From: Matt Arsenault <Matthew.Arsenault@amd.com>
Date: Mon, 18 Nov 2013 20:09:37 +0000 (+0000)
Subject: R600/SI: Match addc to S_ADD_U32.
X-Git-Url: http://demsky.eecs.uci.edu/git/?a=commitdiff_plain;h=b2254e603d12be6673d58f098cd384203f964e55;p=oota-llvm.git

R600/SI: Match addc to S_ADD_U32.

The carry always goes to SCC.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@195037 91177308-0d34-0410-b5e6-96231b3b80d8
---

diff --git a/lib/Target/R600/SIInstructions.td b/lib/Target/R600/SIInstructions.td
index 3ac96d8ea82..10126f6da04 100644
--- a/lib/Target/R600/SIInstructions.td
+++ b/lib/Target/R600/SIInstructions.td
@@ -2054,6 +2054,13 @@ def : Pat <
   (EXTRACT_SUBREG $a, sub0)
 >;
 
+// V_ADD_I32_e32/S_ADD_I32 produces carry in VCC/SCC. For the vector
+// case, the sgpr-copies pass will fix this to use the vector version.
+def : Pat <
+  (i32 (addc i32:$src0, i32:$src1)),
+  (S_ADD_I32 $src0, $src1)
+>;
+
 def : Pat <
   (or i64:$a, i64:$b),
   (INSERT_SUBREG