Igt-dev Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Lucas De Marchi <lucas.demarchi@intel.com>
To: igt-dev@lists.freedesktop.org
Cc: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>,
	Tvrtko Ursulin <tursulin@ursulin.net>,
	Lucas De Marchi <lucas.demarchi@intel.com>
Subject: [PATCH i-g-t v2 00/21] gputop: Add support for xe
Date: Tue, 23 Apr 2024 16:44:10 -0700	[thread overview]
Message-ID: <20240423234431.1959354-1-lucas.demarchi@intel.com> (raw)

v2 of https://lore.kernel.org/all/20240405060056.59379-1-lucas.demarchi@intel.com/

This contains all the patches I had pending for igt to add support in
gputop for the xe driver.

Not all of the patches are critical, some are more on the category
"drive by improvements/fixes". Let me split them for ease of review:

Patches 1-6: already reviewed in previous version, but needed to
refactor them a little bit to detangle them from other patches. I kept
the previous R-b for those.

Patches 7-10: these were already in the previous version as well, but in
a more noisy approach. This new version tries not to change multiple
times the same part of the code.

Tvrtko, worth noting that patches 7-10 still contain the strstartwith()
rather than find_kv(). I did try adding find_kv() back but due to that
function not inlining the strlen() was being computed again and again. I
decided to keep as is so could move to the more important part. Maybe we
can try something else on top.

Patches 11-16 implement the new interface from the kernel. Difference
from v1 is that it's now using the drm-cycles + drm-total-cycles
suggested in the kernel side review. The state tracking is a little bit
heavier due to having to handle 2 keys, possibly appearing in different
order and at the same time decide what strategy to use (drm-engine vs
drm-cycles). However with the additional renames I think this is much
cleaner than v1.

Patches 17-21 are gputop improvements to the visualization and
performance: last patch with lazy stat is back from an older series, now
with proper performance measurements.

Lucas De Marchi (21):
  lib/igt_drm_fdinfo: Fix wrong name len assert
  lib/igt_drm_fdinfo: Assert pdev is not truncated
  lib/igt_drm_fdinfo: Detect invalid drm-client-id
  lib/igt_drm_fdinfo: Stop ignoring space where not needed
  gputop: Extract method to update console size
  gputop: Extract clrscr()
  lib/igt_drm_fdinfo: Extract ignore_space()
  lib/igt_drm_fdinfo: Allow any number of spaces before unit
  lib/igt_drm_fdinfo: Stop passing key twice
  lib/igt_drm_fdinfo: Remove prefix arg from parse functions
  lib/igt_drm_fdinfo: Parse drm-cycles
  lib/igt_drm_fdinfo: Parse drm-total-cycles
  lib/igt_drm_clients: Rename engine busyness variables
  lib/igt_drm_clients: Move engine fields to substruct
  lib/igt_drm_clients: Record total cycles
  gputop: Add support to drm-cycles/drm-total-cycles
  gputop: print percentage number
  gputop: Add option parsing
  gputop: Allow to limit number of executions
  gputop: Allow to set delay interval
  lib/igt_drm_clients: lazy stat process

 lib/igt_core.h        |   1 +
 lib/igt_drm_clients.c | 106 ++++++++++++++++----------
 lib/igt_drm_clients.h |   9 ++-
 lib/igt_drm_fdinfo.c  | 172 ++++++++++++++++++++++++------------------
 lib/igt_drm_fdinfo.h  |  12 +++
 tools/gputop.c        | 142 +++++++++++++++++++++++++++++-----
 tools/intel_gpu_top.c |  15 ++--
 7 files changed, 315 insertions(+), 142 deletions(-)

-- 
2.43.0


             reply	other threads:[~2024-04-23 23:43 UTC|newest]

Thread overview: 36+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-04-23 23:44 Lucas De Marchi [this message]
2024-04-23 23:44 ` [PATCH i-g-t v2 01/21] lib/igt_drm_fdinfo: Fix wrong name len assert Lucas De Marchi
2024-04-23 23:44 ` [PATCH i-g-t v2 02/21] lib/igt_drm_fdinfo: Assert pdev is not truncated Lucas De Marchi
2024-04-23 23:44 ` [PATCH i-g-t v2 03/21] lib/igt_drm_fdinfo: Detect invalid drm-client-id Lucas De Marchi
2024-04-23 23:44 ` [PATCH i-g-t v2 04/21] lib/igt_drm_fdinfo: Stop ignoring space where not needed Lucas De Marchi
2024-04-23 23:44 ` [PATCH i-g-t v2 05/21] gputop: Extract method to update console size Lucas De Marchi
2024-04-23 23:44 ` [PATCH i-g-t v2 06/21] gputop: Extract clrscr() Lucas De Marchi
2024-04-23 23:44 ` [PATCH i-g-t v2 07/21] lib/igt_drm_fdinfo: Extract ignore_space() Lucas De Marchi
2024-04-23 23:44 ` [PATCH i-g-t v2 08/21] lib/igt_drm_fdinfo: Allow any number of spaces before unit Lucas De Marchi
2024-04-23 23:44 ` [PATCH i-g-t v2 09/21] lib/igt_drm_fdinfo: Stop passing key twice Lucas De Marchi
2024-04-25 21:27   ` Lucas De Marchi
2024-04-23 23:44 ` [PATCH i-g-t v2 10/21] lib/igt_drm_fdinfo: Remove prefix arg from parse functions Lucas De Marchi
2024-04-23 23:44 ` [PATCH i-g-t v2 11/21] lib/igt_drm_fdinfo: Parse drm-cycles Lucas De Marchi
2024-04-23 23:44 ` [PATCH i-g-t v2 12/21] lib/igt_drm_fdinfo: Parse drm-total-cycles Lucas De Marchi
2024-04-26 11:03   ` Tvrtko Ursulin
2024-04-23 23:44 ` [PATCH i-g-t v2 13/21] lib/igt_drm_clients: Rename engine busyness variables Lucas De Marchi
2024-04-23 23:44 ` [PATCH i-g-t v2 14/21] lib/igt_drm_clients: Move engine fields to substruct Lucas De Marchi
2024-04-26 11:05   ` Tvrtko Ursulin
2024-04-23 23:44 ` [PATCH i-g-t v2 15/21] lib/igt_drm_clients: Record total cycles Lucas De Marchi
2024-04-26 11:11   ` Tvrtko Ursulin
2024-04-26 13:28     ` Lucas De Marchi
2024-04-23 23:44 ` [PATCH i-g-t v2 16/21] gputop: Add support to drm-cycles/drm-total-cycles Lucas De Marchi
2024-04-23 23:44 ` [PATCH i-g-t v2 17/21] gputop: print percentage number Lucas De Marchi
2024-04-26 11:16   ` Tvrtko Ursulin
2024-04-23 23:44 ` [PATCH i-g-t v2 18/21] gputop: Add option parsing Lucas De Marchi
2024-04-23 23:44 ` [PATCH i-g-t v2 19/21] gputop: Allow to limit number of executions Lucas De Marchi
2024-04-23 23:44 ` [PATCH i-g-t v2 20/21] gputop: Allow to set delay interval Lucas De Marchi
2024-04-23 23:44 ` [PATCH i-g-t v2 21/21] lib/igt_drm_clients: lazy stat process Lucas De Marchi
2024-04-24  0:32 ` ✗ GitLab.Pipeline: warning for gputop: Add support for xe (rev2) Patchwork
2024-04-24  0:53 ` ✓ Fi.CI.BAT: success " Patchwork
2024-04-24  1:29 ` ✓ CI.xeBAT: " Patchwork
2024-04-24 11:47 ` ✗ Fi.CI.IGT: failure " Patchwork
2024-04-24 15:23 ` ✗ CI.xeFULL: " Patchwork
2024-04-25 22:17 ` ✓ CI.xeBAT: success for gputop: Add support for xe (rev3) Patchwork
2024-04-25 22:20 ` ✓ Fi.CI.BAT: " Patchwork
2024-04-26  1:38 ` ✗ Fi.CI.IGT: failure " 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=20240423234431.1959354-1-lucas.demarchi@intel.com \
    --to=lucas.demarchi@intel.com \
    --cc=igt-dev@lists.freedesktop.org \
    --cc=tursulin@ursulin.net \
    --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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox