[PATCH] D88783: [InstCombine] matchFunnelShift - fold or(shl(a,x),lshr(b,sub(bw,x))) -> fshl(a,b,x) iff x < bw

[Martin Storsjö via Phabricator via llvm-commits]

mstorsjo added a comment.

This broke armv7 (tested with both armv7-w64-mingw32 and armv7-linux-gnueabihf) builds of compiler-rt's `__udivmoddi4` function; after this commit, it miscalculates `__udivmoddi4(883547321287490176, 128)`.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D88783/new/

https://reviews.llvm.org/D88783