Date: Tue, 14 Jan 2003 08:30:33 +1100 (EST) From: Bruce Evans <bde@zeta.org.au> To: John Baldwin <jhb@FreeBSD.ORG> Cc: Walter <wjontofs@berkom.de>, <freebsd-hardware@FreeBSD.ORG> Subject: RE: can't boot/install freebsd from cdrom/floppy - serverworks g Message-ID: <20030114080356.J14218-100000@gamplex.bde.org> In-Reply-To: <XFMail.20030113143538.jhb@FreeBSD.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, 13 Jan 2003, John Baldwin wrote: > On 13-Jan-2003 Walter wrote: > > i've problems installing freebsd on System with ServerWorks GC SL -Chipset, > > Pentium 4, LSI Logic MPT (53c1030) SCSI-Controller > > > > Booting with floppy kern.flp from 4.7-release, not even the bootstrap loader > > works, then i tried from 5.0rc3 - i got > > > > fatal trap 10: trace trap while in vm86 mode > > instruction pointer=0xf000:0xf842 > > stack pointer = 0x0:0xff8 > > frame pointer = 0x0:0x0 > > code segment = base 0x0, linit 0x0, type 0x0 > > = DPL 0, pres 0, def32 0, gran 0 > > processor eflags = interupt enabled, vm86 IOPL = 0 > > current process = 0 () > > trap number = 10 > > panic: trace trap > > Uptime: 1s > > > > I thought probably a hardware failure, > > but i tried to install Linux Redhat 8.0, without problems ?! > > > > Any hints ? Try setting hw.hasbrokenint12 in the loader. > You have a buggy BIOS it seems. I'm not sure why you are getting > that kernel trap though. That is a really weird trap. int 0x12 is reported to be broken for some new BIOSes. I think the bug is just that getmemsize() uses vm86_intcall(0x12, ...) before enough is initialized for vm86_intcall(0x12, ...) to work. It's not clear what is needed for it to work, but getmemsize() clearly doesn't do enough in RELENG_4 -- it doesn't map pages into the vm86 page table. Old BIOSes implement int 0x12 by just returning the result of reading the word at 0x40:0x13 or thereabouts. I'm not sure how even that much worked (do we map the first page specially?). Newer BIOSes may well do much more complicated things. Hopefully there is a standard that prevents them arbitrarily using things that don't work in vm86 mode. Rev.1.544 has some too-large changes which attempt to fix this automatically, but it caused panics and was replaced by simpler changes and a non-automatically set loader flag in rev.1.549. Bruce To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20030114080356.J14218-100000>