* [PATCH v2] dma-buf/dma-fence: Prevent misordering for error checks
@ 2026-11-18 11:09 Philipp Stanner
2026-11-18 12:59 ` Gary Guo
0 siblings, 1 reply; 2+ messages in thread
From: Philipp Stanner @ 2026-11-18 11:09 UTC (permalink / raw)
To: Sumit Semwal, Christian König, Philipp Stanner,
Simona Vetter, Chris Wilson, Danilo Krummrich, Gary Guo,
Alice Ryhl, Daniel Almeida, Tvrtko Ursulin
Cc: linux-media, dri-devel, linux-kernel, stable
Many users of dma_fence do things like:
// Thread A
dma_fence_set_error(f, -EINVAL);
dma_fence_signal(f); // <- Takes lock, which orders against the above.
// Thread B
if (dma_fence_is_signaled(f)) // <- Does test_bit(), no ordering.
pr_info("err: %i\n", f->error); // <- Can be loaded first.
There is no memory ordering between reading of the signaled-bit and
reading of the error code. This can result in outdated, wrong error
codes being evaluated by the user.
Though dma_fence_get_status() does take the fence lock, it is equally
broken because the misordering can simply happen within the critical
section.
It's not entirely clear since when this bug exists, but at least since
dma_fence_get_status() was added, which is why its birth commit is
marked as the Fixes commit here.
Ensure ordering for everyone evaluating the error code centrally by
using test_bit_acquire() in dma_fence_test_signaled_flag().
Cc: stable@vger.kernel.org
Fixes: d6c99f4bf093 ("dma-fence: Wrap querying the fence->status")
Signed-off-by: Philipp Stanner <phasta@kernel.org>
---
Changes in v2:
- Use classic C style comments. (Christian)
---
include/linux/dma-fence.h | 13 ++++++++++++-
1 file changed, 12 insertions(+), 1 deletion(-)
diff --git a/include/linux/dma-fence.h b/include/linux/dma-fence.h
index dd07d128adc3..a3fa1341765e 100644
--- a/include/linux/dma-fence.h
+++ b/include/linux/dma-fence.h
@@ -496,7 +496,18 @@ const char __rcu *dma_fence_timeline_name(struct dma_fence *fence);
static inline bool
dma_fence_test_signaled_flag(struct dma_fence *fence)
{
- return test_bit(DMA_FENCE_FLAG_SIGNALED_BIT, &fence->flags);
+ /*
+ * Best-effort ordering of the locklessly evaluated signaled-bit,
+ * especially so that reads of fence->error are not misordered.
+ *
+ * The release counterpart is typically ensured like this:
+ * a) Error codes must only be set before a fence signals and
+ * b) The fence's spinlock's barriers effective in dma_fence_signal()
+ * ensure ordering.
+ * c) For users of dma_fence_signal_locked(), the barrier implied in
+ * test_and_set_bit() should do the trick.
+ */
+ return test_bit_acquire(DMA_FENCE_FLAG_SIGNALED_BIT, &fence->flags);
}
/**
--
2.56.0
^ permalink raw reply related [flat|nested] 2+ messages in thread
* Re: [PATCH v2] dma-buf/dma-fence: Prevent misordering for error checks
2026-11-18 11:09 [PATCH v2] dma-buf/dma-fence: Prevent misordering for error checks Philipp Stanner
@ 2026-11-18 12:59 ` Gary Guo
0 siblings, 0 replies; 2+ messages in thread
From: Gary Guo @ 2026-11-18 12:59 UTC (permalink / raw)
To: Philipp Stanner, Sumit Semwal, Christian König,
Simona Vetter, Chris Wilson, Danilo Krummrich, Gary Guo,
Alice Ryhl, Daniel Almeida, Tvrtko Ursulin
Cc: linux-media, dri-devel, linux-kernel, stable
On Wed Nov 18, 2026 at 11:09 AM GMT, Philipp Stanner wrote:
> Many users of dma_fence do things like:
>
> // Thread A
> dma_fence_set_error(f, -EINVAL);
> dma_fence_signal(f); // <- Takes lock, which orders against the above.
>
> // Thread B
> if (dma_fence_is_signaled(f)) // <- Does test_bit(), no ordering.
> pr_info("err: %i\n", f->error); // <- Can be loaded first.
>
> There is no memory ordering between reading of the signaled-bit and
> reading of the error code. This can result in outdated, wrong error
> codes being evaluated by the user.
>
> Though dma_fence_get_status() does take the fence lock, it is equally
> broken because the misordering can simply happen within the critical
> section.
>
> It's not entirely clear since when this bug exists, but at least since
> dma_fence_get_status() was added, which is why its birth commit is
> marked as the Fixes commit here.
>
> Ensure ordering for everyone evaluating the error code centrally by
> using test_bit_acquire() in dma_fence_test_signaled_flag().
>
> Cc: stable@vger.kernel.org
> Fixes: d6c99f4bf093 ("dma-fence: Wrap querying the fence->status")
> Signed-off-by: Philipp Stanner <phasta@kernel.org>
> ---
> Changes in v2:
> - Use classic C style comments. (Christian)
> ---
> include/linux/dma-fence.h | 13 ++++++++++++-
> 1 file changed, 12 insertions(+), 1 deletion(-)
>
> diff --git a/include/linux/dma-fence.h b/include/linux/dma-fence.h
> index dd07d128adc3..a3fa1341765e 100644
> --- a/include/linux/dma-fence.h
> +++ b/include/linux/dma-fence.h
> @@ -496,7 +496,18 @@ const char __rcu *dma_fence_timeline_name(struct dma_fence *fence);
> static inline bool
> dma_fence_test_signaled_flag(struct dma_fence *fence)
> {
> - return test_bit(DMA_FENCE_FLAG_SIGNALED_BIT, &fence->flags);
> + /*
> + * Best-effort ordering of the locklessly evaluated signaled-bit,
> + * especially so that reads of fence->error are not misordered.
I am not sure if this should focus on the error, or it should just provide a
general guarantee that anything happens before dma_fence_signal is visible when
signal test returns true. I would lean on the latter?
> + *
> + * The release counterpart is typically ensured like this:
> + * a) Error codes must only be set before a fence signals and
> + * b) The fence's spinlock's barriers effective in dma_fence_signal()
> + * ensure ordering.
Spinlock has nothing to do with this. The only thing that guarantees the
ordering is test_and_set_bit in (c).
So probably easier just to say something like
/* Pairs with test_and_set_bit() in dma_fence_signal_timestamp_locked */
Best,
Gary
> + * c) For users of dma_fence_signal_locked(), the barrier implied in
> + * test_and_set_bit() should do the trick.
> + */
> + return test_bit_acquire(DMA_FENCE_FLAG_SIGNALED_BIT, &fence->flags);
> }
>
> /**
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2026-11-18 12:59 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-11-18 11:09 [PATCH v2] dma-buf/dma-fence: Prevent misordering for error checks Philipp Stanner
2026-11-18 12:59 ` Gary Guo
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox