All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Dixit, Ashutosh" <ashutosh.dixit@intel.com>
To: "Sebinraj, S" <s.sebinraj@intel.com>
Cc: "De Marchi, Lucas" <lucas.demarchi@intel.com>,
	"intel-xe@lists.freedesktop.org" <intel-xe@lists.freedesktop.org>,
	"Badrappan, Jeevaka" <jeevaka.badrappan@intel.com>,
	Rodrigo Vivi <rodrigo.vivi@intel.com>
Subject: Re: [PATCH v2] drm/xe: GPU frequency tracing support
Date: Fri, 22 Aug 2025 13:58:33 -0700	[thread overview]
Message-ID: <87wm6vxedi.wl-ashutosh.dixit@intel.com> (raw)
In-Reply-To: <CH0PR11MB5427BE3DA4B3078A7514BCB79833A@CH0PR11MB5427.namprd11.prod.outlook.com>

On Tue, 19 Aug 2025 21:25:43 -0700, Sebinraj, S wrote:
>
> We went for this addition rather than the one in xe_pmu (via perf stat - user space tool),
> as we have a requirement to have the trace point available at d/evens/power/gpu_frequency for Android xTS.
> In this case the user won't be triggering the perf subsystem for gpu frequency for gt0, gt1.. even if we add the trace points in xe_pmu.

Why not? Which userspace tool is consuming this ftrace and they can't
enable the PMU event?

The way the ftrace event is implemented here, it really doesn't provide any
more information than what is provided by the PMU. IMO, a ftrace might be
more acceptable if the freq change event emanated from GuC itself where the
freq change is actually implemented. That would need a change to GuC FW, so
maybe not an option at present.

Also, similar to Rodrigo's comment on the work period series, if we need
d/events/power/gpu_frequency (rather than d/events/power/xe_gpu_frequency),
that should be added at the drm level rather than at xe level? Would that
be acceptable?

Thanks.
--
Ashutosh

  parent reply	other threads:[~2025-08-22 20:58 UTC|newest]

Thread overview: 16+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-08-11  8:34 [PATCH v2] drm/xe: GPU frequency tracing support S Sebinraj
2025-08-11  9:39 ` ✗ CI.checkpatch: warning for drm/xe: GPU frequency tracing support (rev2) Patchwork
2025-08-11  9:40 ` ✓ CI.KUnit: success " Patchwork
2025-08-11 10:42 ` ✓ Xe.CI.BAT: " Patchwork
2025-08-11 11:38 ` ✗ Xe.CI.Full: failure " Patchwork
2025-08-11 20:05 ` [PATCH v2] drm/xe: GPU frequency tracing support Matthew Brost
2025-08-11 20:40 ` Matthew Brost
2025-08-11 21:03   ` Matthew Brost
2025-08-11 21:13 ` Lucas De Marchi
2025-08-11 21:28   ` Matthew Brost
2025-08-20  4:25   ` Sebinraj, S
2025-08-22 19:30     ` Lucas De Marchi
2025-08-27  4:35       ` Matthew Brost
2025-09-03  5:54       ` Sebinraj, S
2025-08-22 20:58     ` Dixit, Ashutosh [this message]
2025-09-03  6:03       ` Sebinraj, S

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=87wm6vxedi.wl-ashutosh.dixit@intel.com \
    --to=ashutosh.dixit@intel.com \
    --cc=intel-xe@lists.freedesktop.org \
    --cc=jeevaka.badrappan@intel.com \
    --cc=lucas.demarchi@intel.com \
    --cc=rodrigo.vivi@intel.com \
    --cc=s.sebinraj@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.