public inbox for intel-xe@lists.freedesktop.org
 help / color / mirror / Atom feed
From: Tejas Upadhyay <tejas.upadhyay@intel.com>
To: intel-xe@lists.freedesktop.org
Cc: matthew.auld@intel.com, matthew.brost@intel.com,
	thomas.hellstrom@linux.intel.com,
	himal.prasad.ghimiray@intel.com,
	Tejas Upadhyay <tejas.upadhyay@intel.com>
Subject: [RFC PATCH V7 02/10] gpu/buddy: Integrate lockdep for gpu buddy manager
Date: Thu, 16 Apr 2026 13:19:51 +0530	[thread overview]
Message-ID: <20260416074958.3722666-14-tejas.upadhyay@intel.com> (raw)
In-Reply-To: <20260416074958.3722666-12-tejas.upadhyay@intel.com>

Integrating lockdep into the gpu_buddy manager as standard practice for
verifying that internal resources are correctly protected by their
associated locks.

Signed-off-by: Tejas Upadhyay <tejas.upadhyay@intel.com>
---
 drivers/gpu/buddy.c                  | 18 ++++++++++--
 drivers/gpu/drm/drm_buddy.c          |  7 +++--
 drivers/gpu/drm/xe/xe_ttm_vram_mgr.c |  3 ++
 include/linux/gpu_buddy.h            | 41 ++++++++++++++++++++++++++++
 4 files changed, 65 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/buddy.c b/drivers/gpu/buddy.c
index 52686672e99f..53ff85ac2105 100644
--- a/drivers/gpu/buddy.c
+++ b/drivers/gpu/buddy.c
@@ -437,6 +437,9 @@ int gpu_buddy_init(struct gpu_buddy *mm, u64 size, u64 chunk_size)
 		root_count++;
 	} while (size);
 
+#ifdef CONFIG_LOCKDEP
+	mm->lock_dep_map = NULL;
+#endif
 	return 0;
 
 out_free_roots:
@@ -464,6 +467,7 @@ void gpu_buddy_fini(struct gpu_buddy *mm)
 	unsigned int order;
 	int i;
 
+	gpu_buddy_driver_lock_held(mm);
 	size = mm->size;
 
 	for (i = 0; i < mm->n_roots; ++i) {
@@ -538,6 +542,7 @@ void gpu_buddy_reset_clear(struct gpu_buddy *mm, bool is_clear)
 	unsigned int order;
 	int i;
 
+	gpu_buddy_driver_lock_held(mm);
 	size = mm->size;
 	for (i = 0; i < mm->n_roots; ++i) {
 		order = ilog2(size) - ilog2(mm->chunk_size);
@@ -580,6 +585,7 @@ EXPORT_SYMBOL(gpu_buddy_reset_clear);
 void gpu_buddy_free_block(struct gpu_buddy *mm,
 			  struct gpu_buddy_block *block)
 {
+	gpu_buddy_driver_lock_held(mm);
 	BUG_ON(!gpu_buddy_block_is_allocated(block));
 	mm->avail += gpu_buddy_block_size(mm, block);
 	if (gpu_buddy_block_is_clear(block))
@@ -633,6 +639,7 @@ void gpu_buddy_free_list(struct gpu_buddy *mm,
 {
 	bool mark_clear = flags & GPU_BUDDY_CLEARED;
 
+	gpu_buddy_driver_lock_held(mm);
 	__gpu_buddy_free_list(mm, objects, mark_clear, !mark_clear);
 }
 EXPORT_SYMBOL(gpu_buddy_free_list);
@@ -1172,6 +1179,8 @@ int gpu_buddy_block_trim(struct gpu_buddy *mm,
 	u64 new_start;
 	int err;
 
+	gpu_buddy_driver_lock_held(mm);
+
 	if (!list_is_singular(blocks))
 		return -EINVAL;
 
@@ -1287,6 +1296,8 @@ int gpu_buddy_alloc_blocks(struct gpu_buddy *mm,
 	unsigned long pages;
 	int err;
 
+	gpu_buddy_driver_lock_held(mm);
+
 	if (size < mm->chunk_size)
 		return -EINVAL;
 
@@ -1458,9 +1469,11 @@ EXPORT_SYMBOL(gpu_buddy_alloc_blocks);
 void gpu_buddy_block_print(struct gpu_buddy *mm,
 			   struct gpu_buddy_block *block)
 {
-	u64 start = gpu_buddy_block_offset(block);
-	u64 size = gpu_buddy_block_size(mm, block);
+	u64 start, size;
 
+	gpu_buddy_driver_lock_held(mm);
+	start = gpu_buddy_block_offset(block);
+	size = gpu_buddy_block_size(mm, block);
 	pr_info("%#018llx-%#018llx: %llu\n", start, start + size, size);
 }
 EXPORT_SYMBOL(gpu_buddy_block_print);
@@ -1475,6 +1488,7 @@ void gpu_buddy_print(struct gpu_buddy *mm)
 {
 	int order;
 
+	gpu_buddy_driver_lock_held(mm);
 	pr_info("chunk_size: %lluKiB, total: %lluMiB, free: %lluMiB, clear_free: %lluMiB\n",
 		mm->chunk_size >> 10, mm->size >> 20, mm->avail >> 20, mm->clear_avail >> 20);
 
diff --git a/drivers/gpu/drm/drm_buddy.c b/drivers/gpu/drm/drm_buddy.c
index 841f3de5f307..f4ad09b8a36e 100644
--- a/drivers/gpu/drm/drm_buddy.c
+++ b/drivers/gpu/drm/drm_buddy.c
@@ -25,9 +25,11 @@ void drm_buddy_block_print(struct gpu_buddy *mm,
 			   struct gpu_buddy_block *block,
 			   struct drm_printer *p)
 {
-	u64 start = gpu_buddy_block_offset(block);
-	u64 size = gpu_buddy_block_size(mm, block);
+	u64 start, size;
 
+	gpu_buddy_driver_lock_held(mm);
+	start = gpu_buddy_block_offset(block);
+	size = gpu_buddy_block_size(mm, block);
 	drm_printf(p, "%#018llx-%#018llx: %llu\n", start, start + size, size);
 }
 EXPORT_SYMBOL(drm_buddy_block_print);
@@ -42,6 +44,7 @@ void drm_buddy_print(struct gpu_buddy *mm, struct drm_printer *p)
 {
 	int order;
 
+	gpu_buddy_driver_lock_held(mm);
 	drm_printf(p, "chunk_size: %lluKiB, total: %lluMiB, free: %lluMiB, clear_free: %lluMiB\n",
 		   mm->chunk_size >> 10, mm->size >> 20, mm->avail >> 20, mm->clear_avail >> 20);
 
diff --git a/drivers/gpu/drm/xe/xe_ttm_vram_mgr.c b/drivers/gpu/drm/xe/xe_ttm_vram_mgr.c
index 01a9b92772f8..935e589dd4b0 100644
--- a/drivers/gpu/drm/xe/xe_ttm_vram_mgr.c
+++ b/drivers/gpu/drm/xe/xe_ttm_vram_mgr.c
@@ -293,7 +293,9 @@ static void xe_ttm_vram_mgr_fini(struct drm_device *dev, void *arg)
 
 	WARN_ON_ONCE(mgr->visible_avail != mgr->visible_size);
 
+	mutex_lock(&mgr->lock);
 	gpu_buddy_fini(&mgr->mm);
+	mutex_unlock(&mgr->lock);
 
 	ttm_resource_manager_cleanup(&mgr->manager);
 
@@ -328,6 +330,7 @@ int __xe_ttm_vram_mgr_init(struct xe_device *xe, struct xe_ttm_vram_mgr *mgr,
 	if (err)
 		return err;
 
+	gpu_buddy_driver_set_lock(&mgr->mm, &mgr->lock);
 	ttm_set_driver_manager(&xe->ttm, mem_type, &mgr->manager);
 	ttm_resource_manager_set_used(&mgr->manager, true);
 
diff --git a/include/linux/gpu_buddy.h b/include/linux/gpu_buddy.h
index 5fa917ba5450..c174de80ad72 100644
--- a/include/linux/gpu_buddy.h
+++ b/include/linux/gpu_buddy.h
@@ -154,6 +154,7 @@ struct gpu_buddy_block {
  * @avail: Total free space currently available for allocation in bytes.
  * @clear_avail: Free space available in the clear tree (zeroed memory) in bytes.
  *               This is a subset of @avail.
+ * @lock_dep_map: Annotates gpu_buddy API with a driver provided lock.
  */
 struct gpu_buddy {
 /* private: */
@@ -179,8 +180,48 @@ struct gpu_buddy {
 	u64 size;
 	u64 avail;
 	u64 clear_avail;
+#ifdef CONFIG_LOCKDEP
+	struct lockdep_map *lock_dep_map;
+#endif
 };
 
+#ifdef CONFIG_LOCKDEP
+/**
+ * gpu_buddy_driver_set_lock() - Set the lock protecting accesses to GPU BUDDY
+ * @mm: Pointer to GPU buddy structure.
+ * @lock: the lock used to protect the gpu buddy. The locking primitive
+ * must contain a dep_map field.
+ *
+ * Call this to annotate gpu_buddy APIs which access/modify gpu_buddy manager
+ */
+#define gpu_buddy_driver_set_lock(mm, lock) \
+	do { \
+		struct gpu_buddy *__mm = (mm); \
+		if (!WARN(__mm->lock_dep_map, "GPU BUDDY MM lock should be set only once.")) \
+			__mm->lock_dep_map = &(lock)->dep_map; \
+	} while (0)
+#else
+#define gpu_buddy_driver_set_lock(mm, lock) do { (void)(mm); (void)(lock); } while (0)
+#endif
+
+#ifdef CONFIG_LOCKDEP
+/**
+ * gpu_buddy_driver_lock_held() - Assert GPU BUDDY manager lock is held
+ * @mm: Pointer to the GPU BUDDY structure.
+ *
+ * Ensure driver lock is held.
+ */
+static inline void gpu_buddy_driver_lock_held(struct gpu_buddy *mm)
+{
+	if ((mm)->lock_dep_map)
+		lockdep_assert(lock_is_held_type((mm)->lock_dep_map, 0));
+}
+#else
+static inline gpu_buddy_driver_lock_held(struct gpu_buddy *mm)
+{
+}
+#endif
+
 static inline u64
 gpu_buddy_block_offset(const struct gpu_buddy_block *block)
 {
-- 
2.52.0


  parent reply	other threads:[~2026-04-16  7:50 UTC|newest]

Thread overview: 27+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-04-16  7:49 [RFC PATCH V7 00/10] Add memory page offlining support Tejas Upadhyay
2026-04-16  7:49 ` [RFC PATCH V7 01/10] drm/xe: Link VRAM object with gpu buddy Tejas Upadhyay
2026-04-16  7:49 ` Tejas Upadhyay [this message]
2026-04-16  8:55   ` [RFC PATCH V7 02/10] gpu/buddy: Integrate lockdep for gpu buddy manager Matthew Auld
2026-04-16  9:43     ` Upadhyay, Tejas
2026-04-16  9:56       ` Matthew Auld
2026-04-16 10:04         ` Upadhyay, Tejas
2026-04-16 10:15           ` Matthew Auld
2026-04-16 10:18             ` Upadhyay, Tejas
2026-04-16  7:49 ` [RFC PATCH V7 03/10] drm/gpu: Add gpu_buddy_allocated_addr_to_block helper Tejas Upadhyay
2026-04-16  7:49 ` [RFC PATCH V7 04/10] drm/xe: Link LRC BO and its execution Queue Tejas Upadhyay
2026-04-30  3:34   ` Matthew Brost
2026-05-04  9:11     ` Upadhyay, Tejas
2026-04-16  7:49 ` [RFC PATCH V7 05/10] drm/xe: Extend BO purge to handle vram pages as well Tejas Upadhyay
2026-04-30  3:44   ` Matthew Brost
2026-04-30 12:08     ` Upadhyay, Tejas
2026-05-05  8:15     ` Yadav, Arvind
2026-04-16  7:49 ` [RFC PATCH V7 06/10] drm/xe: Handle physical memory address error Tejas Upadhyay
2026-04-16  7:49 ` [RFC PATCH V7 07/10] drm/xe/cri: Add debugfs to inject faulty vram address Tejas Upadhyay
2026-04-16  7:49 ` [RFC PATCH V7 08/10] gpu/buddy: Add routine to dump allocated buddy blocks Tejas Upadhyay
2026-04-16  7:49 ` [RFC PATCH V7 09/10] drm/xe/configfs: Add vram bad page reservation policy Tejas Upadhyay
2026-04-16  7:49 ` [RFC PATCH V7 10/10] drm/xe/cri: Add sysfs interface for bad gpu vram pages Tejas Upadhyay
2026-04-30 13:53   ` Matthew Auld
2026-05-04  9:02     ` Upadhyay, Tejas
2026-05-05  8:44       ` Matthew Auld
2026-04-16  7:56 ` ✗ CI.checkpatch: warning for Add memory page offlining support (rev8) Patchwork
2026-04-16  7:57 ` ✗ CI.KUnit: failure " Patchwork

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=20260416074958.3722666-14-tejas.upadhyay@intel.com \
    --to=tejas.upadhyay@intel.com \
    --cc=himal.prasad.ghimiray@intel.com \
    --cc=intel-xe@lists.freedesktop.org \
    --cc=matthew.auld@intel.com \
    --cc=matthew.brost@intel.com \
    --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