diff --git a/drivers/gpu/drm/radeon/r600.c b/drivers/gpu/drm/radeon/r600.c
index 923f936..537e259 100644
--- a/drivers/gpu/drm/radeon/r600.c
+++ b/drivers/gpu/drm/radeon/r600.c
@@ -2646,7 +2646,7 @@
  * @num_gpu_pages: number of GPU pages to xfer
  * @fence: radeon fence object
  *
- * Copy GPU paging using the DMA engine (r6xx-r7xx).
+ * Copy GPU paging using the DMA engine (r6xx).
  * Used by the radeon ttm implementation to move pages if
  * registered as the asic copy callback.
  */
@@ -2669,8 +2669,8 @@
 	}
 
 	size_in_dw = (num_gpu_pages << RADEON_GPU_PAGE_SHIFT) / 4;
-	num_loops = DIV_ROUND_UP(size_in_dw, 0xffff);
-	r = radeon_ring_lock(rdev, ring, num_loops * 5 + 8);
+	num_loops = DIV_ROUND_UP(size_in_dw, 0xFFFE);
+	r = radeon_ring_lock(rdev, ring, num_loops * 4 + 8);
 	if (r) {
 		DRM_ERROR("radeon: moving bo (%d).\n", r);
 		radeon_semaphore_free(rdev, &sem, NULL);
@@ -2693,8 +2693,8 @@
 		radeon_ring_write(ring, DMA_PACKET(DMA_PACKET_COPY, 0, 0, cur_size_in_dw));
 		radeon_ring_write(ring, dst_offset & 0xfffffffc);
 		radeon_ring_write(ring, src_offset & 0xfffffffc);
-		radeon_ring_write(ring, upper_32_bits(dst_offset) & 0xff);
-		radeon_ring_write(ring, upper_32_bits(src_offset) & 0xff);
+		radeon_ring_write(ring, (((upper_32_bits(dst_offset) & 0xff) << 16) |
+					 (upper_32_bits(src_offset) & 0xff)));
 		src_offset += cur_size_in_dw * 4;
 		dst_offset += cur_size_in_dw * 4;
 	}
diff --git a/drivers/gpu/drm/radeon/r600_cs.c b/drivers/gpu/drm/radeon/r600_cs.c
index 9ea13d0..03191a5 100644
--- a/drivers/gpu/drm/radeon/r600_cs.c
+++ b/drivers/gpu/drm/radeon/r600_cs.c
@@ -2677,16 +2677,29 @@
 				}
 				p->idx += 7;
 			} else {
-				src_offset = ib[idx+2];
-				src_offset |= ((u64)(ib[idx+4] & 0xff)) << 32;
-				dst_offset = ib[idx+1];
-				dst_offset |= ((u64)(ib[idx+3] & 0xff)) << 32;
+				if (p->family >= CHIP_RV770) {
+					src_offset = ib[idx+2];
+					src_offset |= ((u64)(ib[idx+4] & 0xff)) << 32;
+					dst_offset = ib[idx+1];
+					dst_offset |= ((u64)(ib[idx+3] & 0xff)) << 32;
 
-				ib[idx+1] += (u32)(dst_reloc->lobj.gpu_offset & 0xfffffffc);
-				ib[idx+2] += (u32)(src_reloc->lobj.gpu_offset & 0xfffffffc);
-				ib[idx+3] += upper_32_bits(dst_reloc->lobj.gpu_offset) & 0xff;
-				ib[idx+4] += upper_32_bits(src_reloc->lobj.gpu_offset) & 0xff;
-				p->idx += 5;
+					ib[idx+1] += (u32)(dst_reloc->lobj.gpu_offset & 0xfffffffc);
+					ib[idx+2] += (u32)(src_reloc->lobj.gpu_offset & 0xfffffffc);
+					ib[idx+3] += upper_32_bits(dst_reloc->lobj.gpu_offset) & 0xff;
+					ib[idx+4] += upper_32_bits(src_reloc->lobj.gpu_offset) & 0xff;
+					p->idx += 5;
+				} else {
+					src_offset = ib[idx+2];
+					src_offset |= ((u64)(ib[idx+3] & 0xff)) << 32;
+					dst_offset = ib[idx+1];
+					dst_offset |= ((u64)(ib[idx+3] & 0xff0000)) << 16;
+
+					ib[idx+1] += (u32)(dst_reloc->lobj.gpu_offset & 0xfffffffc);
+					ib[idx+2] += (u32)(src_reloc->lobj.gpu_offset & 0xfffffffc);
+					ib[idx+3] += upper_32_bits(src_reloc->lobj.gpu_offset) & 0xff;
+					ib[idx+3] += (upper_32_bits(dst_reloc->lobj.gpu_offset) & 0xff) << 16;
+					p->idx += 4;
+				}
 			}
 			if ((src_offset + (count * 4)) > radeon_bo_size(src_reloc->robj)) {
 				dev_warn(p->dev, "DMA copy src buffer too small (%llu %lu)\n",
diff --git a/drivers/gpu/drm/radeon/radeon_asic.c b/drivers/gpu/drm/radeon/radeon_asic.c
index 596bcbe..9056faf 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.c
+++ b/drivers/gpu/drm/radeon/radeon_asic.c
@@ -1140,9 +1140,9 @@
 	.copy = {
 		.blit = &r600_copy_blit,
 		.blit_ring_index = RADEON_RING_TYPE_GFX_INDEX,
-		.dma = &r600_copy_dma,
+		.dma = &rv770_copy_dma,
 		.dma_ring_index = R600_RING_TYPE_DMA_INDEX,
-		.copy = &r600_copy_dma,
+		.copy = &rv770_copy_dma,
 		.copy_ring_index = R600_RING_TYPE_DMA_INDEX,
 	},
 	.surface = {
diff --git a/drivers/gpu/drm/radeon/radeon_asic.h b/drivers/gpu/drm/radeon/radeon_asic.h
index 5f4882c..15d70e6 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.h
+++ b/drivers/gpu/drm/radeon/radeon_asic.h
@@ -403,6 +403,10 @@
 void r700_vram_gtt_location(struct radeon_device *rdev, struct radeon_mc *mc);
 void r700_cp_stop(struct radeon_device *rdev);
 void r700_cp_fini(struct radeon_device *rdev);
+int rv770_copy_dma(struct radeon_device *rdev,
+		  uint64_t src_offset, uint64_t dst_offset,
+		  unsigned num_gpu_pages,
+		   struct radeon_fence **fence);
 
 /*
  * evergreen
diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c
index cd75626..edfc54e 100644
--- a/drivers/gpu/drm/radeon/radeon_device.c
+++ b/drivers/gpu/drm/radeon/radeon_device.c
@@ -897,6 +897,25 @@
 }
 
 /**
+ * radeon_switcheroo_quirk_long_wakeup - return true if longer d3 delay is
+ * needed for waking up.
+ *
+ * @pdev: pci dev pointer
+ */
+static bool radeon_switcheroo_quirk_long_wakeup(struct pci_dev *pdev)
+{
+
+	/* 6600m in a macbook pro */
+	if (pdev->subsystem_vendor == PCI_VENDOR_ID_APPLE &&
+	    pdev->subsystem_device == 0x00e2) {
+		printk(KERN_INFO "radeon: quirking longer d3 wakeup delay\n");
+		return true;
+	}
+
+	return false;
+}
+
+/**
  * radeon_switcheroo_set_state - set switcheroo state
  *
  * @pdev: pci dev pointer
@@ -910,10 +929,19 @@
 	struct drm_device *dev = pci_get_drvdata(pdev);
 	pm_message_t pmm = { .event = PM_EVENT_SUSPEND };
 	if (state == VGA_SWITCHEROO_ON) {
+		unsigned d3_delay = dev->pdev->d3_delay;
+
 		printk(KERN_INFO "radeon: switched on\n");
 		/* don't suspend or resume card normally */
 		dev->switch_power_state = DRM_SWITCH_POWER_CHANGING;
+
+		if (d3_delay < 20 && radeon_switcheroo_quirk_long_wakeup(pdev))
+			dev->pdev->d3_delay = 20;
+
 		radeon_resume_kms(dev);
+
+		dev->pdev->d3_delay = d3_delay;
+
 		dev->switch_power_state = DRM_SWITCH_POWER_ON;
 		drm_kms_helper_poll_enable(dev);
 	} else {
diff --git a/drivers/gpu/drm/radeon/rv770.c b/drivers/gpu/drm/radeon/rv770.c
index 87c979c..1b2444f 100644
--- a/drivers/gpu/drm/radeon/rv770.c
+++ b/drivers/gpu/drm/radeon/rv770.c
@@ -887,6 +887,80 @@
 	return 0;
 }
 
+/**
+ * rv770_copy_dma - copy pages using the DMA engine
+ *
+ * @rdev: radeon_device pointer
+ * @src_offset: src GPU address
+ * @dst_offset: dst GPU address
+ * @num_gpu_pages: number of GPU pages to xfer
+ * @fence: radeon fence object
+ *
+ * Copy GPU paging using the DMA engine (r7xx).
+ * Used by the radeon ttm implementation to move pages if
+ * registered as the asic copy callback.
+ */
+int rv770_copy_dma(struct radeon_device *rdev,
+		  uint64_t src_offset, uint64_t dst_offset,
+		  unsigned num_gpu_pages,
+		  struct radeon_fence **fence)
+{
+	struct radeon_semaphore *sem = NULL;
+	int ring_index = rdev->asic->copy.dma_ring_index;
+	struct radeon_ring *ring = &rdev->ring[ring_index];
+	u32 size_in_dw, cur_size_in_dw;
+	int i, num_loops;
+	int r = 0;
+
+	r = radeon_semaphore_create(rdev, &sem);
+	if (r) {
+		DRM_ERROR("radeon: moving bo (%d).\n", r);
+		return r;
+	}
+
+	size_in_dw = (num_gpu_pages << RADEON_GPU_PAGE_SHIFT) / 4;
+	num_loops = DIV_ROUND_UP(size_in_dw, 0xFFFF);
+	r = radeon_ring_lock(rdev, ring, num_loops * 5 + 8);
+	if (r) {
+		DRM_ERROR("radeon: moving bo (%d).\n", r);
+		radeon_semaphore_free(rdev, &sem, NULL);
+		return r;
+	}
+
+	if (radeon_fence_need_sync(*fence, ring->idx)) {
+		radeon_semaphore_sync_rings(rdev, sem, (*fence)->ring,
+					    ring->idx);
+		radeon_fence_note_sync(*fence, ring->idx);
+	} else {
+		radeon_semaphore_free(rdev, &sem, NULL);
+	}
+
+	for (i = 0; i < num_loops; i++) {
+		cur_size_in_dw = size_in_dw;
+		if (cur_size_in_dw > 0xFFFF)
+			cur_size_in_dw = 0xFFFF;
+		size_in_dw -= cur_size_in_dw;
+		radeon_ring_write(ring, DMA_PACKET(DMA_PACKET_COPY, 0, 0, cur_size_in_dw));
+		radeon_ring_write(ring, dst_offset & 0xfffffffc);
+		radeon_ring_write(ring, src_offset & 0xfffffffc);
+		radeon_ring_write(ring, upper_32_bits(dst_offset) & 0xff);
+		radeon_ring_write(ring, upper_32_bits(src_offset) & 0xff);
+		src_offset += cur_size_in_dw * 4;
+		dst_offset += cur_size_in_dw * 4;
+	}
+
+	r = radeon_fence_emit(rdev, fence, ring->idx);
+	if (r) {
+		radeon_ring_unlock_undo(rdev, ring);
+		return r;
+	}
+
+	radeon_ring_unlock_commit(rdev, ring);
+	radeon_semaphore_free(rdev, &sem, *fence);
+
+	return r;
+}
+
 static int rv770_startup(struct radeon_device *rdev)
 {
 	struct radeon_ring *ring;