drm/radeon/kms/r600: reduce gpu cache flushing

Only flush the gpu caches before we submit a fence.
This leads to a small performance boost when we take
the extra gpu cache flushes out of the ddx and mesa.
Once this patch is in and the drm version is bumped,
we can remove the flushes from the ddx and drm.

Also, remove the extra cache flushes from the blit
routine.
Signed-off-by: Alex Deucher <alexdeucher@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>

drivers/gpu/drm/radeon/r600.c

@@ -1783,6 +1783,13 @@ void r600_fence_ring_emit(struct radeon_device *rdev,
 			  struct radeon_fence *fence)
 {
 	/* Also consider EVENT_WRITE_EOP.  it handles the interrupts + timestamps + events */
+
+	radeon_ring_write(rdev, PACKET3(PACKET3_EVENT_WRITE, 0));
+	radeon_ring_write(rdev, CACHE_FLUSH_AND_INV_EVENT);
+	/* wait for 3D idle clean */
+	radeon_ring_write(rdev, PACKET3(PACKET3_SET_CONFIG_REG, 1));
+	radeon_ring_write(rdev, (WAIT_UNTIL - PACKET3_SET_CONFIG_REG_OFFSET) >> 2);
+	radeon_ring_write(rdev, WAIT_3D_IDLE_bit | WAIT_3D_IDLECLEAN_bit);
 	/* Emit fence sequence & fire IRQ */
 	radeon_ring_write(rdev, PACKET3(PACKET3_SET_CONFIG_REG, 1));
 	radeon_ring_write(rdev, ((rdev->fence_drv.scratch_reg - PACKET3_SET_CONFIG_REG_OFFSET) >> 2));

drivers/gpu/drm/radeon/r600_blit_kms.c

@@ -403,8 +403,6 @@ set_default_state(struct radeon_device *rdev)
 	radeon_ring_write(rdev, upper_32_bits(gpu_addr) & 0xFF);
 	radeon_ring_write(rdev, dwords);
 
-	radeon_ring_write(rdev, PACKET3(PACKET3_EVENT_WRITE, 0));
-	radeon_ring_write(rdev, CACHE_FLUSH_AND_INV_EVENT);
 	/* SQ config */
 	radeon_ring_write(rdev, PACKET3(PACKET3_SET_CONFIG_REG, 6));
 	radeon_ring_write(rdev, (SQ_CONFIG - PACKET3_SET_CONFIG_REG_OFFSET) >> 2);
@@ -581,9 +579,9 @@ int r600_blit_prepare_copy(struct radeon_device *rdev, int size_bytes)
 	ring_size = num_loops * dwords_per_loop;
 	/* set default  + shaders */
 	ring_size += 40; /* shaders + def state */
-	ring_size += 7; /* fence emit for VB IB */
+	ring_size += 12; /* fence emit for VB IB */
 	ring_size += 5; /* done copy */
-	ring_size += 7; /* fence emit for done copy */
+	ring_size += 12; /* fence emit for done copy */
 	r = radeon_ring_lock(rdev, ring_size);
 	if (r)
 		return r;
@@ -597,13 +595,6 @@ void r600_blit_done_copy(struct radeon_device *rdev, struct radeon_fence *fence)
 {
 	int r;
 
-	radeon_ring_write(rdev, PACKET3(PACKET3_EVENT_WRITE, 0));
-	radeon_ring_write(rdev, CACHE_FLUSH_AND_INV_EVENT);
-	/* wait for 3D idle clean */
-	radeon_ring_write(rdev, PACKET3(PACKET3_SET_CONFIG_REG, 1));
-	radeon_ring_write(rdev, (WAIT_UNTIL - PACKET3_SET_CONFIG_REG_OFFSET) >> 2);
-	radeon_ring_write(rdev, WAIT_3D_IDLE_bit | WAIT_3D_IDLECLEAN_bit);
-
 	if (rdev->r600_blit.vb_ib)
 		r600_vb_ib_put(rdev);
 

drivers/gpu/drm/radeon/r600_blit_shaders.c

@@ -9,11 +9,6 @@ const u32 r6xx_default_state[] =
 	0xc0012800,
 	0x80000000,
 	0x80000000,
-	0xc0004600,
-	0x00000016,
-	0xc0016800,
-	0x00000010,
-	0x00028000,
 	0xc0016800,
 	0x00000010,
 	0x00008000,
@@ -531,11 +526,6 @@ const u32 r7xx_default_state[] =
 	0xc0012800,
 	0x80000000,
 	0x80000000,
-	0xc0004600,
-	0x00000016,
-	0xc0016800,
-	0x00000010,
-	0x00028000,
 	0xc0016800,
 	0x00000010,
 	0x00008000,