[PATCH] Allow i965 compositing to target A8 buffers

[Carl Worth]

Without this, software fallbacks were being forced for glyph
rendering. This change appears to eliminate the fallbacks,
(but slows glyph rendering down overall, presumably due to
slow compositing currently in the i965 driver).
---
 src/i965_render.c |    9 +--------
 1 files changed, 1 insertions(+), 8 deletions(-)

 But this does completely eliminate software fallbacks from i965 text
 rendering. So if we can fix that slow, synchronous compositing, I
 think we'll finally get some decent text rendering performance.

 More details (and profiles) here:

	http://cworth.org/exa/i965/eliminating_glyph_fallbacks/

-Carl

diff --git a/src/i965_render.c b/src/i965_render.c
index 744501a..c528813 100644
--- a/src/i965_render.c
+++ b/src/i965_render.c
@@ -161,16 +161,9 @@ static Bool i965_get_dest_format(PicturePtr pDstPicture, CARD32 *dst_format)
     case PICT_x1r5g5b5:
         *dst_format = BRW_SURFACEFORMAT_B5G5R5X1_UNORM;
         break;
-    /* COLR_BUF_8BIT is special for YUV surfaces.  While we may end up being
-     * able to use it depending on how the hardware implements it, disable it
-     * for now while we don't know what exactly it does (what channel does it
-     * read from?
-     */
-    /*
     case PICT_a8:
-        *dst_format = COLR_BUF_8BIT;
+        *dst_format = BRW_SURFACEFORMAT_A8_UNORM;
         break;
-    */
     case PICT_a4r4g4b4:
     case PICT_x4r4g4b4:
 	*dst_format = BRW_SURFACEFORMAT_B4G4R4A4_UNORM;
--
1.5.2.4

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: <http://lists.x.org/archives/xorg/attachments/20070807/194603ce/attachment.pgp>