From: Brian Paul <brianp@vmware.com>
Date: Thu, 2 May 2013 01:15:32 +0000 (-0600)
Subject: mesa: fix save_ShadeModel() logic and add new comments
X-Git-Url: https://git.libre-soc.org/?a=commitdiff_plain;h=aa782f260d09bbc10db039340358b3518e8a5a38;p=mesa.git

mesa: fix save_ShadeModel() logic and add new comments

This removes the test for _mesa_inside_dlist_begin_end().
If ctx->Driver.CurrentSavePrimitive==PRIM_UNKNOWN (the initial value),
_mesa_inside_dlist_begin_end() will, confusingly, return TRUE.
So we didn't set the ctx->ListState.Current.ShadeModel value and it
remained in its indeterminate state.

This didn't effect correctness, but it defeated the intended optimization
of dropping redundant glShadeModel() state changes in order to
coalesce sequences of drawing commands.

Verified with new piglit gl-1.0-dlist-shademodel test.

Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
---

diff --git a/src/mesa/main/dlist.c b/src/mesa/main/dlist.c
index bc0c92a278a..3071a37e14b 100644
--- a/src/mesa/main/dlist.c
+++ b/src/mesa/main/dlist.c
@@ -3771,15 +3771,16 @@ save_ShadeModel(GLenum mode)
       CALL_ShadeModel(ctx->Exec, (mode));
    }
 
+   /* Don't compile this call if it's a no-op.
+    * By avoiding this state change we have a better chance of
+    * coalescing subsequent drawing commands into one batch.
+    */
    if (ctx->ListState.Current.ShadeModel == mode)
       return;
 
    SAVE_FLUSH_VERTICES(ctx);
 
-   /* Only save the value if we know the statechange will take effect:
-    */
-   if (!_mesa_inside_dlist_begin_end(ctx))
-      ctx->ListState.Current.ShadeModel = mode;
+   ctx->ListState.Current.ShadeModel = mode;
 
    n = alloc_instruction(ctx, OPCODE_SHADE_MODEL, 1);
    if (n) {