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, dri-devel@lists.freedesktop.org
Cc: Sashiko <sashiko-bot@kernel.org>, stable@vger.kernel.org
Subject: [PATCH v2] drm/pagemap: Clear driver-provided PFNs from migration PFN array
Date: Tue, 21 Jul 2026 13:43:53 -0700	[thread overview]
Message-ID: <20260721204353.1082632-1-matthew.brost@intel.com> (raw)

DRM pagemap overloads the migration PFN array to store driver-provided
PFNs before calling migrate_vma_*() to finalize the migration. If an
error occurs during the incremental copy phase, the migration PFN
entries are reverted to their original state. After reverting the
device-folio mutations, clear any remaining driver-provided PFNs to
avoid confusing the migrate_vma_*() helpers.

Also clear any driver-provided PFNs if populate_devmem_pfn() fails, as
a precaution against stale entries being interpreted as migration PFNs.

Reported-by: Sashiko <sashiko-bot@kernel.org>
Fixes: 3902846af36b ("drm/pagemap Fix error paths in drm_pagemap_migrate_to_devmem")
Fixes: ec265e1f1cfc ("drm/pagemap: Support source migration over interconnect")
Cc: stable@vger.kernel.org
Signed-off-by: Matthew Brost <matthew.brost@intel.com>

---
v2:
 - Be paranoid and also clear driver-provided on populate_devmem_pfn()
   failure (Sashiko)
---
 drivers/gpu/drm/drm_pagemap.c | 11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/drm_pagemap.c b/drivers/gpu/drm/drm_pagemap.c
index 15c78eca180b..892b325fa99b 100644
--- a/drivers/gpu/drm/drm_pagemap.c
+++ b/drivers/gpu/drm/drm_pagemap.c
@@ -727,8 +727,10 @@ int drm_pagemap_migrate_to_devmem(struct drm_pagemap_devmem *devmem_allocation,
 	}
 
 	err = ops->populate_devmem_pfn(devmem_allocation, npages, migrate.dst);
-	if (err)
-		goto err_aborted_migration;
+	if (err) {
+		npages = 0;
+		goto err_finalize;
+	}
 
 	own_pages = 0;
 
@@ -807,8 +809,11 @@ int drm_pagemap_migrate_to_devmem(struct drm_pagemap_devmem *devmem_allocation,
 		msecs_to_jiffies(mdetails->timeslice_ms);
 
 err_finalize:
-	if (err)
+	if (err) {
 		drm_pagemap_migration_unlock_put_pages(npages, migrate.dst);
+		for (i = npages; i < npages_in_range(start, end); ++i)
+			migrate.dst[i] = 0;
+	}
 err_aborted_migration:
 	migrate_vma_pages(&migrate);
 
-- 
2.34.1


             reply	other threads:[~2026-07-21 20:43 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-07-21 20:43 Matthew Brost [this message]
2026-07-21 21:20 ` ✗ CI.checkpatch: warning for drm/pagemap: Clear driver-provided PFNs from migration PFN array (rev2) Patchwork
2026-07-21 21:21 ` ✓ CI.KUnit: success " Patchwork
2026-07-21 22:16 ` ✓ Xe.CI.BAT: " Patchwork
2026-07-22  5:32 ` [PATCH v2] drm/pagemap: Clear driver-provided PFNs from migration PFN array Ghimiray, Himal Prasad
2026-07-22 10:09 ` ✓ Xe.CI.FULL: success for drm/pagemap: Clear driver-provided PFNs from migration PFN array (rev2) 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=20260721204353.1082632-1-matthew.brost@intel.com \
    --to=matthew.brost@intel.com \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=intel-xe@lists.freedesktop.org \
    --cc=sashiko-bot@kernel.org \
    --cc=stable@vger.kernel.org \
    /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