Date: Wed, 04 Dec 2002 17:22:39 -0800 From: Terry Lambert <tlambert2@mindspring.com> To: John Baldwin <jhb@FreeBSD.org> Cc: smp@FreeBSD.ORG, beemern <beemern@ksu.edu>, Oliver Blasnik <oliver.blasnik@de.tiscali.com> Subject: Re: Intel SE7500CW2 narrowed down... Message-ID: <3DEEAA5F.DDC9536F@mindspring.com> References: <XFMail.20021204152437.jhb@FreeBSD.org>
next in thread | previous in thread | raw e-mail | index | archive | help
John Baldwin wrote: > >> Apparently both Linux and Windows use a different algorithm for starting > >> up AP's on i386 in that (I think) they send broadcast IPI's instead of > >> IPI'ing each CPU individually. > > > > This is why they work on old ASUS Pentium 90 boxes, and FreeBSD > > does not. > > Well, considering FreeBSD started out working on Pentium 90 machines > and has used the same algorithm since it first got SMP support, I > am kind of skeptical that this is the source of the recent breakage > of support for dual Pentiums. I own an ASUS Dual P90 box that was broken by the mpinit changes. > > It's not related to the P4 SE7500CW2-type problems, which are > > incorrect APIC description and chipset programming. > > In which way exactly? The chip isn't programmed to be able to direct interrupts to other than the BP, unless it's in virtual wire mode. You can deduce this, just by knowning the difference between the OS's that do work on the board, and FReeBSD. -- Terry To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3DEEAA5F.DDC9536F>