From 0ef701e6ae816b0360e0a66e8b815ff875fe2a32 Mon Sep 17 00:00:00 2001
From: Dan Gohman <gohman@apple.com>
Date: Thu, 4 Mar 2010 19:58:08 +0000
Subject: [PATCH] Fix recognition of 16-bit bswap for C front-ends which emit
 the clobber registers in a different order.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97741 91177308-0d34-0410-b5e6-96231b3b80d8
---
 lib/Target/X86/X86ISelLowering.cpp   | 15 +++++--
 test/CodeGen/X86/bswap-inline-asm.ll | 67 +++++++++++++++++++++++++++-
 2 files changed, 77 insertions(+), 5 deletions(-)

diff --git a/lib/Target/X86/X86ISelLowering.cpp b/lib/Target/X86/X86ISelLowering.cpp
index e2b81930aec..7ca6f70d139 100644
--- a/lib/Target/X86/X86ISelLowering.cpp
+++ b/lib/Target/X86/X86ISelLowering.cpp
@@ -9838,11 +9838,20 @@ bool X86TargetLowering::ExpandInlineAsm(CallInst *CI) const {
     // rorw $$8, ${0:w}  -->  llvm.bswap.i16
     if (CI->getType()->isIntegerTy(16) &&
         AsmPieces.size() == 3 &&
-        AsmPieces[0] == "rorw" &&
+        (AsmPieces[0] == "rorw" || AsmPieces[0] == "rolw") &&
         AsmPieces[1] == "$$8," &&
         AsmPieces[2] == "${0:w}" &&
-        IA->getConstraintString() == "=r,0,~{dirflag},~{fpsr},~{flags},~{cc}") {
-      return LowerToBSwap(CI);
+        IA->getConstraintString().compare(0, 5, "=r,0,") == 0) {
+      AsmPieces.clear();
+      SplitString(IA->getConstraintString().substr(5), AsmPieces, ",");
+      std::sort(AsmPieces.begin(), AsmPieces.end());
+      if (AsmPieces.size() == 4 &&
+          AsmPieces[0] == "~{cc}" &&
+          AsmPieces[1] == "~{dirflag}" &&
+          AsmPieces[2] == "~{flags}" &&
+          AsmPieces[3] == "~{fpsr}") {
+        return LowerToBSwap(CI);
+      }
     }
     break;
   case 3:
diff --git a/test/CodeGen/X86/bswap-inline-asm.ll b/test/CodeGen/X86/bswap-inline-asm.ll
index 5bf58fa1d50..2b7019371a1 100644
--- a/test/CodeGen/X86/bswap-inline-asm.ll
+++ b/test/CodeGen/X86/bswap-inline-asm.ll
@@ -1,17 +1,80 @@
 ; RUN: llc < %s -march=x86-64 > %t
 ; RUN: not grep APP %t
-; RUN: grep bswapq %t | count 2
-; RUN: grep bswapl %t | count 1
+; RUN: FileCheck %s < %t
 
+; CHECK: foo:
+; CHECK: bswapq
 define i64 @foo(i64 %x) nounwind {
 	%asmtmp = tail call i64 asm "bswap $0", "=r,0,~{dirflag},~{fpsr},~{flags}"(i64 %x) nounwind
 	ret i64 %asmtmp
 }
+
+; CHECK: bar:
+; CHECK: bswapq
 define i64 @bar(i64 %x) nounwind {
 	%asmtmp = tail call i64 asm "bswapq ${0:q}", "=r,0,~{dirflag},~{fpsr},~{flags}"(i64 %x) nounwind
 	ret i64 %asmtmp
 }
+
+; CHECK: pen:
+; CHECK: bswapl
 define i32 @pen(i32 %x) nounwind {
 	%asmtmp = tail call i32 asm "bswapl ${0:q}", "=r,0,~{dirflag},~{fpsr},~{flags}"(i32 %x) nounwind
 	ret i32 %asmtmp
 }
+
+; CHECK: s16:
+; CHECK: rolw    $8,
+define zeroext i16 @s16(i16 zeroext %x) nounwind {
+  %asmtmp = tail call i16 asm "rorw $$8, ${0:w}", "=r,0,~{dirflag},~{fpsr},~{flags},~{cc}"(i16 %x) nounwind
+  ret i16 %asmtmp
+}
+
+; CHECK: t16:
+; CHECK: rolw    $8,
+define zeroext i16 @t16(i16 zeroext %x) nounwind {
+  %asmtmp = tail call i16 asm "rorw $$8, ${0:w}", "=r,0,~{cc},~{dirflag},~{fpsr},~{flags}"(i16 %x) nounwind
+  ret i16 %asmtmp
+}
+
+; CHECK: u16:
+; CHECK: rolw    $8,
+define zeroext i16 @u16(i16 zeroext %x) nounwind {
+  %asmtmp = tail call i16 asm "rolw $$8, ${0:w}", "=r,0,~{dirflag},~{fpsr},~{flags},~{cc}"(i16 %x) nounwind
+  ret i16 %asmtmp
+}
+
+; CHECK: v16:
+; CHECK: rolw    $8,
+define zeroext i16 @v16(i16 zeroext %x) nounwind {
+  %asmtmp = tail call i16 asm "rolw $$8, ${0:w}", "=r,0,~{cc},~{dirflag},~{fpsr},~{flags}"(i16 %x) nounwind
+  ret i16 %asmtmp
+}
+
+; CHECK: s32:
+; CHECK: bswapl
+define i32 @s32(i32 %x) nounwind {
+  %asmtmp = tail call i32 asm "bswap $0", "=r,0,~{dirflag},~{fpsr},~{flags}"(i32 %x) nounwind
+  ret i32 %asmtmp
+}
+
+; CHECK: t32:
+; CHECK: bswapl
+define i32 @t32(i32 %x) nounwind {
+  %asmtmp = tail call i32 asm "bswap $0", "=r,0,~{dirflag},~{flags},~{fpsr}"(i32 %x) nounwind
+  ret i32 %asmtmp
+}
+
+; CHECK: s64:
+; CHECK: bswapq
+define i64 @s64(i64 %x) nounwind {
+  %asmtmp = tail call i64 asm "bswap ${0:q}", "=r,0,~{dirflag},~{fpsr},~{flags}"(i64 %x) nounwind
+  ret i64 %asmtmp
+}
+
+; CHECK: t64:
+; CHECK: bswapq
+define i64 @t64(i64 %x) nounwind {
+  %asmtmp = tail call i64 asm "bswap ${0:q}", "=r,0,~{fpsr},~{dirflag},~{flags}"(i64 %x) nounwind
+  ret i64 %asmtmp
+}
-- 
2.34.1