forked from OSchip/llvm-project
cdcbeb4997
reduceBuildVecConvertToConvertBuildVec vectorizes int2float in the DAGCombiner, which means that even if the LV/SLP has decided to keep scalar code using the cost models, this will override this. While there are cases where vectorization is necessary in the DAG (mainly due to legalization artefacts), I don't think this is the case here, we should assume that the vectorizers know what they are doing. Differential Revision: https://reviews.llvm.org/D53712 llvm-svn: 345964 |
||
|---|---|---|
| .. | ||
| arguments-float.ll | ||
| arguments-fp128.ll | ||
| arguments-hard-float-varargs.ll | ||
| arguments-hard-float.ll | ||
| arguments-hard-fp128.ll | ||
| arguments-small-structures-bigger-than-32bits.ll | ||
| arguments-struct.ll | ||
| arguments-varargs-small-structs-byte.ll | ||
| arguments-varargs-small-structs-combinations.ll | ||
| arguments-varargs-small-structs-multiple-args.ll | ||
| arguments-varargs.ll | ||
| arguments.ll | ||
| byval.ll | ||
| callee-saved-float.ll | ||
| callee-saved-fpxx.ll | ||
| callee-saved-fpxx1.ll | ||
| callee-saved.ll | ||
| fmaxl_call.ll | ||
| memory-layout.ll | ||
| pr33883.ll | ||
| reserved-space.ll | ||
| return-float.ll | ||
| return-hard-float.ll | ||
| return-hard-fp128.ll | ||
| return-hard-struct-f128.ll | ||
| return-struct.ll | ||
| return.ll | ||
| roundl-call.ll | ||
| stack-alignment.ll | ||
| vector.ll | ||