@@ -555,6 +555,11 @@ struct intel_engine_cs {
* Idle is defined as active == 0, active is active > 0.
*/
ktime_t start;
+
+ /**
+ * @rps: Utilisation at last RPS sampling.
+ */
+ ktime_t rps;
} stats;
struct {
@@ -15,6 +15,10 @@
#include "intel_sideband.h"
#include "../../../platform/x86/intel_ips.h"
+#define BUSY_UP 95 /* % utilisation */
+#define BUSY_DOWN 85 /* % utilisation */
+#define BUSY_MAX_EI 20u /* ms */
+
/*
* Lock protecting IPS related data structures
*/
@@ -45,6 +49,99 @@ static inline void set(struct intel_uncore *uncore, i915_reg_t reg, u32 val)
intel_uncore_write_fw(uncore, reg, val);
}
+static void rps_timer(struct timer_list *t)
+{
+ struct intel_rps *rps = from_timer(rps, t, timer);
+ struct intel_engine_cs *engine;
+ enum intel_engine_id id;
+ s64 max_busy[3] = {};
+ ktime_t dt, last;
+
+ for_each_engine(engine, rps_to_gt(rps), id) {
+ s64 busy;
+ int i;
+
+ dt = intel_engine_get_busy_time(engine);
+ last = engine->stats.rps;
+ engine->stats.rps = dt;
+
+ busy = ktime_to_ns(ktime_sub(dt, last));
+ for (i = 0; i < ARRAY_SIZE(max_busy); i++) {
+ if (busy > max_busy[i])
+ swap(busy, max_busy[i]);
+ }
+ }
+
+ dt = ktime_get();
+ last = rps->pm_timestamp;
+ rps->pm_timestamp = dt;
+
+ if (intel_rps_is_active(rps)) {
+ s64 busy;
+ int i;
+
+ dt = ktime_sub(dt, last);
+
+ /*
+ * Our goal is to evaluate each engine independently, so we run
+ * at the lowest clocks required to sustain the heaviest
+ * workload. However, a task may be split into sequential
+ * dependent operations across a set of engines, such that
+ * the independent contributions do not account for high load,
+ * but overall the task is GPU bound. For example, consider
+ * video decode on vcs followed by colour post-processing
+ * on vecs, followed by general post-processing on rcs.
+ * Since multi-engines being active does imply a single
+ * continuous workload across all engines, we hedge our
+ * bets by only contributing a factor of the distributed
+ * load into our busyness calculation.
+ */
+ busy = max_busy[0];
+ for (i = 1; i < ARRAY_SIZE(max_busy); i++) {
+ if (!max_busy[i])
+ break;
+
+ busy += div_u64(max_busy[i], 1 << i);
+ }
+ GT_TRACE(rps_to_gt(rps),
+ "busy:%lld [%d%%], max:[%lld, %lld, %lld]\n",
+ busy, (int)div64_u64(100 * busy, dt),
+ max_busy[0], max_busy[1], max_busy[2]);
+
+ if (100 * busy > BUSY_UP * dt &&
+ rps->cur_freq < rps->max_freq_softlimit) {
+ rps->pm_iir |= GEN6_PM_RP_UP_THRESHOLD;
+ rps->pm_interval = 1;
+ schedule_work(&rps->work);
+ } else if (100 * busy < BUSY_DOWN * dt &&
+ rps->cur_freq > rps->min_freq_softlimit) {
+ rps->pm_iir |= GEN6_PM_RP_DOWN_THRESHOLD;
+ rps->pm_interval = 1;
+ schedule_work(&rps->work);
+ } else {
+ rps->last_adj = 0;
+ }
+
+ mod_timer(&rps->timer,
+ jiffies + msecs_to_jiffies(rps->pm_interval));
+ rps->pm_interval = min(rps->pm_interval * 2, BUSY_MAX_EI);
+ }
+}
+
+static void rps_start_timer(struct intel_rps *rps)
+{
+ rps->pm_timestamp = ktime_sub(ktime_get(), rps->pm_timestamp);
+ rps->pm_interval = 1;
+ mod_timer(&rps->timer, jiffies + 1);
+}
+
+static void rps_stop_timer(struct intel_rps *rps)
+{
+ del_timer_sync(&rps->timer);
+ rps->pm_timestamp = ktime_sub(ktime_get(), rps->pm_timestamp);
+ cancel_work_sync(&rps->work);
+}
+
static u32 rps_pm_mask(struct intel_rps *rps, u8 val)
{
u32 mask = 0;
@@ -742,8 +839,11 @@ void intel_rps_unpark(struct intel_rps *rps)
mutex_unlock(&rps->lock);
+ rps->pm_iir = 0;
if (intel_rps_has_interrupts(rps))
rps_enable_interrupts(rps);
+ if (intel_rps_uses_timer(rps))
+ rps_start_timer(rps);
if (IS_GEN(rps_to_i915(rps), 5))
gen5_rps_update(rps);
@@ -754,6 +854,8 @@ void intel_rps_park(struct intel_rps *rps)
if (!intel_rps_clear_active(rps))
return;
+ if (intel_rps_uses_timer(rps))
+ rps_stop_timer(rps);
if (intel_rps_has_interrupts(rps))
rps_disable_interrupts(rps);
@@ -1211,6 +1313,19 @@ static unsigned long __ips_gfx_val(struct intel_ips *ips)
return ips->gfx_power + state2;
}
+static bool has_busy_stats(struct intel_rps *rps)
+{
+ struct intel_engine_cs *engine;
+ enum intel_engine_id id;
+
+ for_each_engine(engine, rps_to_gt(rps), id) {
+ if (!intel_engine_supports_stats(engine))
+ return false;
+ }
+
+ return true;
+}
+
void intel_rps_enable(struct intel_rps *rps)
{
struct drm_i915_private *i915 = rps_to_i915(rps);
@@ -1255,7 +1370,9 @@ void intel_rps_enable(struct intel_rps *rps)
GEM_BUG_ON(rps->efficient_freq < rps->min_freq);
GEM_BUG_ON(rps->efficient_freq > rps->max_freq);
- if (INTEL_GEN(i915) >= 6)
+ if (has_busy_stats(rps))
+ intel_rps_set_timer(rps);
+ else if (INTEL_GEN(i915) >= 6)
intel_rps_set_interrupts(rps);
else
/* Ironlake currently uses intel_ips.ko */ {}
@@ -1274,6 +1391,7 @@ void intel_rps_disable(struct intel_rps *rps)
intel_rps_clear_enabled(rps);
intel_rps_clear_interrupts(rps);
+ intel_rps_clear_timer(rps);
if (INTEL_GEN(i915) >= 6)
gen6_rps_disable(rps);
@@ -1689,6 +1807,7 @@ void intel_rps_init_early(struct intel_rps *rps)
mutex_init(&rps->power.mutex);
INIT_WORK(&rps->work, rps_work);
+ timer_setup(&rps->timer, rps_timer, 0);
atomic_set(&rps->num_waiters, 0);
}
@@ -81,4 +81,19 @@ static inline void intel_rps_clear_interrupts(struct intel_rps *rps)
clear_bit(INTEL_RPS_INTERRUPTS, &rps->flags);
}
+static inline bool intel_rps_uses_timer(const struct intel_rps *rps)
+{
+ return test_bit(INTEL_RPS_TIMER, &rps->flags);
+}
+
+static inline void intel_rps_set_timer(struct intel_rps *rps)
+{
+ set_bit(INTEL_RPS_TIMER, &rps->flags);
+}
+
+static inline void intel_rps_clear_timer(struct intel_rps *rps)
+{
+ clear_bit(INTEL_RPS_TIMER, &rps->flags);
+}
+
#endif /* INTEL_RPS_H */
@@ -35,6 +35,7 @@ enum {
INTEL_RPS_ENABLED = 0,
INTEL_RPS_ACTIVE,
INTEL_RPS_INTERRUPTS,
+ INTEL_RPS_TIMER,
};
struct intel_rps {
@@ -44,8 +45,12 @@ struct intel_rps {
* work, interrupts_enabled and pm_iir are protected by
* dev_priv->irq_lock
*/
+ struct timer_list timer;
struct work_struct work;
unsigned long flags;
+
+ ktime_t pm_timestamp;
+ u32 pm_interval;
u32 pm_iir;
/* PM interrupt bits that should never be masked */
As with the realisation for soft-rc6, we respond to idling the engines within microseconds, far faster than the response times for HW RC6 and RPS. Furthermore, our fast parking upon idle, prevents HW RPS from running for many desktop workloads, as the RPS evaluation intervals are on the order of tens of milliseconds, but the typical workload is just a couple of milliseconds, but yet we still need to determine the best frequency for user latency versus power. Recognising that the HW evaluation intervals are a poor fit, and that they were deprecated [in bspec at least] from gen10, start to wean ourselves off them and replace the EI with a timer and our accurate busy-stats. The principle benefit of manually evaluating RPS intervals is that we can be more responsive for better performance and powersaving for both spiky workloads and steady-state. Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/1698 Fixes: 98479ada421a ("drm/i915/gt: Treat idling as a RPS downclock event") 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> --- drivers/gpu/drm/i915/gt/intel_engine_types.h | 5 + drivers/gpu/drm/i915/gt/intel_rps.c | 121 ++++++++++++++++++- drivers/gpu/drm/i915/gt/intel_rps.h | 15 +++ drivers/gpu/drm/i915/gt/intel_rps_types.h | 5 + 4 files changed, 145 insertions(+), 1 deletion(-)