public inbox for intel-gfx@lists.freedesktop.org
 help / color / mirror / Atom feed
From: Tvrtko Ursulin <tursulin@ursulin.net>
To: Intel-gfx@lists.freedesktop.org
Cc: Ben Widawsky <benjamin.widawsky@intel.com>
Subject: [RFC 00/14] i915 PMU and engine busy stats
Date: Tue, 18 Jul 2017 15:36:04 +0100	[thread overview]
Message-ID: <20170718143618.12254-1-tvrtko.ursulin@linux.intel.com> (raw)

From: Tvrtko Ursulin <tvrtko.ursulin@intel.com>

Rough sketch of the idea I mentioned a few times to various people - merging
the engine busyness tracking with Chris i915 PMU RFC.

First patch is the actual PMU RFC by Chris. It is followed by some cleanup
patches, then come a few improvements, cheap execlists engine busyness tracking,
debugfs view for the same, and finally the i915 PMU is extended to use this
instead of timer based mmio sampling.

This makes it cheaper and also more accurate since engine busyness is not
derived via sampling.

But I haven't figure out the perf API yet. For example is it possible to access
our events in an usable fashion via perf top/stat or something? Do we want to
make the events discoverable as I did (patch 8).

I could not find much (any?) kernel API level documentation for perf.

Btw patch series actually works since intel-gpu-overlay can use these events
when they are available.

Chris Wilson (1):
  RFC drm/i915: Expose a PMU interface for perf queries

Tvrtko Ursulin (13):
  drm/i915/pmu: Add VCS2 engine to the PMU uAPI
  drm/i915/pmu: Add queued samplers to the PMU uAPI
  drm/i915/pmu: Decouple uAPI engine ids
  drm/i915/pmu: Helper to extract engine and sampler from PMU config
  drm/i915/pmu: Only sample enabled samplers
  drm/i915/pmu: Add fake regs
  drm/i915/pmu: Expose events in sysfs
  drm/i915/pmu: Suspend sampling when GPU is idle
  drm/i915: Wrap context schedule notification
  drm/i915: Engine busy time tracking
  drm/i915: Interface for controling engine stats collection
  drm/i915: Export engine busy stats in debugfs
  drm/i915/pmu: Wire up engine busy stats to PMU

 drivers/gpu/drm/i915/Makefile           |   1 +
 drivers/gpu/drm/i915/i915_debugfs.c     |  89 ++++
 drivers/gpu/drm/i915/i915_drv.c         |   2 +
 drivers/gpu/drm/i915/i915_drv.h         |  32 ++
 drivers/gpu/drm/i915/i915_gem.c         |   1 +
 drivers/gpu/drm/i915/i915_gem_request.c |   1 +
 drivers/gpu/drm/i915/i915_pmu.c         | 697 ++++++++++++++++++++++++++++++++
 drivers/gpu/drm/i915/intel_engine_cs.c  |  59 +++
 drivers/gpu/drm/i915/intel_lrc.c        |  19 +-
 drivers/gpu/drm/i915/intel_ringbuffer.h |  46 +++
 include/uapi/drm/i915_drm.h             |  51 +++
 kernel/events/core.c                    |   1 +
 12 files changed, 996 insertions(+), 3 deletions(-)
 create mode 100644 drivers/gpu/drm/i915/i915_pmu.c

-- 
2.9.4

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

             reply	other threads:[~2017-07-18 14:36 UTC|newest]

Thread overview: 38+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-07-18 14:36 Tvrtko Ursulin [this message]
2017-07-18 14:36 ` [RFC 01/14] RFC drm/i915: Expose a PMU interface for perf queries Tvrtko Ursulin
2017-07-19  9:53   ` Kamble, Sagar A
2017-07-20  8:55     ` Tvrtko Ursulin
2017-07-25  1:09   ` Ben Widawsky
2017-07-18 14:36 ` [RFC 02/14] drm/i915/pmu: Add VCS2 engine to the PMU uAPI Tvrtko Ursulin
2017-07-18 14:36 ` [RFC 03/14] drm/i915/pmu: Add queued samplers " Tvrtko Ursulin
2017-07-18 14:36 ` [RFC 04/14] drm/i915/pmu: Decouple uAPI engine ids Tvrtko Ursulin
2017-07-25  1:18   ` Ben Widawsky
2017-07-26  9:04     ` Tvrtko Ursulin
2017-07-18 14:36 ` [RFC 05/14] drm/i915/pmu: Helper to extract engine and sampler from PMU config Tvrtko Ursulin
2017-07-18 14:36 ` [RFC 06/14] drm/i915/pmu: Only sample enabled samplers Tvrtko Ursulin
2017-07-18 14:36 ` [RFC 07/14] drm/i915/pmu: Add fake regs Tvrtko Ursulin
2017-07-25  1:20   ` Ben Widawsky
2017-07-26  9:07     ` Tvrtko Ursulin
2017-07-18 14:36 ` [RFC 08/14] drm/i915/pmu: Expose events in sysfs Tvrtko Ursulin
2017-07-18 14:36 ` [RFC 09/14] drm/i915/pmu: Suspend sampling when GPU is idle Tvrtko Ursulin
2017-07-18 14:36 ` [RFC 10/14] drm/i915: Wrap context schedule notification Tvrtko Ursulin
2017-07-18 14:36 ` [RFC 11/14] drm/i915: Engine busy time tracking Tvrtko Ursulin
2017-07-18 15:19   ` Chris Wilson
2017-07-19  9:12     ` Tvrtko Ursulin
2017-07-19 10:46       ` Chris Wilson
2017-07-18 14:36 ` [RFC 12/14] drm/i915: Interface for controling engine stats collection Tvrtko Ursulin
2017-07-18 15:22   ` Chris Wilson
2017-07-19  9:30     ` Tvrtko Ursulin
2017-07-19 11:04       ` Chris Wilson
2017-07-20  9:07         ` Tvrtko Ursulin
2017-07-18 15:43   ` Chris Wilson
2017-07-18 18:43   ` Chris Wilson
2017-07-19  9:34   ` Tvrtko Ursulin
2017-07-25  1:28     ` Ben Widawsky
2017-07-18 14:36 ` [RFC 13/14] drm/i915: Export engine busy stats in debugfs Tvrtko Ursulin
2017-07-18 14:36 ` [RFC 14/14] drm/i915/pmu: Wire up engine busy stats to PMU Tvrtko Ursulin
2017-07-18 14:58 ` ✓ Fi.CI.BAT: success for i915 PMU and engine busy stats Patchwork
2017-07-19 12:05 ` [RFC 00/14] " Chris Wilson
2017-07-20  9:03   ` Tvrtko Ursulin
2017-07-26 10:34     ` Tvrtko Ursulin
2017-07-26 10:55       ` Chris Wilson

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=20170718143618.12254-1-tvrtko.ursulin@linux.intel.com \
    --to=tursulin@ursulin.net \
    --cc=Intel-gfx@lists.freedesktop.org \
    --cc=benjamin.widawsky@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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox