2023-04-24 06:05:11

by Sukrut Bellary

Subject: [PATCH] drm:amd:amdgpu: Fix missing bo unlock in failure path

smatch warning - inconsistent handling of buffer object reserve
and unreserve.

Signed-off-by: Sukrut Bellary <[email protected]>
---
drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
index 278416acf060..5de44d7e92de 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
@@ -4686,8 +4686,10 @@ static int gfx_v8_0_kiq_resume(struct amdgpu_device *adev)
return r;

r = amdgpu_bo_kmap(ring->mqd_obj, &ring->mqd_ptr);
- if (unlikely(r != 0))
+ if (unlikely(r != 0)) {
+ amdgpu_bo_unreserve(ring->mqd_obj);
return r;
+ }

gfx_v8_0_kiq_init_queue(ring);
amdgpu_bo_kunmap(ring->mqd_obj);
--
2.34.1

2023-04-24 07:07:47

by Christian König

Subject: Re: [PATCH] drm:amd:amdgpu: Fix missing bo unlock in failure path

Am 24.04.23 um 07:59 schrieb Sukrut Bellary:
> smatch warning - inconsistent handling of buffer object reserve
> and unreserve.
>
> Signed-off-by: Sukrut Bellary <[email protected]>

For now that patch is Reviewed-by: Christian König
<[email protected]>.

But for the record mapping/unmapping the MQD like this is a very bad
idea in the first place.

We could need to shuffle memory around for that during resume and that
is not something we really want to do.

Christian.

> ---
> drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c | 4 +++-
> 1 file changed, 3 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
> index 278416acf060..5de44d7e92de 100644
> --- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
> +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
> @@ -4686,8 +4686,10 @@ static int gfx_v8_0_kiq_resume(struct amdgpu_device *adev)
> return r;
>
> r = amdgpu_bo_kmap(ring->mqd_obj, &ring->mqd_ptr);
> - if (unlikely(r != 0))
> + if (unlikely(r != 0)) {
> + amdgpu_bo_unreserve(ring->mqd_obj);
> return r;
> + }
>
> gfx_v8_0_kiq_init_queue(ring);
> amdgpu_bo_kunmap(ring->mqd_obj);

2023-04-24 15:12:47

by Alex Deucher

Subject: Re: [PATCH] drm:amd:amdgpu: Fix missing bo unlock in failure path

On Mon, Apr 24, 2023 at 3:07 AM Christian König
<[email protected]> wrote:
>
> Am 24.04.23 um 07:59 schrieb Sukrut Bellary:
> > smatch warning - inconsistent handling of buffer object reserve
> > and unreserve.
> >
> > Signed-off-by: Sukrut Bellary <[email protected]>
>
> For now that patch is Reviewed-by: Christian König
> <[email protected]>.

Applied. Thanks.

>
> But for the record mapping/unmapping the MQD like this is a very bad
> idea in the first place.
>
> We could need to shuffle memory around for that during resume and that
> is not something we really want to do.

We should probably just keep the MQDs mapped. On suspend we need to
save out the MQD state so it can be restored on resume when the MQDs
are in vram.

Alex

>
> Christian.
>
> > ---
> > drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c | 4 +++-
> > 1 file changed, 3 insertions(+), 1 deletion(-)
> >
> > diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
> > index 278416acf060..5de44d7e92de 100644
> > --- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
> > +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
> > @@ -4686,8 +4686,10 @@ static int gfx_v8_0_kiq_resume(struct amdgpu_device *adev)
> > return r;
> >
> > r = amdgpu_bo_kmap(ring->mqd_obj, &ring->mqd_ptr);
> > - if (unlikely(r != 0))
> > + if (unlikely(r != 0)) {
> > + amdgpu_bo_unreserve(ring->mqd_obj);
> > return r;
> > + }
> >
> > gfx_v8_0_kiq_init_queue(ring);
> > amdgpu_bo_kunmap(ring->mqd_obj);
>