[Intel-xe] [PATCH 1/2] drm/xe: Stash GMD_ID value in xe_gt
Matt Roper
matthew.d.roper at intel.com
Wed Dec 6 23:09:53 UTC 2023
On Wed, Dec 06, 2023 at 04:55:10PM -0600, Lucas De Marchi wrote:
> On Wed, Dec 06, 2023 at 12:50:36PM -0800, Matt Roper wrote:
> > Although we've stored the major and minor versions for graphics/media in
> > xe_device, it will be simpler to implement the uapi version query if we
> > also stash the raw register value in the GT itself.
> >
> > Signed-off-by: Matt Roper <matthew.d.roper at intel.com>
> > ---
> > drivers/gpu/drm/xe/xe_gt.c | 6 ++++++
> > drivers/gpu/drm/xe/xe_gt_types.h | 2 ++
> > 2 files changed, 8 insertions(+)
> >
> > diff --git a/drivers/gpu/drm/xe/xe_gt.c b/drivers/gpu/drm/xe/xe_gt.c
> > index 154d6c7072b9..c6f7885ffdc0 100644
> > --- a/drivers/gpu/drm/xe/xe_gt.c
> > +++ b/drivers/gpu/drm/xe/xe_gt.c
> > @@ -381,6 +381,12 @@ static int gt_fw_domain_init(struct xe_gt *gt)
> > /* Initialize CCS mode sysfs after early initialization of HW engines */
> > xe_gt_ccs_mode_sysfs_init(gt);
> >
> > + /*
> > + * Stash hardware-reported version. Since this register does not exist
> > + * on pre-MTL platforms, reading it there will (correctly) return 0.
> > + */
> > + gt->info.gmdid = xe_mmio_read32(gt, GMD_ID);
>
> I was coincidentally reading the PICA_DEVICE_ID for a debug (that we
> currently don't read) and stumbled upon this.
>
> I'm wondering what's the correct place to read and save the GMDIDs. We
> already read it on xe_pci.c:handle_gmdid(). Why don't we simply stash it
> there? With the current design, it doesn't seem we allow 2 GTs with
> different graphics / media version since it still on the xe side
> initialization.
>
> So, why don't we just stash it on xe rather than GT?
I mainly did it this way because it makes the actual uapi change a bit
simpler to actually keep this in the GT itself. Plus it's a bit more
future-proof in case we ever do start allowing multiple GTs of the same
type with different versions, or if we ever add new types of GTs beyond
just primary/media.
Matt
>
> Lucas De Marchi
>
> > +
> > err = xe_force_wake_put(gt_to_fw(gt), XE_FW_GT);
> > XE_WARN_ON(err);
> > xe_device_mem_access_put(gt_to_xe(gt));
> > diff --git a/drivers/gpu/drm/xe/xe_gt_types.h b/drivers/gpu/drm/xe/xe_gt_types.h
> > index 4e48c4643163..4e7a4494a554 100644
> > --- a/drivers/gpu/drm/xe/xe_gt_types.h
> > +++ b/drivers/gpu/drm/xe/xe_gt_types.h
> > @@ -117,6 +117,8 @@ struct xe_gt {
> > * hwconfig blob.
> > */
> > u64 __engine_mask;
> > + /** @gmdid: raw GMD_ID value from hardware */
> > + u32 gmdid;
> > } info;
> >
> > /**
> > --
> > 2.43.0
> >
--
Matt Roper
Graphics Software Engineer
Linux GPU Platform Enablement
Intel Corporation
More information about the Intel-xe
mailing list