Non-GEM allocations incorrectly force TILE_NONE (bug 20797)

With -intel 2.6.3 performance was very bad when using a non gem enabled kernel
(2.6.27) and EXA. For example sauerbraten ran with 4 fps and screensaver GLBlur
with 1 fps. With -intel 2.6.1 performance was good using the same kernel.

Git bisecting led me to commit f1ed73c1ef3e3daa9f695194dcc813167cbcb53d (in 2.6
branch) "Make i830_allocate_memory take tiling parameters" as first bad commit.

Using gdb I found tiling was set exactly the same in 2.6.3 as in 2.6.1, so that
was good (TILE_XMAJOR for front, back and depth buffers).
Looking further I found the line mem->tiling = TILE_NONE; (line 961 in
src/i830_memory.c) at the end of i830_allocate_memory suspicious, as
mem->tiling now already gets set via i830_allocate_aperture and some buffers do
have tiling. Removing that line indeed fixed the performance issue. Now
sauerbraten runs with 30+ fps and GLBlur runs smoothly.
(cherry picked from commit e964d4e53af3a47de6d09c884be1cc0044d03bea)

1 files changed, 0 insertions, 2 deletions

diff --git a/src/i830_memory.c b/src/i830_memory.c
index 052d906e..e3314c53 100644
--- a/src/i830_memory.c
+++ b/src/i830_memory.c
@@ -958,8 +958,6 @@ i830_allocate_memory(ScrnInfoPtr pScrn, const char *name,
 	}
     }
 
-    mem->tiling = TILE_NONE;
-
     return mem;
 }