| Message ID | 20220929021813.2172701-5-John.C.Harrison@Intel.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | Improve anti-pre-emption w/a for compute workloads | expand |
On 29/09/2022 03:18, John.C.Harrison@Intel.com wrote: > From: John Harrison <John.C.Harrison@Intel.com> > > A workaround was added to the driver to allow compute workloads to run > 'forever' by disabling pre-emption on the RCS engine for Gen12. > It is not totally unbound as the heartbeat will kick in eventually > and cause a reset of the hung engine. > > However, this does not work well in GuC submission mode. In GuC mode, > the pre-emption timeout is how GuC detects hung contexts and triggers > a per engine reset. Thus, disabling the timeout means also losing all > per engine reset ability. A full GT reset will still occur when the > heartbeat finally expires, but that is a much more destructive and > undesirable mechanism. > > The purpose of the workaround is actually to give compute tasks longer > to reach a pre-emption point after a pre-emption request has been > issued. This is necessary because Gen12 does not support mid-thread > pre-emption and compute tasks can have long running threads. > > So, rather than disabling the timeout completely, just set it to a > 'long' value. > > v2: Review feedback from Tvrtko - must hard code the 'long' value > instead of determining it algorithmically. So make it an extra CONFIG > definition. Also, remove the execlist centric comment from the > existing pre-emption timeout CONFIG option given that it applies to > more than just execlists. > > Signed-off-by: John Harrison <John.C.Harrison@Intel.com> > Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com> (v1) > Acked-by: Michal Mrozek <michal.mrozek@intel.com> Acked-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Regards, Tvrtko > --- > drivers/gpu/drm/i915/Kconfig.profile | 26 +++++++++++++++++++---- > drivers/gpu/drm/i915/gt/intel_engine_cs.c | 9 ++++++-- > 2 files changed, 29 insertions(+), 6 deletions(-) > > diff --git a/drivers/gpu/drm/i915/Kconfig.profile b/drivers/gpu/drm/i915/Kconfig.profile > index 39328567c2007..7cc38d25ee5c8 100644 > --- a/drivers/gpu/drm/i915/Kconfig.profile > +++ b/drivers/gpu/drm/i915/Kconfig.profile > @@ -57,10 +57,28 @@ config DRM_I915_PREEMPT_TIMEOUT > default 640 # milliseconds > help > How long to wait (in milliseconds) for a preemption event to occur > - when submitting a new context via execlists. If the current context > - does not hit an arbitration point and yield to HW before the timer > - expires, the HW will be reset to allow the more important context > - to execute. > + when submitting a new context. If the current context does not hit > + an arbitration point and yield to HW before the timer expires, the > + HW will be reset to allow the more important context to execute. > + > + This is adjustable via > + /sys/class/drm/card?/engine/*/preempt_timeout_ms > + > + May be 0 to disable the timeout. > + > + The compiled in default may get overridden at driver probe time on > + certain platforms and certain engines which will be reflected in the > + sysfs control. > + > +config DRM_I915_PREEMPT_TIMEOUT_COMPUTE > + int "Preempt timeout for compute engines (ms, jiffy granularity)" > + default 7500 # milliseconds > + help > + How long to wait (in milliseconds) for a preemption event to occur > + when submitting a new context to a compute capable engine. If the > + current context does not hit an arbitration point and yield to HW > + before the timer expires, the HW will be reset to allow the more > + important context to execute. > > This is adjustable via > /sys/class/drm/card?/engine/*/preempt_timeout_ms > diff --git a/drivers/gpu/drm/i915/gt/intel_engine_cs.c b/drivers/gpu/drm/i915/gt/intel_engine_cs.c > index fcbccd8d244e9..c1257723d1949 100644 > --- a/drivers/gpu/drm/i915/gt/intel_engine_cs.c > +++ b/drivers/gpu/drm/i915/gt/intel_engine_cs.c > @@ -508,9 +508,14 @@ static int intel_engine_setup(struct intel_gt *gt, enum intel_engine_id id, > engine->props.timeslice_duration_ms = > CONFIG_DRM_I915_TIMESLICE_DURATION; > > - /* Override to uninterruptible for OpenCL workloads. */ > + /* > + * Mid-thread pre-emption is not available in Gen12. Unfortunately, > + * some compute workloads run quite long threads. That means they get > + * reset due to not pre-empting in a timely manner. So, bump the > + * pre-emption timeout value to be much higher for compute engines. > + */ > if (GRAPHICS_VER(i915) == 12 && (engine->flags & I915_ENGINE_HAS_RCS_REG_STATE)) > - engine->props.preempt_timeout_ms = 0; > + engine->props.preempt_timeout_ms = CONFIG_DRM_I915_PREEMPT_TIMEOUT_COMPUTE; > > /* Cap properties according to any system limits */ > #define CLAMP_PROP(field) \
diff --git a/drivers/gpu/drm/i915/Kconfig.profile b/drivers/gpu/drm/i915/Kconfig.profile index 39328567c2007..7cc38d25ee5c8 100644 --- a/drivers/gpu/drm/i915/Kconfig.profile +++ b/drivers/gpu/drm/i915/Kconfig.profile @@ -57,10 +57,28 @@ config DRM_I915_PREEMPT_TIMEOUT default 640 # milliseconds help How long to wait (in milliseconds) for a preemption event to occur - when submitting a new context via execlists. If the current context - does not hit an arbitration point and yield to HW before the timer - expires, the HW will be reset to allow the more important context - to execute. + when submitting a new context. If the current context does not hit + an arbitration point and yield to HW before the timer expires, the + HW will be reset to allow the more important context to execute. + + This is adjustable via + /sys/class/drm/card?/engine/*/preempt_timeout_ms + + May be 0 to disable the timeout. + + The compiled in default may get overridden at driver probe time on + certain platforms and certain engines which will be reflected in the + sysfs control. + +config DRM_I915_PREEMPT_TIMEOUT_COMPUTE + int "Preempt timeout for compute engines (ms, jiffy granularity)" + default 7500 # milliseconds + help + How long to wait (in milliseconds) for a preemption event to occur + when submitting a new context to a compute capable engine. If the + current context does not hit an arbitration point and yield to HW + before the timer expires, the HW will be reset to allow the more + important context to execute. This is adjustable via /sys/class/drm/card?/engine/*/preempt_timeout_ms diff --git a/drivers/gpu/drm/i915/gt/intel_engine_cs.c b/drivers/gpu/drm/i915/gt/intel_engine_cs.c index fcbccd8d244e9..c1257723d1949 100644 --- a/drivers/gpu/drm/i915/gt/intel_engine_cs.c +++ b/drivers/gpu/drm/i915/gt/intel_engine_cs.c @@ -508,9 +508,14 @@ static int intel_engine_setup(struct intel_gt *gt, enum intel_engine_id id, engine->props.timeslice_duration_ms = CONFIG_DRM_I915_TIMESLICE_DURATION; - /* Override to uninterruptible for OpenCL workloads. */ + /* + * Mid-thread pre-emption is not available in Gen12. Unfortunately, + * some compute workloads run quite long threads. That means they get + * reset due to not pre-empting in a timely manner. So, bump the + * pre-emption timeout value to be much higher for compute engines. + */ if (GRAPHICS_VER(i915) == 12 && (engine->flags & I915_ENGINE_HAS_RCS_REG_STATE)) - engine->props.preempt_timeout_ms = 0; + engine->props.preempt_timeout_ms = CONFIG_DRM_I915_PREEMPT_TIMEOUT_COMPUTE; /* Cap properties according to any system limits */ #define CLAMP_PROP(field) \