All of lore.kernel.org
 help / color / mirror / Atom feed
* [PATCH v3] drm/amdgpu/gfx9: drop unnecessary 64-bit fence flag check in KIQ
@ 2026-04-28 16:35 John B. Moore
  2026-04-29  8:22 ` Christian König
  0 siblings, 1 reply; 3+ messages in thread
From: John B. Moore @ 2026-04-28 16:35 UTC (permalink / raw)
  To: Christian König, Alexander Deucher; +Cc: amd-gfx, John B. Moore

Remove the BUG_ON(flags & AMDGPU_FENCE_FLAG_64BIT) assertion from
gfx_v9_0_ring_emit_fence_kiq().  The KIQ hardware supports 64-bit
fence writes; the 32-bit writeback address constraint is an
upper-layer convention, not a hardware limitation.  The check serves
no purpose and should not be present.

Found by code inspection while investigating related BUG_ON
assertions in the GFX and compute ring emission paths.

Signed-off-by: John B. Moore <jbmoore61@gmail.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
index 47e81c33d..fb2a0f1af 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -5679,9 +5679,6 @@ static void gfx_v9_0_ring_emit_fence_kiq(struct amdgpu_ring *ring, u64 addr,
 	struct amdgpu_device *adev = ring->adev;
 
-	/* we only allocate 32bit for each seq wb address */
-	BUG_ON(flags & AMDGPU_FENCE_FLAG_64BIT);
-
 	/* write fence seq to the "addr" */
 	amdgpu_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3));
 	amdgpu_ring_write(ring, (WRITE_DATA_ENGINE_SEL(0) |
-- 
2.43.0


^ permalink raw reply related	[flat|nested] 3+ messages in thread

* Re: [PATCH v3] drm/amdgpu/gfx9: drop unnecessary 64-bit fence flag check in KIQ
  2026-04-28 16:35 [PATCH v3] drm/amdgpu/gfx9: drop unnecessary 64-bit fence flag check in KIQ John B. Moore
@ 2026-04-29  8:22 ` Christian König
  2026-04-29 13:57   ` Alex Deucher
  0 siblings, 1 reply; 3+ messages in thread
From: Christian König @ 2026-04-29  8:22 UTC (permalink / raw)
  To: John B. Moore, Alexander Deucher; +Cc: amd-gfx

On 4/28/26 18:35, John B. Moore wrote:
> Remove the BUG_ON(flags & AMDGPU_FENCE_FLAG_64BIT) assertion from
> gfx_v9_0_ring_emit_fence_kiq().  The KIQ hardware supports 64-bit
> fence writes; the 32-bit writeback address constraint is an
> upper-layer convention, not a hardware limitation.  The check serves
> no purpose and should not be present.
> 
> Found by code inspection while investigating related BUG_ON
> assertions in the GFX and compute ring emission paths.
> 
> Signed-off-by: John B. Moore <jbmoore61@gmail.com>

Reviewed-by: Christian König <christian.koenig@amd.com>

> ---
>  drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c | 3 ---
>  1 file changed, 3 deletions(-)
> 
> diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
> index 47e81c33d..fb2a0f1af 100644
> --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
> +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
> @@ -5679,9 +5679,6 @@ static void gfx_v9_0_ring_emit_fence_kiq(struct amdgpu_ring *ring, u64 addr,
>  	struct amdgpu_device *adev = ring->adev;
>  
> -	/* we only allocate 32bit for each seq wb address */
> -	BUG_ON(flags & AMDGPU_FENCE_FLAG_64BIT);
> -
>  	/* write fence seq to the "addr" */
>  	amdgpu_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3));
>  	amdgpu_ring_write(ring, (WRITE_DATA_ENGINE_SEL(0) |


^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [PATCH v3] drm/amdgpu/gfx9: drop unnecessary 64-bit fence flag check in KIQ
  2026-04-29  8:22 ` Christian König
@ 2026-04-29 13:57   ` Alex Deucher
  0 siblings, 0 replies; 3+ messages in thread
From: Alex Deucher @ 2026-04-29 13:57 UTC (permalink / raw)
  To: Christian König; +Cc: John B. Moore, Alexander Deucher, amd-gfx

Applied.  Please double check your mail client.  It seems to have
mangled the patch, so I had to manually fix it up.

Thanks!

Alex

On Wed, Apr 29, 2026 at 4:22 AM Christian König
<christian.koenig@amd.com> wrote:
>
> On 4/28/26 18:35, John B. Moore wrote:
> > Remove the BUG_ON(flags & AMDGPU_FENCE_FLAG_64BIT) assertion from
> > gfx_v9_0_ring_emit_fence_kiq().  The KIQ hardware supports 64-bit
> > fence writes; the 32-bit writeback address constraint is an
> > upper-layer convention, not a hardware limitation.  The check serves
> > no purpose and should not be present.
> >
> > Found by code inspection while investigating related BUG_ON
> > assertions in the GFX and compute ring emission paths.
> >
> > Signed-off-by: John B. Moore <jbmoore61@gmail.com>
>
> Reviewed-by: Christian König <christian.koenig@amd.com>
>
> > ---
> >  drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c | 3 ---
> >  1 file changed, 3 deletions(-)
> >
> > diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
> > index 47e81c33d..fb2a0f1af 100644
> > --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
> > +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
> > @@ -5679,9 +5679,6 @@ static void gfx_v9_0_ring_emit_fence_kiq(struct amdgpu_ring *ring, u64 addr,
> >       struct amdgpu_device *adev = ring->adev;
> >
> > -     /* we only allocate 32bit for each seq wb address */
> > -     BUG_ON(flags & AMDGPU_FENCE_FLAG_64BIT);
> > -
> >       /* write fence seq to the "addr" */
> >       amdgpu_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3));
> >       amdgpu_ring_write(ring, (WRITE_DATA_ENGINE_SEL(0) |
>

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2026-04-29 13:57 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-04-28 16:35 [PATCH v3] drm/amdgpu/gfx9: drop unnecessary 64-bit fence flag check in KIQ John B. Moore
2026-04-29  8:22 ` Christian König
2026-04-29 13:57   ` Alex Deucher

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.