diff options
| author | Rafael Espindola <rafael.espindola@gmail.com> | 2012-03-31 18:14:00 +0000 |
|---|---|---|
| committer | Rafael Espindola <rafael.espindola@gmail.com> | 2012-03-31 18:14:00 +0000 |
| commit | 80c540e656c536918b7c3d6966cf745358ce18b3 (patch) | |
| tree | 51cf37feed6062f5d3d73331e1393c806e9c436c /llvm/test | |
| parent | 51861b4855853158b9b35d78d46000af8cc1f500 (diff) | |
| download | bcm5719-llvm-80c540e656c536918b7c3d6966cf745358ce18b3.tar.gz bcm5719-llvm-80c540e656c536918b7c3d6966cf745358ce18b3.zip | |
Teach CodeGen's version of computeMaskedBits to understand the range metadata.
This is the CodeGen equivalent of r153747. I tested that there is not noticeable
performance difference with any combination of -O0/-O2 /-g when compiling
gcc as a single compilation unit.
llvm-svn: 153817
Diffstat (limited to 'llvm/test')
| -rw-r--r-- | llvm/test/CodeGen/X86/pr12360.ll | 14 |
1 files changed, 14 insertions, 0 deletions
diff --git a/llvm/test/CodeGen/X86/pr12360.ll b/llvm/test/CodeGen/X86/pr12360.ll new file mode 100644 index 00000000000..c0af432cc8b --- /dev/null +++ b/llvm/test/CodeGen/X86/pr12360.ll @@ -0,0 +1,14 @@ +; RUN: llc < %s -march=x86-64 | FileCheck %s + +define zeroext i1 @f1(i8* %x) { +entry: + %0 = load i8* %x, align 1, !range !0 + %tobool = trunc i8 %0 to i1 + ret i1 %tobool +} + +; CHECK: f1: +; CHECK: movb (%rdi), %al +; CHECK-NEXT: ret + +!0 = metadata !{i8 0, i8 2} |

