Date: Mon, 26 Oct 1998 08:37:05 +0000 (GMT) From: Doug Rabson <dfr@nlsystems.com> To: Mike Smith <mike@smith.net.au> Cc: "'freebsd-alpha@FreeBSD.ORG'" <freebsd-alpha@FreeBSD.ORG> Subject: Re: PCI mapping error Message-ID: <Pine.BSF.4.01.9810260834430.366-100000@herring.nlsystems.com> In-Reply-To: <199810252014.MAA01058@dingo.cdrom.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, 25 Oct 1998, Mike Smith wrote: > > On Sun, 25 Oct 1998, Paul Richards wrote: > > > > > I've just committed a fix for the problem with interrupts not working on > > > the Multia (and any other architecture using the Intel 82738). > > > > > > The mask used to enable/disable interrupts 8 to 15 wasn't correct. Since > > > all interrupts get disabled initially, any device on interrupts > 7 was > > > never having its irq enabled. > > > > > > My ethernet card now works fine :-) > > > > Hehe. I just sent some mail with a suggestion on how to debug it but you > > are way ahead of me :-). > > 8) Now for the next question; does anyone know if it's "normal" for > the world build to sometimes fall over in the middle with a bogus > pointer free in make? > > I figured it probably wasn't, and that this poor little box was just > overheating, but if this is a known heisenbug then I'll warm it up and > try again... I believe that this is caused by make's misuse of execvp after vfork. This normally works but a while ago, malloc was using the wrong value for pagesize on the alpha and this caused problems. I fixed malloc but left the vfork alone; perhaps there are still problems (its in compat.c afaik). -- Doug Rabson Mail: dfr@nlsystems.com Nonlinear Systems Ltd. Phone: +44 181 951 1891 Fax: +44 181 381 1039 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.4.01.9810260834430.366-100000>