[PATCH] D21235: [x86, SSE] change patterns for CMPP to float types to allow matching with SSE1 (PR28044)
Simon Pilgrim via llvm-commits
llvm-commits at lists.llvm.org
Fri Jun 10 13:27:48 PDT 2016
RKSimon added a comment.
Is already a number of SSE1/SSE2+ test cases covering all the float comparisons? All I know of is vector-compare-results.ll
================
Comment at: lib/Target/X86/X86ISelLowering.cpp:15146
@@ -15137,2 +15145,3 @@
if (SetCCOpcode == ISD::SETUEQ) {
- CC0 = 3; CC1 = 0; CombineOpc = ISD::OR;
+ CC0 = 3; CC1 = 0;
+ CombineOpc = Opc == X86ISD::CMPP ? X86ISD::FOR : ISD::OR;
----------------
Please can you place comments near each SSECC hardcoded constant explaining what they are?
http://reviews.llvm.org/D21235
More information about the llvm-commits
mailing list