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 v3 13/13] gputop: Add support to drm-cycles/drm-total-cycles
Date: Fri,  3 May 2024 23:46:43 -0700	[thread overview]
Message-ID: <20240504064643.25863-14-lucas.demarchi@intel.com> (raw)
In-Reply-To: <20240504064643.25863-1-lucas.demarchi@intel.com>

Add support for using only GPU timestamps to calculate the utilization.
It uses drm-cycles and drm-total-cycles read from fdinfo. For any
2 samples, the utilization is calculated as:

	         u[i+1] - u[i]
	pct  = -----------------
	        Gt[i+1] - Gt[i]

with u == drm-cycles and Gt == drm-total-cycles.

Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 tools/gputop.c | 41 +++++++++++++++++++++++++++++++++++++----
 1 file changed, 37 insertions(+), 4 deletions(-)

diff --git a/tools/gputop.c b/tools/gputop.c
index aa88a8021..11e0db482 100644
--- a/tools/gputop.c
+++ b/tools/gputop.c
@@ -30,6 +30,11 @@
 #include "igt_drm_fdinfo.h"
 #include "drmtest.h"
 
+enum utilization_type {
+	UTILIZATION_TYPE_ENGINE_TIME,
+	UTILIZATION_TYPE_TOTAL_CYCLES,
+};
+
 static const char *bars[] = { " ", "▏", "▎", "▍", "▌", "▋", "▊", "▉", "█" };
 
 static void n_spaces(const unsigned int n)
@@ -171,13 +176,33 @@ print_client(struct igt_drm_client *c, struct igt_drm_client **prevc,
 	     double t, int lines, int con_w, int con_h,
 	     unsigned int period_us, int *engine_w)
 {
+	enum utilization_type utilization_type;
 	unsigned int i;
 	uint64_t sz;
 	int len;
 
+	if (c->utilization_mask & IGT_DRM_CLIENT_UTILIZATION_TOTAL_CYCLES &&
+	    c->utilization_mask & IGT_DRM_CLIENT_UTILIZATION_CYCLES)
+		utilization_type = UTILIZATION_TYPE_TOTAL_CYCLES;
+	else if (c->utilization_mask & IGT_DRM_CLIENT_UTILIZATION_ENGINE_TIME)
+		utilization_type = UTILIZATION_TYPE_ENGINE_TIME;
+	else
+		return 0;
+
+	if (c->samples < 2)
+		return 0;
+
 	/* Filter out idle clients. */
-	if (!c->total_engine_time || c->samples < 2)
-		return lines;
+	switch (utilization_type) {
+	case UTILIZATION_TYPE_ENGINE_TIME:
+	       if (!c->total_engine_time)
+		       return 0;
+	       break;
+	case UTILIZATION_TYPE_TOTAL_CYCLES:
+	       if (!c->total_total_cycles)
+		       return 0;
+	       break;
+	}
 
 	/* Print header when moving to a different DRM card. */
 	if (newheader(c, *prevc)) {
@@ -208,8 +233,16 @@ print_client(struct igt_drm_client *c, struct igt_drm_client **prevc,
 		if (!c->engines->capacity[i])
 			continue;
 
-		pct = (double)c->utilization[i].delta_engine_time / period_us / 1e3 * 100 /
-		      c->engines->capacity[i];
+		switch (utilization_type) {
+		case UTILIZATION_TYPE_ENGINE_TIME:
+			pct = (double)c->utilization[i].delta_engine_time / period_us / 1e3 * 100 /
+				c->engines->capacity[i];
+			break;
+		case UTILIZATION_TYPE_TOTAL_CYCLES:
+			pct = (double)c->utilization[i].delta_cycles / c->utilization[i].delta_total_cycles * 100 /
+				c->engines->capacity[i];
+			break;
+		}
 
 		/*
 		 * Guard against fluctuations between our scanning period and
-- 
2.43.0


  parent reply	other threads:[~2024-05-04  6:46 UTC|newest]

Thread overview: 36+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-05-04  6:46 [PATCH i-g-t v3 00/12] gputop: Add support for xe Lucas De Marchi
2024-05-04  6:46 ` [PATCH i-g-t v3 01/13] lib/igt_drm_fdinfo: Extract ignore_space() Lucas De Marchi
2024-05-06 20:21   ` Umesh Nerlige Ramappa
2024-05-04  6:46 ` [PATCH i-g-t v3 02/13] lib/igt_drm_fdinfo: Allow any number of spaces before unit Lucas De Marchi
2024-05-06 20:28   ` Umesh Nerlige Ramappa
2024-05-06 20:32     ` Umesh Nerlige Ramappa
2024-05-06 20:38       ` Lucas De Marchi
2024-05-04  6:46 ` [PATCH i-g-t v3 03/13] fixup! " Lucas De Marchi
2024-05-06 20:33   ` Umesh Nerlige Ramappa
2024-05-04  6:46 ` [PATCH i-g-t v3 04/13] lib/igt_drm_fdinfo: Stop passing key twice Lucas De Marchi
2024-05-04  6:46 ` [PATCH i-g-t v3 05/13] lib/igt_drm_fdinfo: Remove prefix arg from parse functions Lucas De Marchi
2024-05-04  6:46 ` [PATCH i-g-t v3 06/13] lib/igt_drm_fdinfo: Parse drm-cycles Lucas De Marchi
2024-05-07  1:11   ` Umesh Nerlige Ramappa
2024-05-04  6:46 ` [PATCH i-g-t v3 07/13] lib/igt_drm_fdinfo: Start tracking available engine keys Lucas De Marchi
2024-05-07  1:14   ` Umesh Nerlige Ramappa
2024-05-04  6:46 ` [PATCH i-g-t v3 08/13] treewide: Rename engine busyness variables Lucas De Marchi
2024-05-08 18:45   ` Umesh Nerlige Ramappa
2024-05-21 13:47     ` Lucas De Marchi
2024-05-04  6:46 ` [PATCH i-g-t v3 09/13] lib/igt_drm_clients: Move engine fields to substruct Lucas De Marchi
2024-05-08 18:45   ` Umesh Nerlige Ramappa
2024-05-04  6:46 ` [PATCH i-g-t v3 10/13] lib/igt_drm_clients: Record drm-cycles Lucas De Marchi
2024-05-08 18:48   ` Umesh Nerlige Ramappa
2024-05-04  6:46 ` [PATCH i-g-t v3 11/13] lib/igt_drm_fdinfo: Parse drm-total-cycles Lucas De Marchi
2024-05-08 18:49   ` Umesh Nerlige Ramappa
2024-05-04  6:46 ` [PATCH i-g-t v3 12/13] lib/igt_drm_clients: Record total cycles Lucas De Marchi
2024-05-08 19:01   ` Umesh Nerlige Ramappa
2024-05-08 19:07     ` Umesh Nerlige Ramappa
2024-05-04  6:46 ` Lucas De Marchi [this message]
2024-05-08 19:11   ` [PATCH i-g-t v3 13/13] gputop: Add support to drm-cycles/drm-total-cycles Umesh Nerlige Ramappa
2024-05-04  7:31 ` ✓ Fi.CI.BAT: success for gputop: Add support for xe (rev4) Patchwork
2024-05-04  7:36 ` ✓ CI.xeBAT: " Patchwork
2024-05-04  8:33 ` ✗ CI.xeFULL: failure " Patchwork
2024-05-04 12:50 ` ✗ Fi.CI.IGT: " Patchwork
2024-05-06 10:37 ` [PATCH i-g-t v3 00/12] gputop: Add support for xe Tvrtko Ursulin
2024-05-06 20:40   ` Lucas De Marchi
2024-05-08 16:25 ` (subset) " Lucas De Marchi

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=20240504064643.25863-14-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