radeonsi: don't count fast clears and prefetches into CP DMA stats

Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
diff --git a/src/gallium/drivers/radeonsi/si_cp_dma.c b/src/gallium/drivers/radeonsi/si_cp_dma.c
index 596e32c..1996e66 100644
--- a/src/gallium/drivers/radeonsi/si_cp_dma.c
+++ b/src/gallium/drivers/radeonsi/si_cp_dma.c
@@ -234,7 +234,9 @@
 	if (tc_l2_flag)
 		rdst->TC_L2_dirty = true;
 
-	sctx->b.num_cp_dma_calls++;
+	/* If it's not a framebuffer fast clear... */
+	if (coher == R600_COHERENCY_SHADER)
+		sctx->b.num_cp_dma_calls++;
 }
 
 /**
@@ -360,7 +362,9 @@
 	if (tc_l2_flag)
 		r600_resource(dst)->TC_L2_dirty = true;
 
-	sctx->b.num_cp_dma_calls++;
+	/* If it's not a prefetch... */
+	if (dst_offset != src_offset)
+		sctx->b.num_cp_dma_calls++;
 }
 
 void si_init_cp_dma_functions(struct si_context *sctx)