[ 146/147] drm/i915: Mark untiled BLT commands as fenced on gen2/3

[Ben Hutchings]

3.2-stable review patch.  If anyone has any objections, please let me know.

------------------

From: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx>

commit 7dd4906586274f3945f2aeaaa5a33b451c3b4bba upstream.

The BLT commands on gen2/3 utilize the fence registers and so we cannot
modify any fences for the object whilst those commands are in flight.
Currently we marked tiled commands as occupying a fence, but forgot to
restrict the untiled commands from preventing a fence being assigned
before they were completed.

One side-effect is that we ten have to double check that a fence was
allocated for a fenced buffer during move-to-active.

Reported-by: Jiri Slaby <jirislaby@xxxxxxxxx>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=43427
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=47990
Reviewed-by: Daniel Vetter <daniel.vetter@xxxxxxxx>
Testcase: i-g-t/tests/gem_tiled_after_untiled_blt
Tested-by: Daniel Vetter <daniel.vetter@xxxxxxxx>
Signed-off-by: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx>
Signed-off-by: Daniel Vetter <daniel.vetter@xxxxxxxx>
[bwh: Backported to 3.2: The nesting of if-statements in the old
 i915_gem_execbuffer_reserve() differs from pin_and_fence_object(),
 so don't move the assignment of obj->pending_fenced_gpu_access but
 adjust the boolean expression as recommended by Daniel Vetter.]
Signed-off-by: Ben Hutchings <ben@xxxxxxxxxxxxxxx>
---
 drivers/gpu/drm/i915/i915_gem.c            |   15 +++++++++------
 drivers/gpu/drm/i915/i915_gem_execbuffer.c |    2 +-
 2 files changed, 10 insertions(+), 7 deletions(-)

--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -1543,16 +1543,19 @@ i915_gem_object_move_to_active(struct dr
 	list_move_tail(&obj->ring_list, &ring->active_list);
 
 	obj->last_rendering_seqno = seqno;
-	if (obj->fenced_gpu_access) {
-		struct drm_i915_fence_reg *reg;
-
-		BUG_ON(obj->fence_reg == I915_FENCE_REG_NONE);
 
+	if (obj->fenced_gpu_access) {
 		obj->last_fenced_seqno = seqno;
 		obj->last_fenced_ring = ring;
 
-		reg = &dev_priv->fence_regs[obj->fence_reg];
-		list_move_tail(&reg->lru_list, &dev_priv->mm.fence_list);
+		/* Bump MRU to take account of the delayed flush */
+		if (obj->fence_reg != I915_FENCE_REG_NONE) {
+			struct drm_i915_fence_reg *reg;
+
+			reg = &dev_priv->fence_regs[obj->fence_reg];
+			list_move_tail(&reg->lru_list,
+				       &dev_priv->mm.fence_list);
+		}
 	}
 }
 
--- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
@@ -574,7 +574,8 @@ i915_gem_execbuffer_reserve(struct intel
 					if (ret)
 						break;
 				}
-				obj->pending_fenced_gpu_access = need_fence;
+				obj->pending_fenced_gpu_access =
+					!!(entry->flags & EXEC_OBJECT_NEEDS_FENCE);
 			}
 
 			entry->offset = obj->gtt_offset;


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/