llvm.org GIT mirror llvm / bfc0134
[X86] Remove alignment from a load in the f16c intrinsic test. The alignment shouldn't be required for load folding. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@317545 91177308-0d34-0410-b5e6-96231b3b80d8 Craig Topper 3 years ago
1 changed file(s) with 1 addition(s) and 1 deletion(s). Raw diff Collapse all Expand all
7676 ; X64-AVX512VL: # BB#0:
7777 ; X64-AVX512VL-NEXT: vcvtph2ps (%rdi), %ymm0 # EVEX TO VEX Compression encoding: [0xc4,0xe2,0x7d,0x13,0x07]
7878 ; X64-AVX512VL-NEXT: retq # encoding: [0xc3]
79 %load = load <8 x i16>, <8 x i16>* %a, align 16
79 %load = load <8 x i16>, <8 x i16>* %a
8080 %res = tail call <8 x float> @llvm.x86.vcvtph2ps.256(<8 x i16> %load)
8181 ret <8 x float> %res
8282 }