[PATCH v4 3/3] drm/amdgpu: Add support for querying the max ibs in a submission. (v3)
Alex Deucher
alexdeucher at gmail.com
Fri Apr 14 21:46:45 UTC 2023
On Fri, Apr 14, 2023 at 5:24 PM Bas Nieuwenhuizen
<bas at basnieuwenhuizen.nl> wrote:
>
> On Fri, Apr 14, 2023 at 11:18 PM Alex Deucher <alexdeucher at gmail.com> wrote:
> >
> > On Thu, Apr 13, 2023 at 10:25 AM Bas Nieuwenhuizen
> > <bas at basnieuwenhuizen.nl> wrote:
> > >
> > > This info would be used by radv to figure out when we need to
> > > split a submission into multiple submissions. radv currently has
> > > a limit of 192 which seems to work for most gfx submissions, but
> > > is way too high for e.g. compute or sdma.
> > >
> > > Userspace is available at
> > > https://gitlab.freedesktop.org/bnieuwenhuizen/mesa/-/commits/ib-rejection-v3
> > >
> > > v3: Completely rewrote based on suggestion of making it a separate query.
> > >
> > > Link: https://gitlab.freedesktop.org/drm/amd/-/issues/2498
> > > Signed-off-by: Bas Nieuwenhuizen <bas at basnieuwenhuizen.nl>
> >
> > Maybe a 4th patch to bump the driver version?
>
> In the userspace I just make the query and put in some guessed values
> if it errors out, which doesn't need the driver version enumeration.
Ok. Applied the series. Thanks!
Alex
>
> >
> > Alex
> >
> > > ---
> > > drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c | 9 +++++++++
> > > include/uapi/drm/amdgpu_drm.h | 2 ++
> > > 2 files changed, 11 insertions(+)
> > >
> > > diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
> > > index 0efb38539d70..1a2e342af1c0 100644
> > > --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
> > > +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
> > > @@ -1140,6 +1140,15 @@ int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
> > > kfree(caps);
> > > return r;
> > > }
> > > + case AMDGPU_INFO_MAX_IBS: {
> > > + uint32_t max_ibs[AMDGPU_HW_IP_NUM];
> > > +
> > > + for (i = 0; i < AMDGPU_HW_IP_NUM; ++i)
> > > + max_ibs[i] = amdgpu_ring_max_ibs(i);
> > > +
> > > + return copy_to_user(out, max_ibs,
> > > + min((size_t)size, sizeof(max_ibs))) ? -EFAULT : 0;
> > > + }
> > > default:
> > > DRM_DEBUG_KMS("Invalid request %d\n", info->query);
> > > return -EINVAL;
> > > diff --git a/include/uapi/drm/amdgpu_drm.h b/include/uapi/drm/amdgpu_drm.h
> > > index b6eb90df5d05..6981e59a9401 100644
> > > --- a/include/uapi/drm/amdgpu_drm.h
> > > +++ b/include/uapi/drm/amdgpu_drm.h
> > > @@ -876,6 +876,8 @@ struct drm_amdgpu_cs_chunk_data {
> > > #define AMDGPU_INFO_VIDEO_CAPS_DECODE 0
> > > /* Subquery id: Encode */
> > > #define AMDGPU_INFO_VIDEO_CAPS_ENCODE 1
> > > +/* Query the max number of IBs per gang per submission */
> > > +#define AMDGPU_INFO_MAX_IBS 0x22
> > >
> > > #define AMDGPU_INFO_MMR_SE_INDEX_SHIFT 0
> > > #define AMDGPU_INFO_MMR_SE_INDEX_MASK 0xff
> > > --
> > > 2.40.0
> > >
More information about the amd-gfx
mailing list