[Intel-gfx] [PATCH] drm/i915: non-interruptible sleeps can't handle -EAGAIN

[Daniel Vetter]

So don't return -EAGAIN, even in the case of a gpu hang. Remap it to
-EIO instead. Note that this isn't really an issue with
interruptability, but more that we have quite a few codepaths (mostly
around kms stuff) that simply can't handle any errors and hence not
even -EAGAIN. Instead of adding proper failure paths so that we could
restart these ioctls we've opted for the cheap way out of sleeping
non-interruptibly.  Which works everywhere but when the gpu dies,
which this patch fixes.

So essentially interruptible == false means 'wait for the gpu or die
trying'.'

This patch is a bit ugly because intel_ring_begin is all non-interruptible
and hence only returns -EIO. But as the comment in there says,
auditing all the callsites would be a pain.

To avoid duplicating code, reuse i915_gem_check_wedge in __wait_seqno
and intel_wait_ring_buffer. Also use the opportunity to clarify the
different cases in i915_gem_check_wedge a bit with comments.

v2: Don't access dev_priv->mm.interruptible from check_wedge - we
might not hold dev->struct_mutex, making this racy. Instead pass
interruptible in as a parameter. I've noticed this because I've hit a
BUG_ON(!mutex_is_locked) at the top of check_wedge. This has been
added in

commit b4aca0106c466b5a0329318203f65bac2d91b682
Author: Ben Widawsky <ben at bwidawsk.net>
Date:   Wed Apr 25 20:50:12 2012 -0700

    drm/i915: extract some common olr+wedge code

although that commit is missing any justification for this. I guess
it's just copy&paste, because the same commit add the same BUG_ON
check to check_olr, where it indeed makes sense.

But in check_wedge everything we access is protected by other means,
so this is superflous. And because it now gets in the way (we add a
new caller in __wait_seqno, which can be called without
dev->struct_mutext) let's just remove it.

v3: Group all the i915_gem_check_wedge refactoring into this patch, so
that this patch here is all about not returning -EAGAIN to callsites
that can't handle syscall restarting.

v4: Add clarification what interuptible == fales means in our code,
requested by Ben Widawsky.

v5: Fix EAGAIN mispell noticed by Chris Wilson.

Reviewed-by: Ben Widawsky <ben at bwidawsk.net>
Signed-Off-by: Daniel Vetter <daniel.vetter at ffwll.ch>
---
 drivers/gpu/drm/i915/i915_drv.h         |    2 ++
 drivers/gpu/drm/i915/i915_gem.c         |   26 ++++++++++++++++++--------
 drivers/gpu/drm/i915/intel_ringbuffer.c |    6 ++++--
 3 files changed, 24 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index d839e4c..6c3a0bb 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -1336,6 +1336,8 @@ i915_gem_object_unpin_fence(struct drm_i915_gem_object *obj)
 
 void i915_gem_retire_requests(struct drm_device *dev);
 void i915_gem_retire_requests_ring(struct intel_ring_buffer *ring);
+int __must_check i915_gem_check_wedge(struct drm_i915_private *dev_priv,
+				      bool interruptible);
 
 void i915_gem_reset(struct drm_device *dev);
 void i915_gem_clflush_object(struct drm_i915_gem_object *obj);
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 6a98c06..af6a510 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -1863,11 +1863,10 @@ i915_gem_retire_work_handler(struct work_struct *work)
 	mutex_unlock(&dev->struct_mutex);
 }
 
-static int
-i915_gem_check_wedge(struct drm_i915_private *dev_priv)
+int
+i915_gem_check_wedge(struct drm_i915_private *dev_priv,
+		     bool interruptible)
 {
-	BUG_ON(!mutex_is_locked(&dev_priv->dev->struct_mutex));
-
 	if (atomic_read(&dev_priv->mm.wedged)) {
 		struct completion *x = &dev_priv->error_completion;
 		bool recovery_complete;
@@ -1878,7 +1877,16 @@ i915_gem_check_wedge(struct drm_i915_private *dev_priv)
 		recovery_complete = x->done > 0;
 		spin_unlock_irqrestore(&x->wait.lock, flags);
 
-		return recovery_complete ? -EIO : -EAGAIN;
+		/* Non-interruptible callers can't handle -EAGAIN, hence return
+		 * -EIO unconditionally for these. */
+		if (!interruptible)
+			return -EIO;
+
+		/* Recovery complete, but still wedged means reset failure. */
+		if (recovery_complete)
+			return -EIO;
+
+		return -EAGAIN;
 	}
 
 	return 0;
@@ -1932,6 +1940,7 @@ static int __wait_seqno(struct intel_ring_buffer *ring, u32 seqno,
 	unsigned long timeout_jiffies;
 	long end;
 	bool wait_forever = true;
+	int ret;
 
 	if (i915_seqno_passed(ring->get_seqno(ring), seqno))
 		return 0;
@@ -1963,8 +1972,9 @@ static int __wait_seqno(struct intel_ring_buffer *ring, u32 seqno,
 			end = wait_event_timeout(ring->irq_queue, EXIT_COND,
 						 timeout_jiffies);
 
-		if (atomic_read(&dev_priv->mm.wedged))
-			end = -EAGAIN;
+		ret = i915_gem_check_wedge(dev_priv, interruptible);
+		if (ret)
+			end = ret;
 	} while (end == 0 && wait_forever);
 
 	getrawmonotonic(&now);
@@ -2004,7 +2014,7 @@ i915_wait_seqno(struct intel_ring_buffer *ring, uint32_t seqno)
 
 	BUG_ON(seqno == 0);
 
-	ret = i915_gem_check_wedge(dev_priv);
+	ret = i915_gem_check_wedge(dev_priv, dev_priv->mm.interruptible);
 	if (ret)
 		return ret;
 
diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
index dce4d1a..cd35ad4 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.c
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
@@ -1228,8 +1228,10 @@ int intel_wait_ring_buffer(struct intel_ring_buffer *ring, int n)
 		}
 
 		msleep(1);
-		if (atomic_read(&dev_priv->mm.wedged))
-			return -EAGAIN;
+
+		ret = i915_gem_check_wedge(dev_priv, dev_priv->mm.interruptible);
+		if (ret)
+			return ret;
 	} while (!time_after(jiffies, end));
 	trace_i915_ring_wait_end(ring);
 	return -EBUSY;
-- 
1.7.10