diff options
author | Chris Wilson <chris@chris-wilson.co.uk> | 2014-12-16 10:02:27 +0000 |
---|---|---|
committer | Jani Nikula <jani.nikula@intel.com> | 2014-12-16 15:07:53 +0200 |
commit | 2c550183476dfa25641309ae9a28d30feed14379 (patch) | |
tree | d7eeec4df7e2e078a9cf3be5020357b13f96cb77 /drivers/gpu/drm | |
parent | add284a3a2481e759d6bec35f6444c32c8ddc383 (diff) |
drm/i915: Disable PSMI sleep messages on all rings around context switches
There exists a current workaround to prevent a hang on context switch
should the ring go to sleep in the middle of the restore,
WaProgramMiArbOnOffAroundMiSetContext (applicable to all gen7+). In
spite of disabling arbitration (which prevents the ring from powering
down during the critical section) we were still hitting hangs that had
the hallmarks of the known erratum. That is we are still seeing hangs
"on the last instruction in the context restore". By comparing -nightly
(broken) with requests (working), we were able to deduce that it was the
semaphore LRI cross-talk that reproduced the original failure. The key
was that requests implemented deferred semaphore signalling, and
disabling that, i.e. emitting the semaphore signal to every other ring
after every batch restored the frequent hang. Explicitly disabling PSMI
sleep on the RCS ring was insufficient, all the rings had to be awake to
prevent the hangs. Fortunately, we can reduce the wakelock to the
MI_SET_CONTEXT operation itself, and so should be able to limit the extra
power implications.
Since the MI_ARB_ON_OFF workaround is listed for all gen7 and above
products, we should apply this extra hammer for all of the same
platforms despite so far that we have only been able to reproduce the
hang on certain ivb and hsw models. The last question is whether we want
to always use the extra hammer or only when we know semaphores are in
operation. At the moment, we only use LRI on non-RCS rings for
semaphores, but that may change in the future with the possibility of
reintroducing this bug under subtle conditions.
v2: Make it explicit that the PSMI LRI are an extension to the original
workaround for the other rings.
v3: Bikeshedding variable names and whitespacing
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=80660
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=83677
Cc: Simon Farnsworth <simon@farnz.org.uk>
Cc: Daniel Vetter <daniel@ffwll.ch>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Tested-by: Peter Frühberger <fritsch@xbmc.org>
Reviewed-by: Daniel Vetter <daniel@ffwll.ch>
Cc: stable@vger.kernel.org
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Diffstat (limited to 'drivers/gpu/drm')
-rw-r--r-- | drivers/gpu/drm/i915/i915_gem_context.c | 48 | ||||
-rw-r--r-- | drivers/gpu/drm/i915/i915_reg.h | 2 |
2 files changed, 42 insertions, 8 deletions
diff --git a/drivers/gpu/drm/i915/i915_gem_context.c b/drivers/gpu/drm/i915/i915_gem_context.c index d17ff435f276..d011ec82ef1e 100644 --- a/drivers/gpu/drm/i915/i915_gem_context.c +++ b/drivers/gpu/drm/i915/i915_gem_context.c @@ -473,7 +473,12 @@ mi_set_context(struct intel_engine_cs *ring, u32 hw_flags) { u32 flags = hw_flags | MI_MM_SPACE_GTT; - int ret; + const int num_rings = + /* Use an extended w/a on ivb+ if signalling from other rings */ + i915_semaphore_is_enabled(ring->dev) ? + hweight32(INTEL_INFO(ring->dev)->ring_mask) - 1 : + 0; + int len, i, ret; /* w/a: If Flush TLB Invalidation Mode is enabled, driver must do a TLB * invalidation prior to MI_SET_CONTEXT. On GEN6 we don't set the value @@ -490,15 +495,31 @@ mi_set_context(struct intel_engine_cs *ring, if (!IS_HASWELL(ring->dev) && INTEL_INFO(ring->dev)->gen < 8) flags |= (MI_SAVE_EXT_STATE_EN | MI_RESTORE_EXT_STATE_EN); - ret = intel_ring_begin(ring, 6); + + len = 4; + if (INTEL_INFO(ring->dev)->gen >= 7) + len += 2 + (num_rings ? 4*num_rings + 2 : 0); + + ret = intel_ring_begin(ring, len); if (ret) return ret; /* WaProgramMiArbOnOffAroundMiSetContext:ivb,vlv,hsw,bdw,chv */ - if (INTEL_INFO(ring->dev)->gen >= 7) + if (INTEL_INFO(ring->dev)->gen >= 7) { intel_ring_emit(ring, MI_ARB_ON_OFF | MI_ARB_DISABLE); - else - intel_ring_emit(ring, MI_NOOP); + if (num_rings) { + struct intel_engine_cs *signaller; + + intel_ring_emit(ring, MI_LOAD_REGISTER_IMM(num_rings)); + for_each_ring(signaller, to_i915(ring->dev), i) { + if (signaller == ring) + continue; + + intel_ring_emit(ring, RING_PSMI_CTL(signaller->mmio_base)); + intel_ring_emit(ring, _MASKED_BIT_ENABLE(GEN6_PSMI_SLEEP_MSG_DISABLE)); + } + } + } intel_ring_emit(ring, MI_NOOP); intel_ring_emit(ring, MI_SET_CONTEXT); @@ -510,10 +531,21 @@ mi_set_context(struct intel_engine_cs *ring, */ intel_ring_emit(ring, MI_NOOP); - if (INTEL_INFO(ring->dev)->gen >= 7) + if (INTEL_INFO(ring->dev)->gen >= 7) { + if (num_rings) { + struct intel_engine_cs *signaller; + + intel_ring_emit(ring, MI_LOAD_REGISTER_IMM(num_rings)); + for_each_ring(signaller, to_i915(ring->dev), i) { + if (signaller == ring) + continue; + + intel_ring_emit(ring, RING_PSMI_CTL(signaller->mmio_base)); + intel_ring_emit(ring, _MASKED_BIT_DISABLE(GEN6_PSMI_SLEEP_MSG_DISABLE)); + } + } intel_ring_emit(ring, MI_ARB_ON_OFF | MI_ARB_ENABLE); - else - intel_ring_emit(ring, MI_NOOP); + } intel_ring_advance(ring); diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h index 58009558cc94..172de3b3433b 100644 --- a/drivers/gpu/drm/i915/i915_reg.h +++ b/drivers/gpu/drm/i915/i915_reg.h @@ -1129,6 +1129,7 @@ enum punit_power_well { #define GEN6_VERSYNC (RING_SYNC_1(VEBOX_RING_BASE)) #define GEN6_VEVSYNC (RING_SYNC_2(VEBOX_RING_BASE)) #define GEN6_NOSYNC 0 +#define RING_PSMI_CTL(base) ((base)+0x50) #define RING_MAX_IDLE(base) ((base)+0x54) #define RING_HWS_PGA(base) ((base)+0x80) #define RING_HWS_PGA_GEN6(base) ((base)+0x2080) @@ -1459,6 +1460,7 @@ enum punit_power_well { #define GEN6_BLITTER_FBC_NOTIFY (1<<3) #define GEN6_RC_SLEEP_PSMI_CONTROL 0x2050 +#define GEN6_PSMI_SLEEP_MSG_DISABLE (1 << 0) #define GEN8_RC_SEMA_IDLE_MSG_DISABLE (1 << 12) #define GEN8_FF_DOP_CLOCK_GATE_DISABLE (1<<10) |