BUG: spinlockup in kacpid of 2.6.15-rc6 during boot
Quoted 9 lines · Adrian Bunk
Quoted 3 lines · Larry Finger
I am getting an intermittent "BUG: spinlockup on CPU #0, kacpid/13,
e3bcef9c" message when booting. I know it occurs from a cold boot, but am not sure about from a warm reboot.I don't know where in the boot sequence it occurs as I only have the text display on the screen, which is as follows:
You can set a higher resolution using the vga= boot parameter (see
Documentation/kernel-parameters.txtin the kernel sources).
Thanks - I'll set as high as I can.
Quoted 26 lines
Quoted 26 lines
...... unknown material that rolled off the top of the screen
_spinlock+ 0x1b/0x30exit_io_context+ 0x25/0x90do_exit+ 0x54/0x430
die + 0x17c/0x180do_page_fault+ 0x209/0x62eerror_code+ 0x4f/0x54show_stack+ 0x9c/0x0eshow_registers+ 0x18f/0x230
die + 0xfa/0x180do_page_fault+ 0x209/0x62eBUG: spinlockup on CPU #0, kacpid/13, e3bcef9c
dump_stack+ 0x1e/0x20__spin_lock_debug+ 0xb6/0xf0_raw_spin_lock+ 0x67/0x90_spinlock+ 0x1b/0x30exit_io_context+ 0x25/0x90do_exit+ 0x54/0x430
die + 0x17c/0x180do_page_fault+ 0x209/0x62eerror_code+ 0x4f/0x54show_stack+ 0x9c/0x0eshow_registers+ 0x18f/0x230
die + 0xfa/0x180do_page_fault+ 0x209/0x62e
Quoted 8 lines
Quoted 4 lines
The computer is an HP ze1115 notebook with a mobile K7 processor. The beginning lines of a normal dmesg are:
Linux version 2.6.15-rc6 (root@linux) (gcc version 4.0.2 20050901 (prerelease) (SUSE Linux)) #4 PREEMPT
...IOW, 2.6.15-rc6 does sometimes boot, and sometimes it doesn't boot?
Yes. I have never had it fail two times in a row. Powering off and restarting has fixed it everytime so far.
Quoted 1 line
Can you give information regarding which kernel versions are affected and which aren't, and what influences whether 2.6.15-rc6 does boot or does not boot?
I have not had this problem (yet?) with 2.6.15-rc7 or 2.6.15; however, it occurs so infrequently that I'm not sure I can rule out these later versions. The case I reported involved a dirty reiserfs file system. I had been debugging the bcm43xx driver and the system had frozen during shutdown because the module could not be unloaded. After I powered off for about 1 second, I restarted. The bug dump occurred fairly quickly and I believe it is early in the reboot sequence.
Larry