[ 3.5.y.z extended stable ] Patch "drm/i915: Workaround incoherence between fences and LLC" has been added to staging queue

[Luis Henriques]

This is a note to let you know that I have just added a patch titled

    drm/i915: Workaround incoherence between fences and LLC

to the linux-3.5.y-queue branch of the 3.5.y.z extended stable tree 
which can be found at:

 http://kernel.ubuntu.com/git?p=ubuntu/linux.git;a=shortlog;h=refs/heads/linux-3.5.y-queue

If you, or anyone else, feels it should not be added to this tree, please 
reply to this email.

For more information about the 3.5.y.z tree, see
https://wiki.ubuntu.com/Kernel/Dev/ExtendedStable

Thanks.
-Luis

------

>From 411d06743e3ca01d0161ede1a4409d96085d61ff Mon Sep 17 00:00:00 2001
From: Chris Wilson <chris at chris-wilson.co.uk>
Date: Thu, 4 Apr 2013 21:31:03 +0100
Subject: [PATCH] drm/i915: Workaround incoherence between fences and LLC
 across multiple CPUs

commit 25ff1195f8a0b3724541ae7bbe331b4296de9c06 upstream.

In order to fully serialize access to the fenced region and the update
to the fence register we need to take extreme measures on SNB+, and
manually flush writes to memory prior to writing the fence register in
conjunction with the memory barriers placed around the register write.

Fixes i-g-t/gem_fence_thrash

v2: Bring a bigger gun
v3: Switch the bigger gun for heavier bullets (Arjan van de Ven)
v4: Remove changes for working generations.
v5: Reduce to a per-cpu wbinvd() call prior to updating the fences.
v6: Rewrite comments to ellide forgotten history.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=62191
Signed-off-by: Chris Wilson <chris at chris-wilson.co.uk>
Cc: Jon Bloomfield <jon.bloomfield at intel.com>
Tested-by: Jon Bloomfield <jon.bloomfield at intel.com> (v2)
Reviewed-by: Jesse Barnes <jbarnes at virtuousgeek.org>
Signed-off-by: Daniel Vetter <daniel.vetter at ffwll.ch>
Signed-off-by: Luis Henriques <luis.henriques at canonical.com>
---
 drivers/gpu/drm/i915/i915_gem.c | 28 +++++++++++++++++++++++-----
 1 file changed, 23 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 7605f40..19e9f33 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -2332,17 +2332,35 @@ static inline int fence_number(struct drm_i915_private *dev_priv,
 	return fence - dev_priv->fence_regs;
 }

+static void i915_gem_write_fence__ipi(void *data)
+{
+	wbinvd();
+}
+
 static void i915_gem_object_update_fence(struct drm_i915_gem_object *obj,
 					 struct drm_i915_fence_reg *fence,
 					 bool enable)
 {
-	struct drm_i915_private *dev_priv = obj->base.dev->dev_private;
-	int reg = fence_number(dev_priv, fence);
-
-	i915_gem_write_fence(obj->base.dev, reg, enable ? obj : NULL);
+	struct drm_device *dev = obj->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	int fence_reg = fence_number(dev_priv, fence);
+
+	/* In order to fully serialize access to the fenced region and
+	 * the update to the fence register we need to take extreme
+	 * measures on SNB+. In theory, the write to the fence register
+	 * flushes all memory transactions before, and coupled with the
+	 * mb() placed around the register write we serialise all memory
+	 * operations with respect to the changes in the tiler. Yet, on
+	 * SNB+ we need to take a step further and emit an explicit wbinvd()
+	 * on each processor in order to manually flush all memory
+	 * transactions before updating the fence register.
+	 */
+	if (HAS_LLC(obj->base.dev))
+		on_each_cpu(i915_gem_write_fence__ipi, NULL, 1);
+	i915_gem_write_fence(dev, fence_reg, enable ? obj : NULL);

 	if (enable) {
-		obj->fence_reg = reg;
+		obj->fence_reg = fence_reg;
 		fence->obj = obj;
 		list_move_tail(&fence->lru_list, &dev_priv->mm.fence_list);
 	} else {
--
1.8.1.2