Probelem with 2.6.13 kernel and latest mtd

[Zoltan Sogor]

Hi All!

We've got a kernel warning with the lates mtd cvs and linux kernel 2.6.13 which was not present earlier (even with 2.6.12.5).

We used nandsim for testing (64 MB, 128KB erase block) using the following steps:

flash_eraseall -jq /dev/mtd0c
mount -o ro /dev/mtd0c /mnt/flash
mount -o remount,rw /mnt/flash
umount /mnt/flash

and got a following warning message:
Badness in do_exit at kernel/exit.c:787
 [do_exit+1032/1040] do_exit+0x408/0x410
 [complete_and_exit+19/32] complete_and_exit+0x13/0x20
 [pg0+949936634/1069405184] jffs2_garbage_collect_thread+0xda/0x180 [jffs2]
 [_spin_unlock_irq+14/48] _spin_unlock_irq+0xe/0x30
 [sigprocmask+108/208] sigprocmask+0x6c/0xd0
 [_spin_unlock_irq+14/48] _spin_unlock_irq+0xe/0x30
 [schedule_tail+59/160] schedule_tail+0x3b/0xa0
 [ret_from_fork+6/20] ret_from_fork+0x6/0x14
 [pg0+949936416/1069405184] jffs2_garbage_collect_thread+0x0/0x180 [jffs2]
 [pg0+949936416/1069405184] jffs2_garbage_collect_thread+0x0/0x180 [jffs2]
 [kernel_thread_helper+5/16] kernel_thread_helper+0x5/0x10

There is a new line in the kernel (kernel/exit.c, in function do_exit()):

WARN_ON(atomic_read(&tsk->fs_excl));

Does anybody know what it is about exactly?

Thanks,
Zoltan