[PATCH][x86] Add more patterns for SSE/AVX scalar single/double-precision fp arithmetic instructions.
nrotem at apple.com
Wed Dec 11 08:54:35 PST 2013
The patterns look okay but I think that they may be fragile. Small changes in the IR passes or legalization may generate sequences that are not matched. I mention this because it looks like this shuffle should be canonicalized into a select instruction (in InstCombine), because what it does is to blend two vectors:
%2 = shufflevector <4 x float> %1, <4 x float> %a, <4 x i32> <i32 0, i32 5, i32 6, i32 7>
Can you please make sure that after canonicalizing the shuffle->blend your patterns still work?
On Dec 11, 2013, at 5:11 AM, Andrea Di Biagio <andrea.dibiagio at gmail.com> wrote:
More information about the llvm-commits