Patch "drm/i915/gem: Allow EXEC_CAPTURE on recoverable contexts on DG1" has been added to the 6.1-stable tree

This is a note to let you know that I've just added the patch titled

drm/i915/gem: Allow EXEC_CAPTURE on recoverable contexts on DG1

to the 6.1-stable tree which can be found at:
http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary

The filename of the patch is:
drm-i915-gem-allow-exec_capture-on-recoverable-contexts-on-dg1.patch
and it can be found in the queue-6.1 subdirectory.

If you, or anyone else, feels it should not be added to the stable tree, please let <[email protected]> know about it.

From 25eeba495b2fc16037647c1a51bcdf6fc157af5c Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <[email protected]>
Date: Mon, 12 May 2025 21:22:15 +0200
Subject: drm/i915/gem: Allow EXEC_CAPTURE on recoverable contexts on DG1
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

From: Ville Syrjälä <[email protected]>

commit 25eeba495b2fc16037647c1a51bcdf6fc157af5c upstream.

The intel-media-driver is currently broken on DG1 because it uses EXEC_CAPTURE with recovarable contexts. Relax the check to allow that.

I've also submitted a fix for the intel-media-driver:
https://github.com/intel/media-driver/pull/1920

Cc: [email protected] # v6.0+
Cc: Matthew Auld <[email protected]>
Cc: Thomas Hellström <[email protected]>
Testcase: igt/gem_exec_capture/capture-invisible

Fixes 71b1669ea9bd ("drm/i915/uapi: tweak error capture on recoverable contexts")

Reviewed-by Andi Shyti <[email protected]>

Signed-off-by Ville Syrjälä <[email protected]>

Signed-off-by Andi Shyti <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

(cherry picked from commit d6e020819612a4a06207af858e0978be4d3e3140)

Signed-off-by Joonas Lahtinen <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c +1 -1

@@ -2001,7 +2001,7 @@ static int eb_capture_stage(struct i915_
                        continue;
 
                if (i915_gem_context_is_recoverable(eb->gem_context) &&
-                   (IS_DGFX(eb->i915) || GRAPHICS_VER_FULL(eb->i915) > IP_VER(12, 0)))
+                   GRAPHICS_VER_FULL(eb->i915) > IP_VER(12, 10))
                        return -EINVAL;
 
                for_each_batch_create_order(eb, j) {

Patches currently in stable-queue which might be from [email protected] are

queue-6.1/drm-dp-change-aux-dpcd-probe-address-from-dpcd_rev-to-lane0_1_status.patch queue-6.1/revert-drm-i915-gem-allow-exec_capture-on-recoverabl.patch queue-6.1/drm-i915-gem-allow-exec_capture-on-recoverable-contexts-on-dg1.patch queue-6.1/drm-i915-gem-allow-exec_capture-on-recoverable-conte.patch

Hi Greg,

Please do note that there is a revert for this patch that was part of the same pull request. That needs to be picked in too in case you are picking the original patch.

I already got the automated mails from Sasha that both the original commit and revert were already picked into 6.1, 6.6 and 6.12 trees. Are now in a perpetual machinery induced loop where the original commit and revert will be picked in alternating fashion to the stable trees? [1]

Regards, Joonas

[1] Originally, I was under the assumption stable machinery would automatically skip over patches that have later been reverted, but that doesn't seem to be the case?