Intel-XE Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Matthew Brost <matthew.brost@intel.com>
To: intel-xe@lists.freedesktop.org
Cc: thomas.hellstrom@linux.intel.com, himal.prasad.ghimiray@intel.com
Subject: [PATCH v3] drm/xe: Make VMA tile_present, tile_invalidated access rules clear
Date: Thu, 29 May 2025 23:14:33 -0700	[thread overview]
Message-ID: <20250530061433.1819041-1-matthew.brost@intel.com> (raw)

Document VMA tile_invalidated access rules, use READ_ONCE / WRITE_ONCE
for opportunistic checks of tile_present and tile_invalidated, move
tile_invalidated state change from page fault handler to PT code under
the correct locks, and add lockdep asserts to TLB invalidation paths.

v2:
 - Assert VM dma-resv lock rather than BO in zap PTEs
v3:
 - Back to BO's dma-resv lock, adjust documentation

Signed-off-by: Matthew Brost <matthew.brost@intel.com>
---
 drivers/gpu/drm/xe/xe_gt_pagefault.c | 11 +++++++----
 drivers/gpu/drm/xe/xe_pt.c           | 16 +++++++++++++---
 drivers/gpu/drm/xe/xe_vm.c           |  1 +
 drivers/gpu/drm/xe/xe_vm_types.h     | 11 +++++++++--
 4 files changed, 30 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_gt_pagefault.c b/drivers/gpu/drm/xe/xe_gt_pagefault.c
index 7a8f87709e39..05fbc83c64b7 100644
--- a/drivers/gpu/drm/xe/xe_gt_pagefault.c
+++ b/drivers/gpu/drm/xe/xe_gt_pagefault.c
@@ -68,8 +68,12 @@ static bool access_is_atomic(enum access_type access_type)
 
 static bool vma_is_valid(struct xe_tile *tile, struct xe_vma *vma)
 {
-	return BIT(tile->id) & vma->tile_present &&
-		!(BIT(tile->id) & vma->tile_invalidated);
+	/*
+	 * Advisory only check whether the VMA currently has a valid mapping,
+	 * READ_ONCE pairs with WRITE_ONCE in xe_pt.c
+	 */
+	return BIT(tile->id) & READ_ONCE(vma->tile_present) &&
+		!(BIT(tile->id) & READ_ONCE(vma->tile_invalidated));
 }
 
 
@@ -121,7 +125,7 @@ static int handle_vma_pagefault(struct xe_gt *gt, struct xe_vma *vma,
 
 	trace_xe_vma_pagefault(vma);
 
-	/* Check if VMA is valid */
+	/* Check if VMA is valid, opportunistic check only */
 	if (vma_is_valid(tile, vma) && !atomic)
 		return 0;
 
@@ -158,7 +162,6 @@ static int handle_vma_pagefault(struct xe_gt *gt, struct xe_vma *vma,
 
 	dma_fence_wait(fence, false);
 	dma_fence_put(fence);
-	vma->tile_invalidated &= ~BIT(tile->id);
 
 unlock_dma_resv:
 	drm_exec_fini(&exec);
diff --git a/drivers/gpu/drm/xe/xe_pt.c b/drivers/gpu/drm/xe/xe_pt.c
index c9c41fbe125c..f39d5cc9f411 100644
--- a/drivers/gpu/drm/xe/xe_pt.c
+++ b/drivers/gpu/drm/xe/xe_pt.c
@@ -907,6 +907,11 @@ bool xe_pt_zap_ptes(struct xe_tile *tile, struct xe_vma *vma)
 	struct xe_pt *pt = xe_vma_vm(vma)->pt_root[tile->id];
 	u8 pt_mask = (vma->tile_present & ~vma->tile_invalidated);
 
+	if (xe_vma_bo(vma))
+		xe_bo_assert_held(xe_vma_bo(vma));
+	else if (xe_vma_is_userptr(vma))
+		lockdep_assert_held(&xe_vma_vm(vma)->userptr.notifier_lock);
+
 	if (!(pt_mask & BIT(tile->id)))
 		return false;
 
@@ -2191,10 +2196,15 @@ static void bind_op_commit(struct xe_vm *vm, struct xe_tile *tile,
 					   DMA_RESV_USAGE_KERNEL :
 					   DMA_RESV_USAGE_BOOKKEEP);
 	}
-	vma->tile_present |= BIT(tile->id);
-	vma->tile_staged &= ~BIT(tile->id);
+	/* All WRITE_ONCE pair with READ_ONCE in xe_gt_pagefault.c */
+	WRITE_ONCE(vma->tile_present, vma->tile_present | BIT(tile->id));
 	if (invalidate_on_bind)
-		vma->tile_invalidated |= BIT(tile->id);
+		WRITE_ONCE(vma->tile_invalidated,
+			   vma->tile_invalidated | BIT(tile->id));
+	else
+		WRITE_ONCE(vma->tile_invalidated,
+			   vma->tile_invalidated & ~BIT(tile->id));
+	vma->tile_staged &= ~BIT(tile->id);
 	if (xe_vma_is_userptr(vma)) {
 		lockdep_assert_held_read(&vm->userptr.notifier_lock);
 		to_userptr_vma(vma)->userptr.initial_bind = true;
diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
index 7140d8856bad..e781e8138f34 100644
--- a/drivers/gpu/drm/xe/xe_vm.c
+++ b/drivers/gpu/drm/xe/xe_vm.c
@@ -3871,6 +3871,7 @@ int xe_vm_invalidate_vma(struct xe_vma *vma)
 	/* Check that we don't race with page-table updates */
 	if (IS_ENABLED(CONFIG_PROVE_LOCKING)) {
 		if (xe_vma_is_userptr(vma)) {
+			lockdep_assert_held_write(&xe_vma_vm(vma)->userptr.notifier_lock);
 			WARN_ON_ONCE(!mmu_interval_check_retry
 				     (&to_userptr_vma(vma)->userptr.notifier,
 				      to_userptr_vma(vma)->userptr.notifier_seq));
diff --git a/drivers/gpu/drm/xe/xe_vm_types.h b/drivers/gpu/drm/xe/xe_vm_types.h
index bfc145baad49..f3fb29b4d525 100644
--- a/drivers/gpu/drm/xe/xe_vm_types.h
+++ b/drivers/gpu/drm/xe/xe_vm_types.h
@@ -100,14 +100,21 @@ struct xe_vma {
 		struct work_struct destroy_work;
 	};
 
-	/** @tile_invalidated: VMA has been invalidated */
+	/**
+	 * @tile_invalidated: Tile mask of binding are invalidated for this VMA.
+	 * protected by BO's resv and for userptrs, vm->userptr.notifier_lock in
+	 * write mode for writing or vm->userptr.notifier_lock in read mode and
+	 * the vm->resv. For stable reading, BO's resv or useptr
+	 * vm->userptr.notifier_lock in read mode is required. Can be
+	 * opportunisticly read with READ_ONCE outside of locks.
+	 */
 	u8 tile_invalidated;
 
 	/** @tile_mask: Tile mask of where to create binding for this VMA */
 	u8 tile_mask;
 
 	/**
-	 * @tile_present: GT mask of binding are present for this VMA.
+	 * @tile_present: Tile mask of binding are present for this VMA.
 	 * protected by vm->lock, vm->resv and for userptrs,
 	 * vm->userptr.notifier_lock for writing. Needs either for reading,
 	 * but if reading is done under the vm->lock only, it needs to be held
-- 
2.34.1


             reply	other threads:[~2025-05-30  6:13 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-05-30  6:14 Matthew Brost [this message]
2025-05-30  7:58 ` ✓ CI.Patch_applied: success for drm/xe: Make VMA tile_present, tile_invalidated access rules clear (rev3) Patchwork
2025-05-30  7:58 ` ✓ CI.checkpatch: " Patchwork
2025-05-30  7:59 ` ✓ CI.KUnit: " Patchwork
2025-05-30  8:09 ` ✓ CI.Build: " Patchwork
2025-05-30  8:12 ` ✓ CI.Hooks: " Patchwork
2025-05-30  8:13 ` ✓ CI.checksparse: " Patchwork
2025-05-30  8:46 ` ✓ Xe.CI.BAT: " Patchwork
2025-05-31 11:18 ` ✗ Xe.CI.Full: failure " Patchwork
2025-06-02  8:57 ` [PATCH v3] drm/xe: Make VMA tile_present, tile_invalidated access rules clear Thomas Hellström

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20250530061433.1819041-1-matthew.brost@intel.com \
    --to=matthew.brost@intel.com \
    --cc=himal.prasad.ghimiray@intel.com \
    --cc=intel-xe@lists.freedesktop.org \
    --cc=thomas.hellstrom@linux.intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox