draw: split off all the extra functionality in the vertex shader

This will at least allow us to make the initial gains to get decent
vertex performance much more quickly & with higher confidence of getting
it right.

At some later point can look again at code-generating all the
fetch/cliptest/viewport extras in the same block as the vertex shader.
For now, just need to get some decent baseline performance.

1 files changed, 175 insertions, 0 deletions

diff --git a/src/gallium/auxiliary/draw/draw_pt_fetch.c b/src/gallium/auxiliary/draw/draw_pt_fetch.c
new file mode 100644
index 00000000000..a7553023b81
--- /dev/null
+++ b/src/gallium/auxiliary/draw/draw_pt_fetch.c
@@ -0,0 +1,175 @@
+/**************************************************************************
+ *
+ * Copyright 2008 Tungsten Graphics, Inc., Cedar Park, Texas.
+ * All Rights Reserved.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the
+ * "Software"), to deal in the Software without restriction, including
+ * without limitation the rights to use, copy, modify, merge, publish,
+ * distribute, sub license, and/or sell copies of the Software, and to
+ * permit persons to whom the Software is furnished to do so, subject to
+ * the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the
+ * next paragraph) shall be included in all copies or substantial portions
+ * of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
+ * OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT.
+ * IN NO EVENT SHALL TUNGSTEN GRAPHICS AND/OR ITS SUPPLIERS BE LIABLE FOR
+ * ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
+ * TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
+ * SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
+ *
+ **************************************************************************/
+
+#include "pipe/p_util.h"
+#include "draw/draw_context.h"
+#include "draw/draw_private.h"
+#include "draw/draw_vbuf.h"
+#include "draw/draw_vertex.h"
+#include "draw/draw_pt.h"
+#include "translate/translate.h"
+
+
+struct pt_fetch {
+   struct draw_context *draw;
+
+   struct translate *translate;
+   
+   unsigned vertex_size;
+};
+
+
+
+/* Perform the fetch from API vertex elements & vertex buffers, to a
+ * contiguous set of float[4] attributes as required for the
+ * vertex_shader->run_linear() method.
+ *
+ * This is used in all cases except pure passthrough
+ * (draw_pt_fetch_emit.c) which has its own version to translate
+ * directly to hw vertices.
+ *
+ */
+void draw_pt_fetch_prepare( struct pt_fetch *fetch,
+			    boolean emit_header,
+			    unsigned vertex_size )
+{
+   struct draw_context *draw = fetch->draw;
+   unsigned i, nr = 0;
+   unsigned dst_offset = 0;
+   struct translate_key key;
+
+   fetch->vertex_size = vertex_size;
+
+   memset(&key, 0, sizeof(key));
+
+   /* If PT_SHADE is not set, then we are creating post-shader
+    * vertices, meaning that we need to emit/leave space for a vertex
+    * header.
+    *
+    * It's worth considering whether the vertex headers should contain
+    * a pointer to the 'data', rather than having it inline.
+    * Something to look at after we've fully switched over to the pt
+    * paths.
+    */
+   if (emit_header)
+   {
+      /* Need to set header->vertex_id = 0xffff somehow.
+       */
+      key.element[nr].input_format = PIPE_FORMAT_R32_FLOAT;
+      key.element[nr].input_buffer = draw->nr_vertex_buffers;
+      key.element[nr].input_offset = 0;
+      key.element[nr].output_format = PIPE_FORMAT_R32_FLOAT;
+      key.element[nr].output_offset = dst_offset;
+      dst_offset += 1 * sizeof(float);
+      nr++;
+
+
+      /* Just leave the clip[] array untouched.
+       */
+      dst_offset += 4 * sizeof(float);
+   }
+      
+
+   for (i = 0; i < draw->nr_vertex_elements; i++) {
+      key.element[nr].input_format = draw->vertex_element[i].src_format;
+      key.element[nr].input_buffer = draw->vertex_element[i].vertex_buffer_index;
+      key.element[nr].input_offset = draw->vertex_element[i].src_offset;
+      key.element[nr].output_format = PIPE_FORMAT_R32G32B32A32_FLOAT;
+      key.element[nr].output_offset = dst_offset;
+
+      dst_offset += 4 * sizeof(float);
+      nr++;
+   }
+
+   assert(dst_offset <= vertex_size);
+
+   key.nr_elements = nr;
+   key.output_stride = vertex_size;
+
+
+   /* Don't bother with caching at this stage:
+    */
+   if (!fetch->translate ||
+       memcmp(&fetch->translate->key, &key, sizeof(key)) != 0) 
+   {
+      if (fetch->translate)
+	 fetch->translate->release(fetch->translate);
+
+      fetch->translate = translate_generic_create( &key );
+
+      if (emit_header) {
+	 static struct vertex_header vh = { 0, 0, 0, 0xffff };
+	 fetch->translate->set_buffer(fetch->translate, 
+				      draw->nr_vertex_buffers, 
+				      &vh,
+				      0);
+      }
+   }
+}
+
+
+
+
+void draw_pt_fetch_run( struct pt_fetch *fetch,
+			const unsigned *elts,
+			unsigned count,
+			char *verts )
+{
+   struct draw_context *draw = fetch->draw;
+   struct translate *translate = fetch->translate;
+   unsigned i;
+
+   for (i = 0; i < draw->nr_vertex_buffers; i++) {
+      translate->set_buffer(translate, 
+			    i, 
+			    ((char *)draw->user.vbuffer[i] + 
+			     draw->vertex_buffer[i].buffer_offset),
+			    draw->vertex_buffer[i].pitch );
+   }
+
+   translate->run_elts( translate,
+			elts, 
+			count,
+			verts );
+}
+
+
+struct pt_fetch *draw_pt_fetch_create( struct draw_context *draw )
+{
+   struct pt_fetch *fetch = CALLOC_STRUCT(pt_fetch);
+   if (!fetch)
+      return NULL;
+	 
+   fetch->draw = draw;
+   return fetch;
+}
+
+void draw_pt_fetch_destroy( struct pt_fetch *fetch )
+{
+   FREE(fetch);
+}
+