i965: Bump Ivybridge's fake MRF range to g112-127 instead of g111-126.

When I originally implemented the hack to use GRFs 111+ as fake MRFs, I
did so purely to avoid rewriting all the code that dealt with MRFs.
However, it turns out that a similar hack is actually required.

Newly discovered language in the BSpec indicates that SEND instructions
with EOT set "should" use g112-g127 as their source registers.  Based on
assertions in the simulator, this is actually a requirement on certain
platforms.

Since we're faking MRFs already, we may as well use the officially
sanctioned range.  My guess is that we avoided this issue because we
seldom use m0: URB writes in the new VS backend start at m1, and RT
writes in the new FS backend start at m2.

NOTE: This is a candidate for stable release branches.

Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Eric Anholt <eric@anholt.net>
(cherry picked from commit 5acc7f38d42859db459567d4442c18764a4072e7)

1 files changed, 9 insertions, 1 deletions

diff --git a/src/mesa/drivers/dri/i965/brw_eu_emit.c b/src/mesa/drivers/dri/i965/brw_eu_emit.c
index 1f4afa0edab..d8ea06f0e6a 100644
--- a/src/mesa/drivers/dri/i965/brw_eu_emit.c
+++ b/src/mesa/drivers/dri/i965/brw_eu_emit.c
@@ -84,10 +84,18 @@ gen6_resolve_implied_move(struct brw_compile *p,
 static void
 gen7_convert_mrf_to_grf(struct brw_compile *p, struct brw_reg *reg)
 {
+   /* From the BSpec / ISA Reference / send - [DevIVB+]:
+    * "The send with EOT should use register space R112-R127 for <src>. This is
+    *  to enable loading of a new thread into the same slot while the message
+    *  with EOT for current thread is pending dispatch."
+    *
+    * Since we're pretending to have 16 MRFs anyway, we may as well use the
+    * registers required for messages with EOT.
+    */
    struct intel_context *intel = &p->brw->intel;
    if (intel->gen == 7 && reg->file == BRW_MESSAGE_REGISTER_FILE) {
       reg->file = BRW_GENERAL_REGISTER_FILE;
-      reg->nr += 111;
+      reg->nr += 112;
    }
 }