public inbox for igt-dev@lists.freedesktop.org
 help / color / mirror / Atom feed
From: Chris Wilson <chris@chris-wilson.co.uk>
To: intel-gfx@lists.freedesktop.org
Cc: igt-dev@lists.freedesktop.org, Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Subject: [igt-dev] [PATCH i-g-t] i915/perf_pmu: Check that while parked, we report min freq or below
Date: Fri, 29 Nov 2019 11:45:52 +0000	[thread overview]
Message-ID: <20191129114552.802185-1-chris@chris-wilson.co.uk> (raw)

While the HW is parked, the GPU should be turned off and clocks stop
(i.e. running at 0Hz). We should report either the last frequency we
program (which should be the minimum legal value) or a more truthful 0.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
---
 tests/perf_pmu.c | 40 ++++++++++++++++++++++++++++++++++++++++
 1 file changed, 40 insertions(+)

diff --git a/tests/perf_pmu.c b/tests/perf_pmu.c
index 54842784c..a1520d2c4 100644
--- a/tests/perf_pmu.c
+++ b/tests/perf_pmu.c
@@ -1477,6 +1477,44 @@ test_frequency(int gem_fd)
 	__assert_within_epsilon(max[0], max_freq, tolerance, 0.15f);
 }
 
+static void
+test_frequency_idle(int gem_fd)
+{
+	uint32_t min_freq;
+	uint64_t val[2], start[2], slept;
+	double idle[2];
+	int fd, sysfs;
+
+	sysfs = igt_sysfs_open(gem_fd);
+	igt_require(sysfs >= 0);
+
+	min_freq = igt_sysfs_get_u32(sysfs, "gt_RPn_freq_mhz");
+	close(sysfs);
+
+	/* While parked, our convention is to report the GPU at 0Hz */
+
+	fd = open_group(I915_PMU_REQUESTED_FREQUENCY, -1);
+	open_group(I915_PMU_ACTUAL_FREQUENCY, fd);
+
+	gem_quiescent_gpu(gem_fd); /* Be idle! */
+	measured_usleep(2000); /* Wait for timers to cease */
+
+	slept = pmu_read_multi(fd, 2, start);
+	measured_usleep(batch_duration_ns / 1000);
+	slept = pmu_read_multi(fd, 2, val) - slept;
+
+	idle[0] = 1e9*(val[0] - start[0]) / slept;
+	idle[1] = 1e9*(val[1] - start[1]) / slept;
+
+	igt_info("Idle frequency: requested %.1f, actual %.1f; HW min %u\n",
+		 idle[0], idle[1], min_freq);
+
+	igt_assert_f(idle[0] <= min_freq,
+		     "Request frequency should be 0 while parked!\n");
+	igt_assert_f(idle[1] <= min_freq,
+		     "Actual frequency should be 0 while parked!\n");
+}
+
 static bool wait_for_rc6(int fd)
 {
 	struct timespec tv = {};
@@ -1967,6 +2005,8 @@ igt_main
 	 */
 	igt_subtest("frequency")
 		test_frequency(fd);
+	igt_subtest("frequency-idle")
+		test_frequency_idle(fd);
 
 	/**
 	 * Test interrupt count reporting.
-- 
2.24.0

_______________________________________________
igt-dev mailing list
igt-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/igt-dev

             reply	other threads:[~2019-11-29 11:45 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-11-29 11:45 Chris Wilson [this message]
2019-11-29 11:56 ` [igt-dev] [Intel-gfx] [PATCH i-g-t] i915/perf_pmu: Check that while parked, we report min freq or below Tvrtko Ursulin
2019-11-29 11:57 ` Tvrtko Ursulin
2019-11-29 12:42 ` [igt-dev] ✗ GitLab.Pipeline: failure for " Patchwork
2019-11-29 12:55 ` [igt-dev] ✓ Fi.CI.BAT: 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=20191129114552.802185-1-chris@chris-wilson.co.uk \
    --to=chris@chris-wilson.co.uk \
    --cc=igt-dev@lists.freedesktop.org \
    --cc=intel-gfx@lists.freedesktop.org \
    --cc=tvrtko.ursulin@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