We monitor the health of the system via periodic heartbeat pulses. The
pulses also provide the opportunity to perform garbage collection.
However, we interpret an incomplete pulse (a missed heartbeat) as an
indication that the system is no longer responsive, i.e. hung, and
perform an engine or full GPU reset. Given that the preemption
granularity can be very coarse on a system, we let the sysadmin override
our legacy timeouts which were "optimised" for desktop applications.
The heartbeat interval can be adjusted per-engine using,
/sys/class/drm/card?/engine/*/heartbeat_interval_ms
Signed-off-by: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx>
Reviewed-by: Steve Carbonari <steven.carbonari@xxxxxxxxx>
Tested-by: Steve Carbonari <steven.carbonari@xxxxxxxxx>
---
drivers/gpu/drm/i915/Kconfig.profile | 3 ++
drivers/gpu/drm/i915/gt/sysfs_engines.c | 47 +++++++++++++++++++++++++
2 files changed, 50 insertions(+)
diff --git a/drivers/gpu/drm/i915/Kconfig.profile b/drivers/gpu/drm/i915/Kconfig.profile
index 1f4e98a8532f..ba8767fc0d6e 100644
--- a/drivers/gpu/drm/i915/Kconfig.profile
+++ b/drivers/gpu/drm/i915/Kconfig.profile
@@ -20,6 +20,9 @@ config DRM_I915_HEARTBEAT_INTERVAL
check the health of the GPU and undertake regular house-keeping of
internal driver state.
+ This is adjustable via
+ /sys/class/drm/card?/engine/*/heartbeat_interval_ms
+
May be 0 to disable heartbeats and therefore disable automatic GPU
hang detection.
diff --git a/drivers/gpu/drm/i915/gt/sysfs_engines.c b/drivers/gpu/drm/i915/gt/sysfs_engines.c
index 1ebb221bd1cb..b39d4a365d73 100644
--- a/drivers/gpu/drm/i915/gt/sysfs_engines.c
+++ b/drivers/gpu/drm/i915/gt/sysfs_engines.c
@@ -8,6 +8,7 @@
#include "i915_drv.h"
#include "intel_engine.h"
+#include "intel_engine_heartbeat.h"
#include "sysfs_engines.h"
struct kobj_engine {
@@ -315,6 +316,49 @@ preempt_timeout_show(struct kobject *kobj, struct kobj_attribute *attr,
static struct kobj_attribute preempt_timeout_attr =
__ATTR(preempt_timeout_ms, 0644, preempt_timeout_show, preempt_timeout_store);
+static ssize_t
+heartbeat_store(struct kobject *kobj, struct kobj_attribute *attr,
+ const char *buf, size_t count)
+{
+ struct intel_engine_cs *engine = kobj_to_engine(kobj);
+ unsigned long long delay;
+ int err;
+
+ /*
+ * We monitor the health of the system via periodic heartbeat pulses.
+ * The pulses also provide the opportunity to perform garbage
+ * collection. However, we interpret an incomplete pulse (a missed
+ * heartbeat) as an indication that the system is no longer responsive,
+ * i.e. hung, and perform an engine or full GPU reset. Given that the
+ * preemption granularity can be very coarse on a system, the optimal
+ * value for any workload is unknowable!
+ */
+
+ err = kstrtoull(buf, 0, &delay);
+ if (err)
+ return err;
+
+ if (delay >= jiffies_to_msecs(MAX_SCHEDULE_TIMEOUT))
+ return -EINVAL;
+
+ err = intel_engine_set_heartbeat(engine, delay);
+ if (err)
+ return err;
+
+ return count;
+}
+
+static ssize_t
+heartbeat_show(struct kobject *kobj, struct kobj_attribute *attr, char *buf)
+{
+ struct intel_engine_cs *engine = kobj_to_engine(kobj);
+
+ return sprintf(buf, "%lu\n", engine->props.heartbeat_interval_ms);
+}
+
+static struct kobj_attribute heartbeat_interval_attr =
+__ATTR(heartbeat_interval_ms, 0644, heartbeat_show, heartbeat_store);
+
static void kobj_engine_release(struct kobject *kobj)
{
kfree(kobj);
@@ -357,6 +401,9 @@ void intel_engines_add_sysfs(struct drm_i915_private *i915)
&all_caps_attr.attr,
&max_spin_attr.attr,
&stop_timeout_attr.attr,
+#if CONFIG_DRM_I915_HEARTBEAT_INTERVAL
+ &heartbeat_interval_attr.attr,
+#endif
NULL
};
--
2.25.0
_______________________________________________
Intel-gfx mailing list
Intel-gfx@xxxxxxxxxxxxxxxxxxxxx
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
