intel: use pwrite for batch

It's faster. Not only is the memcpy more efficiently performed in the
kernel (making up for the system call overhead), but by not using mmap
we remove the greater overhead of tracking the vma of every batch.

And it means we can read back from the batch buffer without incurring
the cost of a uncached read through the GTT.

Signed-off-by: Chris Wilson <[email protected]>

1 files changed, 7 insertions, 12 deletions

diff --git a/src/mesa/drivers/dri/i965/brw_draw.c b/src/mesa/drivers/dri/i965/brw_draw.c
index 57d1d7bbe07..99c1b2b359c 100644
--- a/src/mesa/drivers/dri/i965/brw_draw.c
+++ b/src/mesa/drivers/dri/i965/brw_draw.c
@@ -155,14 +155,14 @@ static void brw_emit_prim(struct brw_context *brw,
     * the besides the draw code.
     */
    if (intel->always_flush_cache) {
-      intel_batchbuffer_emit_mi_flush(intel->batch);
+      intel_batchbuffer_emit_mi_flush(intel);
    }
    if (prim_packet.verts_per_instance) {
-      intel_batchbuffer_data( brw->intel.batch, &prim_packet,
+      intel_batchbuffer_data(&brw->intel, &prim_packet,
 			      sizeof(prim_packet), false);
    }
    if (intel->always_flush_cache) {
-      intel_batchbuffer_emit_mi_flush(intel->batch);
+      intel_batchbuffer_emit_mi_flush(intel);
    }
 }
 
@@ -303,7 +303,6 @@ static GLboolean brw_try_draw_prims( struct gl_context *ctx,
    struct brw_context *brw = brw_context(ctx);
    GLboolean retval = GL_FALSE;
    GLboolean warn = GL_FALSE;
-   GLboolean first_time = GL_TRUE;
    GLuint i;
 
    if (ctx->NewState)
@@ -351,14 +350,10 @@ static GLboolean brw_try_draw_prims( struct gl_context *ctx,
        * an upper bound of how much we might emit in a single
        * brw_try_draw_prims().
        */
-      intel_batchbuffer_require_space(intel->batch, intel->batch->size / 4,
-				      false);
+      intel_batchbuffer_require_space(intel, 1024, false);
 
       hw_prim = brw_set_prim(brw, &prim[i]);
-
-      if (first_time || (brw->state.dirty.brw & BRW_NEW_PRIMITIVE)) {
-	 first_time = GL_FALSE;
-
+      if (brw->state.dirty.brw) {
 	 brw_validate_state(brw);
 
 	 /* Various fallback checks:  */
@@ -371,7 +366,7 @@ static GLboolean brw_try_draw_prims( struct gl_context *ctx,
 	 if (dri_bufmgr_check_aperture_space(brw->state.validated_bos,
 					     brw->state.validated_bo_count)) {
 	    static GLboolean warned;
-	    intel_batchbuffer_flush(intel->batch);
+	    intel_batchbuffer_flush(intel);
 
 	    /* Validate the state after we flushed the batch (which would have
 	     * changed the set of dirty state).  If we still fail to
@@ -400,7 +395,7 @@ static GLboolean brw_try_draw_prims( struct gl_context *ctx,
    }
 
    if (intel->always_flush_batch)
-      intel_batchbuffer_flush(intel->batch);
+      intel_batchbuffer_flush(intel);
  out:
 
    brw_state_cache_check_size(brw);