From: Tvrtko Ursulin <tursulin@ursulin.net>
To: Intel-gfx@lists.freedesktop.org
Subject: [PATCH v3 i-g-t 0/7] IGT PMU support
Date: Fri, 29 Sep 2017 13:39:32 +0100 [thread overview]
Message-ID: <20170929123939.3312-1-tvrtko.ursulin@linux.intel.com> (raw)
From: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
1.
Fixes for intel-gpu-overlay to work on top of the proposed i915 PMU perf API.
2.
New test to exercise the same API.
3.
Update to gem_wsim and media-bench.pl to be able to use engine busyness via PMU
for making balancing decisions.
v2:
* Added gem_wsim and media-bench.pl patches.
* Comments and fixes for the perf_pmu test.
v3:
* A bunch of review feedback implemented.
TODO:
* Tests for wait/sema metrics.
* RAPL PMU access for intel-gpu-overlay.
Tvrtko Ursulin (7):
intel-gpu-overlay: Move local perf implementation to a library
intel-gpu-overlay: Consolidate perf PMU access to library
intel-gpu-overlay: Fix interrupts PMU readout
intel-gpu-overlay: Catch-up to new i915 PMU
tests/perf_pmu: Tests for i915 PMU API
gem_wsim: Busy stats balancers
media-bench.pl: Add busy balancers to the list
benchmarks/Makefile.am | 2 +-
benchmarks/gem_wsim.c | 142 +++++++
lib/Makefile.am | 6 +-
lib/igt_gt.c | 50 +++
lib/igt_gt.h | 38 ++
lib/igt_perf.c | 58 +++
lib/igt_perf.h | 96 +++++
overlay/Makefile.am | 6 +-
overlay/gem-interrupts.c | 25 +-
overlay/gpu-freq.c | 29 +-
overlay/gpu-perf.c | 3 +-
overlay/gpu-top.c | 87 ++---
overlay/perf.c | 26 --
overlay/perf.h | 64 ----
overlay/power.c | 22 +-
overlay/rc6.c | 41 +-
scripts/media-bench.pl | 5 +-
tests/Makefile.am | 1 +
tests/Makefile.sources | 1 +
tests/perf_pmu.c | 957 +++++++++++++++++++++++++++++++++++++++++++++++
20 files changed, 1425 insertions(+), 234 deletions(-)
create mode 100644 lib/igt_perf.c
create mode 100644 lib/igt_perf.h
delete mode 100644 overlay/perf.c
delete mode 100644 overlay/perf.h
create mode 100644 tests/perf_pmu.c
--
2.9.5
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
next reply other threads:[~2017-09-29 12:39 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-09-29 12:39 Tvrtko Ursulin [this message]
2017-09-29 12:39 ` [PATCH i-g-t 1/7] intel-gpu-overlay: Move local perf implementation to a library Tvrtko Ursulin
2017-09-29 13:43 ` Petri Latvala
2017-10-06 15:25 ` Tvrtko Ursulin
2017-10-09 9:22 ` Petri Latvala
2017-10-09 9:54 ` Tvrtko Ursulin
2017-10-09 10:25 ` Petri Latvala
2017-09-29 12:39 ` [PATCH i-g-t 2/7] intel-gpu-overlay: Consolidate perf PMU access to library Tvrtko Ursulin
2017-09-29 12:39 ` [PATCH i-g-t 3/7] intel-gpu-overlay: Fix interrupts PMU readout Tvrtko Ursulin
2017-09-29 12:39 ` [PATCH i-g-t 4/7] intel-gpu-overlay: Catch-up to new i915 PMU Tvrtko Ursulin
2017-09-29 13:04 ` Chris Wilson
2017-09-29 12:39 ` [PATCH i-g-t 5/7] tests/perf_pmu: Tests for i915 PMU API Tvrtko Ursulin
2017-09-29 12:58 ` Petri Latvala
2017-09-29 12:39 ` [PATCH i-g-t 6/7] gem_wsim: Busy stats balancers Tvrtko Ursulin
2017-09-29 12:39 ` [PATCH i-g-t 7/7] media-bench.pl: Add busy balancers to the list Tvrtko Ursulin
2017-09-29 14:09 ` ✓ Fi.CI.BAT: success for IGT PMU support (rev5) Patchwork
2017-09-29 17:02 ` ✗ Fi.CI.IGT: warning " Patchwork
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=20170929123939.3312-1-tvrtko.ursulin@linux.intel.com \
--to=tursulin@ursulin.net \
--cc=Intel-gfx@lists.freedesktop.org \
/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.