From: "Dixit, Ashutosh" <ashutosh.dixit@intel.com>
To: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: <intel-xe@lists.freedesktop.org>,
Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Subject: Re: [PATCH 1/3] drm/xe/oa/uapi: Expose MERT OA unit
Date: Tue, 16 Dec 2025 17:12:58 -0800 [thread overview]
Message-ID: <87h5tpx65h.wl-ashutosh.dixit@intel.com> (raw)
In-Reply-To: <87tsxr8rp7.wl-ashutosh.dixit@intel.com>
On Mon, 15 Dec 2025 11:34:28 -0800, Dixit, Ashutosh wrote:
>
> On Mon, 08 Dec 2025 11:53:26 -0800, Rodrigo Vivi wrote:
> >
>
> Hi Rodrigo,
>
> > On Mon, Dec 08, 2025 at 11:10:13AM -0800, Dixit, Ashutosh wrote:
> > > On Mon, 08 Dec 2025 04:48:13 -0800, Rodrigo Vivi wrote:
> > > >
> > >
> > > Hi Rodrigo,
> > >
> > > > On Fri, Dec 05, 2025 at 01:26:11PM -0800, Ashutosh Dixit wrote:
> > > > > A MERT OA unit is available in the SoC on some platforms. Add support
> > > > > for this OA unit and expose it to userspace. The MERT OA unit does not
> > > > > have any HW engines attached, but is otherwise similar to an OAM unit.
> > > > >
> > > > > Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
> > > > > Reviewed-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
> > > > > Signed-off-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
> > > > > diff --git a/include/uapi/drm/xe_drm.h b/include/uapi/drm/xe_drm.h
> > > > > index 876a076fa6c0c..f520462c28643 100644
> > > > > --- a/include/uapi/drm/xe_drm.h
> > > > > +++ b/include/uapi/drm/xe_drm.h
> > > > > @@ -1675,6 +1675,9 @@ enum drm_xe_oa_unit_type {
> > > > >
> > > > > /** @DRM_XE_OA_UNIT_TYPE_OAM_SAG: OAM_SAG OA unit */
> > > > > DRM_XE_OA_UNIT_TYPE_OAM_SAG,
> > > > > +
> > > > > + /** @DRM_XE_OA_UNIT_TYPE_MERT: MERT OA unit */
> > > > > + DRM_XE_OA_UNIT_TYPE_MERT,
> > > >
> > > > And who&where is the usage here?
> > >
> > > Sorry I just merged this.
> >
> > No worries, I just force pushed it out.
> >
> > > I will provide the MDAPI UMD pull request for
> > > this when it is available. Hope this is ok.
> >
> > Let's re-merge when we see the usage out there.
>
> We now have a MDAPI PR for this feature:
>
> https://github.com/intel/metrics-discovery/pull/37
>
> I am still waiting for the PR to be reviewed, it is expected to be reviewed
> by EOW, at which point I am planning to re-merge this series (after
> updating this email thread). MDAPI will merge their PR after the holidays.
>
> Hope this plan is ok.
The PR is now reviewed. I went ahead and re-merged the series. Thanks.
>
> >
> > >
> > > This functionality (memory access counters) was part of OAG OA unit but is
> > > now separated out into a separate OA unit on some platforms.
> >
> > Yeap, but any flag or anything that changes the uapi we do in the defined
> > order:
> >
> > 1. both kernel and user space patches are sent out together and reviewed
> > in paralell.
> > 2. kernel code is acked by user-space developers (apparently not relevant
> > in this case)
> > 3. kernel patch lands first. (Here we have the user-space review already
> > finished and ready to get merged)
> > 4. user space patch lands right after.
> >
> > Thanks,
> > Rodrigo.
> >
> > >
> > > Thanks.
> > > --
> > > Ashutosh
next prev parent reply other threads:[~2025-12-17 1:13 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-12-05 21:26 [PATCH 0/3] drm/xe/oa: Expose MERT OA unit Ashutosh Dixit
2025-12-05 21:26 ` [PATCH 1/3] drm/xe/oa/uapi: " Ashutosh Dixit
2025-12-08 12:48 ` Rodrigo Vivi
2025-12-08 19:10 ` Dixit, Ashutosh
2025-12-08 19:53 ` Rodrigo Vivi
2025-12-15 19:34 ` Dixit, Ashutosh
2025-12-17 1:12 ` Dixit, Ashutosh [this message]
2025-12-17 22:36 ` Rodrigo Vivi
2025-12-05 21:26 ` [PATCH 2/3] drm/xe/rtp: Whitelist OAMERT MMIO trigger registers Ashutosh Dixit
2025-12-05 23:34 ` Umesh Nerlige Ramappa
2025-12-05 21:26 ` [PATCH 3/3] drm/xe/oa: Always set OAG_OAGLBCTXCTRL_COUNTER_RESUME Ashutosh Dixit
2025-12-05 23:36 ` Umesh Nerlige Ramappa
2025-12-05 23:15 ` ✓ CI.KUnit: success for drm/xe/oa: Expose MERT OA unit Patchwork
2025-12-06 0:27 ` ✓ Xe.CI.BAT: " Patchwork
2025-12-06 12:07 ` ✗ Xe.CI.Full: failure " Patchwork
2025-12-15 19:37 ` [PATCH 0/3] " Dixit, Ashutosh
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=87h5tpx65h.wl-ashutosh.dixit@intel.com \
--to=ashutosh.dixit@intel.com \
--cc=intel-xe@lists.freedesktop.org \
--cc=rodrigo.vivi@intel.com \
--cc=umesh.nerlige.ramappa@intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
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.