[PATCH v4 0/5] Add a minimum age gate to the shrinker + LRU fixes

[PATCH v4 0/5] Add a minimum age gate to the shrinker + LRU fixes

[Matthew Brost]

Fix misuse of the TTM LRU in Xe + add a minimum age gate to the shrinker
which is hidden behind a Kconfig for shrinker tuning.

Matt  

Matthew Brost (5):
  drm/xe: Bump the LRU of every BO an exec touches
  drm/xe: Bump the LRU on a successful page fault
  drm/xe: Leave the VM bulk move on BO release, not on GEM close
  drm/ttm: Allow skipping to the next priority in the bo LRU cursor
  drm/xe: Add a minimum age gate to the shrinker

 drivers/gpu/drm/ttm/ttm_bo_util.c    |   3 +
 drivers/gpu/drm/ttm/ttm_resource.c   |  35 +++++-
 drivers/gpu/drm/xe/Kconfig.profile   |  15 +++
 drivers/gpu/drm/xe/xe_bo.c           | 106 ++++++++++++++---
 drivers/gpu/drm/xe/xe_bo.h           |   5 +
 drivers/gpu/drm/xe/xe_bo_types.h     |   8 ++
 drivers/gpu/drm/xe/xe_device_sysfs.c |  35 ++++++
 drivers/gpu/drm/xe/xe_pagefault.c    |   2 +
 drivers/gpu/drm/xe/xe_shrinker.c     | 164 +++++++++++++++++++++++++--
 drivers/gpu/drm/xe/xe_shrinker.h     |   3 +
 drivers/gpu/drm/xe/xe_vm.c           |  19 +++-
 drivers/gpu/drm/xe/xe_vm_types.h     |   8 ++
 include/drm/ttm/ttm_bo.h             |  19 ++++
 include/drm/ttm/ttm_resource.h       |   2 +
 14 files changed, 390 insertions(+), 34 deletions(-)

-- 
2.34.1

[PATCH v4 1/5] drm/xe: Bump the LRU of every BO an exec touches

[Matthew Brost]

xe_vm_bump_lru() bumps the VM's bulk move and then, individually, every
locked object drm_gpuvm_is_extobj() reports. That misses BOs which are
neither: a VM in fault mode is created with a NULL bulk move, so its
private BOs ride along with nothing and are never bumped at all. They
just sit wherever they first landed on the LRU, however hot they are,
and are the first thing eviction and the shrinker pick.

Bump everything locked by the exec that the bulk move doesn't already
cover instead. The VM's dummy resv object is skipped explicitly since
it isn't a BO; it was previously filtered out by the extobj test.

Fixes: 7ba4c5f02763 ("drm/xe: VM LRU bulk move")
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
---
 drivers/gpu/drm/xe/xe_vm.c | 17 +++++++++++------
 1 file changed, 11 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
index 86c3b4fd232f..25d5e164d260 100644
--- a/drivers/gpu/drm/xe/xe_vm.c
+++ b/drivers/gpu/drm/xe/xe_vm.c
@@ -453,10 +453,8 @@ int xe_vm_validate_rebind(struct xe_vm *vm, struct drm_exec *exec,
  * @vm: The vm for which we are bumping LRUs.
  * @exec: The struct drm_exec with the locked GEM objects.
  *
- * Move the VM's private BOs, and all external objects locked by @exec,
- * to the LRU tail. Private BOs are moved in bulk using @vm's
- * &xe_vm.lru_bulk_move. External objects, which may be shared with other
- * VMs, are not part of that bulk move and are bumped individually.
+ * Move all BOs locked by @exec to the LRU tail. Those covered by @vm's
+ * &xe_vm.lru_bulk_move are moved in bulk, the rest individually.
  */
 void xe_vm_bump_lru(struct xe_vm *vm, struct drm_exec *exec)
 {
@@ -466,10 +464,17 @@ void xe_vm_bump_lru(struct xe_vm *vm, struct drm_exec *exec)
 	spin_lock(lru_lock);
 	ttm_lru_bulk_move_tail(&vm->lru_bulk_move);
 	drm_exec_for_each_locked_object(exec, obj) {
-		if (!drm_gpuvm_is_extobj(&vm->gpuvm, obj))
+		struct xe_bo *bo;
+
+		/* The VM's dummy resv object, not a BO. */
+		if (obj == xe_vm_obj(vm))
+			continue;
+
+		bo = gem_to_xe_bo(obj);
+		if (bo->ttm.bulk_move == &vm->lru_bulk_move)
 			continue;
 
-		ttm_bo_move_to_lru_tail(&gem_to_xe_bo(obj)->ttm);
+		ttm_bo_move_to_lru_tail(&bo->ttm);
 		cond_resched_lock(lru_lock);
 	}
 	spin_unlock(lru_lock);
-- 
2.34.1

[PATCH v4 2/5] drm/xe: Bump the LRU on a successful page fault

[Matthew Brost]

Servicing a page fault validates the VMA's backing store and binds it,
making the BO part of the VM's active working set, but unlike exec it
never moves anything on the TTM LRU. In fault mode, where a VM may run
for a long time without a single exec, the BOs the GPU actually touches
therefore drift towards the LRU head and are picked by eviction and by
the shrinker ahead of genuinely cold BOs.

Bump the VM's bulk move, and any external objects locked for the fault,
once the rebind has succeeded and while the objects are still locked,
the same way xe_exec_ioctl() does.

Fixes: dd08ebf6c352 ("drm/xe: Introduce a new DRM driver for Intel GPUs")
Assisted-by: LLM
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
---
 drivers/gpu/drm/xe/xe_pagefault.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/gpu/drm/xe/xe_pagefault.c b/drivers/gpu/drm/xe/xe_pagefault.c
index 5247321c805e..3719258aa6f0 100644
--- a/drivers/gpu/drm/xe/xe_pagefault.c
+++ b/drivers/gpu/drm/xe/xe_pagefault.c
@@ -221,6 +221,8 @@ static int xe_pagefault_handle_vma(struct xe_gt *gt, struct xe_vma *vma,
 				xe_pagefault_set_error(pf, XE_PAGEFAULT_ERROR_VMA_REBIND);
 				break;
 			}
+
+			xe_vm_bump_lru(vm, &exec);
 		}
 		xe_validation_ctx_fini(&ctx);
 	} while (err == -EAGAIN);
-- 
2.34.1

[PATCH v4 3/5] drm/xe: Leave the VM bulk move on BO release, not on GEM close

[Matthew Brost]

xe_gem_object_close() drops a VM private BO out of the VM's LRU bulk
move as soon as the GEM handle goes away. A BO can easily outlive its
handle though, since every VM bind holds a reference, and such a BO is
still very much part of the VM's working set.

Once detached, it no longer rides along with the bulk move on exec, so
nothing ever bumps its LRU position again. It just sits where it was
left, orphaned ahead of genuinely colder BOs, and is evicted or shrunk
first even while the VM keeps using it.

Do the detach in xe_ttm_bo_release_notify() instead, where the BO
really is going away. The BO holds a reference to its VM, so the bulk
move is still alive at that point.

Assisted-by: LLM
Fixes: dd08ebf6c352 ("drm/xe: Introduce a new DRM driver for Intel GPUs")
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
---
 drivers/gpu/drm/xe/xe_bo.c | 23 ++++++++---------------
 1 file changed, 8 insertions(+), 15 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_bo.c b/drivers/gpu/drm/xe/xe_bo.c
index 42a6aa1a2e4a..ba16b260dbcb 100644
--- a/drivers/gpu/drm/xe/xe_bo.c
+++ b/drivers/gpu/drm/xe/xe_bo.c
@@ -1869,6 +1869,14 @@ static void xe_ttm_bo_release_notify(struct ttm_buffer_object *ttm_bo)
 	if (!xe_ttm_bo_lock_in_destructor(ttm_bo))
 		return;
 
+	/*
+	 * Leave the VM's bulk move, which is only kept alive by the VM, and
+	 * whose LRU position no longer says anything about a BO on its way
+	 * out.
+	 */
+	if (ttm_bo->bulk_move)
+		ttm_bo_set_bulk_move(ttm_bo, NULL);
+
 	/*
 	 * Scrub the preempt fences if any. The unbind fence is already
 	 * attached to the resv.
@@ -2060,20 +2068,6 @@ static void xe_gem_object_free(struct drm_gem_object *obj)
 	ttm_bo_fini(container_of(obj, struct ttm_buffer_object, base));
 }
 
-static void xe_gem_object_close(struct drm_gem_object *obj,
-				struct drm_file *file_priv)
-{
-	struct xe_bo *bo = gem_to_xe_bo(obj);
-
-	if (bo->vm && !xe_vm_in_fault_mode(bo->vm)) {
-		xe_assert(xe_bo_device(bo), xe_bo_is_user(bo));
-
-		xe_bo_lock(bo, false);
-		ttm_bo_set_bulk_move(&bo->ttm, NULL);
-		xe_bo_unlock(bo);
-	}
-}
-
 static bool should_migrate_to_smem(struct xe_bo *bo)
 {
 	/*
@@ -2408,7 +2402,6 @@ static int xe_gem_object_mmap(struct drm_gem_object *obj, struct vm_area_struct
 
 static const struct drm_gem_object_funcs xe_gem_object_funcs = {
 	.free = xe_gem_object_free,
-	.close = xe_gem_object_close,
 	.mmap = xe_gem_object_mmap,
 	.export = xe_gem_prime_export,
 	.vm_ops = &xe_gem_vm_ops,
-- 
2.34.1

[PATCH v4 4/5] drm/ttm: Allow skipping to the next priority in the bo LRU cursor

[Matthew Brost]

Drivers walking a manager's LRU with ttm_bo_lru_for_each_reserved_guarded()
may decide that the rest of the current priority isn't worth walking,
for example when the priority is roughly age ordered and the walk hit
a BO that is still too hot to evict. Today the only option is to stop
the walk altogether, which also gives up on the remaining priorities.

Add ttm_resource_manager_next_priority() to move a resource cursor to
the start of the next priority, and ttm_bo_lru_cursor_skip_priority()
to make the next ttm_bo_lru_cursor_next() use it.

While at it, stop incrementing the cursor priority past the last valid
priority once the iteration is done.

Assisted-by: LLM
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
---
 drivers/gpu/drm/ttm/ttm_bo_util.c  |  3 +++
 drivers/gpu/drm/ttm/ttm_resource.c | 35 ++++++++++++++++++++++++++----
 include/drm/ttm/ttm_bo.h           | 19 ++++++++++++++++
 include/drm/ttm/ttm_resource.h     |  2 ++
 4 files changed, 55 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/ttm/ttm_bo_util.c b/drivers/gpu/drm/ttm/ttm_bo_util.c
index f9458a4d5a09..d626f014e61f 100644
--- a/drivers/gpu/drm/ttm/ttm_bo_util.c
+++ b/drivers/gpu/drm/ttm/ttm_bo_util.c
@@ -997,6 +997,9 @@ __ttm_bo_lru_cursor_next(struct ttm_bo_lru_cursor *curs)
 		if (first) {
 			res = ttm_resource_manager_first(&curs->res_curs);
 			first = false;
+		} else if (curs->skip_priority) {
+			res = ttm_resource_manager_next_priority(&curs->res_curs);
+			curs->skip_priority = false;
 		} else {
 			res = ttm_resource_manager_next(&curs->res_curs);
 		}
diff --git a/drivers/gpu/drm/ttm/ttm_resource.c b/drivers/gpu/drm/ttm/ttm_resource.c
index 4e6d3d658d85..e631507d7aa1 100644
--- a/drivers/gpu/drm/ttm/ttm_resource.c
+++ b/drivers/gpu/drm/ttm/ttm_resource.c
@@ -674,6 +674,17 @@ ttm_resource_cursor_check_bulk(struct ttm_resource_cursor *cursor,
 	}
 }
 
+static bool ttm_resource_cursor_next_priority(struct ttm_resource_cursor *cursor)
+{
+	if (cursor->priority + 1 >= TTM_MAX_BO_PRIORITY)
+		return false;
+
+	list_move(&cursor->hitch.link, &cursor->man->lru[++cursor->priority]);
+	ttm_resource_cursor_clear_bulk(cursor);
+
+	return true;
+}
+
 /**
  * ttm_resource_manager_first() - Start iterating over the resources
  * of a resource manager
@@ -723,16 +734,32 @@ ttm_resource_manager_next(struct ttm_resource_cursor *cursor)
 			}
 		}
 
-		if (++cursor->priority >= TTM_MAX_BO_PRIORITY)
+		if (!ttm_resource_cursor_next_priority(cursor))
 			break;
-
-		list_move(&cursor->hitch.link, &man->lru[cursor->priority]);
-		ttm_resource_cursor_clear_bulk(cursor);
 	}
 
 	return NULL;
 }
 
+/**
+ * ttm_resource_manager_next_priority() - Skip the rest of the current
+ * priority and continue iterating at the next one
+ * @cursor: cursor to record the position
+ *
+ * Return: the first resource of the next non-empty priority, or NULL if
+ * there is none.
+ */
+struct ttm_resource *
+ttm_resource_manager_next_priority(struct ttm_resource_cursor *cursor)
+{
+	lockdep_assert_held(&cursor->man->bdev->lru_lock);
+
+	if (!ttm_resource_cursor_next_priority(cursor))
+		return NULL;
+
+	return ttm_resource_manager_next(cursor);
+}
+
 /**
  * ttm_lru_first_res_or_null() - Return the first resource on an lru list
  * @head: The list head of the lru list.
diff --git a/include/drm/ttm/ttm_bo.h b/include/drm/ttm/ttm_bo.h
index 32791c4db2a9..2fc517dbf03d 100644
--- a/include/drm/ttm/ttm_bo.h
+++ b/include/drm/ttm/ttm_bo.h
@@ -507,10 +507,29 @@ struct ttm_bo_lru_cursor {
 	 * unlock before the next iteration or after loop exit.
 	 */
 	bool needs_unlock;
+	/**
+	 * @skip_priority: Continue at the next LRU priority on the next
+	 * iteration. See ttm_bo_lru_cursor_skip_priority().
+	 */
+	bool skip_priority;
 	/** @arg: Pointer to common BO LRU walk arguments. */
 	struct ttm_lru_walk_arg *arg;
 };
 
+/**
+ * ttm_bo_lru_cursor_skip_priority() - Skip the rest of the current priority
+ * @curs: The cursor.
+ *
+ * Make the next ttm_bo_lru_cursor_next() skip the remaining buffer objects
+ * of the current LRU priority and continue at the next priority, for
+ * example when the caller decides the rest of the priority isn't worth
+ * walking.
+ */
+static inline void ttm_bo_lru_cursor_skip_priority(struct ttm_bo_lru_cursor *curs)
+{
+	curs->skip_priority = true;
+}
+
 void ttm_bo_lru_cursor_fini(struct ttm_bo_lru_cursor *curs);
 
 struct ttm_bo_lru_cursor *
diff --git a/include/drm/ttm/ttm_resource.h b/include/drm/ttm/ttm_resource.h
index 714c533622c2..8cd70d1e5481 100644
--- a/include/drm/ttm/ttm_resource.h
+++ b/include/drm/ttm/ttm_resource.h
@@ -499,6 +499,8 @@ struct ttm_resource *
 ttm_resource_manager_first(struct ttm_resource_cursor *cursor);
 struct ttm_resource *
 ttm_resource_manager_next(struct ttm_resource_cursor *cursor);
+struct ttm_resource *
+ttm_resource_manager_next_priority(struct ttm_resource_cursor *cursor);
 
 struct ttm_resource *
 ttm_lru_first_res_or_null(struct list_head *head);
-- 
2.34.1

[PATCH v4 5/5] drm/xe: Add a minimum age gate to the shrinker

[Matthew Brost]

Backing up a BO costs a GPU blit plus a page copy, and if the BO is part
of an active working set it is faulted straight back in on the next exec.
Evicting a BO that was used moments ago is therefore almost always a
loss, yet the shrinker has no notion of recency beyond LRU order.

Timestamp BOs wherever their LRU position is updated, and read the stamp
back from whatever owns that position. A BO in a VM's bulk move is
anchored inside its range, so it is aged by the VM-wide stamp taken in
xe_vm_bump_lru() alongside the bulk move itself. Any other BO carries
its own, stamped there too and on a successful move into TT in
xe_bo_move(), where ttm_resource_alloc() has already placed the new
resource at the LRU tail.

Add a per-device shrinker_min_age_ms sysfs attribute, defaulting to
CONFIG_DRM_XE_SHRINKER_MIN_AGE_MS (0, disabled). While set, the shrinker
won't back up BOs younger than the delay. Each LRU priority is roughly
age ordered, so the gate is applied per priority: xe_shrinker_count()
snoops the first shrinkable BO of each priority and reports nothing only
if all of them are young, and the backup walk skips to the next priority
at the first young BO, both as a backstop for the count sample and so a
hot priority doesn't shield colder BOs in another one.

Only BOs this shrinker would actually back up have a say. BOs in
XE_PL_SYSTEM, which are not GPU mapped and hence not part of a working
set, and BOs xe_bo_shrink() rejects outright are never young. Purgeable
BOs and those advised DONTNEED are neither: they are dropped rather than
backed up and accounted to the purgeable shrinker, so they are never
young and the count snoop skips their XE_BO_PRIORITY_LOW band entirely.

The gate is hard and applies at every reclaim priority, so a working set
kept younger than the delay is never backed up and can lead to OOM. This
is deliberate, and why it is off by default.

Assisted-by: LLM
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
---
 drivers/gpu/drm/xe/Kconfig.profile   |  15 +++
 drivers/gpu/drm/xe/xe_bo.c           |  83 ++++++++++++++
 drivers/gpu/drm/xe/xe_bo.h           |   5 +
 drivers/gpu/drm/xe/xe_bo_types.h     |   8 ++
 drivers/gpu/drm/xe/xe_device_sysfs.c |  35 ++++++
 drivers/gpu/drm/xe/xe_shrinker.c     | 164 +++++++++++++++++++++++++--
 drivers/gpu/drm/xe/xe_shrinker.h     |   3 +
 drivers/gpu/drm/xe/xe_vm.c           |   4 +-
 drivers/gpu/drm/xe/xe_vm_types.h     |   8 ++
 9 files changed, 315 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/xe/Kconfig.profile b/drivers/gpu/drm/xe/Kconfig.profile
index 23045ed2aa94..dd3985e502b8 100644
--- a/drivers/gpu/drm/xe/Kconfig.profile
+++ b/drivers/gpu/drm/xe/Kconfig.profile
@@ -74,6 +74,7 @@ config DRM_XE_ENABLE_SCHEDTIMEOUT_LIMIT
 	  to apply to applicable user. For elevated user, all above MIN
 	  and MAX values will apply when this configuration is enable to
 	  apply limitation. By default limitation is applied.
+
 config DRM_XE_SHRINKER_NON_PURGEABLE_SEEKS
 	int "Default seeks value of the non-purgeable bo shrinker"
 	default 8
@@ -86,3 +87,17 @@ config DRM_XE_SHRINKER_NON_PURGEABLE_SEEKS
 	  reclaim pressure on non-purgeable buffer objects, which need
 	  GPU idling and writeback to be reclaimed, in favour of purgeable
 	  ones and other kernel caches.
+
+config DRM_XE_SHRINKER_MIN_AGE_MS
+	int "Default minimum BO age before shrinker backup (ms)"
+	default 0
+	range 0 1000000
+	help
+	  Default for the per-device shrinker_min_age_ms sysfs attribute. The
+	  shrinker won't back up a BO that was moved into TT, or bumped on the
+	  LRU by an exec, less than this many milliseconds ago. BOs already
+	  evicted to system memory are not part of a working set and are
+	  never protected. This applies at every reclaim priority: a working
+	  set kept younger than this is never backed up and can cause OOM.
+	  Only set if that risk is preferred over thrashing a hot working
+	  set. 0 disables.
diff --git a/drivers/gpu/drm/xe/xe_bo.c b/drivers/gpu/drm/xe/xe_bo.c
index ba16b260dbcb..59874f9a1ed7 100644
--- a/drivers/gpu/drm/xe/xe_bo.c
+++ b/drivers/gpu/drm/xe/xe_bo.c
@@ -1357,9 +1357,92 @@ static int xe_bo_move(struct ttm_buffer_object *ttm_bo, bool evict,
 		xe_tt_unmap_sg(xe, ttm_bo->ttm);
 	}
 
+	/*
+	 * ttm_resource_alloc() already placed the new resource at the tail of
+	 * its LRU, or of @bo's bulk move, so only record when that happened.
+	 */
+	if (!ret && ttm_bo->resource && ttm_bo->resource->mem_type == XE_PL_TT)
+		WRITE_ONCE(bo->lru_stamp, jiffies);
+
 	return ret;
 }
 
+/**
+ * xe_bo_bump_lru_locked() - Bump a BO on the LRU and stamp its last use
+ * @bo: The buffer object, reserved by the caller.
+ *
+ * Move @bo to the tail of its LRU and record the time for the shrinker's
+ * minimum age gate. Only meaningful for BOs outside a bulk move, since a
+ * BO in one is anchored inside its range and is bumped, and aged, with it.
+ * The caller must hold the device LRU lock.
+ */
+void xe_bo_bump_lru_locked(struct xe_bo *bo)
+{
+	WRITE_ONCE(bo->lru_stamp, jiffies);
+	ttm_bo_move_to_lru_tail(&bo->ttm);
+}
+
+/**
+ * xe_bo_shrink_backup_candidate() - Whether the shrinker may back up a BO
+ * @bo: The TTM buffer object, reserved by the caller.
+ *
+ * A BO is a backup candidate if backing it up is the only way to reclaim it.
+ * BOs advised DONTNEED, which sit at XE_BO_PRIORITY_LOW, and purgeable ones
+ * are not: their pages are simply dropped, which is cheap, and they are
+ * accounted to the purgeable shrinker. Neither are the BOs xe_bo_shrink()
+ * hands to xe_bo_shrink_purge() outright: ghost BOs, which aren't xe BOs,
+ * and BOs already freed and waiting to be reaped. Nor are BOs whose tt
+ * isn't TTM_TT_FLAG_EXTERNAL_MAPPABLE, which xe_bo_shrink() rejects.
+ *
+ * Return: true if @bo would be backed up by a non-purge shrink pass.
+ */
+bool xe_bo_shrink_backup_candidate(struct ttm_buffer_object *bo)
+{
+	struct xe_ttm_tt *xe_tt;
+
+	if (!xe_bo_is_xe_bo(bo) || !kref_read(&bo->base.refcount) || !bo->ttm ||
+	    !(bo->ttm->page_flags & TTM_TT_FLAG_EXTERNAL_MAPPABLE) ||
+	    bo->priority <= XE_BO_PRIORITY_LOW)
+		return false;
+
+	xe_tt = container_of(bo->ttm, struct xe_ttm_tt, ttm);
+
+	return !xe_tt->purgeable;
+}
+
+/**
+ * xe_bo_shrink_too_young() - Whether a BO was used too recently to shrink
+ * @bo: The TTM buffer object, reserved by the caller.
+ * @min_age: Minimum age in jiffies. 0 disables the check.
+ *
+ * The age is taken from whatever owns @bo's LRU position: &xe_vm.lru_stamp
+ * if @bo rides along with a VM's bulk move, its own &xe_bo.lru_stamp if not.
+ *
+ * BOs that the shrink walk would not back up anyway are never considered
+ * young, see xe_bo_shrink_backup_candidate(). Neither are BOs in
+ * XE_PL_SYSTEM: they are not GPU mapped, hence not part of a working set,
+ * and must be moved back to TT, refreshing the stamp, before use.
+ *
+ * Return: true if @bo was touched less than @min_age jiffies ago.
+ */
+bool xe_bo_shrink_too_young(struct ttm_buffer_object *bo, unsigned long min_age)
+{
+	struct xe_bo *xe_bo;
+	unsigned long stamp;
+
+	if (!min_age || bo->resource->mem_type == XE_PL_SYSTEM ||
+	    !xe_bo_shrink_backup_candidate(bo))
+		return false;
+
+	xe_bo = ttm_to_xe_bo(bo);
+	if (bo->bulk_move && xe_bo->vm)
+		stamp = READ_ONCE(xe_bo->vm->lru_stamp);
+	else
+		stamp = READ_ONCE(xe_bo->lru_stamp);
+
+	return stamp && time_before(jiffies, stamp + min_age);
+}
+
 static long xe_bo_shrink_purge(struct ttm_operation_ctx *ctx,
 			       struct ttm_buffer_object *bo,
 			       unsigned long *scanned)
diff --git a/drivers/gpu/drm/xe/xe_bo.h b/drivers/gpu/drm/xe/xe_bo.h
index 77b3a02e9b8e..65d4c638e8d8 100644
--- a/drivers/gpu/drm/xe/xe_bo.h
+++ b/drivers/gpu/drm/xe/xe_bo.h
@@ -623,6 +623,11 @@ struct xe_bo_shrink_flags {
 	u32 writeback : 1;
 };
 
+void xe_bo_bump_lru_locked(struct xe_bo *bo);
+
+bool xe_bo_shrink_backup_candidate(struct ttm_buffer_object *bo);
+bool xe_bo_shrink_too_young(struct ttm_buffer_object *bo, unsigned long min_age);
+
 long xe_bo_shrink(struct ttm_operation_ctx *ctx, struct ttm_buffer_object *bo,
 		  const struct xe_bo_shrink_flags flags,
 		  unsigned long *scanned);
diff --git a/drivers/gpu/drm/xe/xe_bo_types.h b/drivers/gpu/drm/xe/xe_bo_types.h
index 0265fb314c77..f63b668c7134 100644
--- a/drivers/gpu/drm/xe/xe_bo_types.h
+++ b/drivers/gpu/drm/xe/xe_bo_types.h
@@ -49,6 +49,14 @@ struct xe_bo {
 	u32 flags;
 	/** @vm: VM this BO is attached to, for extobj this will be NULL */
 	struct xe_vm *vm;
+	/**
+	 * @lru_stamp: jiffies at the last LRU bump, done on a move into TT
+	 * and at exec time. Only describes the LRU position of a BO outside
+	 * a bulk move; BOs in one are aged by &xe_vm.lru_stamp instead.
+	 * Consulted by the shrinker's minimum age gate. Accessed with
+	 * READ_ONCE()/WRITE_ONCE().
+	 */
+	unsigned long lru_stamp;
 	/**
 	 * @q: Queue this BO is attached to, mostly for LRC BO, NULL otherwise.
 	 * Protected by the BO dma_resv: readers must hold it across both the
diff --git a/drivers/gpu/drm/xe/xe_device_sysfs.c b/drivers/gpu/drm/xe/xe_device_sysfs.c
index ff5d25dd867c..26b1f71ed522 100644
--- a/drivers/gpu/drm/xe/xe_device_sysfs.c
+++ b/drivers/gpu/drm/xe/xe_device_sysfs.c
@@ -38,6 +38,12 @@
  * reclaimed by the shrinker. Higher values reduce reclaim pressure on
  * non-purgeable buffer objects. Valid values are 1 to 64. Defaults to
  * CONFIG_DRM_XE_SHRINKER_NON_PURGEABLE_SEEKS.
+ *
+ * shrinker_min_age_ms - Report/change the minimum time, in ms, since a BO was
+ * last moved to TT or bumped on the LRU by an exec before the shrinker may
+ * back it up. BOs evicted to system memory are never protected. Defaults to
+ * CONFIG_DRM_XE_SHRINKER_MIN_AGE_MS, 0 disables. This applies at every
+ * reclaim priority, so a non-zero value can lead to OOM.
  */
 
 static ssize_t
@@ -331,8 +337,37 @@ shrinker_non_purgeable_seeks_store(struct device *dev,
 
 static DEVICE_ATTR_RW(shrinker_non_purgeable_seeks);
 
+static ssize_t
+shrinker_min_age_ms_show(struct device *dev, struct device_attribute *attr,
+			 char *buf)
+{
+	struct xe_device *xe = pdev_to_xe_device(to_pci_dev(dev));
+
+	return sysfs_emit(buf, "%u\n", xe_shrinker_min_age_ms(xe->mem.shrinker));
+}
+
+static ssize_t
+shrinker_min_age_ms_store(struct device *dev, struct device_attribute *attr,
+			  const char *buff, size_t count)
+{
+	struct xe_device *xe = pdev_to_xe_device(to_pci_dev(dev));
+	unsigned int ms;
+	int ret;
+
+	ret = kstrtouint(buff, 0, &ms);
+	if (ret)
+		return ret;
+
+	xe_shrinker_set_min_age_ms(xe->mem.shrinker, ms);
+
+	return count;
+}
+
+static DEVICE_ATTR_RW(shrinker_min_age_ms);
+
 static struct attribute *shrinker_attrs[] = {
 	&dev_attr_shrinker_non_purgeable_seeks.attr,
+	&dev_attr_shrinker_min_age_ms.attr,
 	NULL
 };
 
diff --git a/drivers/gpu/drm/xe/xe_shrinker.c b/drivers/gpu/drm/xe/xe_shrinker.c
index c86455ea08d1..c650726ff6de 100644
--- a/drivers/gpu/drm/xe/xe_shrinker.c
+++ b/drivers/gpu/drm/xe/xe_shrinker.c
@@ -39,6 +39,8 @@
  * @shrink: Pointer to the mm shrinker handling non-purgeable (shrinkable) objects.
  * @purge_shrink: Pointer to the mm shrinker handling purgeable objects.
  * @pm_worker: Worker to wake up the device if required.
+ * @min_age_ms: Minimum BO age before backup, in ms. 0 disables. Defaults to
+ * CONFIG_DRM_XE_SHRINKER_MIN_AGE_MS, overridable through sysfs.
  */
 struct xe_shrinker {
 	struct xe_device *xe;
@@ -50,6 +52,7 @@ struct xe_shrinker {
 	struct shrinker *shrink;
 	struct shrinker *purge_shrink;
 	struct work_struct pm_worker;
+	unsigned int min_age_ms;
 };
 
 static struct xe_shrinker *to_xe_shrinker(struct shrinker *shrink)
@@ -104,6 +107,7 @@ static void xe_shrinker_runtime_pm_put(struct xe_shrinker *shrinker, bool runtim
 static int __xe_shrinker_walk(struct xe_shrinker *shrinker,
 			      struct ttm_operation_ctx *ctx,
 			      const struct xe_bo_shrink_flags flags,
+			      unsigned long min_age,
 			      unsigned long to_scan, unsigned long *scanned,
 			      unsigned long *freed)
 {
@@ -139,6 +143,16 @@ static int __xe_shrinker_walk(struct xe_shrinker *shrinker,
 			if (!ttm_bo_shrink_suitable(ttm_bo, ctx))
 				continue;
 
+			/*
+			 * Each LRU priority is roughly age ordered, so the
+			 * rest of this priority is most likely young too.
+			 * Move on to the next priority.
+			 */
+			if (xe_bo_shrink_too_young(ttm_bo, min_age)) {
+				ttm_bo_lru_cursor_skip_priority(&curs);
+				continue;
+			}
+
 			lret = xe_bo_shrink(ctx, ttm_bo, flags, scanned);
 			/* Like the TTM LRU walk, skip bos we can't shrink now. */
 			if (lret == -EBUSY)
@@ -171,6 +185,7 @@ static int __xe_shrinker_walk(struct xe_shrinker *shrinker,
 static int xe_shrinker_walk(struct xe_shrinker *shrinker,
 			    struct ttm_operation_ctx *ctx,
 			    const struct xe_bo_shrink_flags flags,
+			    unsigned long min_age,
 			    unsigned long to_scan, unsigned long *scanned,
 			    unsigned long *freed)
 {
@@ -180,26 +195,126 @@ static int xe_shrinker_walk(struct xe_shrinker *shrinker,
 
 	swap(no_wait_gpu, ctx->no_wait_gpu);
 	save_flags.writeback = false;
-	ret = __xe_shrinker_walk(shrinker, ctx, save_flags, to_scan, scanned,
-				 freed);
+	ret = __xe_shrinker_walk(shrinker, ctx, save_flags, min_age,
+				 to_scan, scanned, freed);
 	swap(no_wait_gpu, ctx->no_wait_gpu);
 	if (ret || *scanned >= to_scan)
 		return ret;
 
 	if (!ctx->no_wait_gpu) {
-		ret = __xe_shrinker_walk(shrinker, ctx, save_flags, to_scan, scanned,
-					 freed);
+		ret = __xe_shrinker_walk(shrinker, ctx, save_flags, min_age,
+					 to_scan, scanned, freed);
 		if (ret || *scanned >= to_scan)
 			return ret;
 	}
 
 	if (flags.writeback)
-		ret = __xe_shrinker_walk(shrinker, ctx, flags, to_scan, scanned,
-					 freed);
+		ret = __xe_shrinker_walk(shrinker, ctx, flags, min_age,
+					 to_scan, scanned, freed);
 
 	return ret;
 }
 
+/**
+ * DOC: Shrinker minimum age
+ *
+ * The per-device shrinker_min_age_ms sysfs attribute, defaulting to
+ * CONFIG_DRM_XE_SHRINKER_MIN_AGE_MS, protects recently used BOs from backup.
+ * A BO's age is taken from whatever owns its position on the LRU, and is
+ * stamped wherever that position is updated: the VM-wide stamp for a BO
+ * riding along with a VM's bulk move, bumped on exec in xe_vm_bump_lru(),
+ * its own stamp otherwise, bumped there as well and on a move into TT in
+ * xe_bo_move(). BOs in XE_PL_SYSTEM are never young: they are not GPU
+ * mapped, so not part of a working set, and will be stamped again when
+ * moved back to TT for use.
+ *
+ * The gate is applied per LRU priority, each of which is roughly age
+ * ordered. xe_shrinker_count() snoops the first shrinkable BO of each
+ * priority on each TT-backed LRU. Only if all of them are young does it stop
+ * reporting shrinkable pages; anything else, including reaching no verdict
+ * at all, lets it report. The backup walk in xe_shrinker_scan() skips to the
+ * next priority at the first young BO, as a backstop since the count snoop
+ * is only a sample. A hot priority thus doesn't shield colder BOs in another
+ * priority. BOs that aren't backed up anyway have no say either way: those
+ * in XE_PL_SYSTEM are never young, and purgeable and DONTNEED ones, which
+ * belong to the purgeable shrinker, are ignored by the snoop entirely.
+ *
+ * The gate is hard and applies at every reclaim priority. A working set kept
+ * younger than the delay is never backed up, which can lead to OOM. This is
+ * intended for users who prefer that risk over thrashing a hot working set,
+ * hence it is disabled by default.
+ */
+static unsigned long xe_shrinker_min_age(struct xe_shrinker *shrinker)
+{
+	unsigned int ms = READ_ONCE(shrinker->min_age_ms);
+
+	return ms ? msecs_to_jiffies(ms) : 0;
+}
+
+/* Max shrink-unsuitable BOs to skip per LRU before giving up the snoop. */
+#define XE_SHRINKER_SNOOP_MAX	16
+
+/*
+ * Whether the first BO of every priority on every TT-backed LRU that this
+ * shrinker would back up is younger than @min_age. BOs are trylocked through
+ * the reserved LRU walker since their tt may be torn down while still on the
+ * LRU. Contended BOs are skipped by the walker, and may be all of them. If no
+ * verdict can be reached, because nothing was inspected or the snoop budget
+ * ran out, report not young and leave it to the scan side check.
+ */
+static bool xe_shrinker_lru_young(struct xe_shrinker *shrinker,
+				  unsigned long min_age)
+{
+	struct ttm_operation_ctx ctx = {
+		.interruptible = false,
+		.no_wait_gpu = false,
+	};
+	struct xe_device *xe = shrinker->xe;
+	unsigned int mem_type;
+	bool young = false;
+
+	for (mem_type = XE_PL_SYSTEM; mem_type <= XE_PL_TT; ++mem_type) {
+		struct ttm_resource_manager *man = ttm_manager_type(&xe->ttm, mem_type);
+		struct ttm_bo_lru_cursor curs;
+		struct ttm_buffer_object *ttm_bo;
+		struct ttm_lru_walk_arg arg = {
+			.ctx = &ctx,
+			.trylock_only = true,
+		};
+		unsigned int budget = XE_SHRINKER_SNOOP_MAX;
+
+		if (!man || !man->use_tt)
+			continue;
+
+		ttm_bo_lru_for_each_reserved_guarded(&curs, man, &arg, ttm_bo) {
+			/*
+			 * Nothing in the DONTNEED band is backed up, and it
+			 * is accounted to the purgeable shrinker, so it has
+			 * no say here. Skip the whole band in one go.
+			 */
+			if (ttm_bo->priority <= XE_BO_PRIORITY_LOW) {
+				ttm_bo_lru_cursor_skip_priority(&curs);
+				continue;
+			}
+
+			if (!ttm_bo_shrink_suitable(ttm_bo, &ctx) ||
+			    !xe_bo_shrink_backup_candidate(ttm_bo)) {
+				if (!--budget)
+					return false;
+				continue;
+			}
+
+			if (!xe_bo_shrink_too_young(ttm_bo, min_age))
+				return false;
+
+			young = true;
+			ttm_bo_lru_cursor_skip_priority(&curs);
+		}
+	}
+
+	return young;
+}
+
 static unsigned long
 xe_shrinker_count_purgeable(struct shrinker *shrink, struct shrink_control *sc)
 {
@@ -239,7 +354,7 @@ static unsigned long
 xe_shrinker_count(struct shrinker *shrink, struct shrink_control *sc)
 {
 	struct xe_shrinker *shrinker = to_xe_shrinker(shrink);
-	unsigned long num_pages;
+	unsigned long num_pages, min_age;
 	bool can_backup = !!(sc->gfp_mask & __GFP_FS);
 
 	if (!can_backup)
@@ -263,6 +378,10 @@ xe_shrinker_count(struct shrinker *shrink, struct shrink_control *sc)
 
 	write_unlock(&shrinker->lock);
 
+	min_age = xe_shrinker_min_age(shrinker);
+	if (num_pages && min_age && xe_shrinker_lru_young(shrinker, min_age))
+		return SHRINK_EMPTY;
+
 	return num_pages ? num_pages : SHRINK_EMPTY;
 }
 
@@ -297,7 +416,8 @@ static unsigned long xe_shrinker_scan_purgeable(struct shrinker *shrink,
 		.purge = true,
 	};
 
-	xe_shrinker_walk(shrinker, &ctx, shrink_flags, nr_to_scan, &nr_scanned, &freed);
+	xe_shrinker_walk(shrinker, &ctx, shrink_flags, 0, nr_to_scan, &nr_scanned,
+			 &freed);
 
 	sc->nr_scanned = nr_scanned;
 
@@ -322,7 +442,9 @@ static unsigned long xe_shrinker_scan(struct shrinker *shrink, struct shrink_con
 	/* Might need runtime PM to access backed-up or CCS-compressed data. */
 	runtime_pm = xe_shrinker_runtime_pm_get(shrinker);
 
-	xe_shrinker_walk(shrinker, &ctx, shrink_flags, nr_to_scan, &nr_scanned, &freed);
+	xe_shrinker_walk(shrinker, &ctx, shrink_flags,
+			 xe_shrinker_min_age(shrinker), nr_to_scan, &nr_scanned,
+			 &freed);
 
 	sc->nr_scanned = nr_scanned;
 	xe_shrinker_runtime_pm_put(shrinker, runtime_pm);
@@ -387,6 +509,29 @@ int xe_shrinker_set_non_purgeable_seeks(struct xe_shrinker *shrinker,
 	return 0;
 }
 
+/**
+ * xe_shrinker_min_age_ms() - Get the shrinker minimum BO age
+ * @shrinker: Pointer to the struct xe_shrinker.
+ *
+ * Return: Minimum age, in ms, a BO must reach before it may be backed up.
+ */
+unsigned int xe_shrinker_min_age_ms(struct xe_shrinker *shrinker)
+{
+	return READ_ONCE(shrinker->min_age_ms);
+}
+
+/**
+ * xe_shrinker_set_min_age_ms() - Set the shrinker minimum BO age
+ * @shrinker: Pointer to the struct xe_shrinker.
+ * @ms: Minimum age, in ms. 0 disables the gate.
+ *
+ * See DOC: Shrinker minimum age. A non-zero value can lead to OOM.
+ */
+void xe_shrinker_set_min_age_ms(struct xe_shrinker *shrinker, unsigned int ms)
+{
+	WRITE_ONCE(shrinker->min_age_ms, ms);
+}
+
 /**
  * xe_shrinker_create() - Create an xe per-device shrinker
  * @xe: Pointer to the xe device.
@@ -415,6 +560,7 @@ int xe_shrinker_create(struct xe_device *xe)
 
 	INIT_WORK(&shrinker->pm_worker, xe_shrinker_pm);
 	shrinker->xe = xe;
+	shrinker->min_age_ms = CONFIG_DRM_XE_SHRINKER_MIN_AGE_MS;
 	rwlock_init(&shrinker->lock);
 
 	shrinker->purge_shrink->count_objects = xe_shrinker_count_purgeable;
diff --git a/drivers/gpu/drm/xe/xe_shrinker.h b/drivers/gpu/drm/xe/xe_shrinker.h
index 4b3c73689619..586d7772bd58 100644
--- a/drivers/gpu/drm/xe/xe_shrinker.h
+++ b/drivers/gpu/drm/xe/xe_shrinker.h
@@ -18,4 +18,7 @@ int xe_shrinker_non_purgeable_seeks(struct xe_shrinker *shrinker);
 int xe_shrinker_set_non_purgeable_seeks(struct xe_shrinker *shrinker,
 					unsigned int seeks);
 
+unsigned int xe_shrinker_min_age_ms(struct xe_shrinker *shrinker);
+void xe_shrinker_set_min_age_ms(struct xe_shrinker *shrinker, unsigned int ms);
+
 #endif
diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
index 25d5e164d260..204bffab9870 100644
--- a/drivers/gpu/drm/xe/xe_vm.c
+++ b/drivers/gpu/drm/xe/xe_vm.c
@@ -461,6 +461,8 @@ void xe_vm_bump_lru(struct xe_vm *vm, struct drm_exec *exec)
 	spinlock_t *lru_lock = &vm->xe->ttm.lru_lock;
 	struct drm_gem_object *obj;
 
+	WRITE_ONCE(vm->lru_stamp, jiffies);
+
 	spin_lock(lru_lock);
 	ttm_lru_bulk_move_tail(&vm->lru_bulk_move);
 	drm_exec_for_each_locked_object(exec, obj) {
@@ -474,7 +476,7 @@ void xe_vm_bump_lru(struct xe_vm *vm, struct drm_exec *exec)
 		if (bo->ttm.bulk_move == &vm->lru_bulk_move)
 			continue;
 
-		ttm_bo_move_to_lru_tail(&bo->ttm);
+		xe_bo_bump_lru_locked(bo);
 		cond_resched_lock(lru_lock);
 	}
 	spin_unlock(lru_lock);
diff --git a/drivers/gpu/drm/xe/xe_vm_types.h b/drivers/gpu/drm/xe/xe_vm_types.h
index fb9305124679..046505a285e9 100644
--- a/drivers/gpu/drm/xe/xe_vm_types.h
+++ b/drivers/gpu/drm/xe/xe_vm_types.h
@@ -267,6 +267,14 @@ struct xe_vm {
 	/** @lru_bulk_move: Bulk LRU move list for this VM's BOs */
 	struct ttm_lru_bulk_move lru_bulk_move;
 
+	/**
+	 * @lru_stamp: jiffies at the last xe_vm_bump_lru(), which is where
+	 * @lru_bulk_move is moved to the LRU tail. Ages every BO in that
+	 * bulk move for the shrinker's minimum age gate. Accessed with
+	 * READ_ONCE()/WRITE_ONCE().
+	 */
+	unsigned long lru_stamp;
+
 	u64 size;
 
 	struct xe_pt *pt_root[XE_MAX_TILES_PER_DEVICE];
-- 
2.34.1

✗ CI.checkpatch: warning for Add a minimum age gate to the shrinker + LRU fixes

[Patchwork]

== Series Details ==

Series: Add a minimum age gate to the shrinker + LRU fixes
URL   : https://patchwork.freedesktop.org/series/176975/
State : warning

== Summary ==

+ KERNEL=/kernel
+ git clone https://gitlab.freedesktop.org/drm/maintainer-tools mt
Cloning into 'mt'...
warning: redirecting to https://gitlab.freedesktop.org/drm/maintainer-tools.git/
+ git -C mt rev-list -n1 origin/master
3009867cd31863246e4afc138520a7a8bd37bb11
+ cd /kernel
+ git config --global --add safe.directory /kernel
+ git log -n1
commit f78b1513f48082ad8dc1c7a46ff68a51c72d8473
Author: Matthew Brost <matthew.brost@intel.com>
Date:   Tue Oct 27 11:36:38 2026 -0700

    drm/xe: Add a minimum age gate to the shrinker
    
    Backing up a BO costs a GPU blit plus a page copy, and if the BO is part
    of an active working set it is faulted straight back in on the next exec.
    Evicting a BO that was used moments ago is therefore almost always a
    loss, yet the shrinker has no notion of recency beyond LRU order.
    
    Timestamp BOs wherever their LRU position is updated, and read the stamp
    back from whatever owns that position. A BO in a VM's bulk move is
    anchored inside its range, so it is aged by the VM-wide stamp taken in
    xe_vm_bump_lru() alongside the bulk move itself. Any other BO carries
    its own, stamped there too and on a successful move into TT in
    xe_bo_move(), where ttm_resource_alloc() has already placed the new
    resource at the LRU tail.
    
    Add a per-device shrinker_min_age_ms sysfs attribute, defaulting to
    CONFIG_DRM_XE_SHRINKER_MIN_AGE_MS (0, disabled). While set, the shrinker
    won't back up BOs younger than the delay. Each LRU priority is roughly
    age ordered, so the gate is applied per priority: xe_shrinker_count()
    snoops the first shrinkable BO of each priority and reports nothing only
    if all of them are young, and the backup walk skips to the next priority
    at the first young BO, both as a backstop for the count sample and so a
    hot priority doesn't shield colder BOs in another one.
    
    Only BOs this shrinker would actually back up have a say. BOs in
    XE_PL_SYSTEM, which are not GPU mapped and hence not part of a working
    set, and BOs xe_bo_shrink() rejects outright are never young. Purgeable
    BOs and those advised DONTNEED are neither: they are dropped rather than
    backed up and accounted to the purgeable shrinker, so they are never
    young and the count snoop skips their XE_BO_PRIORITY_LOW band entirely.
    
    The gate is hard and applies at every reclaim priority, so a working set
    kept younger than the delay is never backed up and can lead to OOM. This
    is deliberate, and why it is off by default.
    
    Assisted-by: LLM
    Signed-off-by: Matthew Brost <matthew.brost@intel.com>
+ /mt/dim checkpatch 5e5b89446954adf0d9b74ba4228862403f71eec4 drm-intel
ba8df199ac94 drm/xe: Bump the LRU of every BO an exec touches
-:19: WARNING:BAD_SIGN_OFF: Non-standard signature: Co-authored-by:
#19: 
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

total: 0 errors, 1 warnings, 0 checks, 31 lines checked
74519a825917 drm/xe: Bump the LRU on a successful page fault
f3969a767705 drm/xe: Leave the VM bulk move on BO release, not on GEM close
4d0c8cdaaeaa drm/ttm: Allow skipping to the next priority in the bo LRU cursor
f78b1513f480 drm/xe: Add a minimum age gate to the shrinker

Re: [PATCH v4 3/5] drm/xe: Leave the VM bulk move on BO release, not on GEM close

[sashiko-bot]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [High] drm/xe: bulk_move detach in xe_ttm_bo_release_notify() violates TTM invariants
--

commit 5d85520c1e2e507cc032c4248bd48c50a734e710
Author: Matthew Brost <matthew.brost@intel.com>

drm/xe: Leave the VM bulk move on BO release, not on GEM close

This commit removes xe_gem_object_close() and moves the detachment of a
VM-private BO from its VM's LRU bulk move into xe_ttm_bo_release_notify()
so that BOs outliving their GEM handles remain part of the VM bulk move
until release.

> diff --git a/drivers/gpu/drm/xe/xe_bo.c b/drivers/gpu/drm/xe/xe_bo.c
> index 42a6aa1a2e4a0..ba16b260dbcb7 100644
> --- a/drivers/gpu/drm/xe/xe_bo.c
> +++ b/drivers/gpu/drm/xe/xe_bo.c
> @@ -1869,6 +1869,14 @@ static void xe_ttm_bo_release_notify(struct ttm_buffer_object *ttm_bo)
>  	if (!xe_ttm_bo_lock_in_destructor(ttm_bo))
>  		return;
>  
> +	/*
> +	 * Leave the VM's bulk move, which is only kept alive by the VM, and
> +	 * whose LRU position no longer says anything about a BO on its way
> +	 * out.
> +	 */
> +	if (ttm_bo->bulk_move)
> +		ttm_bo_set_bulk_move(ttm_bo, NULL);

[Severity: High]
Does detaching from bulk_move in xe_ttm_bo_release_notify() happen too late
in the TTM teardown sequence?

When the final reference to a VM-private user BO on a non-fault-mode VM is
dropped via xe_gem_object_free() -> ttm_bo_fini() -> ttm_bo_put() ->
ttm_bo_release(), ttm_bo_release() checks bo->bulk_move and individualizes
bo->base.resv before invoking bdev->funcs->release_notify():

drivers/gpu/drm/ttm/ttm_bo.c:ttm_bo_release() {
    ...
	WARN_ON_ONCE(bo->pin_count);
	WARN_ON_ONCE(bo->bulk_move);

	if (!bo->deleted) {
		ret = ttm_bo_individualize_resv(bo);
		if (ret) {
			/* Last resort, if we fail to allocate memory for the
			 * fences block for the BO to become idle
			 */
			dma_resv_wait_timeout(bo->base.resv,
					      DMA_RESV_USAGE_BOOKKEEP, false,
					      30 * HZ);
		}

		if (bdev->funcs->release_notify)
			bdev->funcs->release_notify(bo);
    ...
}

Because bo->bulk_move is still set upon entering ttm_bo_release(), releasing
such a BO will immediately trigger WARN_ON_ONCE(bo->bulk_move).

In addition, ttm_bo_individualize_resv() switches bo->base.resv away from
xe_vm_resv(vm) to &bo->base._resv under bo->bdev->lru_lock and then drops
lru_lock while the BO's resource is still linked in vm->lru_bulk_move:

drivers/gpu/drm/ttm/ttm_bo.c:ttm_bo_individualize_resv() {
    ...
		spin_lock(&bo->bdev->lru_lock);
		bo->base.resv = &bo->base._resv;
		spin_unlock(&bo->bdev->lru_lock);
    ...
}

In the window between ttm_bo_individualize_resv() and
ttm_bo_set_bulk_move(ttm_bo, NULL) in xe_ttm_bo_release_notify(), another
thread holding xe_vm_resv(vm) can access vm->lru_bulk_move while the dying
BO is still pos->first or pos->last:

- Calling xe_vm_bump_lru() -> ttm_lru_bulk_move_tail() will fail
  dma_resv_assert_held(pos->first->bo->base.resv) or
  dma_resv_assert_held(pos->last->bo->base.resv) in
  drivers/gpu/drm/ttm/ttm_resource.c:

drivers/gpu/drm/ttm/ttm_resource.c:ttm_lru_bulk_move_tail() {
    ...
			lockdep_assert_held(&pos->first->bo->bdev->lru_lock);
			dma_resv_assert_held(pos->first->bo->base.resv);
			dma_resv_assert_held(pos->last->bo->base.resv);
    ...
}

- Adding or moving another resource in the same bucket via
  ttm_resource_add_bulk_move() -> ttm_lru_bulk_move_add() will trigger
  WARN_ON(pos->first->bo->base.resv != res->bo->base.resv):

drivers/gpu/drm/ttm/ttm_resource.c:ttm_lru_bulk_move_add() {
    ...
	} else {
		WARN_ON(pos->first->bo->base.resv != res->bo->base.resv);
		ttm_lru_bulk_move_pos_tail(pos, res);
	}
}

Conversely, if ttm_bo_individualize_resv() fails (for example, if
dma_resv_copy_fences() returns -ENOMEM), bo->base.resv remains
xe_vm_resv(vm), while xe_ttm_bo_lock_in_destructor() only trylocks
&ttm_bo->base._resv. Calling ttm_bo_set_bulk_move(ttm_bo, NULL) in
xe_ttm_bo_release_notify() will then fail dma_resv_assert_held(bo->base.resv)
in ttm_bo_set_bulk_move() and ttm_resource_is_swapped(), and mutate
vm->lru_bulk_move without holding xe_vm_resv(vm).

Should bulk_move be cleared before ttm_bo_release() runs (or before
ttm_bo_individualize_resv() detaches bo->base.resv from the VM reservation
object) while holding the VM reservation lock?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20261027183638.1031299-1-matthew.brost@intel.com?part=3

Re: [PATCH v4 2/5] drm/xe: Bump the LRU on a successful page fault

[sashiko-bot]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] drm/xe: xe_vm_bump_lru() in xe_pagefault_handle_vma() misses private BOs
--

commit a6d00986d9f12583f84415e17fe0f940a218b2e7
Author: Matthew Brost <matthew.brost@intel.com>

drm/xe: Bump the LRU on a successful page fault

This commit calls xe_vm_bump_lru() in xe_pagefault_handle_vma() after a
successful VMA rebind so that faulted BOs are moved to the tail of the TTM
LRU.

> diff --git a/drivers/gpu/drm/xe/xe_pagefault.c b/drivers/gpu/drm/xe/xe_pagefault.c
> index 5247321c805ec..3719258aa6f09 100644
> --- a/drivers/gpu/drm/xe/xe_pagefault.c
> +++ b/drivers/gpu/drm/xe/xe_pagefault.c
> @@ -221,6 +221,8 @@ static int xe_pagefault_handle_vma(struct xe_gt *gt, struct xe_vma *vma,
>  				xe_pagefault_set_error(pf, XE_PAGEFAULT_ERROR_VMA_REBIND);
>  				break;
>  			}
> +
> +			xe_vm_bump_lru(vm, &exec);

[Severity: Medium]
Does xe_vm_bump_lru() actually move VM-private BOs to the LRU tail when
handling a page fault?

When a VM-private BO is created for a fault-mode VM in
__xe_bo_create_locked(), it is not added to vm->lru_bulk_move:

drivers/gpu/drm/xe/xe_bo.c:__xe_bo_create_locked() {
	...
	bo = xe_bo_init_locked(xe, bo, tile, vm ? xe_vm_resv(vm) : NULL,
			       vm && !xe_vm_in_fault_mode(vm) &&
			       flags & XE_BO_FLAG_USER ?
			       &vm->lru_bulk_move : NULL, size,
			       cpu_caching, type, flags, NULL, exec);
	...
}

Because page faults only occur on fault-mode VMs, vm->lru_bulk_move is empty
and ttm_lru_bulk_move_tail(&vm->lru_bulk_move) in xe_vm_bump_lru() is a
no-op.

Furthermore, during xe_pagefault_begin() -> xe_vm_lock_vma(), only external
BOs (!bo->vm) are locked individually in exec, whereas VM-private BOs share
the VM's reservation object and only lock xe_vm_obj(vm):

drivers/gpu/drm/xe/xe_vm.c:xe_vm_lock_vma() {
	...
	err = drm_exec_lock_obj(exec, xe_vm_obj(vm));
	if (!err && bo && !bo->vm)
		err = drm_exec_lock_obj(exec, &bo->ttm.base);
	...
}

When xe_vm_bump_lru() then iterates over the locked objects in exec, it skips
xe_vm_obj(vm):

drivers/gpu/drm/xe/xe_vm.c:xe_vm_bump_lru() {
	...
	drm_exec_for_each_locked_object(exec, obj) {
		struct xe_bo *bo;

		/* The VM's dummy resv object, not a BO. */
		if (obj == xe_vm_obj(vm))
			continue;
	...
}

If a faulted VM-private BO's backing store is already in a compatible
placement so that ttm_bo_validate() returns early without moving the resource,
will the BO's TTM LRU position remain completely unchanged here?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20261027183638.1031299-1-matthew.brost@intel.com?part=2

✓ CI.KUnit: success for Add a minimum age gate to the shrinker + LRU fixes

[Patchwork]

== Series Details ==

Series: Add a minimum age gate to the shrinker + LRU fixes
URL   : https://patchwork.freedesktop.org/series/176975/
State : success

== Summary ==

+ trap cleanup EXIT
+ kunitconfigs=('/kernel/drivers/gpu/tests/.kunitconfig' '/kernel/drivers/gpu/drm/xe/.kunitconfig' '/kernel/drivers/gpu/drm/tests/.kunitconfig' '/kernel/drivers/gpu/drm/ttm/tests/.kunitconfig' '/kernel/drivers/dma-buf/.kunitconfig')
+ for kcfg in "${kunitconfigs[@]}"
+ [[ ! -f /kernel/drivers/gpu/tests/.kunitconfig ]]
+ /kernel/tools/testing/kunit/kunit.py run --kunitconfig /kernel/drivers/gpu/tests/.kunitconfig
[18:47:42] Configuring KUnit Kernel ...
Generating .config ...
Populating config with:
$ make ARCH=um O=.kunit olddefconfig
[18:47:46] Building KUnit Kernel ...
Populating config with:
$ make ARCH=um O=.kunit olddefconfig
Building with:
$ make all compile_commands.json scripts_gdb ARCH=um O=.kunit --jobs=48
[18:48:06] Starting KUnit Kernel (1/1)...
[18:48:06] ============================================================
Running tests with:
$ .kunit/linux kunit.enable=1 mem=1G console=tty kunit_shutdown=halt
[18:48:06] ================= gpu_buddy (14 subtests) ==================
[18:48:06] [PASSED] gpu_test_buddy_alloc_limit
[18:48:06] [PASSED] gpu_test_buddy_alloc_optimistic
[18:48:06] [PASSED] gpu_test_buddy_alloc_pessimistic
[18:48:06] [PASSED] gpu_test_buddy_alloc_pathological
[18:48:06] [PASSED] gpu_test_buddy_alloc_contiguous
[18:48:06] [PASSED] gpu_test_buddy_alloc_clear
[18:48:06] [PASSED] gpu_test_buddy_alloc_range
[18:48:07] [PASSED] gpu_test_buddy_alloc_range_bias
[18:48:08] [PASSED] gpu_test_buddy_fragmentation_performance
[18:48:08] [PASSED] gpu_test_buddy_dirty_tracker_performance
[18:48:08] [PASSED] gpu_test_buddy_alloc_exceeds_max_order
[18:48:08] [PASSED] gpu_test_buddy_offset_aligned_allocation
[18:48:08] [PASSED] gpu_test_buddy_subtree_offset_alignment_stress
[18:48:08] [PASSED] gpu_test_buddy_addr_to_block
[18:48:08] ==================== [PASSED] gpu_buddy ====================
[18:48:08] ============================================================
[18:48:08] Testing complete. Ran 14 tests: passed: 14
[18:48:08] Elapsed time: 26.573s total, 4.412s configuring, 20.243s building, 1.859s running

+ for kcfg in "${kunitconfigs[@]}"
+ [[ ! -f /kernel/drivers/gpu/drm/xe/.kunitconfig ]]
+ /kernel/tools/testing/kunit/kunit.py run --kunitconfig /kernel/drivers/gpu/drm/xe/.kunitconfig
[18:48:08] Configuring KUnit Kernel ...
Regenerating .config ...
Populating config with:
$ make ARCH=um O=.kunit olddefconfig
[18:48:10] Building KUnit Kernel ...
Populating config with:
$ make ARCH=um O=.kunit olddefconfig
Building with:
$ make all compile_commands.json scripts_gdb ARCH=um O=.kunit --jobs=48
[18:48:49] Starting KUnit Kernel (1/1)...
[18:48:49] ============================================================
Running tests with:
$ .kunit/linux kunit.enable=1 mem=1G console=tty kunit_shutdown=halt
[18:48:49] ================== guc_buf (11 subtests) ===================
[18:48:49] [PASSED] test_smallest
[18:48:49] [PASSED] test_largest
[18:48:49] [PASSED] test_granular
[18:48:49] [PASSED] test_unique
[18:48:49] [PASSED] test_overlap
[18:48:49] [PASSED] test_reusable
[18:48:49] [PASSED] test_too_big
[18:48:49] [PASSED] test_flush
[18:48:49] [PASSED] test_lookup
[18:48:49] [PASSED] test_data
[18:48:49] [PASSED] test_class
[18:48:49] ===================== [PASSED] guc_buf =====================
[18:48:49] =================== guc_dbm (7 subtests) ===================
[18:48:49] [PASSED] test_empty
[18:48:49] [PASSED] test_default
[18:48:49] ======================== test_size  ========================
[18:48:49] [PASSED] 4
[18:48:49] [PASSED] 8
[18:48:49] [PASSED] 32
[18:48:49] [PASSED] 256
[18:48:49] ==================== [PASSED] test_size ====================
[18:48:49] ======================= test_reuse  ========================
[18:48:49] [PASSED] 4
[18:48:49] [PASSED] 8
[18:48:49] [PASSED] 32
[18:48:49] [PASSED] 256
[18:48:49] =================== [PASSED] test_reuse ====================
[18:48:49] =================== test_range_overlap  ====================
[18:48:49] [PASSED] 4
[18:48:49] [PASSED] 8
[18:48:49] [PASSED] 32
[18:48:49] [PASSED] 256
[18:48:49] =============== [PASSED] test_range_overlap ================
[18:48:49] =================== test_range_compact  ====================
[18:48:49] [PASSED] 4
[18:48:49] [PASSED] 8
[18:48:49] [PASSED] 32
[18:48:49] [PASSED] 256
[18:48:49] =============== [PASSED] test_range_compact ================
[18:48:49] ==================== test_range_spare  =====================
[18:48:49] [PASSED] 4
[18:48:49] [PASSED] 8
[18:48:49] [PASSED] 32
[18:48:49] [PASSED] 256
[18:48:49] ================ [PASSED] test_range_spare =================
[18:48:49] ===================== [PASSED] guc_dbm =====================
[18:48:49] ================== guc_idm (22 subtests) ===================
[18:48:49] [PASSED] bad_init
[18:48:49] [PASSED] no_init
[18:48:49] [PASSED] init_fini
[18:48:49] [PASSED] check_init_small
[18:48:49] [PASSED] check_init_shared
[18:48:49] [PASSED] check_used
[18:48:49] [PASSED] check_quota
[18:48:49] [PASSED] check_all
[18:48:49] [PASSED] check_limits
[18:48:49] [PASSED] check_overlap
[18:48:49] [PASSED] check_overlap_alloc
[18:48:49] [PASSED] check_overlap_release
[18:48:49] [PASSED] check_overlap_partial_init
[18:48:49] [PASSED] check_overlap_partial_shared_range_blocking
[18:48:49] [PASSED] check_used_usable
[18:48:49] [PASSED] check_used_shareable
[18:48:49] [PASSED] check_used_mixed
[18:48:49] [PASSED] check_release_usable_range
[18:48:49] [PASSED] check_release_shareable_range
[18:48:49] [PASSED] check_split_boundaries
[18:48:49] [PASSED] check_shareable_spare
[18:48:49] [PASSED] check_shareable_spare_fragmented_tail
[18:48:49] ===================== [PASSED] guc_idm =====================
[18:48:49] ============== guc_klv_helpers (13 subtests) ===============
[18:48:49] [PASSED] test_count
[18:48:49] [PASSED] test_encode_u32
[18:48:49] [PASSED] test_encode_u64
[18:48:49] [PASSED] test_encode_string
[18:48:49] [PASSED] test_encode_object_raw
[18:48:49] [PASSED] test_encode_object_klv
[18:48:49] [PASSED] test_encode_object_nested
[18:48:49] [PASSED] test_encode_object_basic
[18:48:49] ===================== test_decode_u16  =====================
[18:48:49] [PASSED] empty
[18:48:49] [PASSED] valid
[18:48:49] [PASSED] max
[18:48:49] [PASSED] big
[18:48:49] [PASSED] long
[18:48:49] ================= [PASSED] test_decode_u16 =================
[18:48:49] ===================== test_decode_u32  =====================
[18:48:49] [PASSED] empty
[18:48:49] [PASSED] valid
[18:48:49] [PASSED] max
[18:48:49] [PASSED] long
[18:48:49] ================= [PASSED] test_decode_u32 =================
[18:48:49] ===================== test_decode_u64  =====================
[18:48:49] [PASSED] empty
[18:48:49] [PASSED] short
[18:48:49] [PASSED] valid
[18:48:49] [PASSED] max
[18:48:49] [PASSED] long
[18:48:49] ================= [PASSED] test_decode_u64 =================
[18:48:49] ===================== test_decode_str  =====================
[18:48:49] [PASSED] empty
[18:48:49] [PASSED] one
[18:48:49] [PASSED] one_trash
[18:48:49] [PASSED] two
[18:48:49] [PASSED] two_trash
[18:48:49] [PASSED] three
[18:48:49] [PASSED] dword
[18:48:49] [PASSED] four
[18:48:49] [PASSED] five
[18:48:49] [PASSED] six
[18:48:49] [PASSED] seven
[18:48:49] [PASSED] qword
[18:48:49] ================= [PASSED] test_decode_str =================
[18:48:49] [PASSED] test_print
[18:48:49] ================= [PASSED] guc_klv_helpers =================
[18:48:49] =================== xe_log (4 subtests) ====================
[18:48:49] [PASSED] demo_cper
[18:48:49] [PASSED] demo_dmesg
[18:48:49] ======================= test_dmesg  ========================
[18:48:49] [PASSED] test_fatal
[18:48:49] [PASSED] test_fatal_tile
[18:48:49] [PASSED] test_fatal_gt
[18:48:49] [PASSED] test_fatal_comp
[18:48:49] [PASSED] test_fatal_comp_tile
[18:48:49] [PASSED] test_fatal_comp_gt
[18:48:49] [PASSED] test_fatal_all
[18:48:49] [PASSED] test_recoverable
[18:48:49] [PASSED] test_recoverable_tile
[18:48:49] [PASSED] test_recoverable_gt
[18:48:49] [PASSED] test_recoverable_comp
[18:48:49] [PASSED] test_recoverable_comp_tile
[18:48:49] [PASSED] test_recoverable_comp_gt
[18:48:49] [PASSED] test_recoverable_all
[18:48:49] [PASSED] test_info
[18:48:49] [PASSED] test_info_tile
[18:48:49] [PASSED] test_info_gt
[18:48:49] [PASSED] test_info_err
[18:48:49] [PASSED] test_info_comp
[18:48:49] [PASSED] test_info_comp_tile
[18:48:49] [PASSED] test_info_comp_gt
[18:48:49] [PASSED] test_info_all
[18:48:49] [PASSED] test_hw_fatal
[18:48:49] [PASSED] test_hw_recoverable
[18:48:49] [PASSED] test_hw_corrected
[18:48:49] [PASSED] test_hw_informational
[18:48:49] =================== [PASSED] test_dmesg ====================
[18:48:49] ====================== test_invalid  =======================
[18:48:49] [SKIPPED] no-component no-location no-warn (requires CONFIG_DRM_XE_DEBUG)
[18:48:49] [SKIPPED] reserved location (requires CONFIG_DRM_XE_DEBUG)
[18:48:49] [SKIPPED] unknown location (requires CONFIG_DRM_XE_DEBUG)
[18:48:49] [SKIPPED] nonzero-device-id location (requires CONFIG_DRM_XE_DEBUG)
[18:48:49] [SKIPPED] invalid-tile-id location (requires CONFIG_DRM_XE_DEBUG)
[18:48:49] [SKIPPED] invalid-gt-id location (requires CONFIG_DRM_XE_DEBUG)
[18:48:49] [SKIPPED] unknown component class (requires CONFIG_DRM_XE_DEBUG)
[18:48:49] [SKIPPED] unknown system component (requires CONFIG_DRM_XE_DEBUG)
[18:48:49] [SKIPPED] unknown hardware component (requires CONFIG_DRM_XE_DEBUG)
[18:48:49] [SKIPPED] unknown component and location (requires CONFIG_DRM_XE_DEBUG)
[18:48:49] ================== [SKIPPED] test_invalid ==================
[18:48:49] ===================== [PASSED] xe_log ======================
[18:48:49] ================== no_relay (3 subtests) ===================
[18:48:49] [PASSED] xe_drops_guc2pf_if_not_ready
[18:48:49] [PASSED] xe_drops_guc2vf_if_not_ready
[18:48:49] [PASSED] xe_rejects_send_if_not_ready
[18:48:49] ==================== [PASSED] no_relay =====================
[18:48:49] ================== pf_relay (14 subtests) ==================
[18:48:49] [PASSED] pf_rejects_guc2pf_too_short
[18:48:49] [PASSED] pf_rejects_guc2pf_too_long
[18:48:49] [PASSED] pf_rejects_guc2pf_no_payload
[18:48:49] [PASSED] pf_fails_no_payload
[18:48:49] [PASSED] pf_fails_bad_origin
[18:48:49] [PASSED] pf_fails_bad_type
[18:48:49] [PASSED] pf_txn_reports_error
[18:48:49] [PASSED] pf_txn_sends_pf2guc
[18:48:49] [PASSED] pf_sends_pf2guc
[18:48:49] [SKIPPED] pf_loopback_nop (requires CONFIG_DRM_XE_DEBUG_SRIOV)
[18:48:49] [SKIPPED] pf_loopback_echo (requires CONFIG_DRM_XE_DEBUG_SRIOV)
[18:48:49] [SKIPPED] pf_loopback_fail (requires CONFIG_DRM_XE_DEBUG_SRIOV)
[18:48:49] [SKIPPED] pf_loopback_busy (requires CONFIG_DRM_XE_DEBUG_SRIOV)
[18:48:49] [SKIPPED] pf_loopback_retry (requires CONFIG_DRM_XE_DEBUG_SRIOV)
[18:48:49] ==================== [PASSED] pf_relay =====================
[18:48:49] ================== vf_relay (3 subtests) ===================
[18:48:49] [PASSED] vf_rejects_guc2vf_too_short
[18:48:49] [PASSED] vf_rejects_guc2vf_too_long
[18:48:49] [PASSED] vf_rejects_guc2vf_no_payload
[18:48:49] ==================== [PASSED] vf_relay =====================
[18:48:49] ================ pf_gt_config (9 subtests) =================
[18:48:49] [PASSED] fair_contexts_1vf
[18:48:49] [PASSED] fair_doorbells_1vf
[18:48:49] [PASSED] fair_ggtt_1vf
[18:48:49] ====================== fair_vram_1vf  ======================
[18:48:49] [PASSED] 3.50 GiB
[18:48:49] [PASSED] 11.5 GiB
[18:48:49] [PASSED] 15.5 GiB
[18:48:49] [PASSED] 31.5 GiB
[18:48:49] [PASSED] 63.5 GiB
[18:48:49] [PASSED] 1.91 GiB
[18:48:49] ================== [PASSED] fair_vram_1vf ==================
[18:48:49] ================ fair_vram_1vf_admin_only  =================
[18:48:49] [PASSED] 3.50 GiB
[18:48:49] [PASSED] 11.5 GiB
[18:48:49] [PASSED] 15.5 GiB
[18:48:49] [PASSED] 31.5 GiB
[18:48:49] [PASSED] 63.5 GiB
[18:48:49] [PASSED] 1.91 GiB
[18:48:49] ============ [PASSED] fair_vram_1vf_admin_only =============
[18:48:49] ====================== fair_contexts  ======================
[18:48:49] [PASSED] 1 VF
[18:48:49] [PASSED] 2 VFs
[18:48:49] [PASSED] 3 VFs
[18:48:49] [PASSED] 4 VFs
[18:48:49] [PASSED] 5 VFs
[18:48:49] [PASSED] 6 VFs
[18:48:49] [PASSED] 7 VFs
[18:48:49] [PASSED] 8 VFs
[18:48:49] [PASSED] 9 VFs
[18:48:49] [PASSED] 10 VFs
[18:48:49] [PASSED] 11 VFs
[18:48:49] [PASSED] 12 VFs
[18:48:49] [PASSED] 13 VFs
[18:48:49] [PASSED] 14 VFs
[18:48:49] [PASSED] 15 VFs
[18:48:49] [PASSED] 16 VFs
[18:48:49] [PASSED] 17 VFs
[18:48:49] [PASSED] 18 VFs
[18:48:49] [PASSED] 19 VFs
[18:48:49] [PASSED] 20 VFs
[18:48:49] [PASSED] 21 VFs
[18:48:49] [PASSED] 22 VFs
[18:48:49] [PASSED] 23 VFs
[18:48:49] [PASSED] 24 VFs
[18:48:49] [PASSED] 25 VFs
[18:48:49] [PASSED] 26 VFs
[18:48:49] [PASSED] 27 VFs
[18:48:49] [PASSED] 28 VFs
[18:48:49] [PASSED] 29 VFs
[18:48:49] [PASSED] 30 VFs
[18:48:49] [PASSED] 31 VFs
[18:48:49] [PASSED] 32 VFs
[18:48:49] [PASSED] 33 VFs
[18:48:49] [PASSED] 34 VFs
[18:48:49] [PASSED] 35 VFs
[18:48:49] [PASSED] 36 VFs
[18:48:49] [PASSED] 37 VFs
[18:48:49] [PASSED] 38 VFs
[18:48:49] [PASSED] 39 VFs
[18:48:49] [PASSED] 40 VFs
[18:48:49] [PASSED] 41 VFs
[18:48:49] [PASSED] 42 VFs
[18:48:49] [PASSED] 43 VFs
[18:48:49] [PASSED] 44 VFs
[18:48:49] [PASSED] 45 VFs
[18:48:49] [PASSED] 46 VFs
[18:48:49] [PASSED] 47 VFs
[18:48:49] [PASSED] 48 VFs
[18:48:49] [PASSED] 49 VFs
[18:48:49] [PASSED] 50 VFs
[18:48:49] [PASSED] 51 VFs
[18:48:49] [PASSED] 52 VFs
[18:48:49] [PASSED] 53 VFs
[18:48:49] [PASSED] 54 VFs
[18:48:49] [PASSED] 55 VFs
[18:48:49] [PASSED] 56 VFs
[18:48:49] [PASSED] 57 VFs
[18:48:49] [PASSED] 58 VFs
[18:48:49] [PASSED] 59 VFs
[18:48:49] [PASSED] 60 VFs
[18:48:49] [PASSED] 61 VFs
[18:48:49] [PASSED] 62 VFs
[18:48:49] [PASSED] 63 VFs
[18:48:49] ================== [PASSED] fair_contexts ==================
[18:48:49] ===================== fair_doorbells  ======================
[18:48:49] [PASSED] 1 VF
[18:48:49] [PASSED] 2 VFs
[18:48:49] [PASSED] 3 VFs
[18:48:49] [PASSED] 4 VFs
[18:48:49] [PASSED] 5 VFs
[18:48:49] [PASSED] 6 VFs
[18:48:49] [PASSED] 7 VFs
[18:48:49] [PASSED] 8 VFs
[18:48:49] [PASSED] 9 VFs
[18:48:49] [PASSED] 10 VFs
[18:48:49] [PASSED] 11 VFs
[18:48:49] [PASSED] 12 VFs
[18:48:49] [PASSED] 13 VFs
[18:48:49] [PASSED] 14 VFs
[18:48:49] [PASSED] 15 VFs
[18:48:49] [PASSED] 16 VFs
[18:48:49] [PASSED] 17 VFs
[18:48:49] [PASSED] 18 VFs
[18:48:49] [PASSED] 19 VFs
[18:48:49] [PASSED] 20 VFs
[18:48:49] [PASSED] 21 VFs
[18:48:49] [PASSED] 22 VFs
[18:48:49] [PASSED] 23 VFs
[18:48:49] [PASSED] 24 VFs
[18:48:49] [PASSED] 25 VFs
[18:48:49] [PASSED] 26 VFs
[18:48:49] [PASSED] 27 VFs
[18:48:49] [PASSED] 28 VFs
[18:48:49] [PASSED] 29 VFs
[18:48:49] [PASSED] 30 VFs
[18:48:49] [PASSED] 31 VFs
[18:48:49] [PASSED] 32 VFs
[18:48:49] [PASSED] 33 VFs
[18:48:49] [PASSED] 34 VFs
[18:48:49] [PASSED] 35 VFs
[18:48:49] [PASSED] 36 VFs
[18:48:49] [PASSED] 37 VFs
[18:48:49] [PASSED] 38 VFs
[18:48:49] [PASSED] 39 VFs
[18:48:49] [PASSED] 40 VFs
[18:48:49] [PASSED] 41 VFs
[18:48:49] [PASSED] 42 VFs
[18:48:49] [PASSED] 43 VFs
[18:48:49] [PASSED] 44 VFs
[18:48:49] [PASSED] 45 VFs
[18:48:49] [PASSED] 46 VFs
[18:48:49] [PASSED] 47 VFs
[18:48:49] [PASSED] 48 VFs
[18:48:49] [PASSED] 49 VFs
[18:48:49] [PASSED] 50 VFs
[18:48:49] [PASSED] 51 VFs
[18:48:49] [PASSED] 52 VFs
[18:48:49] [PASSED] 53 VFs
[18:48:49] [PASSED] 54 VFs
[18:48:49] [PASSED] 55 VFs
[18:48:49] [PASSED] 56 VFs
[18:48:49] [PASSED] 57 VFs
[18:48:49] [PASSED] 58 VFs
[18:48:49] [PASSED] 59 VFs
[18:48:49] [PASSED] 60 VFs
[18:48:49] [PASSED] 61 VFs
[18:48:49] [PASSED] 62 VFs
[18:48:49] [PASSED] 63 VFs
[18:48:49] ================= [PASSED] fair_doorbells ==================
[18:48:49] ======================== fair_ggtt  ========================
[18:48:49] [PASSED] 1 VF
[18:48:49] [PASSED] 2 VFs
[18:48:49] [PASSED] 3 VFs
[18:48:49] [PASSED] 4 VFs
[18:48:49] [PASSED] 5 VFs
[18:48:49] [PASSED] 6 VFs
[18:48:49] [PASSED] 7 VFs
[18:48:49] [PASSED] 8 VFs
[18:48:49] [PASSED] 9 VFs
[18:48:49] [PASSED] 10 VFs
[18:48:49] [PASSED] 11 VFs
[18:48:49] [PASSED] 12 VFs
[18:48:49] [PASSED] 13 VFs
[18:48:49] [PASSED] 14 VFs
[18:48:49] [PASSED] 15 VFs
[18:48:49] [PASSED] 16 VFs
[18:48:49] [PASSED] 17 VFs
[18:48:49] [PASSED] 18 VFs
[18:48:49] [PASSED] 19 VFs
[18:48:49] [PASSED] 20 VFs
[18:48:49] [PASSED] 21 VFs
[18:48:49] [PASSED] 22 VFs
[18:48:49] [PASSED] 23 VFs
[18:48:49] [PASSED] 24 VFs
[18:48:49] [PASSED] 25 VFs
[18:48:49] [PASSED] 26 VFs
[18:48:49] [PASSED] 27 VFs
[18:48:49] [PASSED] 28 VFs
[18:48:49] [PASSED] 29 VFs
[18:48:49] [PASSED] 30 VFs
[18:48:49] [PASSED] 31 VFs
[18:48:49] [PASSED] 32 VFs
[18:48:49] [PASSED] 33 VFs
[18:48:49] [PASSED] 34 VFs
[18:48:49] [PASSED] 35 VFs
[18:48:49] [PASSED] 36 VFs
[18:48:49] [PASSED] 37 VFs
[18:48:49] [PASSED] 38 VFs
[18:48:49] [PASSED] 39 VFs
[18:48:49] [PASSED] 40 VFs
[18:48:49] [PASSED] 41 VFs
[18:48:49] [PASSED] 42 VFs
[18:48:49] [PASSED] 43 VFs
[18:48:49] [PASSED] 44 VFs
[18:48:49] [PASSED] 45 VFs
[18:48:49] [PASSED] 46 VFs
[18:48:49] [PASSED] 47 VFs
[18:48:49] [PASSED] 48 VFs
[18:48:49] [PASSED] 49 VFs
[18:48:49] [PASSED] 50 VFs
[18:48:49] [PASSED] 51 VFs
[18:48:49] [PASSED] 52 VFs
[18:48:49] [PASSED] 53 VFs
[18:48:49] [PASSED] 54 VFs
[18:48:49] [PASSED] 55 VFs
[18:48:49] [PASSED] 56 VFs
[18:48:49] [PASSED] 57 VFs
[18:48:49] [PASSED] 58 VFs
[18:48:49] [PASSED] 59 VFs
[18:48:49] [PASSED] 60 VFs
[18:48:49] [PASSED] 61 VFs
[18:48:49] [PASSED] 62 VFs
[18:48:49] [PASSED] 63 VFs
[18:48:49] ==================== [PASSED] fair_ggtt ====================
[18:48:49] ======================== fair_vram  ========================
[18:48:49] [PASSED] 1 VF
[18:48:49] [PASSED] 2 VFs
[18:48:49] [PASSED] 3 VFs
[18:48:49] [PASSED] 4 VFs
[18:48:49] [PASSED] 5 VFs
[18:48:49] [PASSED] 6 VFs
[18:48:49] [PASSED] 7 VFs
[18:48:49] [PASSED] 8 VFs
[18:48:49] [PASSED] 9 VFs
[18:48:49] [PASSED] 10 VFs
[18:48:49] [PASSED] 11 VFs
[18:48:49] [PASSED] 12 VFs
[18:48:49] [PASSED] 13 VFs
[18:48:49] [PASSED] 14 VFs
[18:48:49] [PASSED] 15 VFs
[18:48:49] [PASSED] 16 VFs
[18:48:49] [PASSED] 17 VFs
[18:48:49] [PASSED] 18 VFs
[18:48:49] [PASSED] 19 VFs
[18:48:49] [PASSED] 20 VFs
[18:48:49] [PASSED] 21 VFs
[18:48:49] [PASSED] 22 VFs
[18:48:49] [PASSED] 23 VFs
[18:48:49] [PASSED] 24 VFs
[18:48:49] [PASSED] 25 VFs
[18:48:49] [PASSED] 26 VFs
[18:48:49] [PASSED] 27 VFs
[18:48:49] [PASSED] 28 VFs
[18:48:49] [PASSED] 29 VFs
[18:48:49] [PASSED] 30 VFs
[18:48:49] [PASSED] 31 VFs
[18:48:49] [PASSED] 32 VFs
[18:48:49] [PASSED] 33 VFs
[18:48:49] [PASSED] 34 VFs
[18:48:49] [PASSED] 35 VFs
[18:48:49] [PASSED] 36 VFs
[18:48:49] [PASSED] 37 VFs
[18:48:49] [PASSED] 38 VFs
[18:48:49] [PASSED] 39 VFs
[18:48:49] [PASSED] 40 VFs
[18:48:49] [PASSED] 41 VFs
[18:48:49] [PASSED] 42 VFs
[18:48:49] [PASSED] 43 VFs
[18:48:49] [PASSED] 44 VFs
[18:48:49] [PASSED] 45 VFs
[18:48:49] [PASSED] 46 VFs
[18:48:49] [PASSED] 47 VFs
[18:48:49] [PASSED] 48 VFs
[18:48:49] [PASSED] 49 VFs
[18:48:49] [PASSED] 50 VFs
[18:48:49] [PASSED] 51 VFs
[18:48:49] [PASSED] 52 VFs
[18:48:49] [PASSED] 53 VFs
[18:48:49] [PASSED] 54 VFs
[18:48:49] [PASSED] 55 VFs
[18:48:49] [PASSED] 56 VFs
[18:48:49] [PASSED] 57 VFs
[18:48:49] [PASSED] 58 VFs
[18:48:49] [PASSED] 59 VFs
[18:48:49] [PASSED] 60 VFs
[18:48:49] [PASSED] 61 VFs
[18:48:49] [PASSED] 62 VFs
[18:48:49] [PASSED] 63 VFs
[18:48:49] ==================== [PASSED] fair_vram ====================
[18:48:49] ================== [PASSED] pf_gt_config ===================
[18:48:49] ===================== lmtt (1 subtest) =====================
[18:48:49] ======================== test_ops  =========================
[18:48:49] [PASSED] 2-level
[18:48:49] [PASSED] multi-level
[18:48:49] ==================== [PASSED] test_ops =====================
[18:48:49] ====================== [PASSED] lmtt =======================
[18:48:49] ================= sriov_packet (1 subtest) =================
[18:48:49] [PASSED] test_descriptor_init
[18:48:49] ================== [PASSED] sriov_packet ===================
[18:48:49] ================= pf_service (11 subtests) =================
[18:48:49] [PASSED] pf_negotiate_any
[18:48:49] [PASSED] pf_negotiate_base_match
[18:48:49] [PASSED] pf_negotiate_base_newer
[18:48:50] [PASSED] pf_negotiate_base_next
[18:48:50] [SKIPPED] pf_negotiate_base_older (no older minor)
[18:48:50] [PASSED] pf_negotiate_base_prev
[18:48:50] [PASSED] pf_negotiate_latest_match
[18:48:50] [PASSED] pf_negotiate_latest_newer
[18:48:50] [PASSED] pf_negotiate_latest_next
[18:48:50] [SKIPPED] pf_negotiate_latest_older (no older minor)
[18:48:50] [SKIPPED] pf_negotiate_latest_prev (no prev major)
[18:48:50] =================== [PASSED] pf_service ====================
[18:48:50] ================= xe_guc_g2g (2 subtests) ==================
[18:48:50] ============== xe_live_guc_g2g_kunit_default  ==============
[18:48:50] ========= [SKIPPED] xe_live_guc_g2g_kunit_default ==========
[18:48:50] ============== xe_live_guc_g2g_kunit_allmem  ===============
[18:48:50] ========== [SKIPPED] xe_live_guc_g2g_kunit_allmem ==========
[18:48:50] =================== [SKIPPED] xe_guc_g2g ===================
[18:48:50] =================== xe_mocs (2 subtests) ===================
[18:48:50] ================ xe_live_mocs_kernel_kunit  ================
[18:48:50] =========== [SKIPPED] xe_live_mocs_kernel_kunit ============
[18:48:50] ================ xe_live_mocs_reset_kunit  =================
[18:48:50] ============ [SKIPPED] xe_live_mocs_reset_kunit ============
[18:48:50] ==================== [SKIPPED] xe_mocs =====================
[18:48:50] ================= xe_migrate (3 subtests) ==================
[18:48:50] ================= xe_migrate_sanity_kunit  =================
[18:48:50] ============ [SKIPPED] xe_migrate_sanity_kunit =============
[18:48:50] ================== xe_validate_ccs_kunit  ==================
[18:48:50] ============= [SKIPPED] xe_validate_ccs_kunit ==============
[18:48:50] ============= xe_migrate_access_memory_kunit  ==============
[18:48:50] ========= [SKIPPED] xe_migrate_access_memory_kunit =========
[18:48:50] =================== [SKIPPED] xe_migrate ===================
[18:48:50] ================== xe_dma_buf (1 subtest) ==================
[18:48:50] ==================== xe_dma_buf_kunit  =====================
[18:48:50] ================ [SKIPPED] xe_dma_buf_kunit ================
[18:48:50] =================== [SKIPPED] xe_dma_buf ===================
[18:48:50] ================= xe_bo_shrink (1 subtest) =================
[18:48:50] =================== xe_bo_shrink_kunit  ====================
[18:48:50] =============== [SKIPPED] xe_bo_shrink_kunit ===============
[18:48:50] ================== [SKIPPED] xe_bo_shrink ==================
[18:48:50] ==================== xe_bo (2 subtests) ====================
[18:48:50] ================== xe_ccs_migrate_kunit  ===================
[18:48:50] ============== [SKIPPED] xe_ccs_migrate_kunit ==============
[18:48:50] ==================== xe_bo_evict_kunit  ====================
[18:48:50] =============== [SKIPPED] xe_bo_evict_kunit ================
[18:48:50] ===================== [SKIPPED] xe_bo ======================
[18:48:50] =================== xe_any (9 subtests) ====================
[18:48:50] [PASSED] test_to_xe
[18:48:50] [PASSED] test_to_dev
[18:48:50] [PASSED] test_to_pdev
[18:48:50] [PASSED] test_to_drm
[18:48:50] [PASSED] test_if_pdev
[18:48:50] [PASSED] test_if_xe
[18:48:50] [PASSED] test_if_tile
[18:48:50] [PASSED] test_if_gt
[18:48:50] [PASSED] test_to_id
[18:48:50] ===================== [PASSED] xe_any ======================
[18:48:50] ==================== args (13 subtests) ====================
[18:48:50] [PASSED] count_args_test
[18:48:50] [PASSED] call_args_example
[18:48:50] [PASSED] call_args_test
[18:48:50] [PASSED] drop_first_arg_example
[18:48:50] [PASSED] drop_first_arg_test
[18:48:50] [PASSED] first_arg_example
[18:48:50] [PASSED] first_arg_test
[18:48:50] [PASSED] last_arg_example
[18:48:50] [PASSED] last_arg_test
[18:48:50] [PASSED] pick_arg_example
[18:48:50] [PASSED] if_args_example
[18:48:50] [PASSED] if_args_test
[18:48:50] [PASSED] sep_comma_example
[18:48:50] ====================== [PASSED] args =======================
[18:48:50] =================== xe_pci (3 subtests) ====================
[18:48:50] ==================== check_graphics_ip  ====================
[18:48:50] [PASSED] 12.00 Xe_LP
[18:48:50] [PASSED] 12.10 Xe_LP+
[18:48:50] [PASSED] 12.55 Xe_HPG
[18:48:50] [PASSED] 12.60 Xe_HPC
[18:48:50] [PASSED] 12.70 Xe_LPG
[18:48:50] [PASSED] 12.71 Xe_LPG
[18:48:50] [PASSED] 12.74 Xe_LPG+
[18:48:50] [PASSED] 20.01 Xe2_HPG
[18:48:50] [PASSED] 20.02 Xe2_HPG
[18:48:50] [PASSED] 20.04 Xe2_LPG
[18:48:50] [PASSED] 30.00 Xe3_LPG
[18:48:50] [PASSED] 30.01 Xe3_LPG
[18:48:50] [PASSED] 30.03 Xe3_LPG
[18:48:50] [PASSED] 30.04 Xe3_LPG
[18:48:50] [PASSED] 30.05 Xe3_LPG
[18:48:50] [PASSED] 35.10 Xe3p_LPG
[18:48:50] [PASSED] 35.11 Xe3p_XPC
[18:48:50] ================ [PASSED] check_graphics_ip ================
[18:48:50] ===================== check_media_ip  ======================
[18:48:50] [PASSED] 12.00 Xe_M
[18:48:50] [PASSED] 12.55 Xe_HPM
[18:48:50] [PASSED] 13.00 Xe_LPM+
[18:48:50] [PASSED] 13.01 Xe2_HPM
[18:48:50] [PASSED] 20.00 Xe2_LPM
[18:48:50] [PASSED] 30.00 Xe3_LPM
[18:48:50] [PASSED] 30.02 Xe3_LPM
[18:48:50] [PASSED] 35.00 Xe3p_LPM
[18:48:50] [PASSED] 35.03 Xe3p_HPM
[18:48:50] ================= [PASSED] check_media_ip ==================
[18:48:50] =================== check_platform_desc  ===================
[18:48:50] [PASSED] 0x9A60 (TIGERLAKE)
[18:48:50] [PASSED] 0x9A68 (TIGERLAKE)
[18:48:50] [PASSED] 0x9A70 (TIGERLAKE)
[18:48:50] [PASSED] 0x9A40 (TIGERLAKE)
[18:48:50] [PASSED] 0x9A49 (TIGERLAKE)
[18:48:50] [PASSED] 0x9A59 (TIGERLAKE)
[18:48:50] [PASSED] 0x9A78 (TIGERLAKE)
[18:48:50] [PASSED] 0x9AC0 (TIGERLAKE)
[18:48:50] [PASSED] 0x9AC9 (TIGERLAKE)
[18:48:50] [PASSED] 0x9AD9 (TIGERLAKE)
[18:48:50] [PASSED] 0x9AF8 (TIGERLAKE)
[18:48:50] [PASSED] 0x4C80 (ROCKETLAKE)
[18:48:50] [PASSED] 0x4C8A (ROCKETLAKE)
[18:48:50] [PASSED] 0x4C8B (ROCKETLAKE)
[18:48:50] [PASSED] 0x4C8C (ROCKETLAKE)
[18:48:50] [PASSED] 0x4C90 (ROCKETLAKE)
[18:48:50] [PASSED] 0x4C9A (ROCKETLAKE)
[18:48:50] [PASSED] 0x4680 (ALDERLAKE_S)
[18:48:50] [PASSED] 0x4682 (ALDERLAKE_S)
[18:48:50] [PASSED] 0x4688 (ALDERLAKE_S)
[18:48:50] [PASSED] 0x468A (ALDERLAKE_S)
[18:48:50] [PASSED] 0x468B (ALDERLAKE_S)
[18:48:50] [PASSED] 0x4690 (ALDERLAKE_S)
[18:48:50] [PASSED] 0x4692 (ALDERLAKE_S)
[18:48:50] [PASSED] 0x4693 (ALDERLAKE_S)
[18:48:50] [PASSED] 0x46A0 (ALDERLAKE_P)
[18:48:50] [PASSED] 0x46A1 (ALDERLAKE_P)
[18:48:50] [PASSED] 0x46A2 (ALDERLAKE_P)
[18:48:50] [PASSED] 0x46A3 (ALDERLAKE_P)
[18:48:50] [PASSED] 0x46A6 (ALDERLAKE_P)
[18:48:50] [PASSED] 0x46A8 (ALDERLAKE_P)
[18:48:50] [PASSED] 0x46AA (ALDERLAKE_P)
[18:48:50] [PASSED] 0x462A (ALDERLAKE_P)
[18:48:50] [PASSED] 0x4626 (ALDERLAKE_P)
[18:48:50] [PASSED] 0x4628 (ALDERLAKE_P)
[18:48:50] [PASSED] 0x46B0 (ALDERLAKE_P)
[18:48:50] [PASSED] 0x46B1 (ALDERLAKE_P)
[18:48:50] [PASSED] 0x46B2 (ALDERLAKE_P)
[18:48:50] [PASSED] 0x46B3 (ALDERLAKE_P)
[18:48:50] [PASSED] 0x46C0 (ALDERLAKE_P)
[18:48:50] [PASSED] 0x46C1 (ALDERLAKE_P)
[18:48:50] [PASSED] 0x46C2 (ALDERLAKE_P)
[18:48:50] [PASSED] 0x46C3 (ALDERLAKE_P)
[18:48:50] [PASSED] 0x46D0 (ALDERLAKE_N)
[18:48:50] [PASSED] 0x46D1 (ALDERLAKE_N)
[18:48:50] [PASSED] 0x46D2 (ALDERLAKE_N)
[18:48:50] [PASSED] 0x46D3 (ALDERLAKE_N)
[18:48:50] [PASSED] 0x46D4 (ALDERLAKE_N)
[18:48:50] [PASSED] 0xA721 (ALDERLAKE_P)
[18:48:50] [PASSED] 0xA7A1 (ALDERLAKE_P)
[18:48:50] [PASSED] 0xA7A9 (ALDERLAKE_P)
[18:48:50] [PASSED] 0xA7AC (ALDERLAKE_P)
[18:48:50] [PASSED] 0xA7AD (ALDERLAKE_P)
[18:48:50] [PASSED] 0xA720 (ALDERLAKE_P)
[18:48:50] [PASSED] 0xA7A0 (ALDERLAKE_P)
[18:48:50] [PASSED] 0xA7A8 (ALDERLAKE_P)
[18:48:50] [PASSED] 0xA7AA (ALDERLAKE_P)
[18:48:50] [PASSED] 0xA7AB (ALDERLAKE_P)
[18:48:50] [PASSED] 0xA780 (ALDERLAKE_S)
[18:48:50] [PASSED] 0xA781 (ALDERLAKE_S)
[18:48:50] [PASSED] 0xA782 (ALDERLAKE_S)
[18:48:50] [PASSED] 0xA783 (ALDERLAKE_S)
[18:48:50] [PASSED] 0xA788 (ALDERLAKE_S)
[18:48:50] [PASSED] 0xA789 (ALDERLAKE_S)
[18:48:50] [PASSED] 0xA78A (ALDERLAKE_S)
[18:48:50] [PASSED] 0xA78B (ALDERLAKE_S)
[18:48:50] [PASSED] 0x4905 (DG1)
[18:48:50] [PASSED] 0x4906 (DG1)
[18:48:50] [PASSED] 0x4907 (DG1)
[18:48:50] [PASSED] 0x4908 (DG1)
[18:48:50] [PASSED] 0x4909 (DG1)
[18:48:50] [PASSED] 0x56C0 (DG2)
[18:48:50] [PASSED] 0x56C2 (DG2)
[18:48:50] [PASSED] 0x56C1 (DG2)
[18:48:50] [PASSED] 0x7D51 (METEORLAKE)
[18:48:50] [PASSED] 0x7DD1 (METEORLAKE)
[18:48:50] [PASSED] 0x7D41 (METEORLAKE)
[18:48:50] [PASSED] 0x7D67 (METEORLAKE)
[18:48:50] [PASSED] 0xB640 (METEORLAKE)
[18:48:50] [PASSED] 0x56A0 (DG2)
[18:48:50] [PASSED] 0x56A1 (DG2)
[18:48:50] [PASSED] 0x56A2 (DG2)
[18:48:50] [PASSED] 0x56BE (DG2)
[18:48:50] [PASSED] 0x56BF (DG2)
[18:48:50] [PASSED] 0x5690 (DG2)
[18:48:50] [PASSED] 0x5691 (DG2)
[18:48:50] [PASSED] 0x5692 (DG2)
[18:48:50] [PASSED] 0x56A5 (DG2)
[18:48:50] [PASSED] 0x56A6 (DG2)
[18:48:50] [PASSED] 0x56B0 (DG2)
[18:48:50] [PASSED] 0x56B1 (DG2)
[18:48:50] [PASSED] 0x56BA (DG2)
[18:48:50] [PASSED] 0x56BB (DG2)
[18:48:50] [PASSED] 0x56BC (DG2)
[18:48:50] [PASSED] 0x56BD (DG2)
[18:48:50] [PASSED] 0x5693 (DG2)
[18:48:50] [PASSED] 0x5694 (DG2)
[18:48:50] [PASSED] 0x5695 (DG2)
[18:48:50] [PASSED] 0x56A3 (DG2)
[18:48:50] [PASSED] 0x56A4 (DG2)
[18:48:50] [PASSED] 0x56B2 (DG2)
[18:48:50] [PASSED] 0x56B3 (DG2)
[18:48:50] [PASSED] 0x5696 (DG2)
[18:48:50] [PASSED] 0x5697 (DG2)
[18:48:50] [PASSED] 0xB69 (PVC)
[18:48:50] [PASSED] 0xB6E (PVC)
[18:48:50] [PASSED] 0xBD4 (PVC)
[18:48:50] [PASSED] 0xBD5 (PVC)
[18:48:50] [PASSED] 0xBD6 (PVC)
[18:48:50] [PASSED] 0xBD7 (PVC)
[18:48:50] [PASSED] 0xBD8 (PVC)
[18:48:50] [PASSED] 0xBD9 (PVC)
[18:48:50] [PASSED] 0xBDA (PVC)
[18:48:50] [PASSED] 0xBDB (PVC)
[18:48:50] [PASSED] 0xBE0 (PVC)
[18:48:50] [PASSED] 0xBE1 (PVC)
[18:48:50] [PASSED] 0xBE5 (PVC)
[18:48:50] [PASSED] 0x7D40 (METEORLAKE)
[18:48:50] [PASSED] 0x7D45 (METEORLAKE)
[18:48:50] [PASSED] 0x7D55 (METEORLAKE)
[18:48:50] [PASSED] 0x7D60 (METEORLAKE)
[18:48:50] [PASSED] 0x7DD5 (METEORLAKE)
[18:48:50] [PASSED] 0x6420 (LUNARLAKE)
[18:48:50] [PASSED] 0x64A0 (LUNARLAKE)
[18:48:50] [PASSED] 0x64B0 (LUNARLAKE)
[18:48:50] [PASSED] 0xE202 (BATTLEMAGE)
[18:48:50] [PASSED] 0xE209 (BATTLEMAGE)
[18:48:50] [PASSED] 0xE20B (BATTLEMAGE)
[18:48:50] [PASSED] 0xE20C (BATTLEMAGE)
[18:48:50] [PASSED] 0xE20D (BATTLEMAGE)
[18:48:50] [PASSED] 0xE210 (BATTLEMAGE)
[18:48:50] [PASSED] 0xE211 (BATTLEMAGE)
[18:48:50] [PASSED] 0xE212 (BATTLEMAGE)
[18:48:50] [PASSED] 0xE216 (BATTLEMAGE)
[18:48:50] [PASSED] 0xE220 (BATTLEMAGE)
[18:48:50] [PASSED] 0xE221 (BATTLEMAGE)
[18:48:50] [PASSED] 0xE222 (BATTLEMAGE)
[18:48:50] [PASSED] 0xE223 (BATTLEMAGE)
[18:48:50] [PASSED] 0xB080 (PANTHERLAKE)
[18:48:50] [PASSED] 0xB081 (PANTHERLAKE)
[18:48:50] [PASSED] 0xB082 (PANTHERLAKE)
[18:48:50] [PASSED] 0xB083 (PANTHERLAKE)
[18:48:50] [PASSED] 0xB084 (PANTHERLAKE)
[18:48:50] [PASSED] 0xB085 (PANTHERLAKE)
[18:48:50] [PASSED] 0xB086 (PANTHERLAKE)
[18:48:50] [PASSED] 0xB087 (PANTHERLAKE)
[18:48:50] [PASSED] 0xB08F (PANTHERLAKE)
[18:48:50] [PASSED] 0xB090 (PANTHERLAKE)
[18:48:50] [PASSED] 0xB0A0 (PANTHERLAKE)
[18:48:50] [PASSED] 0xB0A1 (PANTHERLAKE)
[18:48:50] [PASSED] 0xB0B0 (PANTHERLAKE)
[18:48:50] [PASSED] 0xFD80 (PANTHERLAKE)
[18:48:50] [PASSED] 0xFD81 (PANTHERLAKE)
[18:48:50] [PASSED] 0xD740 (NOVALAKE_S)
[18:48:50] [PASSED] 0xD741 (NOVALAKE_S)
[18:48:50] [PASSED] 0xD742 (NOVALAKE_S)
[18:48:50] [PASSED] 0xD743 (NOVALAKE_S)
[18:48:50] [PASSED] 0xD745 (NOVALAKE_S)
[18:48:50] [PASSED] 0xD74A (NOVALAKE_S)
[18:48:50] [PASSED] 0xD74B (NOVALAKE_S)
[18:48:50] [PASSED] 0x674C (CRESCENTISLAND)
[18:48:50] [PASSED] 0x674D (CRESCENTISLAND)
[18:48:50] [PASSED] 0x674E (CRESCENTISLAND)
[18:48:50] [PASSED] 0x674F (CRESCENTISLAND)
[18:48:50] [PASSED] 0x6750 (CRESCENTISLAND)
[18:48:50] [PASSED] 0x6751 (CRESCENTISLAND)
[18:48:50] [PASSED] 0xD750 (NOVALAKE_P)
[18:48:50] [PASSED] 0xD751 (NOVALAKE_P)
[18:48:50] [PASSED] 0xD752 (NOVALAKE_P)
[18:48:50] [PASSED] 0xD753 (NOVALAKE_P)
[18:48:50] [PASSED] 0xD754 (NOVALAKE_P)
[18:48:50] [PASSED] 0xD755 (NOVALAKE_P)
[18:48:50] [PASSED] 0xD756 (NOVALAKE_P)
[18:48:50] [PASSED] 0xD757 (NOVALAKE_P)
[18:48:50] [PASSED] 0xD75F (NOVALAKE_P)
[18:48:50] =============== [PASSED] check_platform_desc ===============
[18:48:50] ===================== [PASSED] xe_pci ======================
[18:48:50] ============= xe_rtp_tables_test (5 subtests) ==============
[18:48:50] ================== xe_rtp_table_gt_test  ===================
[18:48:50] [PASSED] gt_was/14011060649
[18:48:50] [PASSED] gt_was/14011059788
[18:48:50] [PASSED] gt_was/14015795083
[18:48:50] [PASSED] gt_was/16021867713
[18:48:50] [PASSED] gt_was/14019449301
[18:48:50] [PASSED] gt_was/16028005424
[18:48:50] [PASSED] gt_was/14026578760
[18:48:50] [PASSED] gt_was/1409420604
[18:48:50] [PASSED] gt_was/1408615072
[18:48:50] [PASSED] gt_was/22010523718
[18:48:50] [PASSED] gt_was/14011006942
[18:48:50] [PASSED] gt_was/14014830051
[18:48:50] [PASSED] gt_was/18018781329
[18:48:50] [PASSED] gt_was/1509235366
[18:48:50] [PASSED] gt_was/18018781329
[18:48:50] [PASSED] gt_was/16016694945
[18:48:50] [PASSED] gt_was/14018575942
[18:48:50] [PASSED] gt_was/22016670082
[18:48:50] [PASSED] gt_was/22016670082
[18:48:50] [PASSED] gt_was/14017421178
[18:48:50] [PASSED] gt_was/16025250150
[18:48:50] [PASSED] gt_was/14021871409
[18:48:50] [PASSED] gt_was/16021865536
[18:48:50] [PASSED] gt_was/14021486841
[18:48:50] [PASSED] gt_was/14025160223
[18:48:50] [PASSED] gt_was/14026144927, 16029437861, 14026127056
[18:48:50] [PASSED] gt_was/14025635424
[18:48:50] [PASSED] gt_was/16028005424
[18:48:50] ============== [PASSED] xe_rtp_table_gt_test ===============
[18:48:50] ================== xe_rtp_table_gt_test  ===================
[18:48:50] [PASSED] gt_tunings/Tuning: Blend Fill Caching Optimization Disable
[18:48:50] [PASSED] gt_tunings/Tuning: 32B Access Enable
[18:48:50] [PASSED] gt_tunings/Tuning: L3 cache
[18:48:50] [PASSED] gt_tunings/Tuning: L3 cache - media
[18:48:50] [PASSED] gt_tunings/Tuning: Compression Overfetch
[18:48:50] [PASSED] gt_tunings/Tuning: Compression Overfetch - media
[18:48:50] [PASSED] gt_tunings/Tuning: Enable compressible partial write overfetch in L3
[18:48:50] [PASSED] gt_tunings/Tuning: Enable compressible partial write overfetch in L3 - media
[18:48:50] [PASSED] gt_tunings/Tuning: L2 Overfetch Compressible Only
[18:48:50] [PASSED] gt_tunings/Tuning: L2 Overfetch Compressible Only - media
[18:48:50] [PASSED] gt_tunings/Tuning: Stateless compression control
[18:48:50] [PASSED] gt_tunings/Tuning: Stateless compression control - media
[18:48:50] [PASSED] gt_tunings/Tuning: L3 RW flush all Cache
[18:48:50] [PASSED] gt_tunings/Tuning: L3 RW flush all cache - media
[18:48:50] [PASSED] gt_tunings/Tuning: Set STLB Bank Hash Mode to 4KB
[18:48:50] [PASSED] gt_tunings/Tuning: Force All Non-Compressible Memory Reads to be 256B Overfetches
[18:48:50] ============== [PASSED] xe_rtp_table_gt_test ===============
[18:48:50] ================== xe_rtp_table_oob_test  ==================
[18:48:50] [PASSED] oob_was/1607983814
[18:48:50] [PASSED] oob_was/16010904313
[18:48:50] [PASSED] oob_was/18022495364
[18:48:50] [PASSED] oob_was/22012773006
[18:48:50] [PASSED] oob_was/14014475959
[18:48:50] [PASSED] oob_was/22011391025
[18:48:50] [PASSED] oob_was/22012727170
[18:48:50] [PASSED] oob_was/22012727685
[18:48:50] [PASSED] oob_was/22016596838
[18:48:50] [PASSED] oob_was/18020744125
[18:48:50] [PASSED] oob_was/1409600907
[18:48:50] [PASSED] oob_was/22014953428
[18:48:50] [PASSED] oob_was/16017236439
[18:48:50] [PASSED] oob_was/14019821291
[18:48:50] [PASSED] oob_was/14015076503
[18:48:50] [PASSED] oob_was/22016122933
[18:48:50] [PASSED] oob_was/14018913170
[18:48:50] [PASSED] oob_was/14018094691
[18:48:50] [PASSED] oob_was/18024947630
[18:48:50] [PASSED] oob_was/16022287689
[18:48:50] [PASSED] oob_was/13011645652
[18:48:50] [PASSED] oob_was/14022293748
[18:48:50] [PASSED] oob_was/22019794406
[18:48:50] [PASSED] oob_was/22019338487
[18:48:50] [PASSED] oob_was/16023588340
[18:48:50] [PASSED] oob_was/14019789679
[18:48:50] [PASSED] oob_was/14022866841
[18:48:50] [PASSED] oob_was/16021333562
[18:48:50] [PASSED] oob_was/14016712196
[18:48:50] [PASSED] oob_was/14015568240
[18:48:50] [PASSED] oob_was/18013179988
[18:48:50] [PASSED] oob_was/1508761755
[18:48:50] [PASSED] oob_was/16023105232
[18:48:50] [PASSED] oob_was/16026508708
[18:48:50] [PASSED] oob_was/14020001231
[18:48:50] [PASSED] oob_was/16023683509
[18:48:50] [PASSED] oob_was/14025515070
[18:48:50] [PASSED] oob_was/15015404425_disable
[18:48:50] [PASSED] oob_was/16026007364
[18:48:50] [PASSED] oob_was/14020316580
[18:48:50] [PASSED] oob_was/14025883347
[18:48:50] [PASSED] oob_was/16029380221
[18:48:50] [PASSED] oob_was/22022079272
[18:48:50] [PASSED] oob_was/16029897822
[18:48:50] [PASSED] oob_was/14027054324
[18:48:50] [PASSED] oob_was/14025941587
[18:48:50] [PASSED] oob_was/16030224309
[18:48:50] ============== [PASSED] xe_rtp_table_oob_test ==============
[18:48:50] ================ xe_rtp_table_dev_oob_test  ================
[18:48:50] [PASSED] device_oob_was/22010954014
[18:48:50] [PASSED] device_oob_was/15015404425
[18:48:50] [PASSED] device_oob_was/22019338487_display
[18:48:50] [PASSED] device_oob_was/14022085890
[18:48:50] [PASSED] device_oob_was/14026539277
[18:48:50] [PASSED] device_oob_was/14026633728
[18:48:50] [PASSED] device_oob_was/14026746987
[18:48:50] [PASSED] device_oob_was/14026779378
[18:48:50] ============ [PASSED] xe_rtp_table_dev_oob_test ============
[18:48:50] ========== xe_rtp_table_missing_upper_bound_test  ==========
[18:48:50] [PASSED] register_whitelist/WaAllowPMDepthAndInvocationCountAccessFromUMD, 1408556865
[18:48:50] [PASSED] register_whitelist/1508744258, 14012131227, 1808121037
[18:48:50] [PASSED] register_whitelist/1806527549
[18:48:50] [PASSED] register_whitelist/allow_read_ctx_timestamp
[18:48:50] [PASSED] register_whitelist/allow_read_queue_timestamp
[18:48:50] [PASSED] register_whitelist/16014440446
[18:48:50] [PASSED] register_whitelist/16017236439
[18:48:50] [PASSED] register_whitelist/16020183090
[18:48:50] [PASSED] register_whitelist/14024997852
[18:48:50] [PASSED] register_whitelist/14024997852
[18:48:50] ====== [PASSED] xe_rtp_table_missing_upper_bound_test ======
[18:48:50] =============== [PASSED] xe_rtp_tables_test ================
[18:48:50] =================== xe_rtp (3 subtests) ====================
[18:48:50] =================== xe_rtp_rules_tests  ====================
[18:48:50] [PASSED] no
[18:48:50] [PASSED] yes
[18:48:50] [PASSED] no-and-no
[18:48:50] [PASSED] no-and-yes
[18:48:50] [PASSED] yes-and-no
[18:48:50] [PASSED] yes-and-yes
[18:48:50] [PASSED] no-or-no
[18:48:50] [PASSED] no-or-yes
[18:48:50] [PASSED] yes-or-no
[18:48:50] [PASSED] yes-or-yes
[18:48:50] [PASSED] no-yes-or-yes-no
[18:48:50] [PASSED] no-yes-or-yes-yes
[18:48:50] [PASSED] yes-yes-or-no-yes
[18:48:50] [PASSED] yes-yes-or-yes-yes
[18:48:50] [PASSED] no-no-or-yes-or-no
[18:48:50] [PASSED] or
[18:48:50] [PASSED] or-yes
[18:48:50] [PASSED] or-no
[18:48:50] [PASSED] yes-or
[18:48:50] [PASSED] no-or
[18:48:50] [PASSED] no-or-or-yes
[18:48:50] [PASSED] yes-or-or-no
[18:48:50] [PASSED] no-or-or-no
[18:48:50] [PASSED] missing-context-engine-class
[18:48:50] [PASSED] missing-context-engine-class-or-yes
[18:48:50] [PASSED] missing-context-engine-class-or-or-yes
[18:48:50] =============== [PASSED] xe_rtp_rules_tests ================
[18:48:50] =============== xe_rtp_process_to_sr_tests  ================
[18:48:50] [PASSED] coalesce-same-reg
[18:48:50] [PASSED] coalesce-same-reg-literal-and-func
[18:48:50] [PASSED] no-match-no-add
[18:48:50] [PASSED] two-regs-two-entries
[18:48:50] [PASSED] clr-one-set-other
[18:48:50] [PASSED] set-field
[18:48:50] [PASSED] conflict-duplicate
[18:48:50] [PASSED] conflict-not-disjoint
[18:48:50] [PASSED] conflict-not-disjoint-literal-and-func
[18:48:50] [PASSED] conflict-reg-type
[18:48:50] [PASSED] bad-mcr-reg-forced-to-regular
[18:48:50] [PASSED] bad-regular-reg-forced-to-mcr
[18:48:50] =========== [PASSED] xe_rtp_process_to_sr_tests ============
[18:48:50] ================== xe_rtp_process_tests  ===================
[18:48:50] [PASSED] active1
[18:48:50] [PASSED] active2
[18:48:50] [PASSED] active-inactive
[18:48:50] [PASSED] inactive-active
[18:48:50] [PASSED] inactive-active-inactive
[18:48:50] [PASSED] inactive-inactive-inactive
[18:48:50] ============== [PASSED] xe_rtp_process_tests ===============
[18:48:50] ===================== [PASSED] xe_rtp ======================
[18:48:50] ==================== xe_wa (1 subtest) =====================
[18:48:50] ======================== xe_wa_gt  =========================
[18:48:50] [PASSED] TIGERLAKE B0
[18:48:50] [PASSED] DG1 A0
[18:48:50] [PASSED] DG1 B0
[18:48:50] [PASSED] ALDERLAKE_S A0
[18:48:50] [PASSED] ALDERLAKE_S B0
[18:48:50] [PASSED] ALDERLAKE_S C0
[18:48:50] [PASSED] ALDERLAKE_S D0
[18:48:50] [PASSED] ALDERLAKE_P A0
[18:48:50] [PASSED] ALDERLAKE_P B0
[18:48:50] [PASSED] ALDERLAKE_P C0
[18:48:50] [PASSED] ALDERLAKE_S RPLS D0
[18:48:50] [PASSED] ALDERLAKE_P RPLU E0
[18:48:50] [PASSED] DG2 G10 C0
[18:48:50] [PASSED] DG2 G11 B1
[18:48:50] [PASSED] DG2 G12 A1
[18:48:50] [PASSED] METEORLAKE 12.70(Xe_LPG) A0 13.00(Xe_LPM+) A0
[18:48:50] [PASSED] METEORLAKE 12.71(Xe_LPG) A0 13.00(Xe_LPM+) A0
[18:48:50] [PASSED] METEORLAKE 12.74(Xe_LPG+) A0 13.00(Xe_LPM+) A0
[18:48:50] [PASSED] LUNARLAKE 20.04(Xe2_LPG) A0 20.00(Xe2_LPM) A0
[18:48:50] [PASSED] LUNARLAKE 20.04(Xe2_LPG) B0 20.00(Xe2_LPM) A0
[18:48:50] [PASSED] BATTLEMAGE 20.01(Xe2_HPG) A0 13.01(Xe2_HPM) A1
[18:48:50] [PASSED] PANTHERLAKE 30.00(Xe3_LPG) A0 30.00(Xe3_LPM) A0
[18:48:50] ==================== [PASSED] xe_wa_gt =====================
[18:48:50] ====================== [PASSED] xe_wa ======================
[18:48:50] ============================================================
[18:48:50] Testing complete. Ran 838 tests: passed: 809, skipped: 29
[18:48:50] Elapsed time: 41.311s total, 1.840s configuring, 38.705s building, 0.723s running

+ for kcfg in "${kunitconfigs[@]}"
+ [[ ! -f /kernel/drivers/gpu/drm/tests/.kunitconfig ]]
+ /kernel/tools/testing/kunit/kunit.py run --kunitconfig /kernel/drivers/gpu/drm/tests/.kunitconfig
[18:48:50] Configuring KUnit Kernel ...
Regenerating .config ...
Populating config with:
$ make ARCH=um O=.kunit olddefconfig
[18:48:52] Building KUnit Kernel ...
Populating config with:
$ make ARCH=um O=.kunit olddefconfig
Building with:
$ make all compile_commands.json scripts_gdb ARCH=um O=.kunit --jobs=48
[18:49:17] Starting KUnit Kernel (1/1)...
[18:49:17] ============================================================
Running tests with:
$ .kunit/linux kunit.enable=1 mem=1G console=tty kunit_shutdown=halt
[18:49:17] ============ drm_test_pick_cmdline (2 subtests) ============
[18:49:17] [PASSED] drm_test_pick_cmdline_res_1920_1080_60
[18:49:17] =============== drm_test_pick_cmdline_named  ===============
[18:49:17] [PASSED] NTSC
[18:49:17] [PASSED] NTSC-J
[18:49:17] [PASSED] PAL
[18:49:17] [PASSED] PAL-M
[18:49:17] =========== [PASSED] drm_test_pick_cmdline_named ===========
[18:49:17] ============== [PASSED] drm_test_pick_cmdline ==============
[18:49:17] == drm_test_atomic_get_connector_for_encoder (1 subtest) ===
[18:49:17] [PASSED] drm_test_drm_atomic_get_connector_for_encoder
[18:49:17] ==== [PASSED] drm_test_atomic_get_connector_for_encoder ====
[18:49:17] =========== drm_validate_clone_mode (2 subtests) ===========
[18:49:17] ============== drm_test_check_in_clone_mode  ===============
[18:49:17] [PASSED] in_clone_mode
[18:49:17] [PASSED] not_in_clone_mode
[18:49:17] ========== [PASSED] drm_test_check_in_clone_mode ===========
[18:49:17] =============== drm_test_check_valid_clones  ===============
[18:49:17] [PASSED] not_in_clone_mode
[18:49:17] [PASSED] valid_clone
[18:49:17] [PASSED] invalid_clone
[18:49:17] =========== [PASSED] drm_test_check_valid_clones ===========
[18:49:17] ============= [PASSED] drm_validate_clone_mode =============
[18:49:17] ============= drm_validate_modeset (1 subtest) =============
[18:49:17] [PASSED] drm_test_check_connector_changed_modeset
[18:49:17] ============== [PASSED] drm_validate_modeset ===============
[18:49:17] ====== drm_test_bridge_get_current_state (1 subtest) =======
[18:49:17] [PASSED] drm_test_drm_bridge_get_current_state_atomic
[18:49:17] ======== [PASSED] drm_test_bridge_get_current_state ========
[18:49:17] ====== drm_test_bridge_helper_reset_crtc (3 subtests) ======
[18:49:17] [PASSED] drm_test_drm_bridge_helper_reset_crtc_atomic
[18:49:17] [PASSED] drm_test_drm_bridge_helper_reset_crtc_atomic_disabled
[18:49:17] [PASSED] drm_test_drm_bridge_helper_hdmi_output_bus_fmts
[18:49:17] ======== [PASSED] drm_test_bridge_helper_reset_crtc ========
[18:49:17] ============== drm_bridge_alloc (2 subtests) ===============
[18:49:17] [PASSED] drm_test_drm_bridge_alloc_basic
[18:49:17] [PASSED] drm_test_drm_bridge_alloc_get_put
[18:49:17] ================ [PASSED] drm_bridge_alloc =================
[18:49:17] ============= drm_bridge_bus_fmt (5 subtests) ==============
[18:49:17] [PASSED] drm_test_bridge_rgb_yuv_rgb
[18:49:17] [PASSED] drm_test_bridge_must_convert_to_yuv444
[18:49:17] [PASSED] drm_test_bridge_hdmi_auto_rgb
[18:49:17] [PASSED] drm_test_bridge_auto_first
[18:49:17] [PASSED] drm_test_bridge_rgb_yuv_no_path
[18:49:17] =============== [PASSED] drm_bridge_bus_fmt ================
[18:49:17] ============= drm_cmdline_parser (40 subtests) =============
[18:49:17] [PASSED] drm_test_cmdline_force_d_only
[18:49:17] [PASSED] drm_test_cmdline_force_D_only_dvi
[18:49:17] [PASSED] drm_test_cmdline_force_D_only_hdmi
[18:49:17] [PASSED] drm_test_cmdline_force_D_only_not_digital
[18:49:17] [PASSED] drm_test_cmdline_force_e_only
[18:49:17] [PASSED] drm_test_cmdline_res
[18:49:17] [PASSED] drm_test_cmdline_res_vesa
[18:49:17] [PASSED] drm_test_cmdline_res_vesa_rblank
[18:49:17] [PASSED] drm_test_cmdline_res_rblank
[18:49:17] [PASSED] drm_test_cmdline_res_bpp
[18:49:17] [PASSED] drm_test_cmdline_res_refresh
[18:49:17] [PASSED] drm_test_cmdline_res_bpp_refresh
[18:49:17] [PASSED] drm_test_cmdline_res_bpp_refresh_interlaced
[18:49:17] [PASSED] drm_test_cmdline_res_bpp_refresh_margins
[18:49:17] [PASSED] drm_test_cmdline_res_bpp_refresh_force_off
[18:49:17] [PASSED] drm_test_cmdline_res_bpp_refresh_force_on
[18:49:17] [PASSED] drm_test_cmdline_res_bpp_refresh_force_on_analog
[18:49:17] [PASSED] drm_test_cmdline_res_bpp_refresh_force_on_digital
[18:49:17] [PASSED] drm_test_cmdline_res_bpp_refresh_interlaced_margins_force_on
[18:49:17] [PASSED] drm_test_cmdline_res_margins_force_on
[18:49:17] [PASSED] drm_test_cmdline_res_vesa_margins
[18:49:17] [PASSED] drm_test_cmdline_name
[18:49:17] [PASSED] drm_test_cmdline_name_bpp
[18:49:17] [PASSED] drm_test_cmdline_name_option
[18:49:17] [PASSED] drm_test_cmdline_name_bpp_option
[18:49:17] [PASSED] drm_test_cmdline_rotate_0
[18:49:17] [PASSED] drm_test_cmdline_rotate_90
[18:49:17] [PASSED] drm_test_cmdline_rotate_180
[18:49:17] [PASSED] drm_test_cmdline_rotate_270
[18:49:17] [PASSED] drm_test_cmdline_hmirror
[18:49:17] [PASSED] drm_test_cmdline_vmirror
[18:49:17] [PASSED] drm_test_cmdline_margin_options
[18:49:17] [PASSED] drm_test_cmdline_multiple_options
[18:49:17] [PASSED] drm_test_cmdline_bpp_extra_and_option
[18:49:17] [PASSED] drm_test_cmdline_extra_and_option
[18:49:17] [PASSED] drm_test_cmdline_freestanding_options
[18:49:17] [PASSED] drm_test_cmdline_freestanding_force_e_and_options
[18:49:17] [PASSED] drm_test_cmdline_panel_orientation
[18:49:17] ================ drm_test_cmdline_invalid  =================
[18:49:17] [PASSED] margin_only
[18:49:17] [PASSED] interlace_only
[18:49:17] [PASSED] res_missing_x
[18:49:17] [PASSED] res_missing_y
[18:49:17] [PASSED] res_bad_y
[18:49:17] [PASSED] res_missing_y_bpp
[18:49:17] [PASSED] res_bad_bpp
[18:49:17] [PASSED] res_bad_refresh
[18:49:17] [PASSED] res_bpp_refresh_force_on_off
[18:49:17] [PASSED] res_invalid_mode
[18:49:17] [PASSED] res_bpp_wrong_place_mode
[18:49:17] [PASSED] name_bpp_refresh
[18:49:17] [PASSED] name_refresh
[18:49:17] [PASSED] name_refresh_wrong_mode
[18:49:17] [PASSED] name_refresh_invalid_mode
[18:49:17] [PASSED] rotate_multiple
[18:49:17] [PASSED] rotate_invalid_val
[18:49:17] [PASSED] rotate_truncated
[18:49:17] [PASSED] invalid_option
[18:49:17] [PASSED] invalid_tv_option
[18:49:17] [PASSED] truncated_tv_option
[18:49:17] ============ [PASSED] drm_test_cmdline_invalid =============
[18:49:17] =============== drm_test_cmdline_tv_options  ===============
[18:49:17] [PASSED] NTSC
[18:49:17] [PASSED] NTSC_443
[18:49:17] [PASSED] NTSC_J
[18:49:17] [PASSED] PAL
[18:49:17] [PASSED] PAL_M
[18:49:17] [PASSED] PAL_N
[18:49:17] [PASSED] SECAM
[18:49:17] [PASSED] MONO_525
[18:49:17] [PASSED] MONO_625
[18:49:17] =========== [PASSED] drm_test_cmdline_tv_options ===========
[18:49:17] =============== [PASSED] drm_cmdline_parser ================
[18:49:17] ========== drmm_connector_hdmi_init (20 subtests) ==========
[18:49:17] [PASSED] drm_test_connector_hdmi_init_valid
[18:49:17] [PASSED] drm_test_connector_hdmi_init_bpc_8
[18:49:17] [PASSED] drm_test_connector_hdmi_init_bpc_10
[18:49:17] [PASSED] drm_test_connector_hdmi_init_bpc_12
[18:49:17] [PASSED] drm_test_connector_hdmi_init_bpc_invalid
[18:49:17] [PASSED] drm_test_connector_hdmi_init_bpc_null
[18:49:17] [PASSED] drm_test_connector_hdmi_init_formats_empty
[18:49:17] [PASSED] drm_test_connector_hdmi_init_formats_no_rgb
[18:49:17] === drm_test_connector_hdmi_init_formats_yuv420_allowed  ===
[18:49:17] [PASSED] supported_formats=0x9 yuv420_allowed=1
[18:49:17] [PASSED] supported_formats=0x9 yuv420_allowed=0
[18:49:17] [PASSED] supported_formats=0x5 yuv420_allowed=1
[18:49:17] [PASSED] supported_formats=0x5 yuv420_allowed=0
[18:49:17] === [PASSED] drm_test_connector_hdmi_init_formats_yuv420_allowed ===
[18:49:17] [PASSED] drm_test_connector_hdmi_init_null_ddc
[18:49:17] [PASSED] drm_test_connector_hdmi_init_null_product
[18:49:17] [PASSED] drm_test_connector_hdmi_init_null_vendor
[18:49:17] [PASSED] drm_test_connector_hdmi_init_product_length_exact
[18:49:17] [PASSED] drm_test_connector_hdmi_init_product_length_too_long
[18:49:17] [PASSED] drm_test_connector_hdmi_init_product_valid
[18:49:17] [PASSED] drm_test_connector_hdmi_init_vendor_length_exact
[18:49:17] [PASSED] drm_test_connector_hdmi_init_vendor_length_too_long
[18:49:17] [PASSED] drm_test_connector_hdmi_init_vendor_valid
[18:49:17] ========= drm_test_connector_hdmi_init_type_valid  =========
[18:49:17] [PASSED] HDMI-A
[18:49:17] [PASSED] HDMI-B
[18:49:17] ===== [PASSED] drm_test_connector_hdmi_init_type_valid =====
[18:49:17] ======== drm_test_connector_hdmi_init_type_invalid  ========
[18:49:17] [PASSED] Unknown
[18:49:17] [PASSED] VGA
[18:49:17] [PASSED] DVI-I
[18:49:17] [PASSED] DVI-D
[18:49:17] [PASSED] DVI-A
[18:49:17] [PASSED] Composite
[18:49:17] [PASSED] SVIDEO
[18:49:17] [PASSED] LVDS
[18:49:17] [PASSED] Component
[18:49:17] [PASSED] DIN
[18:49:17] [PASSED] DP
[18:49:17] [PASSED] TV
[18:49:17] [PASSED] eDP
[18:49:17] [PASSED] Virtual
[18:49:17] [PASSED] DSI
[18:49:17] [PASSED] DPI
[18:49:17] [PASSED] Writeback
[18:49:17] [PASSED] SPI
[18:49:17] [PASSED] USB
[18:49:17] ==== [PASSED] drm_test_connector_hdmi_init_type_invalid ====
[18:49:17] ============ [PASSED] drmm_connector_hdmi_init =============
[18:49:17] ============= drmm_connector_init (3 subtests) =============
[18:49:17] [PASSED] drm_test_drmm_connector_init
[18:49:17] [PASSED] drm_test_drmm_connector_init_null_ddc
[18:49:17] ========= drm_test_drmm_connector_init_type_valid  =========
[18:49:17] [PASSED] Unknown
[18:49:17] [PASSED] VGA
[18:49:17] [PASSED] DVI-I
[18:49:17] [PASSED] DVI-D
[18:49:17] [PASSED] DVI-A
[18:49:17] [PASSED] Composite
[18:49:17] [PASSED] SVIDEO
[18:49:17] [PASSED] LVDS
[18:49:17] [PASSED] Component
[18:49:17] [PASSED] DIN
[18:49:17] [PASSED] DP
[18:49:17] [PASSED] HDMI-A
[18:49:17] [PASSED] HDMI-B
[18:49:17] [PASSED] TV
[18:49:17] [PASSED] eDP
[18:49:17] [PASSED] Virtual
[18:49:17] [PASSED] DSI
[18:49:17] [PASSED] DPI
[18:49:17] [PASSED] Writeback
[18:49:17] [PASSED] SPI
[18:49:17] [PASSED] USB
[18:49:17] ===== [PASSED] drm_test_drmm_connector_init_type_valid =====
[18:49:17] =============== [PASSED] drmm_connector_init ===============
[18:49:17] ========= drm_connector_dynamic_init (6 subtests) ==========
[18:49:17] [PASSED] drm_test_drm_connector_dynamic_init
[18:49:17] [PASSED] drm_test_drm_connector_dynamic_init_null_ddc
[18:49:17] [PASSED] drm_test_drm_connector_dynamic_init_not_added
[18:49:17] [PASSED] drm_test_drm_connector_dynamic_init_properties
[18:49:17] ===== drm_test_drm_connector_dynamic_init_type_valid  ======
[18:49:17] [PASSED] Unknown
[18:49:17] [PASSED] VGA
[18:49:17] [PASSED] DVI-I
[18:49:17] [PASSED] DVI-D
[18:49:17] [PASSED] DVI-A
[18:49:17] [PASSED] Composite
[18:49:17] [PASSED] SVIDEO
[18:49:17] [PASSED] LVDS
[18:49:17] [PASSED] Component
[18:49:17] [PASSED] DIN
[18:49:17] [PASSED] DP
[18:49:17] [PASSED] HDMI-A
[18:49:17] [PASSED] HDMI-B
[18:49:17] [PASSED] TV
[18:49:17] [PASSED] eDP
[18:49:17] [PASSED] Virtual
[18:49:17] [PASSED] DSI
[18:49:17] [PASSED] DPI
[18:49:17] [PASSED] Writeback
[18:49:17] [PASSED] SPI
[18:49:17] [PASSED] USB
[18:49:17] = [PASSED] drm_test_drm_connector_dynamic_init_type_valid ==
[18:49:17] ======== drm_test_drm_connector_dynamic_init_name  =========
[18:49:17] [PASSED] Unknown
[18:49:17] [PASSED] VGA
[18:49:17] [PASSED] DVI-I
[18:49:17] [PASSED] DVI-D
[18:49:17] [PASSED] DVI-A
[18:49:17] [PASSED] Composite
[18:49:17] [PASSED] SVIDEO
[18:49:17] [PASSED] LVDS
[18:49:17] [PASSED] Component
[18:49:17] [PASSED] DIN
[18:49:17] [PASSED] DP
[18:49:17] [PASSED] HDMI-A
[18:49:17] [PASSED] HDMI-B
[18:49:17] [PASSED] TV
[18:49:17] [PASSED] eDP
[18:49:17] [PASSED] Virtual
[18:49:17] [PASSED] DSI
[18:49:17] [PASSED] DPI
[18:49:17] [PASSED] Writeback
[18:49:17] [PASSED] SPI
[18:49:17] [PASSED] USB
[18:49:17] ==== [PASSED] drm_test_drm_connector_dynamic_init_name =====
[18:49:17] =========== [PASSED] drm_connector_dynamic_init ============
[18:49:17] ==== drm_connector_dynamic_register_early (4 subtests) =====
[18:49:17] [PASSED] drm_test_drm_connector_dynamic_register_early_on_list
[18:49:17] [PASSED] drm_test_drm_connector_dynamic_register_early_defer
[18:49:17] [PASSED] drm_test_drm_connector_dynamic_register_early_no_init
[18:49:17] [PASSED] drm_test_drm_connector_dynamic_register_early_no_mode_object
[18:49:17] ====== [PASSED] drm_connector_dynamic_register_early =======
[18:49:17] ======= drm_connector_dynamic_register (7 subtests) ========
[18:49:17] [PASSED] drm_test_drm_connector_dynamic_register_on_list
[18:49:17] [PASSED] drm_test_drm_connector_dynamic_register_no_defer
[18:49:17] [PASSED] drm_test_drm_connector_dynamic_register_no_init
[18:49:17] [PASSED] drm_test_drm_connector_dynamic_register_mode_object
[18:49:17] [PASSED] drm_test_drm_connector_dynamic_register_sysfs
[18:49:17] [PASSED] drm_test_drm_connector_dynamic_register_sysfs_name
[18:49:17] [PASSED] drm_test_drm_connector_dynamic_register_debugfs
[18:49:17] ========= [PASSED] drm_connector_dynamic_register ==========
[18:49:17] = drm_connector_attach_broadcast_rgb_property (2 subtests) =
[18:49:17] [PASSED] drm_test_drm_connector_attach_broadcast_rgb_property
[18:49:17] [PASSED] drm_test_drm_connector_attach_broadcast_rgb_property_hdmi_connector
[18:49:17] === [PASSED] drm_connector_attach_broadcast_rgb_property ===
[18:49:17] ========== drm_get_tv_mode_from_name (2 subtests) ==========
[18:49:17] ========== drm_test_get_tv_mode_from_name_valid  ===========
[18:49:17] [PASSED] NTSC
[18:49:17] [PASSED] NTSC-443
[18:49:17] [PASSED] NTSC-J
[18:49:17] [PASSED] PAL
[18:49:17] [PASSED] PAL-M
[18:49:17] [PASSED] PAL-N
[18:49:17] [PASSED] SECAM
[18:49:17] [PASSED] Mono
[18:49:17] ====== [PASSED] drm_test_get_tv_mode_from_name_valid =======
[18:49:17] [PASSED] drm_test_get_tv_mode_from_name_truncated
[18:49:17] ============ [PASSED] drm_get_tv_mode_from_name ============
[18:49:17] = drm_test_connector_hdmi_compute_mode_clock (12 subtests) =
[18:49:17] [PASSED] drm_test_drm_hdmi_compute_mode_clock_rgb
[18:49:17] [PASSED] drm_test_drm_hdmi_compute_mode_clock_rgb_10bpc
[18:49:17] [PASSED] drm_test_drm_hdmi_compute_mode_clock_rgb_10bpc_vic_1
[18:49:17] [PASSED] drm_test_drm_hdmi_compute_mode_clock_rgb_12bpc
[18:49:17] [PASSED] drm_test_drm_hdmi_compute_mode_clock_rgb_12bpc_vic_1
[18:49:17] [PASSED] drm_test_drm_hdmi_compute_mode_clock_rgb_double
[18:49:17] = drm_test_connector_hdmi_compute_mode_clock_yuv420_valid  =
[18:49:17] [PASSED] VIC 96
[18:49:17] [PASSED] VIC 97
[18:49:17] [PASSED] VIC 101
[18:49:17] [PASSED] VIC 102
[18:49:17] [PASSED] VIC 106
[18:49:17] [PASSED] VIC 107
[18:49:17] === [PASSED] drm_test_connector_hdmi_compute_mode_clock_yuv420_valid ===
[18:49:17] [PASSED] drm_test_connector_hdmi_compute_mode_clock_yuv420_10_bpc
[18:49:17] [PASSED] drm_test_connector_hdmi_compute_mode_clock_yuv420_12_bpc
[18:49:17] [PASSED] drm_test_connector_hdmi_compute_mode_clock_yuv422_8_bpc
[18:49:17] [PASSED] drm_test_connector_hdmi_compute_mode_clock_yuv422_10_bpc
[18:49:17] [PASSED] drm_test_connector_hdmi_compute_mode_clock_yuv422_12_bpc
[18:49:17] === [PASSED] drm_test_connector_hdmi_compute_mode_clock ====
[18:49:17] == drm_hdmi_connector_get_broadcast_rgb_name (2 subtests) ==
[18:49:17] === drm_test_drm_hdmi_connector_get_broadcast_rgb_name  ====
[18:49:17] [PASSED] Automatic
[18:49:17] [PASSED] Full
[18:49:17] [PASSED] Limited 16:235
[18:49:17] === [PASSED] drm_test_drm_hdmi_connector_get_broadcast_rgb_name ===
[18:49:17] [PASSED] drm_test_drm_hdmi_connector_get_broadcast_rgb_name_invalid
[18:49:17] ==== [PASSED] drm_hdmi_connector_get_broadcast_rgb_name ====
[18:49:17] == drm_hdmi_connector_get_output_format_name (2 subtests) ==
[18:49:17] === drm_test_drm_hdmi_connector_get_output_format_name  ====
[18:49:17] [PASSED] RGB
[18:49:17] [PASSED] YUV 4:2:0
[18:49:17] [PASSED] YUV 4:2:2
[18:49:17] [PASSED] YUV 4:4:4
[18:49:17] === [PASSED] drm_test_drm_hdmi_connector_get_output_format_name ===
[18:49:17] [PASSED] drm_test_drm_hdmi_connector_get_output_format_name_invalid
[18:49:17] ==== [PASSED] drm_hdmi_connector_get_output_format_name ====
[18:49:17] ============= drm_damage_helper (21 subtests) ==============
[18:49:17] [PASSED] drm_test_damage_iter_no_damage
[18:49:17] [PASSED] drm_test_damage_iter_no_damage_fractional_src
[18:49:17] [PASSED] drm_test_damage_iter_no_damage_src_moved
[18:49:17] [PASSED] drm_test_damage_iter_no_damage_fractional_src_moved
[18:49:17] [PASSED] drm_test_damage_iter_no_damage_not_visible
[18:49:17] [PASSED] drm_test_damage_iter_no_damage_no_crtc
[18:49:17] [PASSED] drm_test_damage_iter_no_damage_no_fb
[18:49:17] [PASSED] drm_test_damage_iter_simple_damage
[18:49:17] [PASSED] drm_test_damage_iter_single_damage
[18:49:17] [PASSED] drm_test_damage_iter_single_damage_intersect_src
[18:49:17] [PASSED] drm_test_damage_iter_single_damage_outside_src
[18:49:17] [PASSED] drm_test_damage_iter_single_damage_fractional_src
[18:49:17] [PASSED] drm_test_damage_iter_single_damage_intersect_fractional_src
[18:49:17] [PASSED] drm_test_damage_iter_single_damage_outside_fractional_src
[18:49:17] [PASSED] drm_test_damage_iter_single_damage_src_moved
[18:49:17] [PASSED] drm_test_damage_iter_single_damage_fractional_src_moved
[18:49:17] [PASSED] drm_test_damage_iter_damage
[18:49:17] [PASSED] drm_test_damage_iter_damage_one_intersect
[18:49:17] [PASSED] drm_test_damage_iter_damage_one_outside
[18:49:17] [PASSED] drm_test_damage_iter_damage_src_moved
[18:49:17] [PASSED] drm_test_damage_iter_damage_not_visible
[18:49:17] ================ [PASSED] drm_damage_helper ================
[18:49:17] ============== drm_dp_mst_helper (3 subtests) ==============
[18:49:17] ============== drm_test_dp_mst_calc_pbn_mode  ==============
[18:49:17] [PASSED] Clock 154000 BPP 30 DSC disabled
[18:49:17] [PASSED] Clock 234000 BPP 30 DSC disabled
[18:49:17] [PASSED] Clock 297000 BPP 24 DSC disabled
[18:49:17] [PASSED] Clock 332880 BPP 24 DSC enabled
[18:49:17] [PASSED] Clock 324540 BPP 24 DSC enabled
[18:49:17] ========== [PASSED] drm_test_dp_mst_calc_pbn_mode ==========
[18:49:17] ============== drm_test_dp_mst_calc_pbn_div  ===============
[18:49:17] [PASSED] Link rate 2000000 lane count 4
[18:49:17] [PASSED] Link rate 2000000 lane count 2
[18:49:17] [PASSED] Link rate 2000000 lane count 1
[18:49:17] [PASSED] Link rate 1350000 lane count 4
[18:49:17] [PASSED] Link rate 1350000 lane count 2
[18:49:17] [PASSED] Link rate 1350000 lane count 1
[18:49:17] [PASSED] Link rate 1000000 lane count 4
[18:49:17] [PASSED] Link rate 1000000 lane count 2
[18:49:17] [PASSED] Link rate 1000000 lane count 1
[18:49:17] [PASSED] Link rate 810000 lane count 4
[18:49:17] [PASSED] Link rate 810000 lane count 2
[18:49:17] [PASSED] Link rate 810000 lane count 1
[18:49:17] [PASSED] Link rate 540000 lane count 4
[18:49:17] [PASSED] Link rate 540000 lane count 2
[18:49:17] [PASSED] Link rate 540000 lane count 1
[18:49:17] [PASSED] Link rate 270000 lane count 4
[18:49:17] [PASSED] Link rate 270000 lane count 2
[18:49:17] [PASSED] Link rate 270000 lane count 1
[18:49:17] [PASSED] Link rate 162000 lane count 4
[18:49:17] [PASSED] Link rate 162000 lane count 2
[18:49:17] [PASSED] Link rate 162000 lane count 1
[18:49:17] ========== [PASSED] drm_test_dp_mst_calc_pbn_div ===========
[18:49:17] ========= drm_test_dp_mst_sideband_msg_req_decode  =========
[18:49:17] [PASSED] DP_ENUM_PATH_RESOURCES with port number
[18:49:17] [PASSED] DP_POWER_UP_PHY with port number
[18:49:17] [PASSED] DP_POWER_DOWN_PHY with port number
[18:49:17] [PASSED] DP_ALLOCATE_PAYLOAD with SDP stream sinks
[18:49:17] [PASSED] DP_ALLOCATE_PAYLOAD with port number
[18:49:17] [PASSED] DP_ALLOCATE_PAYLOAD with VCPI
[18:49:17] [PASSED] DP_ALLOCATE_PAYLOAD with PBN
[18:49:17] [PASSED] DP_QUERY_PAYLOAD with port number
[18:49:17] [PASSED] DP_QUERY_PAYLOAD with VCPI
[18:49:17] [PASSED] DP_REMOTE_DPCD_READ with port number
[18:49:17] [PASSED] DP_REMOTE_DPCD_READ with DPCD address
[18:49:17] [PASSED] DP_REMOTE_DPCD_READ with max number of bytes
[18:49:17] [PASSED] DP_REMOTE_DPCD_WRITE with port number
[18:49:17] [PASSED] DP_REMOTE_DPCD_WRITE with DPCD address
[18:49:17] [PASSED] DP_REMOTE_DPCD_WRITE with data array
[18:49:17] [PASSED] DP_REMOTE_I2C_READ with port number
[18:49:17] [PASSED] DP_REMOTE_I2C_READ with I2C device ID
[18:49:17] [PASSED] DP_REMOTE_I2C_READ with transactions array
[18:49:17] [PASSED] DP_REMOTE_I2C_WRITE with port number
[18:49:17] [PASSED] DP_REMOTE_I2C_WRITE with I2C device ID
[18:49:17] [PASSED] DP_REMOTE_I2C_WRITE with data array
[18:49:17] [PASSED] DP_QUERY_STREAM_ENC_STATUS with stream ID
[18:49:17] [PASSED] DP_QUERY_STREAM_ENC_STATUS with client ID
[18:49:17] [PASSED] DP_QUERY_STREAM_ENC_STATUS with stream event
[18:49:17] [PASSED] DP_QUERY_STREAM_ENC_STATUS with valid stream event
[18:49:17] [PASSED] DP_QUERY_STREAM_ENC_STATUS with stream behavior
[18:49:17] [PASSED] DP_QUERY_STREAM_ENC_STATUS with a valid stream behavior
[18:49:17] ===== [PASSED] drm_test_dp_mst_sideband_msg_req_decode =====
[18:49:17] ================ [PASSED] drm_dp_mst_helper ================
[18:49:17] ================== drm_exec (7 subtests) ===================
[18:49:17] [PASSED] sanitycheck
[18:49:17] [PASSED] test_lock
[18:49:17] [PASSED] test_lock_unlock
[18:49:17] [PASSED] test_duplicates
[18:49:17] [PASSED] test_prepare
[18:49:17] [PASSED] test_prepare_array
[18:49:17] [PASSED] test_multiple_loops
[18:49:17] ==================== [PASSED] drm_exec =====================
[18:49:17] =========== drm_format_helper_test (17 subtests) ===========
[18:49:17] ============== drm_test_fb_xrgb8888_to_gray8  ==============
[18:49:17] [PASSED] single_pixel_source_buffer
[18:49:17] [PASSED] single_pixel_clip_rectangle
[18:49:17] [PASSED] well_known_colors
[18:49:17] [PASSED] destination_pitch
[18:49:17] ========== [PASSED] drm_test_fb_xrgb8888_to_gray8 ==========
[18:49:17] ============= drm_test_fb_xrgb8888_to_rgb332  ==============
[18:49:17] [PASSED] single_pixel_source_buffer
[18:49:17] [PASSED] single_pixel_clip_rectangle
[18:49:17] [PASSED] well_known_colors
[18:49:17] [PASSED] destination_pitch
[18:49:17] ========= [PASSED] drm_test_fb_xrgb8888_to_rgb332 ==========
[18:49:17] ============= drm_test_fb_xrgb8888_to_rgb565  ==============
[18:49:17] [PASSED] single_pixel_source_buffer
[18:49:17] [PASSED] single_pixel_clip_rectangle
[18:49:17] [PASSED] well_known_colors
[18:49:17] [PASSED] destination_pitch
[18:49:17] ========= [PASSED] drm_test_fb_xrgb8888_to_rgb565 ==========
[18:49:17] ============ drm_test_fb_xrgb8888_to_xrgb1555  =============
[18:49:17] [PASSED] single_pixel_source_buffer
[18:49:17] [PASSED] single_pixel_clip_rectangle
[18:49:17] [PASSED] well_known_colors
[18:49:17] [PASSED] destination_pitch
[18:49:17] ======== [PASSED] drm_test_fb_xrgb8888_to_xrgb1555 =========
[18:49:17] ============ drm_test_fb_xrgb8888_to_argb1555  =============
[18:49:17] [PASSED] single_pixel_source_buffer
[18:49:17] [PASSED] single_pixel_clip_rectangle
[18:49:17] [PASSED] well_known_colors
[18:49:17] [PASSED] destination_pitch
[18:49:17] ======== [PASSED] drm_test_fb_xrgb8888_to_argb1555 =========
[18:49:17] ============ drm_test_fb_xrgb8888_to_rgba5551  =============
[18:49:17] [PASSED] single_pixel_source_buffer
[18:49:17] [PASSED] single_pixel_clip_rectangle
[18:49:17] [PASSED] well_known_colors
[18:49:17] [PASSED] destination_pitch
[18:49:17] ======== [PASSED] drm_test_fb_xrgb8888_to_rgba5551 =========
[18:49:17] ============= drm_test_fb_xrgb8888_to_rgb888  ==============
[18:49:17] [PASSED] single_pixel_source_buffer
[18:49:17] [PASSED] single_pixel_clip_rectangle
[18:49:17] [PASSED] well_known_colors
[18:49:17] [PASSED] destination_pitch
[18:49:17] ========= [PASSED] drm_test_fb_xrgb8888_to_rgb888 ==========
[18:49:17] ============= drm_test_fb_xrgb8888_to_bgr888  ==============
[18:49:17] [PASSED] single_pixel_source_buffer
[18:49:17] [PASSED] single_pixel_clip_rectangle
[18:49:17] [PASSED] well_known_colors
[18:49:17] [PASSED] destination_pitch
[18:49:17] ========= [PASSED] drm_test_fb_xrgb8888_to_bgr888 ==========
[18:49:17] ============ drm_test_fb_xrgb8888_to_argb8888  =============
[18:49:17] [PASSED] single_pixel_source_buffer
[18:49:17] [PASSED] single_pixel_clip_rectangle
[18:49:17] [PASSED] well_known_colors
[18:49:17] [PASSED] destination_pitch
[18:49:17] ======== [PASSED] drm_test_fb_xrgb8888_to_argb8888 =========
[18:49:17] =========== drm_test_fb_xrgb8888_to_xrgb2101010  ===========
[18:49:17] [PASSED] single_pixel_source_buffer
[18:49:17] [PASSED] single_pixel_clip_rectangle
[18:49:17] [PASSED] well_known_colors
[18:49:17] [PASSED] destination_pitch
[18:49:17] ======= [PASSED] drm_test_fb_xrgb8888_to_xrgb2101010 =======
[18:49:17] =========== drm_test_fb_xrgb8888_to_argb2101010  ===========
[18:49:17] [PASSED] single_pixel_source_buffer
[18:49:17] [PASSED] single_pixel_clip_rectangle
[18:49:17] [PASSED] well_known_colors
[18:49:17] [PASSED] destination_pitch
[18:49:17] ======= [PASSED] drm_test_fb_xrgb8888_to_argb2101010 =======
[18:49:17] ============== drm_test_fb_xrgb8888_to_mono  ===============
[18:49:17] [PASSED] single_pixel_source_buffer
[18:49:17] [PASSED] single_pixel_clip_rectangle
[18:49:17] [PASSED] well_known_colors
[18:49:17] [PASSED] destination_pitch
[18:49:17] ========== [PASSED] drm_test_fb_xrgb8888_to_mono ===========
[18:49:17] ==================== drm_test_fb_swab  =====================
[18:49:17] [PASSED] single_pixel_source_buffer
[18:49:17] [PASSED] single_pixel_clip_rectangle
[18:49:17] [PASSED] well_known_colors
[18:49:17] [PASSED] destination_pitch
[18:49:17] ================ [PASSED] drm_test_fb_swab =================
[18:49:17] ============ drm_test_fb_xrgb8888_to_xbgr8888  =============
[18:49:17] [PASSED] single_pixel_source_buffer
[18:49:17] [PASSED] single_pixel_clip_rectangle
[18:49:17] [PASSED] well_known_colors
[18:49:17] [PASSED] destination_pitch
[18:49:17] ======== [PASSED] drm_test_fb_xrgb8888_to_xbgr8888 =========
[18:49:17] ============ drm_test_fb_xrgb8888_to_abgr8888  =============
[18:49:17] [PASSED] single_pixel_source_buffer
[18:49:17] [PASSED] single_pixel_clip_rectangle
[18:49:17] [PASSED] well_known_colors
[18:49:17] [PASSED] destination_pitch
[18:49:17] ======== [PASSED] drm_test_fb_xrgb8888_to_abgr8888 =========
[18:49:17] ================= drm_test_fb_clip_offset  =================
[18:49:17] [PASSED] pass through
[18:49:17] [PASSED] horizontal offset
[18:49:17] [PASSED] vertical offset
[18:49:17] [PASSED] horizontal and vertical offset
[18:49:17] [PASSED] horizontal offset (custom pitch)
[18:49:17] [PASSED] vertical offset (custom pitch)
[18:49:17] [PASSED] horizontal and vertical offset (custom pitch)
[18:49:17] ============= [PASSED] drm_test_fb_clip_offset =============
[18:49:17] =================== drm_test_fb_memcpy  ====================
[18:49:17] [PASSED] single_pixel_source_buffer: XR24 little-endian (0x34325258)
[18:49:17] [PASSED] single_pixel_source_buffer: XRA8 little-endian (0x38415258)
[18:49:17] [PASSED] single_pixel_source_buffer: YU24 little-endian (0x34325559)
[18:49:17] [PASSED] single_pixel_clip_rectangle: XB24 little-endian (0x34324258)
[18:49:17] [PASSED] single_pixel_clip_rectangle: XRA8 little-endian (0x38415258)
[18:49:17] [PASSED] single_pixel_clip_rectangle: YU24 little-endian (0x34325559)
[18:49:17] [PASSED] well_known_colors: XB24 little-endian (0x34324258)
[18:49:17] [PASSED] well_known_colors: XRA8 little-endian (0x38415258)
[18:49:17] [PASSED] well_known_colors: YU24 little-endian (0x34325559)
[18:49:17] [PASSED] destination_pitch: XB24 little-endian (0x34324258)
[18:49:17] [PASSED] destination_pitch: XRA8 little-endian (0x38415258)
[18:49:17] [PASSED] destination_pitch: YU24 little-endian (0x34325559)
[18:49:17] =============== [PASSED] drm_test_fb_memcpy ================
[18:49:17] ============= [PASSED] drm_format_helper_test ==============
[18:49:17] ================= drm_format (18 subtests) =================
[18:49:17] [PASSED] drm_test_format_block_width_invalid
[18:49:17] [PASSED] drm_test_format_block_width_one_plane
[18:49:17] [PASSED] drm_test_format_block_width_two_plane
[18:49:17] [PASSED] drm_test_format_block_width_three_plane
[18:49:17] [PASSED] drm_test_format_block_width_tiled
[18:49:17] [PASSED] drm_test_format_block_height_invalid
[18:49:17] [PASSED] drm_test_format_block_height_one_plane
[18:49:17] [PASSED] drm_test_format_block_height_two_plane
[18:49:17] [PASSED] drm_test_format_block_height_three_plane
[18:49:17] [PASSED] drm_test_format_block_height_tiled
[18:49:17] [PASSED] drm_test_format_min_pitch_invalid
[18:49:17] [PASSED] drm_test_format_min_pitch_one_plane_8bpp
[18:49:17] [PASSED] drm_test_format_min_pitch_one_plane_16bpp
[18:49:17] [PASSED] drm_test_format_min_pitch_one_plane_24bpp
[18:49:17] [PASSED] drm_test_format_min_pitch_one_plane_32bpp
[18:49:17] [PASSED] drm_test_format_min_pitch_two_plane
[18:49:17] [PASSED] drm_test_format_min_pitch_three_plane_8bpp
[18:49:17] [PASSED] drm_test_format_min_pitch_tiled
[18:49:17] =================== [PASSED] drm_format ====================
[18:49:17] ============== drm_framebuffer (10 subtests) ===============
[18:49:17] ========== drm_test_framebuffer_check_src_coords  ==========
[18:49:17] [PASSED] Success: source fits into fb
[18:49:17] [PASSED] Fail: overflowing fb with x-axis coordinate
[18:49:17] [PASSED] Fail: overflowing fb with y-axis coordinate
[18:49:17] [PASSED] Fail: overflowing fb with source width
[18:49:17] [PASSED] Fail: overflowing fb with source height
[18:49:17] ====== [PASSED] drm_test_framebuffer_check_src_coords ======
[18:49:17] [PASSED] drm_test_framebuffer_cleanup
[18:49:17] =============== drm_test_framebuffer_create  ===============
[18:49:17] [PASSED] ABGR8888 normal sizes
[18:49:17] [PASSED] ABGR8888 max sizes
[18:49:17] [PASSED] ABGR8888 pitch greater than min required
[18:49:17] [PASSED] ABGR8888 pitch less than min required
[18:49:17] [PASSED] ABGR8888 Invalid width
[18:49:17] [PASSED] ABGR8888 Invalid buffer handle
[18:49:17] [PASSED] No pixel format
[18:49:17] [PASSED] ABGR8888 Width 0
[18:49:17] [PASSED] ABGR8888 Height 0
[18:49:17] [PASSED] ABGR8888 Out of bound height * pitch combination
[18:49:17] [PASSED] ABGR8888 Large buffer offset
[18:49:17] [PASSED] ABGR8888 Buffer offset for inexistent plane
[18:49:17] [PASSED] ABGR8888 Invalid flag
[18:49:17] [PASSED] ABGR8888 Set DRM_MODE_FB_MODIFIERS without modifiers
[18:49:17] [PASSED] ABGR8888 Valid buffer modifier
[18:49:17] [PASSED] ABGR8888 Invalid buffer modifier(DRM_FORMAT_MOD_SAMSUNG_64_32_TILE)
[18:49:17] [PASSED] ABGR8888 Extra pitches without DRM_MODE_FB_MODIFIERS
[18:49:17] [PASSED] ABGR8888 Extra pitches with DRM_MODE_FB_MODIFIERS
[18:49:17] [PASSED] NV12 Normal sizes
[18:49:17] [PASSED] NV12 Max sizes
[18:49:17] [PASSED] NV12 Invalid pitch
[18:49:17] [PASSED] NV12 Invalid modifier/missing DRM_MODE_FB_MODIFIERS flag
[18:49:17] [PASSED] NV12 different  modifier per-plane
[18:49:17] [PASSED] NV12 with DRM_FORMAT_MOD_SAMSUNG_64_32_TILE
[18:49:17] [PASSED] NV12 Valid modifiers without DRM_MODE_FB_MODIFIERS
[18:49:17] [PASSED] NV12 Modifier for inexistent plane
[18:49:17] [PASSED] NV12 Handle for inexistent plane
[18:49:17] [PASSED] NV12 Handle for inexistent plane without DRM_MODE_FB_MODIFIERS
[18:49:17] [PASSED] YVU420 DRM_MODE_FB_MODIFIERS set without modifier
[18:49:17] [PASSED] YVU420 Normal sizes
[18:49:17] [PASSED] YVU420 Max sizes
[18:49:17] [PASSED] YVU420 Invalid pitch
[18:49:17] [PASSED] YVU420 Different pitches
[18:49:17] [PASSED] YVU420 Different buffer offsets/pitches
[18:49:17] [PASSED] YVU420 Modifier set just for plane 0, without DRM_MODE_FB_MODIFIERS
[18:49:17] [PASSED] YVU420 Modifier set just for planes 0, 1, without DRM_MODE_FB_MODIFIERS
[18:49:17] [PASSED] YVU420 Modifier set just for plane 0, 1, with DRM_MODE_FB_MODIFIERS
[18:49:17] [PASSED] YVU420 Valid modifier
[18:49:17] [PASSED] YVU420 Different modifiers per plane
[18:49:17] [PASSED] YVU420 Modifier for inexistent plane
[18:49:17] [PASSED] YUV420_10BIT Invalid modifier(DRM_FORMAT_MOD_LINEAR)
[18:49:17] [PASSED] X0L2 Normal sizes
[18:49:17] [PASSED] X0L2 Max sizes
[18:49:17] [PASSED] X0L2 Invalid pitch
[18:49:17] [PASSED] X0L2 Pitch greater than minimum required
[18:49:17] [PASSED] X0L2 Handle for inexistent plane
[18:49:17] [PASSED] X0L2 Offset for inexistent plane, without DRM_MODE_FB_MODIFIERS set
[18:49:17] [PASSED] X0L2 Modifier without DRM_MODE_FB_MODIFIERS set
[18:49:17] [PASSED] X0L2 Valid modifier
[18:49:17] [PASSED] X0L2 Modifier for inexistent plane
[18:49:17] =========== [PASSED] drm_test_framebuffer_create ===========
[18:49:17] [PASSED] drm_test_framebuffer_free
[18:49:17] [PASSED] drm_test_framebuffer_init
[18:49:17] [PASSED] drm_test_framebuffer_init_bad_format
[18:49:17] [PASSED] drm_test_framebuffer_init_dev_mismatch
[18:49:17] [PASSED] drm_test_framebuffer_lookup
[18:49:17] [PASSED] drm_test_framebuffer_lookup_inexistent
[18:49:17] [PASSED] drm_test_framebuffer_modifiers_not_supported
[18:49:17] ================= [PASSED] drm_framebuffer =================
[18:49:17] ================ drm_gem_shmem (8 subtests) ================
[18:49:17] [PASSED] drm_gem_shmem_test_obj_create
[18:49:17] [PASSED] drm_gem_shmem_test_obj_create_private
[18:49:17] [PASSED] drm_gem_shmem_test_pin_pages
[18:49:17] [PASSED] drm_gem_shmem_test_vmap
[18:49:17] [PASSED] drm_gem_shmem_test_get_sg_table
[18:49:17] [PASSED] drm_gem_shmem_test_get_pages_sgt
[18:49:17] [PASSED] drm_gem_shmem_test_madvise
[18:49:17] [PASSED] drm_gem_shmem_test_purge
[18:49:17] ================== [PASSED] drm_gem_shmem ==================
[18:49:17] === drm_atomic_helper_connector_hdmi_check (29 subtests) ===
[18:49:17] [PASSED] drm_test_check_broadcast_rgb_auto_cea_mode
[18:49:17] [PASSED] drm_test_check_broadcast_rgb_auto_cea_mode_vic_1
[18:49:17] [PASSED] drm_test_check_broadcast_rgb_full_cea_mode
[18:49:17] [PASSED] drm_test_check_broadcast_rgb_full_cea_mode_vic_1
[18:49:17] [PASSED] drm_test_check_broadcast_rgb_limited_cea_mode
[18:49:17] [PASSED] drm_test_check_broadcast_rgb_limited_cea_mode_vic_1
[18:49:17] ====== drm_test_check_broadcast_rgb_cea_mode_yuv420  =======
[18:49:17] [PASSED] Automatic
[18:49:17] [PASSED] Full
[18:49:17] [PASSED] Limited 16:235
[18:49:17] == [PASSED] drm_test_check_broadcast_rgb_cea_mode_yuv420 ===
[18:49:17] [PASSED] drm_test_check_broadcast_rgb_crtc_mode_changed
[18:49:17] [PASSED] drm_test_check_broadcast_rgb_crtc_mode_not_changed
[18:49:17] [PASSED] drm_test_check_disable_connector
[18:49:17] [PASSED] drm_test_check_hdmi_funcs_reject_rate
[18:49:17] [PASSED] drm_test_check_max_tmds_rate_bpc_fallback_rgb
[18:49:17] [PASSED] drm_test_check_max_tmds_rate_bpc_fallback_yuv420
[18:49:17] [PASSED] drm_test_check_max_tmds_rate_bpc_fallback_ignore_yuv422
[18:49:17] [PASSED] drm_test_check_max_tmds_rate_bpc_fallback_ignore_yuv420
[18:49:17] [PASSED] drm_test_check_driver_unsupported_fallback_yuv420
[18:49:17] [PASSED] drm_test_check_output_bpc_crtc_mode_changed
[18:49:17] [PASSED] drm_test_check_output_bpc_crtc_mode_not_changed
[18:49:17] [PASSED] drm_test_check_output_bpc_dvi
[18:49:17] [PASSED] drm_test_check_output_bpc_format_vic_1
[18:49:17] [PASSED] drm_test_check_output_bpc_format_display_8bpc_only
[18:49:17] [PASSED] drm_test_check_output_bpc_format_display_rgb_only
[18:49:17] [PASSED] drm_test_check_output_bpc_format_driver_8bpc_only
[18:49:17] [PASSED] drm_test_check_output_bpc_format_driver_rgb_only
[18:49:17] [PASSED] drm_test_check_tmds_char_rate_rgb_8bpc
[18:49:17] [PASSED] drm_test_check_tmds_char_rate_rgb_10bpc
[18:49:17] [PASSED] drm_test_check_tmds_char_rate_rgb_12bpc
[18:49:17] ============ drm_test_check_hdmi_color_format  =============
[18:49:17] [PASSED] AUTO -> RGB
[18:49:17] [PASSED] YCBCR422 -> YUV422
[18:49:17] [PASSED] YCBCR420 -> YUV420
[18:49:17] [PASSED] YCBCR444 -> YUV444
[18:49:17] [PASSED] RGB -> RGB
[18:49:17] ======== [PASSED] drm_test_check_hdmi_color_format =========
[18:49:17] ======== drm_test_check_hdmi_color_format_420_only  ========
[18:49:17] [PASSED] RGB should fail
[18:49:17] [PASSED] YUV444 should fail
[18:49:17] [PASSED] YUV422 should fail
[18:49:17] [PASSED] YUV420 should work
[18:49:17] ==== [PASSED] drm_test_check_hdmi_color_format_420_only ====
[18:49:17] ===== [PASSED] drm_atomic_helper_connector_hdmi_check ======
[18:49:17] === drm_atomic_helper_connector_hdmi_reset (6 subtests) ====
[18:49:17] [PASSED] drm_test_check_broadcast_rgb_value
[18:49:17] [PASSED] drm_test_check_bpc_8_value
[18:49:17] [PASSED] drm_test_check_bpc_10_value
[18:49:17] [PASSED] drm_test_check_bpc_12_value
[18:49:17] [PASSED] drm_test_check_format_value
[18:49:17] [PASSED] drm_test_check_tmds_char_value
[18:49:17] ===== [PASSED] drm_atomic_helper_connector_hdmi_reset ======
[18:49:17] = drm_atomic_helper_connector_hdmi_mode_valid (7 subtests) =
[18:49:17] [PASSED] drm_test_check_mode_valid
[18:49:17] [PASSED] drm_test_check_mode_valid_reject
[18:49:17] [PASSED] drm_test_check_mode_valid_reject_rate
[18:49:17] [PASSED] drm_test_check_mode_valid_reject_max_clock
[18:49:17] [PASSED] drm_test_check_mode_valid_yuv420_only_max_clock
[18:49:17] [PASSED] drm_test_check_mode_valid_reject_yuv420_only_connector
[18:49:17] [PASSED] drm_test_check_mode_valid_accept_yuv420_also_connector_rgb
[18:49:17] === [PASSED] drm_atomic_helper_connector_hdmi_mode_valid ===
[18:49:17] = drm_atomic_helper_connector_hdmi_infoframes (5 subtests) =
[18:49:17] [PASSED] drm_test_check_infoframes
[18:49:17] [PASSED] drm_test_check_reject_avi_infoframe
[18:49:17] [PASSED] drm_test_check_reject_hdr_infoframe_bpc_8
[18:49:17] [PASSED] drm_test_check_reject_hdr_infoframe_bpc_10
[18:49:17] [PASSED] drm_test_check_reject_audio_infoframe
[18:49:17] === [PASSED] drm_atomic_helper_connector_hdmi_infoframes ===
[18:49:17] ================= drm_managed (2 subtests) =================
[18:49:17] [PASSED] drm_test_managed_release_action
[18:49:17] [PASSED] drm_test_managed_run_action
[18:49:17] =================== [PASSED] drm_managed ===================
[18:49:17] =================== drm_mm (6 subtests) ====================
[18:49:17] [PASSED] drm_test_mm_init
[18:49:17] [PASSED] drm_test_mm_debug
[18:49:17] [PASSED] drm_test_mm_align32
[18:49:17] [PASSED] drm_test_mm_align64
[18:49:17] [PASSED] drm_test_mm_lowest
[18:49:17] [PASSED] drm_test_mm_highest
[18:49:17] ===================== [PASSED] drm_mm ======================
[18:49:17] ============= drm_modes_analog_tv (5 subtests) =============
[18:49:17] [PASSED] drm_test_modes_analog_tv_mono_576i
[18:49:17] [PASSED] drm_test_modes_analog_tv_ntsc_480i
[18:49:17] [PASSED] drm_test_modes_analog_tv_ntsc_480i_inlined
[18:49:17] [PASSED] drm_test_modes_analog_tv_pal_576i
[18:49:17] [PASSED] drm_test_modes_analog_tv_pal_576i_inlined
[18:49:17] =============== [PASSED] drm_modes_analog_tv ===============
[18:49:17] ============== drm_panic_helper (3 subtests) ===============
[18:49:17] ============= drm_test_panic_screen_user_map  ==============
[18:49:17] [PASSED] Panic screen user, mode: 1024 x 768 	XR24 little-endian (0x34325258)
[18:49:17] [PASSED] Panic screen user, mode: 494 x 494 	XR24 little-endian (0x34325258)
[18:49:17] [PASSED] Panic screen user, mode: 1920 x 1080 	XR24 little-endian (0x34325258)
[18:49:17] [PASSED] Panic screen user, mode: 1024 x 768 	RG16 little-endian (0x36314752)
[18:49:17] [PASSED] Panic screen user, mode: 1024 x 768 	RG24 little-endian (0x34324752)
[18:49:17] [PASSED] Panic screen kmsg, mode: 1024 x 768 	XR24 little-endian (0x34325258)
[18:49:17] [PASSED] Panic screen kmsg, mode: 494 x 494 	XR24 little-endian (0x34325258)
[18:49:17] [PASSED] Panic screen kmsg, mode: 1920 x 1080 	XR24 little-endian (0x34325258)
[18:49:17] [PASSED] Panic screen kmsg, mode: 1024 x 768 	RG16 little-endian (0x36314752)
[18:49:17] [PASSED] Panic screen kmsg, mode: 1024 x 768 	RG24 little-endian (0x34324752)
[18:49:17] ========= [PASSED] drm_test_panic_screen_user_map ==========
[18:49:17] ============= drm_test_panic_screen_user_page  =============
[18:49:17] [PASSED] Panic screen user, mode: 1024 x 768 	XR24 little-endian (0x34325258)
[18:49:17] [PASSED] Panic screen user, mode: 494 x 494 	XR24 little-endian (0x34325258)
[18:49:17] [PASSED] Panic screen user, mode: 1920 x 1080 	XR24 little-endian (0x34325258)
[18:49:17] [PASSED] Panic screen user, mode: 1024 x 768 	RG16 little-endian (0x36314752)
[18:49:17] [PASSED] Panic screen user, mode: 1024 x 768 	RG24 little-endian (0x34324752)
[18:49:17] [PASSED] Panic screen kmsg, mode: 1024 x 768 	XR24 little-endian (0x34325258)
[18:49:17] [PASSED] Panic screen kmsg, mode: 494 x 494 	XR24 little-endian (0x34325258)
[18:49:17] [PASSED] Panic screen kmsg, mode: 1920 x 1080 	XR24 little-endian (0x34325258)
[18:49:17] [PASSED] Panic screen kmsg, mode: 1024 x 768 	RG16 little-endian (0x36314752)
[18:49:17] [PASSED] Panic screen kmsg, mode: 1024 x 768 	RG24 little-endian (0x34324752)
[18:49:17] ========= [PASSED] drm_test_panic_screen_user_page =========
[18:49:17] ========== drm_test_panic_screen_user_set_pixel  ===========
[18:49:17] [PASSED] Panic screen user, mode: 1024 x 768 	XR24 little-endian (0x34325258)
[18:49:17] [PASSED] Panic screen user, mode: 494 x 494 	XR24 little-endian (0x34325258)
[18:49:17] [PASSED] Panic screen user, mode: 1920 x 1080 	XR24 little-endian (0x34325258)
[18:49:17] [PASSED] Panic screen user, mode: 1024 x 768 	RG16 little-endian (0x36314752)
[18:49:17] [PASSED] Panic screen user, mode: 1024 x 768 	RG24 little-endian (0x34324752)
[18:49:17] [PASSED] Panic screen kmsg, mode: 1024 x 768 	XR24 little-endian (0x34325258)
[18:49:17] [PASSED] Panic screen kmsg, mode: 494 x 494 	XR24 little-endian (0x34325258)
[18:49:17] [PASSED] Panic screen kmsg, mode: 1920 x 1080 	XR24 little-endian (0x34325258)
[18:49:17] [PASSED] Panic screen kmsg, mode: 1024 x 768 	RG16 little-endian (0x36314752)
[18:49:17] [PASSED] Panic screen kmsg, mode: 1024 x 768 	RG24 little-endian (0x34324752)
[18:49:17] ====== [PASSED] drm_test_panic_screen_user_set_pixel =======
[18:49:17] ================ [PASSED] drm_panic_helper =================
[18:49:17] ============== drm_plane_helper (2 subtests) ===============
[18:49:17] =============== drm_test_check_plane_state  ================
[18:49:17] [PASSED] clipping_simple
[18:49:17] [PASSED] clipping_rotate_reflect
[18:49:17] [PASSED] positioning_simple
[18:49:17] [PASSED] upscaling
[18:49:17] [PASSED] downscaling
[18:49:17] [PASSED] rounding1
[18:49:17] [PASSED] rounding2
[18:49:17] [PASSED] rounding3
[18:49:17] [PASSED] rounding4
[18:49:17] =========== [PASSED] drm_test_check_plane_state ============
[18:49:17] =========== drm_test_check_invalid_plane_state  ============
[18:49:17] [PASSED] positioning_invalid
[18:49:17] [PASSED] upscaling_invalid
[18:49:17] [PASSED] downscaling_invalid
[18:49:17] ======= [PASSED] drm_test_check_invalid_plane_state ========
[18:49:17] ================ [PASSED] drm_plane_helper =================
[18:49:17] ====== drm_connector_helper_tv_get_modes (1 subtest) =======
[18:49:17] ====== drm_test_connector_helper_tv_get_modes_check  =======
[18:49:17] [PASSED] None
[18:49:17] [PASSED] PAL
[18:49:17] [PASSED] NTSC
[18:49:17] [PASSED] Both, NTSC Default
[18:49:17] [PASSED] Both, PAL Default
[18:49:17] [PASSED] Both, NTSC Default, with PAL on command-line
[18:49:17] [PASSED] Both, PAL Default, with NTSC on command-line
[18:49:17] == [PASSED] drm_test_connector_helper_tv_get_modes_check ===
[18:49:17] ======== [PASSED] drm_connector_helper_tv_get_modes ========
[18:49:17] ================== drm_rect (9 subtests) ===================
[18:49:17] [PASSED] drm_test_rect_clip_scaled_div_by_zero
[18:49:17] [PASSED] drm_test_rect_clip_scaled_not_clipped
[18:49:17] [PASSED] drm_test_rect_clip_scaled_clipped
[18:49:17] [PASSED] drm_test_rect_clip_scaled_signed_vs_unsigned
[18:49:17] ================= drm_test_rect_intersect  =================
[18:49:17] [PASSED] top-left x bottom-right: 2x2+1+1 x 2x2+0+0
[18:49:17] [PASSED] top-right x bottom-left: 2x2+0+0 x 2x2+1-1
[18:49:17] [PASSED] bottom-left x top-right: 2x2+1-1 x 2x2+0+0
[18:49:17] [PASSED] bottom-right x top-left: 2x2+0+0 x 2x2+1+1
[18:49:17] [PASSED] right x left: 2x1+0+0 x 3x1+1+0
[18:49:17] [PASSED] left x right: 3x1+1+0 x 2x1+0+0
[18:49:17] [PASSED] up x bottom: 1x2+0+0 x 1x3+0-1
[18:49:17] [PASSED] bottom x up: 1x3+0-1 x 1x2+0+0
[18:49:17] [PASSED] touching corner: 1x1+0+0 x 2x2+1+1
[18:49:17] [PASSED] touching side: 1x1+0+0 x 1x1+1+0
[18:49:17] [PASSED] equal rects: 2x2+0+0 x 2x2+0+0
[18:49:17] [PASSED] inside another: 2x2+0+0 x 1x1+1+1
[18:49:17] [PASSED] far away: 1x1+0+0 x 1x1+3+6
[18:49:17] [PASSED] points intersecting: 0x0+5+10 x 0x0+5+10
[18:49:17] [PASSED] points not intersecting: 0x0+0+0 x 0x0+5+10
[18:49:17] ============= [PASSED] drm_test_rect_intersect =============
[18:49:17] ================ drm_test_rect_calc_hscale  ================
[18:49:17] [PASSED] normal use
[18:49:17] [PASSED] out of max range
[18:49:17] [PASSED] out of min range
[18:49:17] [PASSED] zero dst
[18:49:17] [PASSED] negative src
[18:49:17] [PASSED] negative dst
[18:49:17] ============ [PASSED] drm_test_rect_calc_hscale ============
[18:49:17] ================ drm_test_rect_calc_vscale  ================
[18:49:17] [PASSED] normal use
[18:49:17] [PASSED] out of max range
[18:49:17] [PASSED] out of min range
[18:49:17] [PASSED] zero dst
[18:49:17] [PASSED] negative src
[18:49:17] [PASSED] negative dst
[18:49:17] ============ [PASSED] drm_test_rect_calc_vscale ============
[18:49:17] ================== drm_test_rect_rotate  ===================
[18:49:17] [PASSED] reflect-x
[18:49:17] [PASSED] reflect-y
[18:49:17] [PASSED] rotate-0
[18:49:17] [PASSED] rotate-90
[18:49:17] [PASSED] rotate-180
[18:49:17] [PASSED] rotate-270
[18:49:17] ============== [PASSED] drm_test_rect_rotate ===============
[18:49:17] ================ drm_test_rect_rotate_inv  =================
[18:49:17] [PASSED] reflect-x
[18:49:17] [PASSED] reflect-y
[18:49:17] [PASSED] rotate-0
[18:49:17] [PASSED] rotate-90
[18:49:17] [PASSED] rotate-180
[18:49:17] [PASSED] rotate-270
[18:49:17] ============ [PASSED] drm_test_rect_rotate_inv =============
[18:49:17] ==================== [PASSED] drm_rect =====================
[18:49:17] ============ drm_sysfb_modeset_test (1 subtest) ============
[18:49:17] ============ drm_test_sysfb_build_fourcc_list  =============
[18:49:17] [PASSED] no native formats
[18:49:17] [PASSED] XRGB8888 as native format
[18:49:17] [PASSED] remove duplicates
[18:49:17] [PASSED] convert alpha formats
[18:49:17] [PASSED] random formats
[18:49:17] ======== [PASSED] drm_test_sysfb_build_fourcc_list =========
[18:49:17] ============= [PASSED] drm_sysfb_modeset_test ==============
[18:49:17] ================== drm_fixp (2 subtests) ===================
[18:49:17] [PASSED] drm_test_int2fixp
[18:49:17] [PASSED] drm_test_sm2fixp
[18:49:17] ==================== [PASSED] drm_fixp =====================
[18:49:17] ============================================================
[18:49:17] Testing complete. Ran 667 tests: passed: 667
[18:49:17] Elapsed time: 27.404s total, 1.850s configuring, 25.087s building, 0.414s running

+ for kcfg in "${kunitconfigs[@]}"
+ [[ ! -f /kernel/drivers/gpu/drm/ttm/tests/.kunitconfig ]]
+ /kernel/tools/testing/kunit/kunit.py run --kunitconfig /kernel/drivers/gpu/drm/ttm/tests/.kunitconfig
[18:49:17] Configuring KUnit Kernel ...
Regenerating .config ...
Populating config with:
$ make ARCH=um O=.kunit olddefconfig
[18:49:19] Building KUnit Kernel ...
Populating config with:
$ make ARCH=um O=.kunit olddefconfig
Building with:
$ make all compile_commands.json scripts_gdb ARCH=um O=.kunit --jobs=48
[18:49:29] Starting KUnit Kernel (1/1)...
[18:49:29] ============================================================
Running tests with:
$ .kunit/linux kunit.enable=1 mem=1G console=tty kunit_shutdown=halt
[18:49:29] ================= ttm_device (5 subtests) ==================
[18:49:29] [PASSED] ttm_device_init_basic
[18:49:29] [PASSED] ttm_device_init_multiple
[18:49:29] [PASSED] ttm_device_fini_basic
[18:49:29] [PASSED] ttm_device_init_no_vma_man
[18:49:29] ================== ttm_device_init_pools  ==================
[18:49:29] [PASSED] No DMA allocations, no DMA32 required
[18:49:29] [PASSED] DMA allocations, DMA32 required
[18:49:29] [PASSED] No DMA allocations, DMA32 required
[18:49:29] [PASSED] DMA allocations, no DMA32 required
[18:49:29] ============== [PASSED] ttm_device_init_pools ==============
[18:49:29] =================== [PASSED] ttm_device ====================
[18:49:29] ================== ttm_pool (8 subtests) ===================
[18:49:29] ================== ttm_pool_alloc_basic  ===================
[18:49:29] [PASSED] One page
[18:49:29] [PASSED] More than one page
[18:49:29] [PASSED] Above the allocation limit
[18:49:29] [PASSED] One page, with coherent DMA mappings enabled
[18:49:29] [PASSED] Above the allocation limit, with coherent DMA mappings enabled
[18:49:29] ============== [PASSED] ttm_pool_alloc_basic ===============
[18:49:29] ============== ttm_pool_alloc_basic_dma_addr  ==============
[18:49:29] [PASSED] One page
[18:49:29] [PASSED] More than one page
[18:49:29] [PASSED] Above the allocation limit
[18:49:29] [PASSED] One page, with coherent DMA mappings enabled
[18:49:29] [PASSED] Above the allocation limit, with coherent DMA mappings enabled
[18:49:29] ========== [PASSED] ttm_pool_alloc_basic_dma_addr ==========
[18:49:29] [PASSED] ttm_pool_alloc_order_caching_match
[18:49:29] [PASSED] ttm_pool_alloc_caching_mismatch
[18:49:29] [PASSED] ttm_pool_alloc_order_mismatch
[18:49:29] [PASSED] ttm_pool_free_dma_alloc
[18:49:29] [PASSED] ttm_pool_free_no_dma_alloc
[18:49:29] [PASSED] ttm_pool_fini_basic
[18:49:29] ==================== [PASSED] ttm_pool =====================
[18:49:29] ================ ttm_resource (8 subtests) =================
[18:49:29] ================= ttm_resource_init_basic  =================
[18:49:29] [PASSED] Init resource in TTM_PL_SYSTEM
[18:49:29] [PASSED] Init resource in TTM_PL_VRAM
[18:49:29] [PASSED] Init resource in a private placement
[18:49:29] [PASSED] Init resource in TTM_PL_SYSTEM, set placement flags
[18:49:30] ============= [PASSED] ttm_resource_init_basic =============
[18:49:30] [PASSED] ttm_resource_init_pinned
[18:49:30] [PASSED] ttm_resource_fini_basic
[18:49:30] [PASSED] ttm_resource_manager_init_basic
[18:49:30] [PASSED] ttm_resource_manager_usage_basic
[18:49:30] [PASSED] ttm_resource_manager_set_used_basic
[18:49:30] [PASSED] ttm_sys_man_alloc_basic
[18:49:30] [PASSED] ttm_sys_man_free_basic
[18:49:30] ================== [PASSED] ttm_resource ===================
[18:49:30] =================== ttm_tt (15 subtests) ===================
[18:49:30] ==================== ttm_tt_init_basic  ====================
[18:49:30] [PASSED] Page-aligned size
[18:49:30] [PASSED] Extra pages requested
[18:49:30] ================ [PASSED] ttm_tt_init_basic ================
[18:49:30] [PASSED] ttm_tt_init_misaligned
[18:49:30] [PASSED] ttm_tt_fini_basic
[18:49:30] [PASSED] ttm_tt_fini_sg
[18:49:30] [PASSED] ttm_tt_fini_shmem
[18:49:30] [PASSED] ttm_tt_create_basic
[18:49:30] [PASSED] ttm_tt_create_invalid_bo_type
[18:49:30] [PASSED] ttm_tt_create_ttm_exists
[18:49:30] [PASSED] ttm_tt_create_failed
[18:49:30] [PASSED] ttm_tt_destroy_basic
[18:49:30] [PASSED] ttm_tt_populate_null_ttm
[18:49:30] [PASSED] ttm_tt_populate_populated_ttm
[18:49:30] [PASSED] ttm_tt_unpopulate_basic
[18:49:30] [PASSED] ttm_tt_unpopulate_empty_ttm
[18:49:30] [PASSED] ttm_tt_swapin_basic
[18:49:30] ===================== [PASSED] ttm_tt ======================
[18:49:30] =================== ttm_bo (14 subtests) ===================
[18:49:30] =========== ttm_bo_reserve_optimistic_no_ticket  ===========
[18:49:30] [PASSED] Cannot be interrupted and sleeps
[18:49:30] [PASSED] Cannot be interrupted, locks straight away
[18:49:30] [PASSED] Can be interrupted, sleeps
[18:49:30] ======= [PASSED] ttm_bo_reserve_optimistic_no_ticket =======
[18:49:30] [PASSED] ttm_bo_reserve_locked_no_sleep
[18:49:30] [PASSED] ttm_bo_reserve_no_wait_ticket
[18:49:30] [PASSED] ttm_bo_reserve_double_resv
[18:49:30] [PASSED] ttm_bo_reserve_interrupted
[18:49:30] [PASSED] ttm_bo_reserve_deadlock
[18:49:30] [PASSED] ttm_bo_unreserve_basic
[18:49:30] [PASSED] ttm_bo_unreserve_pinned
[18:49:30] [PASSED] ttm_bo_unreserve_bulk
[18:49:30] [PASSED] ttm_bo_fini_basic
[18:49:30] [PASSED] ttm_bo_fini_shared_resv
[18:49:30] [PASSED] ttm_bo_pin_basic
[18:49:30] [PASSED] ttm_bo_pin_unpin_resource
[18:49:30] [PASSED] ttm_bo_multiple_pin_one_unpin
[18:49:30] ===================== [PASSED] ttm_bo ======================
[18:49:30] ============== ttm_bo_validate (22 subtests) ===============
[18:49:30] ============== ttm_bo_init_reserved_sys_man  ===============
[18:49:30] [PASSED] Buffer object for userspace
[18:49:30] [PASSED] Kernel buffer object
[18:49:30] [PASSED] Shared buffer object
[18:49:30] ========== [PASSED] ttm_bo_init_reserved_sys_man ===========
[18:49:30] ============== ttm_bo_init_reserved_mock_man  ==============
[18:49:30] [PASSED] Buffer object for userspace
[18:49:30] [PASSED] Kernel buffer object
[18:49:30] [PASSED] Shared buffer object
[18:49:30] ========== [PASSED] ttm_bo_init_reserved_mock_man ==========
[18:49:30] [PASSED] ttm_bo_init_reserved_resv
[18:49:30] ================== ttm_bo_validate_basic  ==================
[18:49:30] [PASSED] Buffer object for userspace
[18:49:30] [PASSED] Kernel buffer object
[18:49:30] [PASSED] Shared buffer object
[18:49:30] ============== [PASSED] ttm_bo_validate_basic ==============
[18:49:30] [PASSED] ttm_bo_validate_invalid_placement
[18:49:30] ============= ttm_bo_validate_same_placement  ==============
[18:49:30] [PASSED] System manager
[18:49:30] [PASSED] VRAM manager
[18:49:30] ========= [PASSED] ttm_bo_validate_same_placement ==========
[18:49:30] [PASSED] ttm_bo_validate_failed_alloc
[18:49:30] [PASSED] ttm_bo_validate_pinned
[18:49:30] [PASSED] ttm_bo_validate_busy_placement
[18:49:30] ================ ttm_bo_validate_multihop  =================
[18:49:30] [PASSED] Buffer object for userspace
[18:49:30] [PASSED] Kernel buffer object
[18:49:30] [PASSED] Shared buffer object
[18:49:30] ============ [PASSED] ttm_bo_validate_multihop =============
[18:49:30] ========== ttm_bo_validate_no_placement_signaled  ==========
[18:49:30] [PASSED] Buffer object in system domain, no page vector
[18:49:30] [PASSED] Buffer object in system domain with an existing page vector
[18:49:30] ====== [PASSED] ttm_bo_validate_no_placement_signaled ======
[18:49:30] ======== ttm_bo_validate_no_placement_not_signaled  ========
[18:49:30] [PASSED] Buffer object for userspace
[18:49:30] [PASSED] Kernel buffer object
[18:49:30] [PASSED] Shared buffer object
[18:49:30] ==== [PASSED] ttm_bo_validate_no_placement_not_signaled ====
[18:49:30] [PASSED] ttm_bo_validate_move_fence_signaled
[18:49:30] ========= ttm_bo_validate_move_fence_not_signaled  =========
[18:49:30] [PASSED] Waits for GPU
[18:49:30] [PASSED] Tries to lock straight away
[18:49:30] ===== [PASSED] ttm_bo_validate_move_fence_not_signaled =====
[18:49:30] [PASSED] ttm_bo_validate_swapout
[18:49:30] [PASSED] ttm_bo_validate_happy_evict
[18:49:30] [PASSED] ttm_bo_validate_all_pinned_evict
[18:49:30] [PASSED] ttm_bo_validate_allowed_only_evict
[18:49:30] [PASSED] ttm_bo_validate_deleted_evict
[18:49:30] [PASSED] ttm_bo_validate_busy_domain_evict
[18:49:30] [PASSED] ttm_bo_validate_evict_gutting
[18:49:30] [PASSED] ttm_bo_validate_recrusive_evict
[18:49:30] ================= [PASSED] ttm_bo_validate =================
[18:49:30] ============================================================
[18:49:30] Testing complete. Ran 102 tests: passed: 102
[18:49:30] Elapsed time: 12.310s total, 1.863s configuring, 10.182s building, 0.229s running

+ for kcfg in "${kunitconfigs[@]}"
+ [[ ! -f /kernel/drivers/dma-buf/.kunitconfig ]]
+ /kernel/tools/testing/kunit/kunit.py run --kunitconfig /kernel/drivers/dma-buf/.kunitconfig
[18:49:30] Configuring KUnit Kernel ...
Regenerating .config ...
Populating config with:
$ make ARCH=um O=.kunit olddefconfig
[18:49:32] Building KUnit Kernel ...
Populating config with:
$ make ARCH=um O=.kunit olddefconfig
Building with:
$ make all compile_commands.json scripts_gdb ARCH=um O=.kunit --jobs=48
[18:49:40] Starting KUnit Kernel (1/1)...
[18:49:40] ============================================================
Running tests with:
$ .kunit/linux kunit.enable=1 mem=1G console=tty kunit_shutdown=halt
[18:49:41] =============== dma-buf-fence (12 subtests) ================
[18:49:41] [PASSED] test_sanitycheck
[18:49:41] [PASSED] test_signaling
[18:49:41] [PASSED] test_add_callback
[18:49:41] [PASSED] test_late_add_callback
[18:49:41] [PASSED] test_rm_callback
[18:49:41] [PASSED] test_late_rm_callback
[18:49:41] [PASSED] test_status
[18:49:41] [PASSED] test_error
[18:49:41] [PASSED] test_wait
[18:49:41] [PASSED] test_wait_timeout
[18:49:41] [PASSED] test_stub
[18:49:41] [SKIPPED] test_race_signal_callback (requires at least 2 CPUs)
[18:49:41] ================== [PASSED] dma-buf-fence ==================
[18:49:41] ============ dma-buf-fence-chain (11 subtests) =============
[18:49:41] [PASSED] test_sanitycheck
[18:49:41] [PASSED] test_find_seqno
[18:49:41] [PASSED] test_find_signaled
[18:49:41] [PASSED] test_find_out_of_order
[18:49:46] [PASSED] test_find_gap
[18:49:46] [PASSED] test_find_race
[18:49:46] [PASSED] test_signal_forward
[18:49:46] [PASSED] test_signal_backward
[18:49:46] [PASSED] test_wait_forward
[18:49:46] [PASSED] test_wait_backward
[18:49:46] [PASSED] test_wait_random
[18:49:46] =============== [PASSED] dma-buf-fence-chain ===============
[18:49:46] ============ dma-buf-fence-unwrap (11 subtests) ============
[18:49:46] [PASSED] test_sanitycheck
[18:49:46] [PASSED] test_unwrap_array
[18:49:46] [PASSED] test_unwrap_chain
[18:49:46] [PASSED] test_unwrap_chain_array
[18:49:46] [PASSED] test_unwrap_merge
[18:49:46] [PASSED] test_unwrap_merge_duplicate
[18:49:46] [PASSED] test_unwrap_merge_seqno
[18:49:46] [PASSED] test_unwrap_merge_order
[18:49:46] [PASSED] test_unwrap_merge_complex
[18:49:46] [PASSED] test_unwrap_merge_complex_seqno
[18:49:46] [PASSED] test_dedup_empty
[18:49:46] ============== [PASSED] dma-buf-fence-unwrap ===============
[18:49:46] ================ dma-buf-resv (5 subtests) =================
[18:49:46] [PASSED] test_sanitycheck
[18:49:46] ===================== test_signaling  ======================
[18:49:46] [PASSED] kernel
[18:49:46] [PASSED] write
[18:49:46] [PASSED] read
[18:49:46] [PASSED] bookkeep
[18:49:46] ================= [PASSED] test_signaling ==================
[18:49:46] ====================== test_for_each  ======================
[18:49:46] [PASSED] kernel
[18:49:46] [PASSED] write
[18:49:46] [PASSED] read
[18:49:46] [PASSED] bookkeep
[18:49:46] ================== [PASSED] test_for_each ==================
[18:49:46] ================= test_for_each_unlocked  ==================
[18:49:46] [PASSED] kernel
[18:49:46] [PASSED] write
[18:49:46] [PASSED] read
[18:49:46] [PASSED] bookkeep
[18:49:46] ============= [PASSED] test_for_each_unlocked ==============
[18:49:46] ===================== test_get_fences  =====================
[18:49:46] [PASSED] kernel
[18:49:46] [PASSED] write
[18:49:46] [PASSED] read
[18:49:46] [PASSED] bookkeep
[18:49:46] ================= [PASSED] test_get_fences =================
[18:49:46] ================== [PASSED] dma-buf-resv ===================
[18:49:46] ============================================================
[18:49:46] Testing complete. Ran 51 tests: passed: 50, skipped: 1
[18:49:46] Elapsed time: 16.065s total, 1.929s configuring, 8.764s building, 5.335s running

+ cleanup
++ stat -c %u:%g /kernel
+ chown -R 1003:1003 /kernel

Re: [PATCH v4 1/5] drm/xe: Bump the LRU of every BO an exec touches

[sashiko-bot]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] drm/xe: xe_vm_bump_lru() never visits private BOs in drm_exec loop
--

commit 035844b594106f3b21b07d586a5574fd0b1e0af5
Author: Matthew Brost <matthew.brost@intel.com>

drm/xe: Bump the LRU of every BO an exec touches

This commit updates xe_vm_bump_lru() to replace the drm_gpuvm_is_extobj()
check in its drm_exec_for_each_locked_object() loop with explicit checks
skipping xe_vm_obj(vm) and BOs whose bulk_move matches vm->lru_bulk_move,
intending to bump non-bulk-moved private BOs such as those in fault-mode VMs.

> diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
> index 1ee72721b2ce3..2b84721804a77 100644
> --- a/drivers/gpu/drm/xe/xe_vm.c
> +++ b/drivers/gpu/drm/xe/xe_vm.c
> @@ -426,10 +426,8 @@ int xe_vm_validate_rebind(struct xe_vm *vm, struct drm_exec *exec,
>   * @vm: The vm for which we are bumping LRUs.
>   * @exec: The struct drm_exec with the locked GEM objects.
>   *
> - * Move the VM's private BOs, and all external objects locked by @exec,
> - * to the LRU tail. Private BOs are moved in bulk using @vm's
> - * &xe_vm.lru_bulk_move. External objects, which may be shared with other
> - * VMs, are not part of that bulk move and are bumped individually.
> + * Move all BOs locked by @exec to the LRU tail. Those covered by @vm's
> + * &xe_vm.lru_bulk_move are moved in bulk, the rest individually.
>   */
>  void xe_vm_bump_lru(struct xe_vm *vm, struct drm_exec *exec)
>  {
> @@ -439,10 +437,17 @@ void xe_vm_bump_lru(struct xe_vm *vm, struct drm_exec *exec)
>  	spin_lock(lru_lock);
>  	ttm_lru_bulk_move_tail(&vm->lru_bulk_move);
>  	drm_exec_for_each_locked_object(exec, obj) {
> -		if (!drm_gpuvm_is_extobj(&vm->gpuvm, obj))
> +		struct xe_bo *bo;
> +
> +		/* The VM's dummy resv object, not a BO. */
> +		if (obj == xe_vm_obj(vm))
> +			continue;
> +
> +		bo = gem_to_xe_bo(obj);
> +		if (bo->ttm.bulk_move == &vm->lru_bulk_move)
>  			continue;
>  
> -		ttm_bo_move_to_lru_tail(&gem_to_xe_bo(obj)->ttm);
> +		ttm_bo_move_to_lru_tail(&bo->ttm);

[Severity: Medium]
Can drm_exec_for_each_locked_object() in xe_vm_bump_lru() ever visit a
private BO of the VM?

In drivers/gpu/drm/xe/xe_bo.c:__xe_bo_create_locked(), private BOs (where
vm is non-null) are initialized with xe_vm_resv(vm) as their reservation
object:

	bo = xe_bo_init_locked(xe, bo, tile, vm ? xe_vm_resv(vm) : NULL,
			       vm && !xe_vm_in_fault_mode(vm) &&
			       flags & XE_BO_FLAG_USER ?
			       &vm->lru_bulk_move : NULL, size,
			       cpu_caching, type, flags, NULL, exec);

Because private BOs share the VM's dma_resv, drm_gpuvm_is_extobj() is false
and they are never added to gpuvm->extobj.list.

When callers prepare the drm_exec context before calling xe_vm_bump_lru()
(via xe_validation_exec_lock() -> drm_gpuvm_exec_lock() in xe_exec_ioctl()
or xe_preempt_work_begin() in preempt_rebind_work_func()),
drm_gpuvm_prepare_vm() only locks xe_vm_obj(vm) and
drm_gpuvm_prepare_objects_locked() only locks external objects from
gpuvm->extobj.list:

	drm_gpuvm_resv_assert_held(gpuvm);
	list_for_each_entry(vm_bo, &gpuvm->extobj.list, list.entry.extobj) {
		if (drm_gpuvm_bo_is_zombie(vm_bo))
			continue;

		ret = exec_prepare_obj(exec, vm_bo->obj, num_fences);

In addition, even when the follow-up commit a6d00986d9f1 ("drm/xe: Bump the
LRU on a successful page fault") calls xe_vm_bump_lru() from
xe_pagefault_handle_vma(), xe_pagefault_begin() -> xe_vm_lock_vma() only
locks xe_vm_obj(vm) and external BOs:

	err = drm_exec_lock_obj(exec, xe_vm_obj(vm));
	if (!err && bo && !bo->vm)
		err = drm_exec_lock_obj(exec, &bo->ttm.base);

Since private BOs are never recorded in exec->objects, the loop in
xe_vm_bump_lru() only ever visits xe_vm_obj(vm) and external BOs (where
bo->vm and bo->ttm.bulk_move are always null). Does this mean the
bo->ttm.bulk_move == &vm->lru_bulk_move check is unreachable and private BOs
without a bulk_move (such as fault-mode private BOs or private BOs detached
from bulk_move in xe_gem_object_close()) still never get their LRU bumped?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20261027183638.1031299-1-matthew.brost@intel.com?part=1

Re: [PATCH v4 5/5] drm/xe: Add a minimum age gate to the shrinker

[sashiko-bot]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] drm/xe: fault-mode VM-private BOs are not stamped or aged in xe_vm_bump_lru()
--

commit 169ae3636dd7a7f18b87dc4d22d9738909855961
Author: Matthew Brost <matthew.brost@intel.com>

drm/xe: Add a minimum age gate to the shrinker

This commit adds a per-device shrinker_min_age_ms sysfs attribute and
timestamps buffer objects on LRU updates so the shrinker avoids backing up
recently used objects. Buffer objects in a VM's bulk move use the VM-wide
timestamp updated in xe_vm_bump_lru(), while other buffer objects use their
own timestamp updated in xe_bo_move() and xe_bo_bump_lru_locked().

> diff --git a/drivers/gpu/drm/xe/xe_bo.c b/drivers/gpu/drm/xe/xe_bo.c
> index ba16b260dbcb7..59874f9a1ed7a 100644
> --- a/drivers/gpu/drm/xe/xe_bo.c
> +++ b/drivers/gpu/drm/xe/xe_bo.c
> @@ -1357,9 +1357,92 @@ static int xe_bo_move(struct ttm_buffer_object *ttm_bo, bool evict,

[ ... ]

> +bool xe_bo_shrink_too_young(struct ttm_buffer_object *bo, unsigned long min_age)
> +{
> +	struct xe_bo *xe_bo;
> +	unsigned long stamp;
> +
> +	if (!min_age || bo->resource->mem_type == XE_PL_SYSTEM ||
> +	    !xe_bo_shrink_backup_candidate(bo))
> +		return false;
> +
> +	xe_bo = ttm_to_xe_bo(bo);
> +	if (bo->bulk_move && xe_bo->vm)
> +		stamp = READ_ONCE(xe_bo->vm->lru_stamp);
> +	else
> +		stamp = READ_ONCE(xe_bo->lru_stamp);
> +
> +	return stamp && time_before(jiffies, stamp + min_age);
> +}

[Severity: Medium]
Does this leave VM-private buffer objects in fault-mode VMs with a stale
lru_stamp when they are faulted and rebound through xe_vm_bump_lru()?

In drivers/gpu/drm/xe/xe_bo.c, __xe_bo_create_locked() initializes
VM-private user buffer objects in fault-mode VMs with bo->vm set, resv set
to xe_vm_resv(vm), and bulk_move set to NULL:

drivers/gpu/drm/xe/xe_bo.c:__xe_bo_create_locked() {
	...
	bo = xe_bo_init_locked(xe, bo, tile, vm ? xe_vm_resv(vm) : NULL,
			       vm && !xe_vm_in_fault_mode(vm) &&
			       flags & XE_BO_FLAG_USER ?
			       &vm->lru_bulk_move : NULL, size,
			       cpu_caching, type, flags, NULL, exec);
	...
}

When a GPU page fault is handled via
xe_pagefault_handle_vma()->xe_pagefault_begin()->xe_vm_lock_vma(),
drivers/gpu/drm/xe/xe_vm.c only adds the buffer object to drm_exec when
bo->vm is NULL, since VM-private objects share xe_vm_obj(vm)'s reservation:

drivers/gpu/drm/xe/xe_vm.c:xe_vm_lock_vma() {
	...
	err = drm_exec_lock_obj(exec, xe_vm_obj(vm));
	if (!err && bo && !bo->vm)
		err = drm_exec_lock_obj(exec, &bo->ttm.base);
	...
}

When xe_pagefault_handle_vma() then calls xe_vm_bump_lru(vm, &exec),
vm->lru_stamp is updated, but drm_exec_for_each_locked_object(exec, obj)
only sees xe_vm_obj(vm) (which is skipped) and any external objects, so
xe_bo_bump_lru_locked(bo) is never called for VM-private objects.

When the shrinker later calls xe_bo_shrink_too_young(), bo->bulk_move is
NULL for fault-mode VM-private objects, so it reads xe_bo->lru_stamp
rather than xe_bo->vm->lru_stamp. If the object was already resident in
XE_PL_TT when faulted (for example, on another tile or after VMA
invalidation), won't its age remain stuck at the initial xe_bo_move()
timestamp rather than reflecting the fault activity?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20261027183638.1031299-1-matthew.brost@intel.com?part=5

Re: [PATCH v4 3/5] drm/xe: Leave the VM bulk move on BO release, not on GEM close

[Matthew Brost]

On Tue, Oct 27, 2026 at 06:47:57PM +0000, sashiko-bot@kernel.org wrote:
> Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
> - [High] drm/xe: bulk_move detach in xe_ttm_bo_release_notify() violates TTM invariants
> --
> 
> commit 5d85520c1e2e507cc032c4248bd48c50a734e710
> Author: Matthew Brost <matthew.brost@intel.com>
> 
> drm/xe: Leave the VM bulk move on BO release, not on GEM close
> 
> This commit removes xe_gem_object_close() and moves the detachment of a
> VM-private BO from its VM's LRU bulk move into xe_ttm_bo_release_notify()
> so that BOs outliving their GEM handles remain part of the VM bulk move
> until release.
> 
> > diff --git a/drivers/gpu/drm/xe/xe_bo.c b/drivers/gpu/drm/xe/xe_bo.c
> > index 42a6aa1a2e4a0..ba16b260dbcb7 100644
> > --- a/drivers/gpu/drm/xe/xe_bo.c
> > +++ b/drivers/gpu/drm/xe/xe_bo.c
> > @@ -1869,6 +1869,14 @@ static void xe_ttm_bo_release_notify(struct ttm_buffer_object *ttm_bo)
> >  	if (!xe_ttm_bo_lock_in_destructor(ttm_bo))
> >  		return;
> >  
> > +	/*
> > +	 * Leave the VM's bulk move, which is only kept alive by the VM, and
> > +	 * whose LRU position no longer says anything about a BO on its way
> > +	 * out.
> > +	 */
> > +	if (ttm_bo->bulk_move)
> > +		ttm_bo_set_bulk_move(ttm_bo, NULL);
> 
> [Severity: High]

Ugh, this is correct - this patch doesn't work but the code prior to
this patch is also incorrect.

I'm not immediately spoting an obvisous solution, so I will just drop
this patch for now.

Matt

> Does detaching from bulk_move in xe_ttm_bo_release_notify() happen too late
> in the TTM teardown sequence?
> 
> When the final reference to a VM-private user BO on a non-fault-mode VM is
> dropped via xe_gem_object_free() -> ttm_bo_fini() -> ttm_bo_put() ->
> ttm_bo_release(), ttm_bo_release() checks bo->bulk_move and individualizes
> bo->base.resv before invoking bdev->funcs->release_notify():
> 
> drivers/gpu/drm/ttm/ttm_bo.c:ttm_bo_release() {
>     ...
> 	WARN_ON_ONCE(bo->pin_count);
> 	WARN_ON_ONCE(bo->bulk_move);
> 
> 	if (!bo->deleted) {
> 		ret = ttm_bo_individualize_resv(bo);
> 		if (ret) {
> 			/* Last resort, if we fail to allocate memory for the
> 			 * fences block for the BO to become idle
> 			 */
> 			dma_resv_wait_timeout(bo->base.resv,
> 					      DMA_RESV_USAGE_BOOKKEEP, false,
> 					      30 * HZ);
> 		}
> 
> 		if (bdev->funcs->release_notify)
> 			bdev->funcs->release_notify(bo);
>     ...
> }
> 
> Because bo->bulk_move is still set upon entering ttm_bo_release(), releasing
> such a BO will immediately trigger WARN_ON_ONCE(bo->bulk_move).
> 
> In addition, ttm_bo_individualize_resv() switches bo->base.resv away from
> xe_vm_resv(vm) to &bo->base._resv under bo->bdev->lru_lock and then drops
> lru_lock while the BO's resource is still linked in vm->lru_bulk_move:
> 
> drivers/gpu/drm/ttm/ttm_bo.c:ttm_bo_individualize_resv() {
>     ...
> 		spin_lock(&bo->bdev->lru_lock);
> 		bo->base.resv = &bo->base._resv;
> 		spin_unlock(&bo->bdev->lru_lock);
>     ...
> }
> 
> In the window between ttm_bo_individualize_resv() and
> ttm_bo_set_bulk_move(ttm_bo, NULL) in xe_ttm_bo_release_notify(), another
> thread holding xe_vm_resv(vm) can access vm->lru_bulk_move while the dying
> BO is still pos->first or pos->last:
> 
> - Calling xe_vm_bump_lru() -> ttm_lru_bulk_move_tail() will fail
>   dma_resv_assert_held(pos->first->bo->base.resv) or
>   dma_resv_assert_held(pos->last->bo->base.resv) in
>   drivers/gpu/drm/ttm/ttm_resource.c:
> 
> drivers/gpu/drm/ttm/ttm_resource.c:ttm_lru_bulk_move_tail() {
>     ...
> 			lockdep_assert_held(&pos->first->bo->bdev->lru_lock);
> 			dma_resv_assert_held(pos->first->bo->base.resv);
> 			dma_resv_assert_held(pos->last->bo->base.resv);
>     ...
> }
> 
> - Adding or moving another resource in the same bucket via
>   ttm_resource_add_bulk_move() -> ttm_lru_bulk_move_add() will trigger
>   WARN_ON(pos->first->bo->base.resv != res->bo->base.resv):
> 
> drivers/gpu/drm/ttm/ttm_resource.c:ttm_lru_bulk_move_add() {
>     ...
> 	} else {
> 		WARN_ON(pos->first->bo->base.resv != res->bo->base.resv);
> 		ttm_lru_bulk_move_pos_tail(pos, res);
> 	}
> }
> 
> Conversely, if ttm_bo_individualize_resv() fails (for example, if
> dma_resv_copy_fences() returns -ENOMEM), bo->base.resv remains
> xe_vm_resv(vm), while xe_ttm_bo_lock_in_destructor() only trylocks
> &ttm_bo->base._resv. Calling ttm_bo_set_bulk_move(ttm_bo, NULL) in
> xe_ttm_bo_release_notify() will then fail dma_resv_assert_held(bo->base.resv)
> in ttm_bo_set_bulk_move() and ttm_resource_is_swapped(), and mutate
> vm->lru_bulk_move without holding xe_vm_resv(vm).
> 
> Should bulk_move be cleared before ttm_bo_release() runs (or before
> ttm_bo_individualize_resv() detaches bo->base.resv from the VM reservation
> object) while holding the VM reservation lock?
> 
> -- 
> Sashiko AI review · https://sashiko.dev/#/patchset/20261027183638.1031299-1-matthew.brost@intel.com?part=3

✗ Xe.CI.BAT: failure for Add a minimum age gate to the shrinker + LRU fixes

[Patchwork]

[-- Attachment #1: Type: text/plain, Size: 6073 bytes --]

== Series Details ==

Series: Add a minimum age gate to the shrinker + LRU fixes
URL   : https://patchwork.freedesktop.org/series/176975/
State : failure

== Summary ==

CI Bug Log - changes from xe-6016-5e5b89446954adf0d9b74ba4228862403f71eec4_BAT -> xe-pw-176975v1_BAT
====================================================

Summary
-------

  **FAILURE**

  Serious unknown changes coming with xe-pw-176975v1_BAT absolutely need to be
  verified manually.
  
  If you think the reported changes have nothing to do with the changes
  introduced in xe-pw-176975v1_BAT, please notify your bug team (I915-ci-infra@lists.freedesktop.org) to allow them
  to document this new failure mode, which will reduce false positives in CI.

  

Participating hosts (15 -> 15)
------------------------------

  No changes in participating hosts

Possible new issues
-------------------

  Here are the unknown changes that may have been introduced in xe-pw-176975v1_BAT:

### IGT changes ###

#### Possible regressions ####

  * igt@xe_compute@compute-square:
    - bat-ptl-1:          [PASS][1] -> [ABORT][2]
   [1]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-6016-5e5b89446954adf0d9b74ba4228862403f71eec4/bat-ptl-1/igt@xe_compute@compute-square.html
   [2]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-176975v1/bat-ptl-1/igt@xe_compute@compute-square.html
    - bat-ptl-vm:         [PASS][3] -> [ABORT][4]
   [3]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-6016-5e5b89446954adf0d9b74ba4228862403f71eec4/bat-ptl-vm/igt@xe_compute@compute-square.html
   [4]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-176975v1/bat-ptl-vm/igt@xe_compute@compute-square.html
    - bat-lnl-1:          [PASS][5] -> [ABORT][6]
   [5]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-6016-5e5b89446954adf0d9b74ba4228862403f71eec4/bat-lnl-1/igt@xe_compute@compute-square.html
   [6]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-176975v1/bat-lnl-1/igt@xe_compute@compute-square.html
    - bat-bmg-vm:         [PASS][7] -> [ABORT][8]
   [7]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-6016-5e5b89446954adf0d9b74ba4228862403f71eec4/bat-bmg-vm/igt@xe_compute@compute-square.html
   [8]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-176975v1/bat-bmg-vm/igt@xe_compute@compute-square.html
    - bat-bmg-2:          [PASS][9] -> [ABORT][10]
   [9]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-6016-5e5b89446954adf0d9b74ba4228862403f71eec4/bat-bmg-2/igt@xe_compute@compute-square.html
   [10]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-176975v1/bat-bmg-2/igt@xe_compute@compute-square.html
    - bat-adlp-7:         [PASS][11] -> [ABORT][12]
   [11]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-6016-5e5b89446954adf0d9b74ba4228862403f71eec4/bat-adlp-7/igt@xe_compute@compute-square.html
   [12]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-176975v1/bat-adlp-7/igt@xe_compute@compute-square.html
    - bat-ptl-2:          [PASS][13] -> [ABORT][14]
   [13]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-6016-5e5b89446954adf0d9b74ba4228862403f71eec4/bat-ptl-2/igt@xe_compute@compute-square.html
   [14]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-176975v1/bat-ptl-2/igt@xe_compute@compute-square.html
    - bat-atsm-2:         [PASS][15] -> [ABORT][16]
   [15]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-6016-5e5b89446954adf0d9b74ba4228862403f71eec4/bat-atsm-2/igt@xe_compute@compute-square.html
   [16]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-176975v1/bat-atsm-2/igt@xe_compute@compute-square.html
    - bat-wcl-2:          [PASS][17] -> [ABORT][18]
   [17]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-6016-5e5b89446954adf0d9b74ba4228862403f71eec4/bat-wcl-2/igt@xe_compute@compute-square.html
   [18]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-176975v1/bat-wcl-2/igt@xe_compute@compute-square.html
    - bat-bmg-1:          [PASS][19] -> [ABORT][20]
   [19]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-6016-5e5b89446954adf0d9b74ba4228862403f71eec4/bat-bmg-1/igt@xe_compute@compute-square.html
   [20]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-176975v1/bat-bmg-1/igt@xe_compute@compute-square.html
    - bat-nvls-1:         [PASS][21] -> [ABORT][22]
   [21]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-6016-5e5b89446954adf0d9b74ba4228862403f71eec4/bat-nvls-1/igt@xe_compute@compute-square.html
   [22]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-176975v1/bat-nvls-1/igt@xe_compute@compute-square.html
    - bat-lnl-2:          [PASS][23] -> [ABORT][24]
   [23]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-6016-5e5b89446954adf0d9b74ba4228862403f71eec4/bat-lnl-2/igt@xe_compute@compute-square.html
   [24]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-176975v1/bat-lnl-2/igt@xe_compute@compute-square.html
    - bat-nvls-2:         [PASS][25] -> [ABORT][26]
   [25]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-6016-5e5b89446954adf0d9b74ba4228862403f71eec4/bat-nvls-2/igt@xe_compute@compute-square.html
   [26]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-176975v1/bat-nvls-2/igt@xe_compute@compute-square.html
    - bat-wcl-1:          [PASS][27] -> [ABORT][28]
   [27]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-6016-5e5b89446954adf0d9b74ba4228862403f71eec4/bat-wcl-1/igt@xe_compute@compute-square.html
   [28]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-176975v1/bat-wcl-1/igt@xe_compute@compute-square.html

  * igt@xe_exec_basic@multigpu-once-basic:
    - bat-bmg-3:          [PASS][29] -> [ABORT][30]
   [29]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-6016-5e5b89446954adf0d9b74ba4228862403f71eec4/bat-bmg-3/igt@xe_exec_basic@multigpu-once-basic.html
   [30]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-176975v1/bat-bmg-3/igt@xe_exec_basic@multigpu-once-basic.html

  


Build changes
-------------

  * Linux: xe-6016-5e5b89446954adf0d9b74ba4228862403f71eec4 -> xe-pw-176975v1

  IGT_9168: 9168
  xe-6016-5e5b89446954adf0d9b74ba4228862403f71eec4: 5e5b89446954adf0d9b74ba4228862403f71eec4
  xe-pw-176975v1: 176975v1

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-176975v1/index.html

[-- Attachment #2: Type: text/html, Size: 6761 bytes --]

Re: [PATCH v4 3/5] drm/xe: Leave the VM bulk move on BO release, not on GEM close

[Matthew Brost]

On Tue, Oct 27, 2026 at 12:01:33PM -0700, Matthew Brost wrote:
> On Tue, Oct 27, 2026 at 06:47:57PM +0000, sashiko-bot@kernel.org wrote:
> > Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
> > - [High] drm/xe: bulk_move detach in xe_ttm_bo_release_notify() violates TTM invariants
> > --
> > 
> > commit 5d85520c1e2e507cc032c4248bd48c50a734e710
> > Author: Matthew Brost <matthew.brost@intel.com>
> > 
> > drm/xe: Leave the VM bulk move on BO release, not on GEM close
> > 
> > This commit removes xe_gem_object_close() and moves the detachment of a
> > VM-private BO from its VM's LRU bulk move into xe_ttm_bo_release_notify()
> > so that BOs outliving their GEM handles remain part of the VM bulk move
> > until release.
> > 
> > > diff --git a/drivers/gpu/drm/xe/xe_bo.c b/drivers/gpu/drm/xe/xe_bo.c
> > > index 42a6aa1a2e4a0..ba16b260dbcb7 100644
> > > --- a/drivers/gpu/drm/xe/xe_bo.c
> > > +++ b/drivers/gpu/drm/xe/xe_bo.c
> > > @@ -1869,6 +1869,14 @@ static void xe_ttm_bo_release_notify(struct ttm_buffer_object *ttm_bo)
> > >  	if (!xe_ttm_bo_lock_in_destructor(ttm_bo))
> > >  		return;
> > >  
> > > +	/*
> > > +	 * Leave the VM's bulk move, which is only kept alive by the VM, and
> > > +	 * whose LRU position no longer says anything about a BO on its way
> > > +	 * out.
> > > +	 */
> > > +	if (ttm_bo->bulk_move)
> > > +		ttm_bo_set_bulk_move(ttm_bo, NULL);
> > 
> > [Severity: High]
> 
> Ugh, this is correct - this patch doesn't work but the code prior to
> this patch is also incorrect.
> 
> I'm not immediately spoting an obvisous solution, so I will just drop
> this patch for now.
> 

So it probably is;

 - If in xe_gem_object_close if bind count == 0, remove from bulk
 - If in a bind operation, bind count == 0 and gem handle closed, remove from bulk

I will do this part in an independent follow up.

Matt

> Matt
> 
> > Does detaching from bulk_move in xe_ttm_bo_release_notify() happen too late
> > in the TTM teardown sequence?
> > 
> > When the final reference to a VM-private user BO on a non-fault-mode VM is
> > dropped via xe_gem_object_free() -> ttm_bo_fini() -> ttm_bo_put() ->
> > ttm_bo_release(), ttm_bo_release() checks bo->bulk_move and individualizes
> > bo->base.resv before invoking bdev->funcs->release_notify():
> > 
> > drivers/gpu/drm/ttm/ttm_bo.c:ttm_bo_release() {
> >     ...
> > 	WARN_ON_ONCE(bo->pin_count);
> > 	WARN_ON_ONCE(bo->bulk_move);
> > 
> > 	if (!bo->deleted) {
> > 		ret = ttm_bo_individualize_resv(bo);
> > 		if (ret) {
> > 			/* Last resort, if we fail to allocate memory for the
> > 			 * fences block for the BO to become idle
> > 			 */
> > 			dma_resv_wait_timeout(bo->base.resv,
> > 					      DMA_RESV_USAGE_BOOKKEEP, false,
> > 					      30 * HZ);
> > 		}
> > 
> > 		if (bdev->funcs->release_notify)
> > 			bdev->funcs->release_notify(bo);
> >     ...
> > }
> > 
> > Because bo->bulk_move is still set upon entering ttm_bo_release(), releasing
> > such a BO will immediately trigger WARN_ON_ONCE(bo->bulk_move).
> > 
> > In addition, ttm_bo_individualize_resv() switches bo->base.resv away from
> > xe_vm_resv(vm) to &bo->base._resv under bo->bdev->lru_lock and then drops
> > lru_lock while the BO's resource is still linked in vm->lru_bulk_move:
> > 
> > drivers/gpu/drm/ttm/ttm_bo.c:ttm_bo_individualize_resv() {
> >     ...
> > 		spin_lock(&bo->bdev->lru_lock);
> > 		bo->base.resv = &bo->base._resv;
> > 		spin_unlock(&bo->bdev->lru_lock);
> >     ...
> > }
> > 
> > In the window between ttm_bo_individualize_resv() and
> > ttm_bo_set_bulk_move(ttm_bo, NULL) in xe_ttm_bo_release_notify(), another
> > thread holding xe_vm_resv(vm) can access vm->lru_bulk_move while the dying
> > BO is still pos->first or pos->last:
> > 
> > - Calling xe_vm_bump_lru() -> ttm_lru_bulk_move_tail() will fail
> >   dma_resv_assert_held(pos->first->bo->base.resv) or
> >   dma_resv_assert_held(pos->last->bo->base.resv) in
> >   drivers/gpu/drm/ttm/ttm_resource.c:
> > 
> > drivers/gpu/drm/ttm/ttm_resource.c:ttm_lru_bulk_move_tail() {
> >     ...
> > 			lockdep_assert_held(&pos->first->bo->bdev->lru_lock);
> > 			dma_resv_assert_held(pos->first->bo->base.resv);
> > 			dma_resv_assert_held(pos->last->bo->base.resv);
> >     ...
> > }
> > 
> > - Adding or moving another resource in the same bucket via
> >   ttm_resource_add_bulk_move() -> ttm_lru_bulk_move_add() will trigger
> >   WARN_ON(pos->first->bo->base.resv != res->bo->base.resv):
> > 
> > drivers/gpu/drm/ttm/ttm_resource.c:ttm_lru_bulk_move_add() {
> >     ...
> > 	} else {
> > 		WARN_ON(pos->first->bo->base.resv != res->bo->base.resv);
> > 		ttm_lru_bulk_move_pos_tail(pos, res);
> > 	}
> > }
> > 
> > Conversely, if ttm_bo_individualize_resv() fails (for example, if
> > dma_resv_copy_fences() returns -ENOMEM), bo->base.resv remains
> > xe_vm_resv(vm), while xe_ttm_bo_lock_in_destructor() only trylocks
> > &ttm_bo->base._resv. Calling ttm_bo_set_bulk_move(ttm_bo, NULL) in
> > xe_ttm_bo_release_notify() will then fail dma_resv_assert_held(bo->base.resv)
> > in ttm_bo_set_bulk_move() and ttm_resource_is_swapped(), and mutate
> > vm->lru_bulk_move without holding xe_vm_resv(vm).
> > 
> > Should bulk_move be cleared before ttm_bo_release() runs (or before
> > ttm_bo_individualize_resv() detaches bo->base.resv from the VM reservation
> > object) while holding the VM reservation lock?
> > 
> > -- 
> > Sashiko AI review · https://sashiko.dev/#/patchset/20261027183638.1031299-1-matthew.brost@intel.com?part=3

Re: [PATCH v4 1/5] drm/xe: Bump the LRU of every BO an exec touches

[Matthew Brost]

On Tue, Oct 27, 2026 at 06:59:16PM +0000, sashiko-bot@kernel.org wrote:
> Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
> - [Medium] drm/xe: xe_vm_bump_lru() never visits private BOs in drm_exec loop
> --
> 
> commit 035844b594106f3b21b07d586a5574fd0b1e0af5
> Author: Matthew Brost <matthew.brost@intel.com>
> 
> drm/xe: Bump the LRU of every BO an exec touches
> 
> This commit updates xe_vm_bump_lru() to replace the drm_gpuvm_is_extobj()
> check in its drm_exec_for_each_locked_object() loop with explicit checks
> skipping xe_vm_obj(vm) and BOs whose bulk_move matches vm->lru_bulk_move,
> intending to bump non-bulk-moved private BOs such as those in fault-mode VMs.
> 
> > diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
> > index 1ee72721b2ce3..2b84721804a77 100644
> > --- a/drivers/gpu/drm/xe/xe_vm.c
> > +++ b/drivers/gpu/drm/xe/xe_vm.c
> > @@ -426,10 +426,8 @@ int xe_vm_validate_rebind(struct xe_vm *vm, struct drm_exec *exec,
> >   * @vm: The vm for which we are bumping LRUs.
> >   * @exec: The struct drm_exec with the locked GEM objects.
> >   *
> > - * Move the VM's private BOs, and all external objects locked by @exec,
> > - * to the LRU tail. Private BOs are moved in bulk using @vm's
> > - * &xe_vm.lru_bulk_move. External objects, which may be shared with other
> > - * VMs, are not part of that bulk move and are bumped individually.
> > + * Move all BOs locked by @exec to the LRU tail. Those covered by @vm's
> > + * &xe_vm.lru_bulk_move are moved in bulk, the rest individually.
> >   */
> >  void xe_vm_bump_lru(struct xe_vm *vm, struct drm_exec *exec)
> >  {
> > @@ -439,10 +437,17 @@ void xe_vm_bump_lru(struct xe_vm *vm, struct drm_exec *exec)
> >  	spin_lock(lru_lock);
> >  	ttm_lru_bulk_move_tail(&vm->lru_bulk_move);
> >  	drm_exec_for_each_locked_object(exec, obj) {
> > -		if (!drm_gpuvm_is_extobj(&vm->gpuvm, obj))
> > +		struct xe_bo *bo;
> > +
> > +		/* The VM's dummy resv object, not a BO. */
> > +		if (obj == xe_vm_obj(vm))
> > +			continue;
> > +
> > +		bo = gem_to_xe_bo(obj);
> > +		if (bo->ttm.bulk_move == &vm->lru_bulk_move)
> >  			continue;
> >  
> > -		ttm_bo_move_to_lru_tail(&gem_to_xe_bo(obj)->ttm);
> > +		ttm_bo_move_to_lru_tail(&bo->ttm);
> 
> [Severity: Medium]
> Can drm_exec_for_each_locked_object() in xe_vm_bump_lru() ever visit a
> private BO of the VM?
> 
> In drivers/gpu/drm/xe/xe_bo.c:__xe_bo_create_locked(), private BOs (where
> vm is non-null) are initialized with xe_vm_resv(vm) as their reservation
> object:
> 
> 	bo = xe_bo_init_locked(xe, bo, tile, vm ? xe_vm_resv(vm) : NULL,
> 			       vm && !xe_vm_in_fault_mode(vm) &&
> 			       flags & XE_BO_FLAG_USER ?
> 			       &vm->lru_bulk_move : NULL, size,
> 			       cpu_caching, type, flags, NULL, exec);
> 
> Because private BOs share the VM's dma_resv, drm_gpuvm_is_extobj() is false
> and they are never added to gpuvm->extobj.list.
> 
> When callers prepare the drm_exec context before calling xe_vm_bump_lru()
> (via xe_validation_exec_lock() -> drm_gpuvm_exec_lock() in xe_exec_ioctl()
> or xe_preempt_work_begin() in preempt_rebind_work_func()),
> drm_gpuvm_prepare_vm() only locks xe_vm_obj(vm) and
> drm_gpuvm_prepare_objects_locked() only locks external objects from
> gpuvm->extobj.list:
> 
> 	drm_gpuvm_resv_assert_held(gpuvm);
> 	list_for_each_entry(vm_bo, &gpuvm->extobj.list, list.entry.extobj) {
> 		if (drm_gpuvm_bo_is_zombie(vm_bo))
> 			continue;
> 
> 		ret = exec_prepare_obj(exec, vm_bo->obj, num_fences);
> 
> In addition, even when the follow-up commit a6d00986d9f1 ("drm/xe: Bump the
> LRU on a successful page fault") calls xe_vm_bump_lru() from
> xe_pagefault_handle_vma(), xe_pagefault_begin() -> xe_vm_lock_vma() only
> locks xe_vm_obj(vm) and external BOs:
> 
> 	err = drm_exec_lock_obj(exec, xe_vm_obj(vm));
> 	if (!err && bo && !bo->vm)
> 		err = drm_exec_lock_obj(exec, &bo->ttm.base);
> 
> Since private BOs are never recorded in exec->objects, the loop in
> xe_vm_bump_lru() only ever visits xe_vm_obj(vm) and external BOs (where
> bo->vm and bo->ttm.bulk_move are always null). Does this mean the
> bo->ttm.bulk_move == &vm->lru_bulk_move check is unreachable and private BOs
> without a bulk_move (such as fault-mode private BOs or private BOs detached
> from bulk_move in xe_gem_object_close()) still never get their LRU bumped?
>

I think this is right. Let me drop this patch + adjust page fault patch
to just direcly call ttm_bo_move_to_lru_tail.

Matt

> -- 
> Sashiko AI review · https://sashiko.dev/#/patchset/20261027183638.1031299-1-matthew.brost@intel.com?part=1