drm/i915/selftests: Measure the energy consumed while in RC6

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>
Reviewed-by: Andi Shyti <andi.shyti@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200325101502.12591-1-chris@chris-wilson.co.uk

diff --git a/drivers/gpu/drm/i915/gt/selftest_rc6.c b/drivers/gpu/drm/i915/gt/selftest_rc6.c
index 95b165faeba78f2fe39dfa92781f1a3c37e73ddf..2f3316c924a35f09a3dd958f789cb787a1b1590c 100644 (file)
--- 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"
 
+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;
+
+       if (rdmsrl_safe(MSR_PP1_ENERGY_STATUS, &power))
+               return 0;
+
+       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;
 
@@ -54,7 +72,11 @@ int live_rc6_manual(void *arg)
        msleep(1); /* wakeup is not immediate, takes about 100us on icl */
 
        res[0] = rc6_residency(rc6);
+       dt = ktime_get();
+       rc0_power = energy_uJ(rc6);
        msleep(250);
+       rc0_power = energy_uJ(rc6) - rc0_power;
+       dt = ktime_sub(ktime_get(), dt);
        res[1] = rc6_residency(rc6);
        if ((res[1] - res[0]) >> 10) {
                pr_err("RC6 residency increased by %lldus while disabled for 250ms!\n",
@@ -63,13 +85,24 @@ int live_rc6_manual(void *arg)
                goto out_unlock;
        }
 
+       rc0_power = div64_u64(NSEC_PER_SEC * rc0_power, ktime_to_ns(dt));
+       if (!rc0_power) {
+               pr_err("No power measured while in RC0\n");
+               err = -EINVAL;
+               goto out_unlock;
+       }
+
        /* Manually enter RC6 */
        intel_rc6_park(rc6);
 
        res[0] = rc6_residency(rc6);
+       intel_uncore_forcewake_flush(rc6_to_uncore(rc6), FORCEWAKE_ALL);
+       dt = ktime_get();
+       rc6_power = energy_uJ(rc6);
        msleep(100);
+       rc6_power = energy_uJ(rc6) - rc6_power;
+       dt = ktime_sub(ktime_get(), dt);
        res[1] = rc6_residency(rc6);
-
        if (res[1] == res[0]) {
                pr_err("Did not enter RC6! RC6_STATE=%08x, RC6_CONTROL=%08x, residency=%lld\n",
                       intel_uncore_read_fw(gt->uncore, GEN6_RC_STATE),
@@ -78,6 +111,15 @@ int live_rc6_manual(void *arg)
                err = -EINVAL;
        }
 
+       rc6_power = div64_u64(NSEC_PER_SEC * rc6_power, ktime_to_ns(dt));
+       pr_info("GPU consumed %llduW in RC0 and %llduW in RC6\n",
+               rc0_power, rc6_power);
+       if (2 * rc6_power > rc0_power) {
+               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);
 

diff --git a/drivers/gpu/drm/i915/intel_uncore.c b/drivers/gpu/drm/i915/intel_uncore.c
index abb18b90d7c392eea4d0d48b50bf48ecc07021b1..013312e9b55c14dd8bba8b62ad6642cc376330f4 100644 (file)
--- a/drivers/gpu/drm/i915/intel_uncore.c
+++ b/drivers/gpu/drm/i915/intel_uncore.c
@@ -734,6 +734,28 @@ void intel_uncore_forcewake_put(struct intel_uncore *uncore,
        spin_unlock_irqrestore(&uncore->lock, irqflags);
 }
 
+/**
+ * intel_uncore_forcewake_flush - flush the delayed release
+ * @uncore: the intel_uncore structure
+ * @fw_domains: forcewake domains to flush
+ */
+void intel_uncore_forcewake_flush(struct intel_uncore *uncore,
+                                 enum forcewake_domains fw_domains)
+{
+       struct intel_uncore_forcewake_domain *domain;
+       unsigned int tmp;
+
+       if (!uncore->funcs.force_wake_put)
+               return;
+
+       fw_domains &= uncore->fw_domains;
+       for_each_fw_domain_masked(domain, fw_domains, uncore, tmp) {
+               WRITE_ONCE(domain->active, false);
+               if (hrtimer_cancel(&domain->timer))
+                       intel_uncore_fw_release_timer(&domain->timer);
+       }
+}
+
 /**
  * intel_uncore_forcewake_put__locked - grab forcewake domain references
  * @uncore: the intel_uncore structure

diff --git a/drivers/gpu/drm/i915/intel_uncore.h b/drivers/gpu/drm/i915/intel_uncore.h
index dcfa243892c67e8c842f5f08c987b08a1dc5fcc1..8d3aa8b9acf9fe9999cb8e7f425d31be31461e50 100644 (file)
--- a/drivers/gpu/drm/i915/intel_uncore.h
+++ b/drivers/gpu/drm/i915/intel_uncore.h
@@ -209,7 +209,11 @@ void intel_uncore_forcewake_get(struct intel_uncore *uncore,
                                enum forcewake_domains domains);
 void intel_uncore_forcewake_put(struct intel_uncore *uncore,
                                enum forcewake_domains domains);
-/* Like above but the caller must manage the uncore.lock itself.
+void intel_uncore_forcewake_flush(struct intel_uncore *uncore,
+                                 enum forcewake_domains fw_domains);
+
+/*
+ * Like above but the caller must manage the uncore.lock itself.
  * Must be used with I915_READ_FW and friends.
  */
 void intel_uncore_forcewake_get__locked(struct intel_uncore *uncore,