On 05/26/2016 07:24 PM, Torbjörn Granlund wrote: > Frantisek Kluknavsky <[email protected]> writes: > > on a Pentium G4400, GMP 6.1.0 configured with --enable-fat crashes > because of an illegal instruction: > > > Program received signal SIGILL, Illegal instruction. > 0x00007ffff7501514 in __gmpn_submul_1_coreihwl () > at tmp-coreihwl_submul_1.s:165 > 165 jmp .Llo3 > (gdb) bt full > #0 0x00007ffff7501514 in __gmpn_submul_1_coreihwl () > at tmp-coreihwl_submul_1.s:165 > > > I can not reproduce it myself, it is a report from a different user. > > > http://www.intel.com/content/dam/www/public/us/en/documents/specification-updates/desktop-6th-gen-core-family-spec-update.pdf > says there is a bug in reporting BMI support. I am not sure how exactly > to read the document if the bug affects G4400 and if this is the real > reason of the crash. > > Are you interested in working around hardware bugs? > > We are, to some extent. > > Please ask that someone else if a recent snapshot helps: > > https://gmplib.org/download/snapshot/ > > And then please report back here! >
Hi again, I sent packaged gmp-6.1.99-20160531 to 'the user' and he says it solves all his crashes. Thank you very much. Fero
signature.asc
Description: OpenPGP digital signature
_______________________________________________ gmp-bugs mailing list [email protected] https://gmplib.org/mailman/listinfo/gmp-bugs
