All of lore.kernel.org
 help / color / mirror / Atom feed
From: Chris Wilson <chris@chris-wilson.co.uk>
To: intel-gfx@lists.freedesktop.org
Subject: [Intel-gfx] [PATCH] drm/i915/selftests: Measure the energy consumed while in RC6
Date: Tue, 24 Mar 2020 20:44:55 +0000	[thread overview]
Message-ID: <20200324204455.2988-1-chris@chris-wilson.co.uk> (raw)
In-Reply-To: <20200324201526.16571-1-chris@chris-wilson.co.uk>

Measure and compare the energy consumed, as reported by the rapl MSR,
by the GPU while in RC0 and RC6 states. Throw an error if RC6 does not
at least halve the energy consumption of RC0, as this more than likely
means we failed to enter RC0 correctly.

If we can't measure the energy draw with the MSR, then it will report 0
for both measurements. Since the measurement works on all gen6+, this seems
worth flagging as an error.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Cc: Andi Shyti <andi.shyti@intel.com>
---
 drivers/gpu/drm/i915/gt/selftest_rc6.c | 39 ++++++++++++++++++++++++++
 1 file changed, 39 insertions(+)

diff --git a/drivers/gpu/drm/i915/gt/selftest_rc6.c b/drivers/gpu/drm/i915/gt/selftest_rc6.c
index 95b165faeba7..3ac9a8925218 100644
--- a/drivers/gpu/drm/i915/gt/selftest_rc6.c
+++ b/drivers/gpu/drm/i915/gt/selftest_rc6.c
@@ -12,6 +12,22 @@
 
 #include "selftests/i915_random.h"
 
+#define MCH_SECP_NRG_STTS              _MMIO(MCHBAR_MIRROR_BASE_SNB + 0x592c)
+
+static u64 energy_uJ(struct intel_rc6 *rc6)
+{
+	unsigned long long power;
+	u32 units;
+
+	if (rdmsrl_safe(MSR_RAPL_POWER_UNIT, &power))
+		return 0;
+
+	units = (power & 0x1f00) >> 8;
+	power = intel_uncore_read_fw(rc6_to_uncore(rc6), MCH_SECP_NRG_STTS);
+
+	return (1000000 * power) >> units; /* convert to uJ */
+}
+
 static u64 rc6_residency(struct intel_rc6 *rc6)
 {
 	u64 result;
@@ -31,7 +47,9 @@ int live_rc6_manual(void *arg)
 {
 	struct intel_gt *gt = arg;
 	struct intel_rc6 *rc6 = &gt->rc6;
+	u64 rc0_power, rc6_power;
 	intel_wakeref_t wakeref;
+	ktime_t dt;
 	u64 res[2];
 	int err = 0;
 
@@ -53,22 +71,35 @@ int live_rc6_manual(void *arg)
 	__intel_rc6_disable(rc6);
 	msleep(1); /* wakeup is not immediate, takes about 100us on icl */
 
+	dt = ktime_get();
+	rc0_power = energy_uJ(rc6);
 	res[0] = rc6_residency(rc6);
 	msleep(250);
 	res[1] = rc6_residency(rc6);
+	rc0_power = div64_u64(energy_uJ(rc6) - rc0_power,
+			      ktime_to_ns(ktime_sub(ktime_get(), dt)));
 	if ((res[1] - res[0]) >> 10) {
 		pr_err("RC6 residency increased by %lldus while disabled for 250ms!\n",
 		       (res[1] - res[0]) >> 10);
 		err = -EINVAL;
 		goto out_unlock;
 	}
+	if (!rc0_power) {
+		pr_err("No power measured while in RC0\n");
+		err = -EINVAL;
+		goto out_unlock;
+	}
 
 	/* Manually enter RC6 */
 	intel_rc6_park(rc6);
 
+	dt = ktime_get();
+	rc6_power = energy_uJ(rc6);
 	res[0] = rc6_residency(rc6);
 	msleep(100);
 	res[1] = rc6_residency(rc6);
+	rc6_power = div64_u64(energy_uJ(rc6) - rc6_power,
+			      ktime_to_ns(ktime_sub(ktime_get(), dt)));
 
 	if (res[1] == res[0]) {
 		pr_err("Did not enter RC6! RC6_STATE=%08x, RC6_CONTROL=%08x, residency=%lld\n",
@@ -78,6 +109,14 @@ int live_rc6_manual(void *arg)
 		err = -EINVAL;
 	}
 
+	pr_info("GPU consumed %llduW in RC0 and %llduW in RC6\n",
+		rc0_power, rc6_power);
+	if ((rc6_power >> 10) > (rc0_power >> 10) / 2) { /* compare mW */
+		pr_err("GPU leaked energy while in RC6!\n");
+		err = -EINVAL;
+		goto out_unlock;
+	}
+
 	/* Restore what should have been the original state! */
 	intel_rc6_unpark(rc6);
 
-- 
2.20.1

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

  reply	other threads:[~2020-03-24 20:45 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-03-24 20:15 [Intel-gfx] [PATCH] drm/i915/selftests: Measure the energy consumed while in RC6 Chris Wilson
2020-03-24 20:44 ` Chris Wilson [this message]
2020-03-24 20:56   ` Chris Wilson
2020-09-28 23:56   ` Lucas De Marchi
2020-09-29  7:59     ` Chris Wilson
2020-03-24 20:58 ` Chris Wilson
2020-03-24 23:48 ` [Intel-gfx] ✗ Fi.CI.BAT: failure for drm/i915/selftests: Measure the energy consumed while in RC6 (rev3) Patchwork
2020-03-25  0:08 ` [Intel-gfx] [PATCH] drm/i915/selftests: Measure the energy consumed while in RC6 Chris Wilson
2020-03-25  0:41 ` [Intel-gfx] ✗ Fi.CI.BAT: failure for drm/i915/selftests: Measure the energy consumed while in RC6 (rev4) Patchwork
2020-03-25  8:10 ` [Intel-gfx] [PATCH] drm/i915/selftests: Measure the energy consumed while in RC6 Chris Wilson
2020-03-25  8:58   ` Andi Shyti
2020-03-25  9:10     ` Chris Wilson
2020-03-25  9:07 ` [Intel-gfx] ✗ Fi.CI.BAT: failure for drm/i915/selftests: Measure the energy consumed while in RC6 (rev5) 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=20200324204455.2988-1-chris@chris-wilson.co.uk \
    --to=chris@chris-wilson.co.uk \
    --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.