drm/nv50: fix iommu errors caused by device reading from address 0

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>

drivers/gpu/drm/nouveau/nv50_fb.c

@@ -6,10 +6,16 @@
 int
 nv50_fb_init(struct drm_device *dev)
 {
-	/* This is needed to get meaningful information from 100c90
-	 * on traps. No idea what these values mean exactly. */
 	struct drm_nouveau_private *dev_priv = dev->dev_private;
 
+	/* Not a clue what this is exactly.  Without pointing it at a
+	 * scratch page, VRAM->GART blits with M2MF (as in DDX DFS)
+	 * cause IOMMU "read from address 0" errors (rh#561267)
+	 */
+	nv_wr32(dev, 0x100c08, dev_priv->gart_info.sg_dummy_bus >> 8);
+
+	/* This is needed to get meaningful information from 100c90
+	 * on traps. No idea what these values mean exactly. */
 	switch (dev_priv->chipset) {
 	case 0x50:
 		nv_wr32(dev, 0x100c90, 0x0707ff);