https://gcc.gnu.org/bugzilla/show_bug.cgi?id=127055
--- Comment #20 from Hongtao Liu <liuhongt at gcc dot gnu.org> --- (In reply to Matthias Kretz (Vir) from comment #18) > But spurious FE_INEXACT is not a problem when using vround* or vrndscale* > instructions. > Quoting the Intel docs: "Precision (signaled only if imm[3] = ‘0; if imm[3] > = ‘1, then the Precision Mask in the MXSCSR is ignored and precision > exception is not signaled.)" Yes, I think we can remove !flag_trapping_math from vector expander of ceil/floor/lceil/lfloor/btrunc, so for patch at #c8, please also remove !flag_trapping_mathb from vector 27017:trunc<mode>2.
