Re: Bug in 2.6.8-rc3 at mm/page_alloc.c:792 and mm/rmap.c:407

[Andrew Morton]

Oskar Berggren <beo@xxxxxxxx> wrote:
>
> I think I've managed to catch one of the missing lines that you
> mention in todays messages file, and then even more in the
> /var/log/syslog file:
> 
> Aug  9 13:07:59 otukt kernel:  <0>Bad page state at free_hot_cold_page
> (in process 'events/0', page c1090200)
> Aug  9 13:07:59 otukt kernel: flags:0x20000080 mapping:00000000
> mapcount:0 count:0
> Aug  9 13:07:59 otukt kernel: Backtrace:
> Aug  9 13:07:59 otukt kernel:  [bad_page+109/153] bad_page+0x6d/0x99
> Aug  9 13:07:59 otukt kernel:  [free_hot_cold_page+81/270]
> free_hot_cold_page+0x51/0x10e
> Aug  9 13:07:59 otukt kernel:  [sk_free+191/258] sk_free+0xbf/0x102
> Aug  9 13:07:59 otukt kernel:  [sk_common_release+87/203]
> sk_common_release+0x57/0xcb
> Aug  9 13:07:59 otukt kernel:  [inet_release+82/96]
> inet_release+0x52/0x60
> Aug  9 13:07:59 otukt kernel:  [sock_release+149/225]
> sock_release+0x95/0xe1
> Aug  9 13:07:59 otukt kernel:  [__crc_bio_get_nr_vecs+4158660/6507043]
> xprt_socket_autoclose+0x26/0x63 [sunrpc]
> Aug  9 13:07:59 otukt kernel:  [worker_thread+464/655]
> worker_thread+0x1d0/0x28f
> Aug  9 13:07:59 otukt kernel:  [__crc_bio_get_nr_vecs+4158622/6507043]
> xprt_socket_autoclose+0x0/0x63 [sunrpc]

hm.  There was a page double-freeing bug in the nfs/networking code
which Dave Miller fixed just a few days ago.  I'd suggest that you retest
using the latest tree from ftp://ftp.kernel.org/pub/linux/kernel/v2.6/snapshots/
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/