Quoting Tvrtko Ursulin (2018-02-08 07:04:37)
From: Tvrtko Ursulin <tvrtko.ursulin@xxxxxxxxx>
We can implement limited RC6 counter wrap-around protection under the
assumption that clients will be reading this value more frequently than
the wrap period on a given platform.
With the typical wrap-around period being ~90 minutes, even with the
exception of Baytrail which wraps every 13 seconds, this sounds like a
reasonable assumption.
Implementation works by storing a 64-bit software copy of a hardware RC6
counter, along with the previous HW counter snapshot. This enables it to
detect wrap is polled frequently enough and keep the software copy
monotonically incrementing.
v2:
* Missed GEN6_GT_GFX_RC6_LOCKED when considering slot sizing and
indexing.
* Fixed off-by-one in wrap-around handling. (Chris Wilson)
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@xxxxxxxxx>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=94852
Cc: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx>
---
drivers/gpu/drm/i915/i915_drv.h | 2 ++
drivers/gpu/drm/i915/intel_pm.c | 57 ++++++++++++++++++++++++++++++++++-------
2 files changed, 50 insertions(+), 9 deletions(-)
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index ad1fc845cd1b..28a2671a26c7 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -946,6 +946,8 @@ struct intel_rps {
struct intel_rc6 {
bool enabled;
+ u64 prev_hw_residency[4];
+ u64 cur_residency[4];
};
struct intel_llc_pstate {
diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index 41f26ab46501..6b01700334a9 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -9417,15 +9417,16 @@ static u64 vlv_residency_raw(struct drm_i915_private *dev_priv,
const i915_reg_t reg)
{
u32 lower, upper, tmp;
- unsigned long flags;
int loop = 2;
- /* The register accessed do not need forcewake. We borrow
+ /*
+ * The register accessed do not need forcewake. We borrow
* uncore lock to prevent concurrent access to range reg.
*/
- spin_lock_irqsave(&dev_priv->uncore.lock, flags);
+ lockdep_assert_held(&dev_priv->uncore.lock);
- /* vlv and chv residency counters are 40 bits in width.
+ /*
+ * vlv and chv residency counters are 40 bits in width.
* With a control bit, we can choose between upper or lower
* 32bit window into this counter.
*
@@ -9449,29 +9450,44 @@ static u64 vlv_residency_raw(struct drm_i915_private *dev_priv,
upper = I915_READ_FW(reg);
} while (upper != tmp && --loop);
- /* Everywhere else we always use VLV_COUNTER_CONTROL with the
+ /*
+ * Everywhere else we always use VLV_COUNTER_CONTROL with the
* VLV_COUNT_RANGE_HIGH bit set - so it is safe to leave it set
* now.
*/
- spin_unlock_irqrestore(&dev_priv->uncore.lock, flags);
-
return lower | (u64)upper << 8;
}
u64 intel_rc6_residency_ns(struct drm_i915_private *dev_priv,
const i915_reg_t reg)
{
- u64 time_hw;
+ u64 time_hw, prev_hw, overflow_hw;
+ unsigned int fw_domains;
+ unsigned long flags;
u32 mul, div;
+ int i;
if (!HAS_RC6(dev_priv))
return 0;
+ /* Counter wrap handling stores previous hw counter values. */
+ i = (i915_mmio_reg_offset(reg) -
+ i915_mmio_reg_offset(GEN6_GT_GFX_RC6_LOCKED)) / sizeof(u32);
+ if (WARN_ON_ONCE(i < 0 ||
+ i >= ARRAY_SIZE(dev_priv->gt_pm.rc6.cur_residency)))
int i, but ARRAY_SIZE is unsigned, so just
if (WARN_ON_ONCE(i >= ARRAY_SIZE(dev_priv->gt_pm.rc6.cur_residency)))
would suffice.
I remain surprised that they are in a neat block across atom/big-core
#define VLV_GT_RENDER_RC6 _MMIO(0x138108)
#define VLV_GT_MEDIA_RC6 _MMIO(0x13810C)
#define GEN6_GT_GFX_RC6_LOCKED _MMIO(0x138104)
#define GEN6_GT_GFX_RC6 _MMIO(0x138108)
#define GEN6_GT_GFX_RC6p _MMIO(0x13810C)
#define GEN6_GT_GFX_RC6pp _MMIO(0x138110)
are indeed just the 4 registers in use.
+ fw_domains = intel_uncore_forcewake_for_reg(dev_priv, reg, FW_REG_READ);
+
+ spin_lock_irqsave(&dev_priv->uncore.lock, flags);
+ intel_uncore_forcewake_get__locked(dev_priv, fw_domains);
+
/* On VLV and CHV, residency time is in CZ units rather than 1.28us */
if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) {
mul = 1000000;
div = dev_priv->czclk_freq;
+ overflow_hw = BIT_ULL(40);
time_hw = vlv_residency_raw(dev_priv, reg);
} else {
/* 833.33ns units on Gen9LP, 1.28us elsewhere. */
@@ -9483,9 +9499,32 @@ u64 intel_rc6_residency_ns(struct drm_i915_private *dev_priv,
div = 1;
}
- time_hw = I915_READ(reg);
+ overflow_hw = BIT_ULL(32);
+ time_hw = I915_READ_FW(reg);
}
+ /*
+ * Counter wrap handling.
+ *
+ * But relying on a sufficient frequency of queries otherwise counters
+ * can still wrap.
+ */
+ prev_hw = dev_priv->gt_pm.rc6.prev_hw_residency[i];
+ dev_priv->gt_pm.rc6.prev_hw_residency[i] = time_hw;
+
+ /* RC6 delta from last sample. */
+ if (time_hw >= prev_hw)
+ time_hw -= prev_hw;
+ else
+ time_hw += overflow_hw - prev_hw;
+
+ /* Add delta to RC6 extended raw driver copy. */
+ time_hw += dev_priv->gt_pm.rc6.cur_residency[i];
+ dev_priv->gt_pm.rc6.cur_residency[i] = time_hw;
Ok.
Reviewed-by: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx>
The other part of the puzzle was perhaps just using a timer to read the
registers every 1/2 wrap period, but there was no clear picture on when
to start and stop that timer for sysfs. However, with perf we can create
and destroy a background timer on PMU open/destroy.
Fancy hooking up a background timer?
