From owner-freebsd-questions Wed Aug 7 08:06:04 1996 Return-Path: owner-questions Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id IAA02801 for questions-outgoing; Wed, 7 Aug 1996 08:06:04 -0700 (PDT) Received: from garfield.cs.mun.ca (paul8@garfield.cs.mun.ca [134.153.1.1]) by freefall.freebsd.org (8.7.5/8.7.3) with SMTP id IAA02786 for ; Wed, 7 Aug 1996 08:06:00 -0700 (PDT) Received: (from paul8@localhost) by garfield.cs.mun.ca (8.6.13/8.6.12) id MAA28468; Wed, 7 Aug 1996 12:35:54 -0230 Message-ID: X-Mailer: XFMail 0.4-prerelease [p0] on Unknown Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 8bit MIME-Version: 1.0 In-Reply-To: <199608062134.VAA04676@jraynard.demon.co.uk> Date: Wed, 07 Aug 1996 12:19:08 -0230 (NDT) Organization: Memorial University of NF From: Paul Manuel To: James Raynard Subject: Re: kernel fails to compile with 64MB RAM Cc: questions@freebsd.org Sender: owner-questions@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On 07-Aug-96 James Raynard wrote: >>> >> Is there a known problem with machines have 64MB or more of memory? > >No, except that you need to specify the amount of memory in the kernel >config file if you want more than the first 64MB of RAM to be used. > >> I just purchased two P166 machines with 64MB EDO RAM and both fail with the >> same errors when trying to compile the kernel. >> >> A couple of the errors are: >> >> ------------- >> /kernel: pid 1009 (cc1), uid 0: exited on signal 6 >> cc: Internal compiler error: program cc1 got fatal signal 6 >> ------------- >> >> ------------- >> assertion "(fragP->ft_next == 0) || ((fregP->fr_next-fr_address - fragP->fr_a ddr >> ess) == fragP->fr_fix)" failed: file "/usr/src/gnu/usr.bin/as/write.c", line 354 >> >> /kernel: pid 1058 (as), uid 0: exited on signal 6 >> cc: Internal compiler error: program as got fatal signal 6 >> ------------- > >Hmm. Not sure what I can say here, except that signal 6 is SIGABRT >and an assertion failure normally signifies a "can't happen" situation. >Something is getting really screwed up here. > >> Fatal trap 12: page fault while in kernel mode >> fault virtual address = 0x88f1163d >> fault code = supervisor read, page not present >> instruction pointer = 0x8:0xf018ad9c >> code segment = base 0x0, limit 0xfffff, type 0x1b >> = DPL 0, pres 1, def32 1, gran 1 >> processor eflags = interrupt enabled, resume, IOPL = 0 >> current process = 1277 (make) >> interrupt mask = >> panic: page fault >> >> syncing disks... etc >> ------------- > >Was this with version 2.1.5? If so, do 'nm /kernel -n | more' >and use send-pr to post the area around f018ad9c to the bugs list, >together with the above description and your kernel config file >(if you've customised the kernel) and someone *may* be able to track >down the problem. > >If it was 2.1.0 or earlier, it's probably better to upgrade to 2.1.5, >as quite a few bugs in the VM system (which is what this appears to >be) were fixed in the last release. > Yes it is 2.1.5. I got the kernel to compile with 64MB of memory but I had to turn off the external cache. With 32 and 16 megs of memory it works fine with the external cache on. I have done what you suggested and a bug report is on the way. Thanks for the reply. Paul Manuel Systems Programmer Memorial University of NF St. John's, NF Canada (709) 737-2661 E-Mail: Paul Manuel Date: 08/07/96 Time: 12:35:54