Hello. Today, when trying to make this crappy legacy nVidia driver to work, i managed to boot in single-user and i wanted screen(1) too. I was afraid kldload nvidia.ko will panic my box, so i was planning to do it while [most of] disks were mounted read-only. (Yes nvidia.ko v.7xxx paniced it finally) So i created following script, which _alone_ paniced my box. ...it loads snd_emu10k1 here... i really miss color beastie. OK boot -vs ...dmesg (verbose) goes here... ...enter path name or /bin/sh: # fsck -y # mount -at ufs -o ro # tcsh > cat /ramfs.sh #!/bin/sh swapon -a mount -uo rw / mkdir /mnt/ramfs 2>&- 1>&- mount -o ro -at ufs mount -ufo ro -at ufs # just to be sure umount /mnt/ramfs 2>&- 1>&- mdconfig -d -u 1 mdconfig -a -t malloc -s 16m -u 1 bsdlabel -w /dev/md1 newfs -O 1 -m 0 /dev/md1a mount /dev/md1a /mnt/ramfs mount_unionfs /mnt/ramfs /var chmod 700 /tmp/screens/S-root > /ramfs.sh ...newfs output goes here, with GEOM messages... > pkg_info |grep screen screen-4.0.2_1 A multi-screen window manager > uname -a FreeBSD Droid.local 6.0-BETA3 FreeBSD 6.0-BETA3 #3: Sun Sep 0 20:49:18 EEST 2005 root_at_Droid.local:/usr/src/sys/i386/compile/Droid.debug i386 > screen panic: can't fifo/vnode bypass -1 KDB: enter: panic [thread pid 146 tid 100053 ] Stopped ad kdb_enter+0x30: leave db> trace Tracing pid 146 tid 100057 td 0xc1932af0 kdb_enter(c07e4ddb,c085bb00,c07dc744,d52c8964,100) at panic+0x30 panic(c07dc744,ffffffff,c07dc716,11b,c1782d00) at panic+0xd5 fifo_open(d52c89e4,c1ae3dd0,d52c8bc4,c1a833d0,d52c8a10) at fifo_open+0x660 VOP_OPEN_APV(c083bba0,d52c89e4,c1932af0,d52c8a54,1) at VOP_OPEN_APV+0xac union_open(d52c8a54,d52c8a04,d52c8bc4,0,d52c8b10) at union_open+0x8e VOP_OPEN_APV(c1b35240,d52c8a54,267,c1932af0,1) at VOP_OPEN_APV+0xac vn_open_cred(d52c8bc4,d52c8cc4,0,c1782d00,4) at vn_open_cred+0x472 vn_open(d52c8bc4,d52c8cc4,0,4,c08a7018,4) at vn_open+0x33 kern_open(c1932af0,808e220,0,5,0) at kern_open+0xc8 open(c1932af0,d52c8d04,c,422,3) at open+0x36 syscall(bfbf003b,3b,bfbf003b,4,3) at syscall+0x2c0 Xint0x80_syscall() at Xint0x80_syscall+0x1f --- syscall(5, FreeBSD ELF32, open), eip = 0x28172e5f, esp = 0xbfbfe00c, ebp = 0xbfbfe158 --- db> ps pid proc uid ppid pgrp flag stat wmesg wchan cmd 146 c1a4720c 0 145 145 000002 [CPU 0] screen 145 c1a47000 0 73 145 0004002 [SLP0 pause 0xc1a47034][SLP] screen 138 c1921418 0 0 0 0000204 [SLP0 mdwait 0xc1a90000][SLP] md1 73 c1a47418 0 54 73 0004002 [SLP0 pause 0xc1a4744c][SLP] tcsh 54 c1a47624 0 1 54 0004002 [SLP0 wait 0xc1a47624][SLP] sh ...kernel threads here... Panic was copied by hand - i have no serial console; but i can set up one. Pleace advice me if i provided right information (or You need more). The dmesg is on http://mclone.org.ua/~mclone/dmesg.6b3.txt (they're from slightly older kernel; i'll reboot and replace ASAP) Kernel config is here: http://mclone.org.ua/~mclone/Droid.debug Nothing fancy (except verbose load) in /boot/loader.conf, and i have only CPUTYPE?=i686 and WITH_OPT_PPRO=yes in /etc/make.conf [coupled with various ports defines, that is]. -- wbr, |\ _,,,---,,_ dog bless ya! ` Zzz /,`.-'`' -. ;-;;,_ McLone at GMail dot com |,4- ) )-,_. ,\ ( `'-' net- and *BSD admin '---''(_/--' `-'\_) ...sorry for translitReceived on Sun Sep 04 2005 - 17:18:08 UTC
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:38:42 UTC