[PATCH 5.10/6.1] drm/amd/display: set new_stream to NULL after release

Roman Demidov posted 1 patch 1 week, 4 days ago
There is a newer version of this series
drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 1 +
1 file changed, 1 insertion(+)
[PATCH 5.10/6.1] drm/amd/display: set new_stream to NULL after release
Posted by Roman Demidov 1 week, 4 days ago
From: WenTao Liang <vulab@iscas.ac.cn>

commit 9fa26b9eed6195bf840f39ac183b9a6237548755 upstream.

In dm_update_crtc_state(), the skip_modeset path releases new_stream
via dc_stream_release() but does not set the pointer to NULL.

If a later error (e.g., color management failure) triggers the fail
label, the error path calls dc_stream_release() again on the same
dangling pointer, causing a double release and potential use-after-free.

Fix this by setting new_stream to NULL after the initial release.

Fixes: 9b690ef3c704 ("drm/amd/display: Avoid full modeset when not required")
Signed-off-by: WenTao Liang <vulab@iscas.ac.cn>
Reviewed-by: George Zhang <george.zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Roman Demidov <roman.demidov.nn@gmail.com>
---
Backport fix for CVE-2026-68236

 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index bb5e3a6086f2..deaf98957cb7 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -9293,6 +9293,7 @@ static int dm_update_crtc_state(struct amdgpu_display_manager *dm,
 	/* Release extra reference */
 	if (new_stream)
 		 dc_stream_release(new_stream);
+	new_stream = NULL;
 
 	/*
 	 * We want to do dc stream updates that do not require a
-- 
2.53.0
Re: [PATCH 5.10/6.1] drm/amd/display: set new_stream to NULL after release
Posted by Sasha Levin 1 week, 3 days ago
> If a later error (e.g., color management failure) triggers the fail
> label, the error path calls dc_stream_release() again on the same
> dangling pointer, causing a double release and potential use-after-free.

5.15 carries the same double release of new_stream in dm_update_crtc_state()
and has no fix for it, so taking 6.1 and 5.10 alone would make a 5.10 to 5.15
upgrade a regression.

Could you send a v2 covering 5.15 as well?

-- 
Thanks,
Sasha