OSDN Git Service

i965/skl: Use 1 register for uniform pull constant payload
authorBen Widawsky <benjamin.widawsky@intel.com>
Tue, 23 Jun 2015 00:18:02 +0000 (17:18 -0700)
committerEmil Velikov <emil.l.velikov@gmail.com>
Wed, 1 Jul 2015 14:22:41 +0000 (15:22 +0100)
commit80e283b0539f284563482e9bf901924ac134c0a5
treef0f43d4c6589e1c3536e55d95f6723af60ba6b8e
parent999b2b27168c85f2895c6574de28c31114fc5b7b
i965/skl: Use 1 register for uniform pull constant payload

When under dispatch_width=16 the previous code would allocate 2 registers for
the payload when only one is needed. This manifested itself through bugs on SKL
which needs to mess with this instruction.

Ken thought this might impact shader-db, but apparently it doesn't

Backported to 10.5.8 from (upstream uses alloc.allocate()):
commit 6e62a52865787362ae1deb9dee80140d3a66c519
Author: Ben Widawsky <benjamin.widawsky@intel.com>
Date:   Thu Feb 19 15:49:34 2015 -0800

    i965/skl: Use 1 register for uniform pull constant payload

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89118
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=88999
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Tested-by: Timo Aaltonen <timo.aaltonen@canonical.com>
Cc: "10.5" <mesa-stable@lists.freedesktop.org>
src/mesa/drivers/dri/i965/brw_fs.cpp