https://sourceware.org/bugzilla/show_bug.cgi?id=32591
--- Comment #8 from H.J. Lu <hjl.tools at gmail dot com> --- (In reply to Jan Beulich from comment #7) > You say "estimate", and since you exclude linker created sections it looks > like it very much is just an estimate. While for the specific case of Xen > this may be sufficient (see below though), in the general case it won't > afaict. > > From looking over the patch I also didn't get the impression that this would > allow to fall back to the MOV -> LEA transformation when the MOV-from-GOT -> > MOV-imm one would result in an overflow. Did I overlook anything? My patch doesn't support fallback which is very intrusive with few benefits. My patch should cover most cases. -- You are receiving this mail because: You are on the CC list for the bug.
