
 
 int xe_sync_entry_parse(struct xe_device *xe, struct xe_file *xef,
@@ -239,7 +248,16 @@ void xe_sync_entry_signal(struct xe_sync_entry *sync, struct dma_fence *fence)
 		err = dma_fence_add_callback(fence, &sync->ufence->cb,
 					     user_fence_cb);
 		if (err == -ENOENT) {
-			kick_ufence(sync->ufence, fence);
+			/*
+			 * use unordered_wq to schedule it faster and to keep
+			 * the ordered_wq less loaded as serialization is not
+			 * needed for when the fence is already signaled.
+			 *
+			 * This needs to be done with a wq here to avoid locking
+			 * issue when a ufence addr is backed by a bo and also
+			 * tsk->mm needs to null to call kthread_use_mm().
+			 */
+			kick_ufence_unordered(sync->ufence, fence);
 		} else if (err) {
 			XE_WARN_ON("failed to add user fence");
 			user_fence_put(sync->ufence);
-- 
2.46.0


^ permalink raw reply related	[flat|nested] 20+ messages in thread
end of thread, other threads:[~2024-11-11 14:49 UTC | newest]

Thread overview: 20+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2024-10-18 15:29 [PATCH v4] drm/xe/ufence: Signal ufence faster when possible Nirmoy Das
2024-10-18 16:16 ` ✓ CI.Patch_applied: success for " Patchwork
2024-10-18 16:17 ` ✓ CI.checkpatch: " Patchwork
2024-10-18 16:18 ` ✓ CI.KUnit: " Patchwork
2024-10-18 16:29 ` ✓ CI.Build: " Patchwork
2024-10-18 16:32 ` ✓ CI.Hooks: " Patchwork
2024-10-18 16:35 ` ✓ CI.checksparse: " Patchwork
2024-10-18 16:58 ` ✓ CI.BAT: " Patchwork
2024-10-19  6:56 ` ✗ CI.FULL: failure " Patchwork
2024-10-22 16:06 ` [PATCH v4] " Cavitt, Jonathan
2024-11-07 11:23 ` ✓ CI.Patch_applied: success for drm/xe/ufence: Signal ufence faster when possible (rev2) Patchwork
2024-11-07 11:23 ` ✓ CI.checkpatch: " Patchwork
2024-11-07 11:24 ` ✓ CI.KUnit: " Patchwork
2024-11-07 11:36 ` ✓ CI.Build: " Patchwork
2024-11-07 11:38 ` ✓ CI.Hooks: " Patchwork
2024-11-07 11:39 ` ✓ CI.checksparse: " Patchwork
2024-11-07 12:04 ` ✓ CI.BAT: " Patchwork
2024-11-08 18:12 ` ✗ CI.FULL: failure " Patchwork
2024-11-10  3:51 ` [PATCH v4] drm/xe/ufence: Signal ufence faster when possible Matthew Brost
2024-11-11 14:49   ` Nirmoy Das

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox