drm/i915: Allow userspace to request no-error-capture upon GPU hangs
igt likes to inject GPU hangs into its command streams. However, as we
expect these hangs, we don't actually want them recorded in the dmesg
output or stored in the i915_error_state (usually). To accommodate this
allow userspace to set a flag on the context that any hang emanating
from that context will not be recorded. We still do the error capture
(otherwise how do we find the guilty context and know its intent?) as
part of the reason for random GPU hang injection is to exercise the race
conditions between the error capture and normal execution.
v2: Split out the request->ringbuf error capture changes.
v3: Move the flag defines next to the intel_context->flags definition
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: Dave Gordon <david.s.gordon@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1467616119-4093-9-git-send-email-chris@chris-wilson.co.uk
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 4888918..251a08d 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -475,6 +475,7 @@
struct timeval time;
char error_msg[128];
+ bool simulated;
int iommu;
u32 reset_count;
u32 suspend_count;
@@ -875,9 +876,10 @@
/* Unique identifier for this context, used by the hw for tracking */
unsigned long flags;
+#define CONTEXT_NO_ZEROMAP BIT(0)
+#define CONTEXT_NO_ERROR_CAPTURE BIT(1)
unsigned hw_id;
u32 user_handle;
-#define CONTEXT_NO_ZEROMAP (1<<0)
u32 ggtt_alignment;
diff --git a/drivers/gpu/drm/i915/i915_gem_context.c b/drivers/gpu/drm/i915/i915_gem_context.c
index 3a6594b..8e952b1 100644
--- a/drivers/gpu/drm/i915/i915_gem_context.c
+++ b/drivers/gpu/drm/i915/i915_gem_context.c
@@ -1026,6 +1026,9 @@
else
args->value = to_i915(dev)->ggtt.base.total;
break;
+ case I915_CONTEXT_PARAM_NO_ERROR_CAPTURE:
+ args->value = !!(ctx->flags & CONTEXT_NO_ERROR_CAPTURE);
+ break;
default:
ret = -EINVAL;
break;
@@ -1071,6 +1074,16 @@
ctx->flags |= args->value ? CONTEXT_NO_ZEROMAP : 0;
}
break;
+ case I915_CONTEXT_PARAM_NO_ERROR_CAPTURE:
+ if (args->size) {
+ ret = -EINVAL;
+ } else {
+ if (args->value)
+ ctx->flags |= CONTEXT_NO_ERROR_CAPTURE;
+ else
+ ctx->flags &= ~CONTEXT_NO_ERROR_CAPTURE;
+ }
+ break;
default:
ret = -EINVAL;
break;
diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c
index 1be6359..c6e05cc 100644
--- a/drivers/gpu/drm/i915/i915_gpu_error.c
+++ b/drivers/gpu/drm/i915/i915_gpu_error.c
@@ -1093,9 +1093,8 @@
struct i915_address_space *vm;
struct intel_ringbuffer *rb;
- vm = request->ctx && request->ctx->ppgtt ?
- &request->ctx->ppgtt->base :
- &ggtt->base;
+ vm = request->ctx->ppgtt ?
+ &request->ctx->ppgtt->base : &ggtt->base;
/* We need to copy these to an anonymous buffer
* as the simplest method to avoid being overwritten
@@ -1123,6 +1122,9 @@
rcu_read_unlock();
}
+ error->simulated |=
+ request->ctx->flags & CONTEXT_NO_ERROR_CAPTURE;
+
rb = request->ringbuf;
error->ring[i].cpu_ring_head = rb->head;
error->ring[i].cpu_ring_tail = rb->tail;
@@ -1422,12 +1424,14 @@
i915_error_capture_msg(dev_priv, error, engine_mask, error_msg);
DRM_INFO("%s\n", error->error_msg);
- spin_lock_irqsave(&dev_priv->gpu_error.lock, flags);
- if (dev_priv->gpu_error.first_error == NULL) {
- dev_priv->gpu_error.first_error = error;
- error = NULL;
+ if (!error->simulated) {
+ spin_lock_irqsave(&dev_priv->gpu_error.lock, flags);
+ if (!dev_priv->gpu_error.first_error) {
+ dev_priv->gpu_error.first_error = error;
+ error = NULL;
+ }
+ spin_unlock_irqrestore(&dev_priv->gpu_error.lock, flags);
}
- spin_unlock_irqrestore(&dev_priv->gpu_error.lock, flags);
if (error) {
i915_error_state_free(&error->ref);