https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=219399
--- Comment #91 from Nils Beyer <[email protected]> --- (In reply to Don Lewis from comment #89) really interessting research you've done there; big thanks for that. If the shifting of the shared page really helps getting the system stable, is it still a hardware problem/bug then or more a software problem? If hardware, is that fixable by a microcode update or is the CPU design per-se "incompatible"? What do you think? Anyways, I've applied your patch and started a poudriere build. Based on your experiences, I'm somewhat optimistic now. I'll let you know how's it going here. The "ryzen_segv_test" still fails despite the patch, the only way to circumvent that is to disable SMT. But that's no final option for me... -- You are receiving this mail because: You are the assignee for the bug. _______________________________________________ [email protected] mailing list https://lists.freebsd.org/mailman/listinfo/freebsd-bugs To unsubscribe, send any mail to "[email protected]"
