From 6a2f9b91379140c36a11ade6c0673bd7490eba32 Mon Sep 17 00:00:00 2001
From: Tom Stellard <thomas.stellard@amd.com>
Date: Wed, 23 Oct 2013 00:44:19 +0000
Subject: [PATCH] R600/SI: Add support for i64 bitwise or

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@193213 91177308-0d34-0410-b5e6-96231b3b80d8
---
 lib/Target/R600/SIInstructions.td | 19 +++++++++++++++++++
 test/CodeGen/R600/or.ll           | 21 +++++++++++++++++----
 2 files changed, 36 insertions(+), 4 deletions(-)

diff --git a/lib/Target/R600/SIInstructions.td b/lib/Target/R600/SIInstructions.td
index 08a8b56fdc0..ab6193fa4cc 100644
--- a/lib/Target/R600/SIInstructions.td
+++ b/lib/Target/R600/SIInstructions.td
@@ -1963,6 +1963,25 @@ def : Pat<
 >;
 
 //============================================================================//
+// Miscellaneous Patterns
+//===----------------------------------------------------------------------===//
+
+def : Pat <
+  (i64 (trunc i128:$x)),
+  (INSERT_SUBREG (INSERT_SUBREG (i64 (IMPLICIT_DEF)),
+    (i32 (EXTRACT_SUBREG $x, sub0)), sub0),
+    (i32 (EXTRACT_SUBREG $x, sub1)), sub1)
+>;
+
+def : Pat <
+  (or i64:$a, i64:$b),
+  (INSERT_SUBREG
+    (INSERT_SUBREG (IMPLICIT_DEF),
+      (V_OR_B32_e32 (EXTRACT_SUBREG $a, sub0), (EXTRACT_SUBREG $b, sub0)), sub0),
+    (V_OR_B32_e32 (EXTRACT_SUBREG $a, sub1), (EXTRACT_SUBREG $b, sub1)), sub1)
+>;
+
+//============================================================================//
 // Miscellaneous Optimization Patterns
 //============================================================================//
 
diff --git a/test/CodeGen/R600/or.ll b/test/CodeGen/R600/or.ll
index 3db892a3624..6950ed0dfb5 100644
--- a/test/CodeGen/R600/or.ll
+++ b/test/CodeGen/R600/or.ll
@@ -1,11 +1,11 @@
 ;RUN: llc < %s -march=r600 -mcpu=redwood | FileCheck --check-prefix=EG-CHECK %s
 ;RUN: llc < %s -march=r600 -mcpu=verde -verify-machineinstrs | FileCheck --check-prefix=SI-CHECK %s
 
-; EG-CHECK: @or_v2i32
+; EG-CHECK-LABEL: @or_v2i32
 ; EG-CHECK: OR_INT {{\*? *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
 ; EG-CHECK: OR_INT {{\*? *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
 
-;SI-CHECK: @or_v2i32
+;SI-CHECK-LABEL: @or_v2i32
 ;SI-CHECK: V_OR_B32_e32 VGPR{{[0-9]+, VGPR[0-9]+, VGPR[0-9]+}}
 ;SI-CHECK: V_OR_B32_e32 VGPR{{[0-9]+, VGPR[0-9]+, VGPR[0-9]+}}
 
@@ -18,13 +18,13 @@ define void @or_v2i32(<2 x i32> addrspace(1)* %out, <2 x i32> addrspace(1)* %in)
   ret void
 }
 
-; EG-CHECK: @or_v4i32
+; EG-CHECK-LABEL: @or_v4i32
 ; EG-CHECK: OR_INT {{\*? *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
 ; EG-CHECK: OR_INT {{\*? *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
 ; EG-CHECK: OR_INT {{\*? *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
 ; EG-CHECK: OR_INT {{\*? *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
 
-;SI-CHECK: @or_v4i32
+;SI-CHECK-LABEL: @or_v4i32
 ;SI-CHECK: V_OR_B32_e32 VGPR{{[0-9]+, VGPR[0-9]+, VGPR[0-9]+}}
 ;SI-CHECK: V_OR_B32_e32 VGPR{{[0-9]+, VGPR[0-9]+, VGPR[0-9]+}}
 ;SI-CHECK: V_OR_B32_e32 VGPR{{[0-9]+, VGPR[0-9]+, VGPR[0-9]+}}
@@ -38,3 +38,16 @@ define void @or_v4i32(<4 x i32> addrspace(1)* %out, <4 x i32> addrspace(1)* %in)
   store <4 x i32> %result, <4 x i32> addrspace(1)* %out
   ret void
 }
+
+; EG-CHECK-LABEL: @or_i64
+; EG-CHECK-DAG: OR_INT * T{{[0-9]\.[XYZW]}}, KC0[2].W, KC0[3].Y
+; EG-CHECK-DAG: OR_INT * T{{[0-9]\.[XYZW]}}, KC0[2].Z, KC0[3].X
+; SI-CHECK-LABEL: @or_i64
+; SI-CHECK: V_OR_B32_e32 VGPR{{[0-9]}}
+; SI-CHECK: V_OR_B32_e32 VGPR{{[0-9]}}
+define void @or_i64(i64 addrspace(1)* %out, i64 %a, i64 %b) {
+entry:
+	%0 = or i64 %a, %b
+	store i64 %0, i64 addrspace(1)* %out
+	ret void
+}
-- 
2.11.0