aboutsummaryrefslogtreecommitdiffstats
path: root/drivers/gpu/drm/i915/intel_ringbuffer.c
diff options
context:
space:
mode:
authorChris Wilson <chris@chris-wilson.co.uk>2011-01-20 17:00:10 +0000
committerChris Wilson <chris@chris-wilson.co.uk>2011-01-20 17:26:57 +0000
commitc7dca47bd6fbb7c215cb1ce6bc40398b4b017752 (patch)
tree0255506bd3b12333addebf13b0b4ee804ea24f8b /drivers/gpu/drm/i915/intel_ringbuffer.c
parent475553de2fc861d53396dd8fd14cc22f30ab97ab (diff)
downloadkernel_samsung_smdk4412-c7dca47bd6fbb7c215cb1ce6bc40398b4b017752.zip
kernel_samsung_smdk4412-c7dca47bd6fbb7c215cb1ce6bc40398b4b017752.tar.gz
kernel_samsung_smdk4412-c7dca47bd6fbb7c215cb1ce6bc40398b4b017752.tar.bz2
drm/i915/ringbuffer: Fix use of stale HEAD position whilst polling for space
During suspend, Linus found that his machine would hang for 3 seconds, and identified that intel_ring_buffer_wait() was the culprit: "Because from looking at the code, I get the notion that "intel_read_status_page()" may not be exact. But what happens if that inexact value matches our cached ring->actual_head, so we never even try to read the exact case? Does it _stay_ inexact for arbitrarily long times? If so, we might wait for the ring to empty forever (well, until the timeout - the behavior I see), even though the ring really _is_ empty." As the reported HEAD position is only updated every time it crosses a 64k boundary, whilst draining the ring it is indeed likely to remain one value. If that value matches the last known HEAD position, we never read the true value from the register and so trigger a timeout. Reported-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Diffstat (limited to 'drivers/gpu/drm/i915/intel_ringbuffer.c')
-rw-r--r--drivers/gpu/drm/i915/intel_ringbuffer.c40
1 files changed, 24 insertions, 16 deletions
diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
index 51fbc5e..6218fa9 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.c
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
@@ -34,6 +34,14 @@
#include "i915_trace.h"
#include "intel_drv.h"
+static inline int ring_space(struct intel_ring_buffer *ring)
+{
+ int space = (ring->head & HEAD_ADDR) - (ring->tail + 8);
+ if (space < 0)
+ space += ring->size;
+ return space;
+}
+
static u32 i915_gem_get_seqno(struct drm_device *dev)
{
drm_i915_private_t *dev_priv = dev->dev_private;
@@ -204,11 +212,9 @@ static int init_ring_common(struct intel_ring_buffer *ring)
if (!drm_core_check_feature(ring->dev, DRIVER_MODESET))
i915_kernel_lost_context(ring->dev);
else {
- ring->head = I915_READ_HEAD(ring) & HEAD_ADDR;
+ ring->head = I915_READ_HEAD(ring);
ring->tail = I915_READ_TAIL(ring) & TAIL_ADDR;
- ring->space = ring->head - (ring->tail + 8);
- if (ring->space < 0)
- ring->space += ring->size;
+ ring->space = ring_space(ring);
}
return 0;
@@ -921,7 +927,7 @@ static int intel_wrap_ring_buffer(struct intel_ring_buffer *ring)
}
ring->tail = 0;
- ring->space = ring->head - 8;
+ ring->space = ring_space(ring);
return 0;
}
@@ -933,20 +939,22 @@ int intel_wait_ring_buffer(struct intel_ring_buffer *ring, int n)
unsigned long end;
u32 head;
+ /* If the reported head position has wrapped or hasn't advanced,
+ * fallback to the slow and accurate path.
+ */
+ head = intel_read_status_page(ring, 4);
+ if (head > ring->head) {
+ ring->head = head;
+ ring->space = ring_space(ring);
+ if (ring->space >= n)
+ return 0;
+ }
+
trace_i915_ring_wait_begin (dev);
end = jiffies + 3 * HZ;
do {
- /* If the reported head position has wrapped or hasn't advanced,
- * fallback to the slow and accurate path.
- */
- head = intel_read_status_page(ring, 4);
- if (head < ring->actual_head)
- head = I915_READ_HEAD(ring);
- ring->actual_head = head;
- ring->head = head & HEAD_ADDR;
- ring->space = ring->head - (ring->tail + 8);
- if (ring->space < 0)
- ring->space += ring->size;
+ ring->head = I915_READ_HEAD(ring);
+ ring->space = ring_space(ring);
if (ring->space >= n) {
trace_i915_ring_wait_end(dev);
return 0;