Date: Sun, 23 Mar 1997 21:35:29 GMT From: root@ultima.thenerve.com To: FreeBSD-gnats-submit@freebsd.org, dsf@frontiernet.net Subject: bin/3073: cc1 crashes machine after triggering an invalid page Message-ID: <199703232135.VAA01636@ultima.thenerve.com> Resent-Message-ID: <199703240240.SAA20054@freefall.freebsd.org>
next in thread | raw e-mail | index | archive | help
>Number: 3073 >Category: bin >Synopsis: cc1 crashes machine after triggering an invalid page >Confidential: no >Severity: serious >Priority: medium >Responsible: freebsd-bugs >State: open >Class: sw-bug >Submitter-Id: current-users >Arrival-Date: Sun Mar 23 18:40:01 PST 1997 >Last-Modified: >Originator: Dan Foster >Organization: The Nerve Internet Service >Release: FreeBSD 2.2-RELEASE i386 >Environment: Hope you don't mind my dmesg output and kernel config file. :) Mar 23 19:43:12 ultima /kernel: Copyright (c) 1992-1996 FreeBSD Inc. Mar 23 19:43:12 ultima /kernel: Copyright (c) 1982, 1986, 1989, 1991, 1993 Mar 23 19:43:12 ultima /kernel: The Regents of the University of Califor nia. All rights reserved. Mar 23 19:43:12 ultima /kernel: Mar 23 19:43:12 ultima /kernel: FreeBSD 2.2-970224-GAMMA #5: Tue Mar 18 02:32:43 GMT 1997 Mar 23 19:43:12 ultima /kernel: root@ultima.thenerve.com:/usr/src/sys/compil e/THENERVE Mar 23 19:43:12 ultima /kernel: Calibrating clock(s) relative to mc146818A clock ... i586 clock: 166211591 Hz, i8254 clock: 1193315 Hz Mar 23 19:43:12 ultima /kernel: CPU: Pentium (166.19-MHz 586-class CPU) Mar 23 19:43:12 ultima /kernel: Origin = "GenuineIntel" Id = 0x52c Stepping= 12 Mar 23 19:43:12 ultima /kernel: Features=0x1bf<FPU,VME,DE,PSE,TSC,MSR,MCE,CX8> Mar 23 19:43:12 ultima /kernel: real memory = 25165824 (24576K bytes) Mar 23 19:43:12 ultima /kernel: avail memory = 22269952 (21748K bytes) Mar 23 19:43:12 ultima /kernel: Probing for devices on PCI bus 0: Mar 23 19:43:12 ultima /kernel: chip0 <Intel 82437VX PCI cache memory controller > rev 1 on pci0:0 Mar 23 19:43:12 ultima /kernel: chip1 <Intel 82371SB PCI-ISA bridge> rev 1 on pc i0:7:0 Mar 23 19:43:12 ultima /kernel: chip2 <Intel 82371SB IDE interface> rev 0 on pci 0:7:1 Mar 23 19:43:12 ultima /kernel: ahc0 <Adaptec 2940 SCSI host adapter> rev 0 int a irq 11 on pci0:17 Mar 23 19:43:12 ultima /kernel: ahc0: aic7870 Single Channel, SCSI Id=7, 16 SCBs Mar 23 19:43:12 ultima /kernel: ahc0 waiting for scsi devices to settle Mar 23 19:43:12 ultima /kernel: (ahc0:1:0): "HP C1533A 9503" type 1 removable SC SI 2 Mar 23 19:43:12 ultima /kernel: st0(ahc0:1:0): Sequential-Access density code 0x 13, variable blocks, write-enabled Mar 23 19:43:12 ultima /kernel: (ahc0:2:0): "QUANTUM FIREBALL_TM2110S 300X" type 0 fixed SCSI 2 Mar 23 19:43:12 ultima /kernel: sd0(ahc0:2:0): Direct-Access 2014MB (4124736 512 byte sectors) Mar 23 19:43:12 ultima /kernel: sd0(ahc0:2:0): with 6810 cyls, 4 heads, and an a verage 151 sectors/track Mar 23 19:43:12 ultima /kernel: vga0 <VGA-compatible display device> rev 252 on pci0:18 Mar 23 19:43:12 ultima /kernel: de0 <Digital 21041 Ethernet> rev 17 int a irq 10 on pci0:19 Mar 23 19:43:12 ultima /kernel: de0: 21041 [10Mb/s] pass 1.1 Mar 23 19:43:12 ultima /kernel: de0: address 00:00:c0:5d:33:c7 Mar 23 19:43:12 ultima /kernel: Probing for devices on the ISA bus: Mar 23 19:43:12 ultima /kernel: sc0 at 0x60-0x6f irq 1 on motherboard Mar 23 19:43:12 ultima /kernel: sc0: VGA color <16 virtual consoles, flags=0x0> Mar 23 19:43:12 ultima /kernel: sio0 at 0x3f8-0x3ff irq 4 on isa Mar 23 19:43:12 ultima /kernel: sio0: type 16550A Mar 23 19:43:12 ultima /kernel: sio3 not found at 0x2e8 Mar 23 19:43:12 ultima /kernel: lpt0 at 0x378-0x37f irq 7 on isa Mar 23 19:43:12 ultima /kernel: lpt0: Interrupt-driven port Mar 23 19:43:12 ultima /kernel: lp0: TCP/IP capable interface Mar 23 19:43:12 ultima /kernel: fdc0 at 0x3f0-0x3f7 irq 6 drq 2 on isa Mar 23 19:43:12 ultima /kernel: fdc0: NEC 72065B Mar 23 19:43:12 ultima /kernel: fd0: 1.44MB 3.5in Mar 23 19:43:13 ultima /kernel: scd0 not found at 0x230 Mar 23 19:43:13 ultima /kernel: npx0 on motherboard Mar 23 19:43:13 ultima /kernel: npx0: INT 16 interface Mar 23 19:43:13 ultima /kernel: ccd0-3: Concatenated disk drivers Mar 23 19:43:13 ultima /kernel: de0: enabling 10baseT port machine "i386" cpu "I586_CPU" ident THENERVE maxusers 80 options INET options FFS options NFS options MSDOSFS options "CD9660" options PROCFS options "COMPAT_43" options "SCSI_DELAY=5" options SCSIDEBUG options BOUNCE_BUFFERS options UCONSOLE options "CHILD_MAX=128" options "OPEN_MAX=512" options DDB options KTRACE options "NSWAPDEV=5" options QUOTA options SCSI_REPORT_GEOMETRY options SYSVSHM options SYSVSEM options SYSVMSG config kernel root on sd0a controller isa0 controller pci0 controller fdc0 at isa? port "IO_FD1" bio irq 6 drq 2 vector fdintr disk fd0 at fdc0 drive 0 controller ahc0 controller scbus0 device sd0 device st0 device cd0 device scd0 at isa? port 0x230 bio device sc0 at isa? port "IO_KBD" tty irq 1 vector scintr device npx0 at isa? port "IO_NPX" irq 13 vector npxintr device sio0 at isa? port "IO_COM1" tty irq 4 vector siointr device sio3 at isa? port "IO_COM4" tty irq 3 vector siointr device lpt0 at isa? port? tty irq 7 vector lptintr device de0 pseudo-device loop pseudo-device ether pseudo-device bpfilter 4 pseudo-device log pseudo-device sl 1 pseudo-device pty 16 pseudo-device gzip pseudo-device vn pseudo-device snp 3 >Description: This is my first PR submission - so please bear with me if I have any errors in doing this (procedural, technical, etc). Whenever I run a C compile on anything that requires a bunch of files built (or a fairly complex file that includes a bunch of files, etc..), then cc1 either crashes with an error message, saying it got a SIGSEGV if it's being nice. If being its usual nasty self, the machine crashes - the only one time I was at the console (it's 350 miles away in NYC), I saw something about an invalid page - machine rebooted before I could write it down. Not sure if it's cc1 at fault, or if it's tickling something in the kernel, or something else entirely. My cc1.core is worth nothing (to me) because cc1 wasn't built with -g for debugging itself. Is there any way for me to get gcc 2.7.2.1 binaries with -g enabled? I would do it myself, but cc trips me up. ;-) Or can someone make available 2.7.2.2 binaries with or without -g? Problem was not present with 2.1.6-RELEASE and the gcc binaries provided. Problem first showed up with the snapshot in 2.2-GAMMA around Feb. 4th and the gcc binaries. Persisted througout the 2.2-RELEASE upgrade. If I can produce *any* other information, please let me know. I also saved the cc1 core dumps just in case, though. P.S. Please adjust any part of this PR if filled out or set incorrectly and let me know? Many thanks! >How-To-Repeat: cd /usr/src/compile/THENERVE ; make /* wait about 2 minutes or less */ >Fix: do 'sync' two or three times for every couple C files built. Do the same thing if cc1 crashes completely to the shell prompt - do 2 or 3 'sync' then restart the make. A bit of a kludge/workaround, but it's what got me to build this kernel along with a few crashes on the way. >Audit-Trail: >Unformatted:
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199703232135.VAA01636>