[InstCombine] fix nsz (fast-math) propagation from fneg-of-select
As discussed in the post-commit comments for: 3cdd05e5 It seems to be safe to propagate all flags from the final fneg except for 'nsz' to the new select: https://alive2.llvm.org/ce/z/J_APDc nsz has unique FMF semantics: it is not poison, it is only "insignificant" in the calculation according to the LangRef.
Loading
Please sign in to comment