diff options
| author | Dan Gohman <gohman@apple.com> | 2009-06-04 22:49:04 +0000 |
|---|---|---|
| committer | Dan Gohman <gohman@apple.com> | 2009-06-04 22:49:04 +0000 |
| commit | a5b9645c4b7a1d8be5e41081b99f27b49b8aa8cf (patch) | |
| tree | 7f5a9f6633be0c4e77a1bb00e5bfcfcca14e219b /llvm/test/CodeGen/X86/2009-01-16-UIntToFP.ll | |
| parent | 72a4d2fec138ad6d2becbc69c6d034246a056d09 (diff) | |
| download | bcm5719-llvm-a5b9645c4b7a1d8be5e41081b99f27b49b8aa8cf.tar.gz bcm5719-llvm-a5b9645c4b7a1d8be5e41081b99f27b49b8aa8cf.zip | |
Split the Add, Sub, and Mul instruction opcodes into separate
integer and floating-point opcodes, introducing
FAdd, FSub, and FMul.
For now, the AsmParser, BitcodeReader, and IRBuilder all preserve
backwards compatability, and the Core LLVM APIs preserve backwards
compatibility for IR producers. Most front-ends won't need to change
immediately.
This implements the first step of the plan outlined here:
http://nondot.org/sabre/LLVMNotes/IntegerOverflow.txt
llvm-svn: 72897
Diffstat (limited to 'llvm/test/CodeGen/X86/2009-01-16-UIntToFP.ll')
| -rw-r--r-- | llvm/test/CodeGen/X86/2009-01-16-UIntToFP.ll | 4 |
1 files changed, 2 insertions, 2 deletions
diff --git a/llvm/test/CodeGen/X86/2009-01-16-UIntToFP.ll b/llvm/test/CodeGen/X86/2009-01-16-UIntToFP.ll index 6de11c94794..340608af35a 100644 --- a/llvm/test/CodeGen/X86/2009-01-16-UIntToFP.ll +++ b/llvm/test/CodeGen/X86/2009-01-16-UIntToFP.ll @@ -22,10 +22,10 @@ bb2: ; preds = %bb1, %bb, %entry %5 = lshr i64 %u_addr.0, 32 ; <i64> [#uses=1] %6 = trunc i64 %5 to i32 ; <i32> [#uses=1] %7 = uitofp i32 %6 to double ; <double> [#uses=1] - %8 = mul double %7, 0x41F0000000000000 ; <double> [#uses=1] + %8 = fmul double %7, 0x41F0000000000000 ; <double> [#uses=1] %9 = trunc i64 %u_addr.0 to i32 ; <i32> [#uses=1] %10 = uitofp i32 %9 to double ; <double> [#uses=1] - %11 = add double %10, %8 ; <double> [#uses=1] + %11 = fadd double %10, %8 ; <double> [#uses=1] %12 = fptrunc double %11 to float ; <float> [#uses=1] ret float %12 } |
