Hi folks-- i can also duplicate this bug on a mixed sarge/sarge-backports machine with a serial console.
I haven't had the chance yet to try the workaround jan wagner has suggested. The errors from the console look like this: ... 00:06: ttyS1 at I/O 0x2f8 (irq = 3) is a 16550A hw_random: RNG not detected pci_hotplug: PCI Hot Plug PCI Core version: 0.5 shpchp: Standard Hot Plug PCI Controller Driver version: 0.4 BUG: unable to handle kernel NULL pointer dereference at virtual address 0000000c printing eip: b02090d4 *pde = 00000000 Oops: 0000 [#1] SMP Modules linked in: shpchp pci_hotplug i2c_i801 psmouse i2c_core 8250_pnp e752x_edac evdev serio_raw edac_mc rtc floppy pcspkr ext3 jbd mbcache dm_mirror dm_snapshot dm_ CPU: 3 EIP: 0060:[<b02090d4>] Not tainted VLI EFLAGS: 00010206 (2.6.17-2-686 #1) EIP is at uart_write_room+0x9/0x16 eax: dfdbfc00 ebx: 00000006 ecx: 00000006 edx: 00000000 esi: dfd87800 edi: dfaedf28 ebp: ef27c800 esp: dfaedecc ds: 007b es: 007b ss: 0068 Process S03udev (pid: 2374, threadinfo=dfaec000 task=dfafca50) Stack: b01fc704 fffffff2 00000000 dfaedf28 00000006 b01fdfb2 ef27c800 ef27c800 ef5c2cc0 dfd87800 00000000 dfafca50 b01193f1 00000000 00000000 b0147b6d 080c2de8 00000012 00000000 dfafca50 b01193f1 dfd87938 dfd87938 ef036a7c Call Trace: <b01fc704> opost_block+0x13/0x107 <b01fdfb2> write_chan+0x118/0x1d7 <b01193f1> default_wake_function+0x0/0xc <b0147b6d> do_wp_page+0x88/0x293 <b01193f1> default_wake_function+0x0/0xc <b01f9d3a> tty_write+0x14b/0x1d8 <b01fde9a> write_chan+0x0/0x1d7 <b01f9de3> redirected_tty_write+0x1c/0x71 <b015607a> vfs_write+0xad/0x14e <b01561b9> sys_write+0x3c/0x62 <b0102b4f> syscall_call+0x7/0xb Code: a9 8b 54 24 04 8b 44 24 08 e8 b7 52 07 00 8b 44 24 10 e8 4b fb ff ff 8b 04 24 83 c4 14 5b 5e 5f 5d c3 8b 80 80 01 00 00 8b 50 10 <8b> 42 0c 2b 42 08 48 25 ff 0f 0 EIP: [<b02090d4>] uart_write_room+0x9/0x16 SS:ESP 0068:dfaedecc The same machine boots fine with a backported 2.6.16-2-686 with a serial console, though that kernel doesn't contain any modules named 8250_pnp. Interestingly, i've also got panic=15 set as a kernel parameter, but it is not triggering a reboot 15 seconds later as i expected it to. Regards, --dkg -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]