Date: Tue, 20 Oct 1998 20:02:31 -0700 From: Julian Elischer <julian@whistle.com> To: sos@FreeBSD.ORG Cc: current@FreeBSD.ORG Subject: Re: softupdates/smp Message-ID: <362D4EC7.167EB0E7@whistle.com> References: <199810180918.LAA12711@sos.freebsd.dk>
index | next in thread | previous in thread | raw e-mail
S?ren Schmidt wrote: > > In reply to Julian Elischer who wrote: > > Well I'm doing it, but I've 'left off' development until the > > ELF/CAM/3.0/etc changes all settle down. > > > > looks to me as if someone has broken it again. > > I'll be in europe for a fortnight. When I get back > > I hope things will have settled down enough for me to look at it again. > > Erhm, I've never seen it work reliably under SMP, did you ever get the > giant lock prims into the syncer ?? > > -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- > Søren Schmidt (sos@FreeBSD.org) FreeBSD Core Team > Even more code to hack -- will it ever end? > .. As the syncer is a kernel thread and always in the kernel, it can only be scheduled if it already has the lock. At least that is my understanding. I have no proof of this however. as I haven't looked at the 'giant lock' code.. If this is not true we have other problems. Who knows about the "GL" code? julian To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the messagehelp
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?362D4EC7.167EB0E7>