From: Tejas Upadhyay <tejas.upadhyay@intel.com>
To: intel-xe@lists.freedesktop.org
Cc: himal.prasad.ghimiray@intel.com, rodrigo.vivi@intel.com,
Matthew Brost <matthew.brost@intel.com>,
Tejas Upadhyay <tejas.upadhyay@intel.com>
Subject: [PATCH V21 06/15] drm/xe: Guard teardown paths against purged BOs
Date: Thu, 3 Sep 2026 21:45:59 +0530 [thread overview]
Message-ID: <20260903161553.528932-23-tejas.upadhyay@intel.com> (raw)
In-Reply-To: <20260903161553.528932-17-tejas.upadhyay@intel.com>
VRAM page offlining can purge BOs that are still referenced by page
tables, exec queues, and DMA-buf exports. Add xe_bo_is_purged()
guards in the teardown paths to prevent unpinning or mapping an
already-purged BO:
- xe_bo_unpin_map_no_vm(): skip unpin if purged
- xe_dma_buf_map(): return -ENOENT early if purged
- xe_exec_queue_update_run_ticks(): skip LRC timestamp read if purged
- xe_pt_destroy(): skip unpin if purged
v4(Sashiko):
- Move bo purge check in xe_bo_unpin() to handle all instances
v3(Sashiko):
- Remove dma_resv is already held
v2(Himal):
- take dma_resv lock before calling xe_bo_is_purged()
Reviewed-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Signed-off-by: Tejas Upadhyay <tejas.upadhyay@intel.com>
---
drivers/gpu/drm/xe/xe_bo.c | 3 +++
drivers/gpu/drm/xe/xe_dma_buf.c | 3 +++
drivers/gpu/drm/xe/xe_exec_queue.c | 8 ++++++--
3 files changed, 12 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/xe/xe_bo.c b/drivers/gpu/drm/xe/xe_bo.c
index c255327669da..b162753cebb7 100644
--- a/drivers/gpu/drm/xe/xe_bo.c
+++ b/drivers/gpu/drm/xe/xe_bo.c
@@ -3267,6 +3267,9 @@ void xe_bo_unpin(struct xe_bo *bo)
struct ttm_place *place = &bo->placements[0];
struct xe_device *xe = xe_bo_device(bo);
+ if (xe_bo_is_purged(bo))
+ return;
+
xe_assert(xe, !bo->ttm.base.import_attach);
xe_assert(xe, xe_bo_is_pinned(bo));
diff --git a/drivers/gpu/drm/xe/xe_dma_buf.c b/drivers/gpu/drm/xe/xe_dma_buf.c
index bf0728838ead..5d9f1cd24b7f 100644
--- a/drivers/gpu/drm/xe/xe_dma_buf.c
+++ b/drivers/gpu/drm/xe/xe_dma_buf.c
@@ -104,6 +104,9 @@ static struct sg_table *xe_dma_buf_map(struct dma_buf_attachment *attach,
struct sg_table *sgt;
int r = 0;
+ if (xe_bo_is_purged(bo))
+ return ERR_PTR(-ENOENT);
+
if (!attach->peer2peer && !xe_bo_can_migrate(bo, XE_PL_TT))
return ERR_PTR(-EOPNOTSUPP);
diff --git a/drivers/gpu/drm/xe/xe_exec_queue.c b/drivers/gpu/drm/xe/xe_exec_queue.c
index f3ea4807ac33..e63559a2f582 100644
--- a/drivers/gpu/drm/xe/xe_exec_queue.c
+++ b/drivers/gpu/drm/xe/xe_exec_queue.c
@@ -1630,8 +1630,12 @@ void xe_exec_queue_update_run_ticks(struct xe_exec_queue *q)
* errors.
*/
lrc = q->lrc[0];
- new_ts = xe_lrc_update_timestamp(lrc, &old_ts);
- q->xef->run_ticks[q->class] += (new_ts - old_ts) * q->width;
+ xe_bo_lock(lrc->bo, false);
+ if (!xe_bo_is_purged(lrc->bo)) {
+ new_ts = xe_lrc_update_timestamp(lrc, &old_ts);
+ q->xef->run_ticks[q->class] += (new_ts - old_ts) * q->width;
+ }
+ xe_bo_unlock(lrc->bo);
drm_dev_exit(idx);
}
--
2.52.0
next prev parent reply other threads:[~2026-09-03 16:16 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-09-03 16:15 [PATCH V21 00/15] Add memory page offlining support Tejas Upadhyay
2026-09-03 16:15 ` [PATCH V21 01/15] drm/xe: Link VRAM resource allocation with gpu buddy Tejas Upadhyay
2026-09-03 16:15 ` [PATCH V21 02/15] drm/xe: Link LRC BO and its execution queue with safe lifetime rules Tejas Upadhyay
2026-09-03 16:15 ` [PATCH V21 03/15] drm/xe: Export xe_ttm_bo_purge() Tejas Upadhyay
2026-09-03 16:15 ` [PATCH V21 04/15] drm/xe: Handle NULL resource and allow purging of VRAM pages Tejas Upadhyay
2026-09-03 17:08 ` sashiko-bot
2026-09-03 16:15 ` [PATCH V21 05/15] drm/xe/bo: Make xe_bo_is_user() public Tejas Upadhyay
2026-09-03 16:15 ` Tejas Upadhyay [this message]
2026-09-03 16:55 ` [PATCH V21 06/15] drm/xe: Guard teardown paths against purged BOs sashiko-bot
2026-09-03 16:16 ` [PATCH V21 07/15] drm/xe/vram: Extract buddy allocation and free helpers Tejas Upadhyay
2026-09-03 16:16 ` [PATCH V21 08/15] drm/xe/vram: Add page offline data structures and lifecycle Tejas Upadhyay
2026-09-03 16:16 ` [PATCH V21 09/15] drm/xe/vram: Add VRAM page offline fault handler Tejas Upadhyay
2026-09-03 17:23 ` sashiko-bot
2026-09-03 16:16 ` [PATCH V21 10/15] drm/xe/configfs: Add disable_vram_page_offline attribute Tejas Upadhyay
2026-09-03 16:16 ` [PATCH V21 11/15] drm/xe/ras: Cache disable_vram_page_offline policy at init Tejas Upadhyay
2026-09-03 16:16 ` [PATCH V21 12/15] drm/xe/vram: Check disable_vram_page_offline policy in fault handler Tejas Upadhyay
2026-09-03 16:16 ` [PATCH V21 13/15] drm/xe: Expose bad VRAM pages via debugfs Tejas Upadhyay
2026-09-03 16:16 ` [PATCH V21 14/15] drm/xe/uapi: Expose ban reason in EXEC_QUEUE_GET_PROPERTY_BAN Tejas Upadhyay
2026-09-03 17:46 ` sashiko-bot
2026-09-03 16:16 ` [PATCH V21 15/15] drm/xe: Add fault-inject based VRAM page offline injection Tejas Upadhyay
2026-09-03 16:23 ` ✗ CI.checkpatch: warning for Add memory page offlining support (rev25) Patchwork
2026-09-03 16:25 ` ✓ CI.KUnit: success " Patchwork
2026-09-03 17:46 ` ✗ Xe.CI.BAT: failure " Patchwork
2026-09-04 4:33 ` [PATCH V21 00/15] Add memory page offlining support Matthew Brost
2026-09-04 5:18 ` ✗ Xe.CI.FULL: failure for Add memory page offlining support (rev25) 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=20260903161553.528932-23-tejas.upadhyay@intel.com \
--to=tejas.upadhyay@intel.com \
--cc=himal.prasad.ghimiray@intel.com \
--cc=intel-xe@lists.freedesktop.org \
--cc=matthew.brost@intel.com \
--cc=rodrigo.vivi@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