[PATCH] D41794: [X86] Improve AVX1 shuffle lowering for v8f32 shuffles where the low half comes from V1 and the high half comes from V2 and the halves do the same operation
Simon Pilgrim via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Sun Jan 14 13:29:02 PST 2018
RKSimon added a comment.
This looks like there is a lot of crossover with lowerVectorShuffleByMerging128BitLanes?
================
Comment at: lib/Target/X86/X86ISelLowering.cpp:10761
+ int Size = Mask.size();
+ SmallVector<int, 8> RepeatMask(Size, -1);
+
----------------
Should't RepeatMask be just LaneSize wide?
================
Comment at: lib/Target/X86/X86ISelLowering.cpp:10791
+ PermuteMask[i] = M;
+ if (PermuteMask[i] < 0)
+ continue;
----------------
Do we gain anything by relaxing this and keeping PermuteMask[i] as UNDEF if the original Mask[i] was UNDEF?
Repository:
rL LLVM
https://reviews.llvm.org/D41794
More information about the llvm-commits
mailing list