Upon hitting ^c, inotify blocks on mutex_lock
2.6.27.8.
Today, having had gthumb running for a while, I hit ^c expecting it to quit. The result was a process with zero VSZ and RSS, and it sitting in run state. Only possible recovery action seems to be a reboot.
A sysrq-t reveals the following backtrace for gthumb:
gthumb R running 2052 26500 25452
ce631030 00200046 00000005 00000058 ce631290 0d179289 000043e8 c0691a7a
c06f007b f6d299a0 f794c5c8 f6d299a0 e3a9def0 c041b36d c06109d2 c0610eda
f794c5c8 f6d29980 c048aa70 00000001 40000010 f6d29400 de0ae880 f6d29418
Call Trace:
[<c041b36d>] __cond_resched+0x22/0x37
[<c06109d2>] _cond_resched+0x21/0x2a
[<c0610eda>] mutex_lock+0x1a/0x28
[<c048aa70>] inotify_destroy+0x13/0xa9
[<c048af3c>] inotify_release+0x14/0x7c
[<c04688f2>] __fput+0xab/0x158
[<c04663c0>] filp_close+0x4d/0x53
[<c041efca>] put_files_struct+0x5f/0xa7
[<c04201c1>] do_exit+0x1b8/0x6c8
[<c0425636>] __dequeue_signal+0xd/0xce
[<c042072c>] do_group_exit+0x5b/0x83
[<c0427842>] get_signal_to_deliver+0x2e2/0x322
[<c0402aa3>] do_notify_resume+0x62/0x691
[<c05fce6c>] unix_ioctl+0xa6/0xaf
[<c05a0c2d>] sock_ioctl+0x1ad/0x1d1
[<c05a0a80>] sock_ioctl+0x0/0x1d1
[<c0406f9c>] read_tsc+0x6/0x22
[<c04301df>] getnstimeofday+0x32/0xb1
[<c04722ea>] sys_poll+0x3b/0x6e
[<c040384a>] work_notifysig+0x13/0x19Signature · 3 lines
Russell King Linux kernel 2.6 ARM Linux - http://www.arm.linux.org.uk/ maintainer of:
[BUG] 2.6.27.8: Reboot fails after previous (^c on gthumb) bug
Might be related to the previous bug, or may not be. Reboot wedged for several minutes with the following as the last line:
sd 1:0:0:0: [sda] Synchronizing SCSI cache
Signature · 3 lines
Russell King Linux kernel 2.6 ARM Linux - http://www.arm.linux.org.uk/ maintainer of:
Quoted 3 lines · Russell King
Today, having had gthumb running for a while, I hit ^c expecting it to quit. The result was a process with zero VSZ and RSS, and it sitting in run state. Only possible recovery action seems to be a reboot.
Al Viro looked into this, and suggested that 711a49a might fix the problem. I've just tested this on top of 2.6.27.8, and it does indeed appear to fix the problem. Suggest it should be included in the next stable kernel.
Signature · 3 lines
Russell King Linux kernel 2.6 ARM Linux - http://www.arm.linux.org.uk/ maintainer of:
Quoted 9 lines · Russell King
Quoted 3 lines · Russell King
Today, having had gthumb running for a while, I hit ^c expecting it to quit. The result was a process with zero VSZ and RSS, and it sitting in run state. Only possible recovery action seems to be a reboot.
Al Viro looked into this, and suggested that 711a49a might fix the problem. I've just tested this on top of 2.6.27.8, and it does indeed appear to fix the problem. Suggest it should be included in the next stable kernel.
It's already queued up for the 2.6.27.10 release to be happening tomorrow :)
thanks,
greg k-h