From: Lucas De Marchi <lucas.demarchi@intel.com>
To: igt-dev@lists.freedesktop.org
Subject: [CI 4/6] lib/igt_drm_fdinfo: Parse drm-total-cycles
Date: Tue, 21 May 2024 09:15:23 -0700 [thread overview]
Message-ID: <20240521161525.159720-4-lucas.demarchi@intel.com> (raw)
In-Reply-To: <20240521161525.159720-1-lucas.demarchi@intel.com>
Parse drm-total-cycles and annotate what are the possible engine
utilization types that can be used. Currently there are 2 possible ways
to calculate utilization:
1) via drm-engine, that uses the time and derive the utilization by
comparing to the CPU time
2) via drm-cycles + drm-total-cycles, that uses the GPU time
igt_drm_fdinfo just marks the presence of the keys, letting the caller
decide which method to use.
It is assumed that if any drm-engine, drm-cycles or drm-total-cycles is
available, it's available for all engines with non-zero utilization.
Reviewed-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Link: https://lore.kernel.org/r/20240504064643.25863-12-lucas.demarchi@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
lib/igt_drm_fdinfo.c | 4 ++++
lib/igt_drm_fdinfo.h | 3 +++
2 files changed, 7 insertions(+)
diff --git a/lib/igt_drm_fdinfo.c b/lib/igt_drm_fdinfo.c
index 9622d05b7..539d15056 100644
--- a/lib/igt_drm_fdinfo.c
+++ b/lib/igt_drm_fdinfo.c
@@ -265,6 +265,10 @@ __igt_parse_drm_fdinfo(int dir, const char *fd, struct drm_client_fdinfo *info,
idx = parse_engine(l + keylen, info,
name_map, map_entries, &val);
UPDATE_ENGINE(idx, cycles, val, DRM_FDINFO_UTILIZATION_CYCLES);
+ } else if (strstartswith(l, "drm-total-cycles-", &keylen)) {
+ idx = parse_engine(l + keylen, info,
+ name_map, map_entries, &val);
+ UPDATE_ENGINE(idx, total_cycles, val, DRM_FDINFO_UTILIZATION_TOTAL_CYCLES);
} else if (strstartswith(l, "drm-total-", &keylen)) {
idx = parse_region(l + keylen, info,
region_map, region_entries, &val);
diff --git a/lib/igt_drm_fdinfo.h b/lib/igt_drm_fdinfo.h
index ea90edd79..35f43305b 100644
--- a/lib/igt_drm_fdinfo.h
+++ b/lib/igt_drm_fdinfo.h
@@ -44,6 +44,7 @@ struct drm_client_meminfo {
enum drm_fdinfo_utilization {
DRM_FDINFO_UTILIZATION_ENGINE_TIME = 1U << 0,
DRM_FDINFO_UTILIZATION_CYCLES = 1U << 1,
+ DRM_FDINFO_UTILIZATION_TOTAL_CYCLES = 1U << 2,
};
struct drm_client_fdinfo {
@@ -63,6 +64,8 @@ struct drm_client_fdinfo {
uint64_t engine_time[DRM_CLIENT_FDINFO_MAX_ENGINES];
/* drm-cycles-<engine> values */
uint64_t cycles[DRM_CLIENT_FDINFO_MAX_ENGINES];
+ /* drm-total-cycles-<engine> values */
+ uint64_t total_cycles[DRM_CLIENT_FDINFO_MAX_ENGINES];
unsigned int num_regions;
unsigned int last_region_index;
--
2.43.0
next prev parent reply other threads:[~2024-05-21 16:15 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-05-21 16:15 [CI 1/6] treewide: Rename engine busyness variables Lucas De Marchi
2024-05-21 16:15 ` [CI 2/6] lib/igt_drm_clients: Move engine fields to substruct Lucas De Marchi
2024-05-21 16:15 ` [CI 3/6] lib/igt_drm_clients: Record drm-cycles Lucas De Marchi
2024-05-21 16:15 ` Lucas De Marchi [this message]
2024-05-21 16:15 ` [CI 5/6] lib/igt_drm_clients: Record total cycles Lucas De Marchi
2024-05-21 16:15 ` [CI 6/6] gputop: Add support to drm-cycles/drm-total-cycles Lucas De Marchi
2024-05-21 19:00 ` ✓ Fi.CI.BAT: success for series starting with [CI,1/6] treewide: Rename engine busyness variables Patchwork
2024-05-21 19:41 ` ✗ CI.xeBAT: failure " Patchwork
2024-05-22 2:15 ` ✗ CI.xeFULL: " Patchwork
2024-05-22 8:01 ` ✓ Fi.CI.IGT: success " 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=20240521161525.159720-4-lucas.demarchi@intel.com \
--to=lucas.demarchi@intel.com \
--cc=igt-dev@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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox