https://bugs.kde.org/show_bug.cgi?id=385411
--- Comment #9 from Andreas Arnez <ar...@linux.ibm.com> --- For the record, Julian Seward commented the following in IRC: * Regarding the fused multiply-add/subs: "I think the *right* fix here is to create new Iops, Iop_MAddF64x2 and IOpMSubF64x2 and use those instead (see libvex_ir.h, Iop_MAddF64 for description) or (ugly slow hack) in s390_vector_fp_mulAddOrSub_operation, for the vector case, split up the each operand into 2 64-bit scalars, use the existing scalar FMA operations, and reconstruct the vector (it will generate worse code, but it is less effort to implement because you don't need to change any code-generator stuff)." * About formal aspects: "...also (1) please get it inside 80 cols [as you requested] and (2) I think it would be clearer to remove the LIKELY/UNLIKELY hints in the new code in guest_s390_toIR.c." -- You are receiving this mail because: You are watching all bug changes.
