Oh OK!! I've been looking at the Core2 file. Sorry. In fact that instruction is used in the K10 and atom files.
However, a) The K10 file passes the try test code on Selmer, which is a K10 b) If the data was not aligned, a segfault would occur, not a wrong answer. Bill. 2009/12/11 Bill Hart <goodwillh...@googlemail.com>: > Right, but the try test would fail in that case if it wasn't handled > properly, as it definitely tests alignments on alternate limbs. I > would consider it a bug if MPIR used this instruction on data which > was not aligned. It should have special code to handle the unaligned > case, or use the MOVDQU instruction instead. > > Bill. > > 2009/12/11 Jason Martin <jason.worth.mar...@gmail.com>: >> I believe that the alignment needs to be 16 bytes, not just 8 bytes >> for the MOVDQA instruction. >> >> Jason Worth Martin >> Asst. Professor of Mathematics >> http://www.math.jmu.edu/~martin >> >> >> >> On Fri, Dec 11, 2009 at 9:31 AM, Cactus <rieman...@googlemail.com> wrote: >>> >>> >>> On Dec 11, 2:11 pm, Bill Hart <goodwillh...@googlemail.com> wrote: >>>> omalloc is easy to fix. Just configure it with --with-align=8 on 64 >>>> bit systems. There are other configure options for how alignment is >>>> done. >>> >>> In fact it should be configured this way by the build system on 64-bit >>> systems: >>> >>> #ifndef OM_ALIGN_8 >>> /* define if you want your memory to be 8-aligned */ >>> #define OM_ALIGN_8 1 >>> >>> so Dan's problem is probably a build configuration error of some >>> kind. >>> >>> Brian >>> >>> -- >>> >>> You received this message because you are subscribed to the Google Groups >>> "mpir-devel" group. >>> To post to this group, send email to mpir-de...@googlegroups.com. >>> To unsubscribe from this group, send email to >>> mpir-devel+unsubscr...@googlegroups.com. >>> For more options, visit this group at >>> http://groups.google.com/group/mpir-devel?hl=en. >>> >>> >>> >> >> -- >> >> You received this message because you are subscribed to the Google Groups >> "mpir-devel" group. >> To post to this group, send email to mpir-de...@googlegroups.com. >> To unsubscribe from this group, send email to >> mpir-devel+unsubscr...@googlegroups.com. >> For more options, visit this group at >> http://groups.google.com/group/mpir-devel?hl=en. >> >> >> > -- You received this message because you are subscribed to the Google Groups "mpir-devel" group. To post to this group, send email to mpir-de...@googlegroups.com. To unsubscribe from this group, send email to mpir-devel+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/mpir-devel?hl=en.
