* [PATCH] dma-fence: use correct callback in dma_fence_timeline_name()
@ 2026-06-18 12:00 André Draszik
2026-06-18 12:46 ` Christian König
0 siblings, 1 reply; 4+ messages in thread
From: André Draszik @ 2026-06-18 12:00 UTC (permalink / raw)
To: Sumit Semwal, Christian König, Tvrtko Ursulin
Cc: linux-media, dri-devel, linaro-mm-sig, linux-kernel,
Peter Griffin, Tudor Ambarus, Juan Yescas, kernel-team,
André Draszik
dma_fence_timeline_name() is a wrapper around
dma_fence_ops::get_timeline_name(). Since the blamed commit below, it
calls an incorrect callback.
Update it to restore functionality by calling the intended callback.
Fixes: 62918542b7bf ("dma-fence: Fix sparse warnings due __rcu annotations")
Signed-off-by: André Draszik <andre.draszik@linaro.org>
---
drivers/dma-buf/dma-fence.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/dma-buf/dma-fence.c b/drivers/dma-buf/dma-fence.c
index b3bfa6943a8e..5292d714419b 100644
--- a/drivers/dma-buf/dma-fence.c
+++ b/drivers/dma-buf/dma-fence.c
@@ -1202,7 +1202,7 @@ const char __rcu *dma_fence_timeline_name(struct dma_fence *fence)
/* RCU protection is required for safe access to returned string */
ops = rcu_dereference(fence->ops);
if (!dma_fence_test_signaled_flag(fence))
- return (const char __rcu *)ops->get_driver_name(fence);
+ return (const char __rcu *)ops->get_timeline_name(fence);
else
return (const char __rcu *)"signaled-timeline";
}
---
base-commit: e2cae00c05d196491c318196792297f2dfbaa02c
change-id: 20260618-linux-drm_crtc_fix-267655a1c89f
Best regards,
--
André Draszik <andre.draszik@linaro.org>
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH] dma-fence: use correct callback in dma_fence_timeline_name()
2026-06-18 12:00 [PATCH] dma-fence: use correct callback in dma_fence_timeline_name() André Draszik
@ 2026-06-18 12:46 ` Christian König
2026-06-18 18:30 ` Tvrtko Ursulin
0 siblings, 1 reply; 4+ messages in thread
From: Christian König @ 2026-06-18 12:46 UTC (permalink / raw)
To: André Draszik, Sumit Semwal, Tvrtko Ursulin
Cc: linux-media, dri-devel, linaro-mm-sig, linux-kernel,
Peter Griffin, Tudor Ambarus, Juan Yescas, kernel-team
On 6/18/26 14:00, André Draszik wrote:
> dma_fence_timeline_name() is a wrapper around
> dma_fence_ops::get_timeline_name(). Since the blamed commit below, it
> calls an incorrect callback.
>
> Update it to restore functionality by calling the intended callback.
>
> Fixes: 62918542b7bf ("dma-fence: Fix sparse warnings due __rcu annotations")
> Signed-off-by: André Draszik <andre.draszik@linaro.org>
I'm like 99% sure we had that already fixed weeks ago, but somehow it looks like that was never merged to drm-misc-fixes.
Anyway Reviewed-by: Christian König <christian.koenig@amd.com>
> ---
> drivers/dma-buf/dma-fence.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/dma-buf/dma-fence.c b/drivers/dma-buf/dma-fence.c
> index b3bfa6943a8e..5292d714419b 100644
> --- a/drivers/dma-buf/dma-fence.c
> +++ b/drivers/dma-buf/dma-fence.c
> @@ -1202,7 +1202,7 @@ const char __rcu *dma_fence_timeline_name(struct dma_fence *fence)
> /* RCU protection is required for safe access to returned string */
> ops = rcu_dereference(fence->ops);
> if (!dma_fence_test_signaled_flag(fence))
> - return (const char __rcu *)ops->get_driver_name(fence);
> + return (const char __rcu *)ops->get_timeline_name(fence);
> else
> return (const char __rcu *)"signaled-timeline";
> }
>
> ---
> base-commit: e2cae00c05d196491c318196792297f2dfbaa02c
> change-id: 20260618-linux-drm_crtc_fix-267655a1c89f
>
> Best regards,
> --
> André Draszik <andre.draszik@linaro.org>
>
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH] dma-fence: use correct callback in dma_fence_timeline_name()
2026-06-18 12:46 ` Christian König
@ 2026-06-18 18:30 ` Tvrtko Ursulin
2026-06-23 13:08 ` pstanner
0 siblings, 1 reply; 4+ messages in thread
From: Tvrtko Ursulin @ 2026-06-18 18:30 UTC (permalink / raw)
To: Christian König, André Draszik, Sumit Semwal, Tvrtko Ursulin
Cc: linux-media, dri-devel, linaro-mm-sig, linux-kernel,
Peter Griffin, Tudor Ambarus, Juan Yescas, kernel-team
On 18/06/2026 14:46, Christian König wrote:
> On 6/18/26 14:00, André Draszik wrote:
>> dma_fence_timeline_name() is a wrapper around
>> dma_fence_ops::get_timeline_name(). Since the blamed commit below, it
>> calls an incorrect callback.
>>
>> Update it to restore functionality by calling the intended callback.
>>
>> Fixes: 62918542b7bf ("dma-fence: Fix sparse warnings due __rcu annotations")
>> Signed-off-by: André Draszik <andre.draszik@linaro.org>
> I'm like 99% sure we had that already fixed weeks ago, but somehow it looks like that was never merged to drm-misc-fixes.
>
> Anyway Reviewed-by: Christian König <christian.koenig@amd.com>
I was also perplexed and took me some digging to realise what had happened.
The copy-and-paste bug was indeed originally fixed in:
commit 033559473dd3b55558b535aa37b8848c207b5cbb
Author: Akash Goel <akash.goel@arm.com>
AuthorDate: Tue Oct 21 17:09:51 2025 +0100
Commit: Tvrtko Ursulin <tursulin@ursulin.net>
CommitDate: Fri Oct 24 16:56:37 2025 +0100
dma-fence: Fix safe access wrapper to call timeline name method
But waaay before that I sent a fix for something else which touched
those lines not noticing the copy-and-paste, fixing a different issue.
That one however was not merged until earlier this year:
commit 62918542b7bf08860a60ebbde7654486e0ac0776
Author: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
AuthorDate: Mon Jun 16 16:59:52 2025 +0100
Commit: Christian König <christian.koenig@amd.com>
CommitDate: Mon Feb 9 14:05:20 2026 +0100
dma-fence: Fix sparse warnings due __rcu annotations
Notice the authored vs commited dates.
I guess it was a silent conflict which overwrote the earlier fix and
restore the copy-and-paste bug.
Regards,
Tvrtko
>
>> ---
>> drivers/dma-buf/dma-fence.c | 2 +-
>> 1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/drivers/dma-buf/dma-fence.c b/drivers/dma-buf/dma-fence.c
>> index b3bfa6943a8e..5292d714419b 100644
>> --- a/drivers/dma-buf/dma-fence.c
>> +++ b/drivers/dma-buf/dma-fence.c
>> @@ -1202,7 +1202,7 @@ const char __rcu *dma_fence_timeline_name(struct dma_fence *fence)
>> /* RCU protection is required for safe access to returned string */
>> ops = rcu_dereference(fence->ops);
>> if (!dma_fence_test_signaled_flag(fence))
>> - return (const char __rcu *)ops->get_driver_name(fence);
>> + return (const char __rcu *)ops->get_timeline_name(fence);
>> else
>> return (const char __rcu *)"signaled-timeline";
>> }
>>
>> ---
>> base-commit: e2cae00c05d196491c318196792297f2dfbaa02c
>> change-id: 20260618-linux-drm_crtc_fix-267655a1c89f
>>
>> Best regards,
>> --
>> André Draszik <andre.draszik@linaro.org>
>>
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH] dma-fence: use correct callback in dma_fence_timeline_name()
2026-06-18 18:30 ` Tvrtko Ursulin
@ 2026-06-23 13:08 ` pstanner
0 siblings, 0 replies; 4+ messages in thread
From: pstanner @ 2026-06-23 13:08 UTC (permalink / raw)
To: Tvrtko Ursulin, Christian König, André Draszik,
Sumit Semwal, Tvrtko Ursulin
Cc: linux-media, dri-devel, linaro-mm-sig, linux-kernel,
Peter Griffin, Tudor Ambarus, Juan Yescas, kernel-team
On Thu, 2026-06-18 at 20:30 +0200, Tvrtko Ursulin wrote:
>
> On 18/06/2026 14:46, Christian König wrote:
> > On 6/18/26 14:00, André Draszik wrote:
> > > dma_fence_timeline_name() is a wrapper around
> > > dma_fence_ops::get_timeline_name(). Since the blamed commit below, it
> > > calls an incorrect callback.
> > >
> > > Update it to restore functionality by calling the intended callback.
> > >
> > > Fixes: 62918542b7bf ("dma-fence: Fix sparse warnings due __rcu annotations")
> > > Signed-off-by: André Draszik <andre.draszik@linaro.org>
Reviewed-by: Philipp Stanner <phasta@kernel.org>
> > I'm like 99% sure we had that already fixed weeks ago, but somehow it looks like that was never merged to drm-misc-fixes.
I'm also very sure that we had shot that bug down not too long in the
past.
> >
> > Anyway Reviewed-by: Christian König <christian.koenig@amd.com>
>
> I was also perplexed and took me some digging to realise what had happened.
>
> The copy-and-paste bug was indeed originally fixed in:
>
> commit 033559473dd3b55558b535aa37b8848c207b5cbb
> Author: Akash Goel <akash.goel@arm.com>
> AuthorDate: Tue Oct 21 17:09:51 2025 +0100
> Commit: Tvrtko Ursulin <tursulin@ursulin.net>
> CommitDate: Fri Oct 24 16:56:37 2025 +0100
>
> dma-fence: Fix safe access wrapper to call timeline name method
>
>
> But waaay before that I sent a fix for something else which touched
> those lines not noticing the copy-and-paste, fixing a different issue.
> That one however was not merged until earlier this year:
>
> commit 62918542b7bf08860a60ebbde7654486e0ac0776
> Author: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
> AuthorDate: Mon Jun 16 16:59:52 2025 +0100
> Commit: Christian König <christian.koenig@amd.com>
> CommitDate: Mon Feb 9 14:05:20 2026 +0100
>
> dma-fence: Fix sparse warnings due __rcu annotations
>
> Notice the authored vs commited dates.
>
> I guess it was a silent conflict which overwrote the earlier fix and
> restore the copy-and-paste bug.
silent conflict? If that can happen with our git workflow, how is it
that we don't see things like that more often?
P.
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2026-06-23 13:08 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2026-06-18 12:00 [PATCH] dma-fence: use correct callback in dma_fence_timeline_name() André Draszik
2026-06-18 12:46 ` Christian König
2026-06-18 18:30 ` Tvrtko Ursulin
2026-06-23 13:08 ` pstanner
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox