DECLARE_MUTEX+wq debug+module?

From: Daniel Kobras (kobras@tat.physik.uni-tuebingen.de)
Date: Thu Jul 13 2000 - 14:33:55 EST

Next message: Chris Wedgwood: "Re: Multithreaded TCP/IP stack"
Previous message: Kai Vehmanen: "Re: [linux-audio-dev] low-latency patch"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

Hi!

2.4.0-test3 + reiserfs patch 3.6.11 (reiserfs built as a module) just spew
out this strange message when I tried to access a reiserfs partition:

bad magic d1005900 (should be cff03e78), wq bug, forcing oops
Kernel BUG at /usr/src/linux/include/asm/semaphore.h:111!
invalid operand: 0000
CPU: 1
EIP: %4x:[<%8lx>]
EFLAGS: %8lx
eax: %8lx edx: %8lx ecx: %8lx edx: %8lx

[The machine locked up afterwards so the message is copied by hand but
yes, it indeed printed out the format tags. Is BUG() buggy??]

modutils logged the following module addresses:

d0853000 __insmod_reiserfs_O/lib/modules/2.4.0-test3/fs/reiserfs.o_M396DE341_V132096 [reiserfs]
d0853060 __insmod_reiserfs_S.text_L116620 [reiserfs]
d086f960 __insmod_reiserfs_S.rodata_L32758 [reiserfs]
d0877960 __insmod_reiserfs_S.data_L764 [reiserfs]
d0877d40 __insmod_reiserfs_S.bss_L6624 [reiserfs]
d084b850 __insmod_de4x5_S.data_L304 [de4x5]
d0842060 de4x5_probe [de4x5]
d084a9a0 __insmod_de4x5_S.rodata_L3702 [de4x5]
d0842060 __insmod_de4x5_S.text_L34941 [de4x5]
d0842000 __insmod_de4x5_O/lib/modules/2.4.0-test3/net/de4x5.o_M396DE341_V132096 [de4x5]
d084ba60 __insmod_de4x5_S.bss_L161 [de4x5]

No further modules were loaded on top, so 0xd1005900 looks
reiserfs-related. Checking the down() calls in reiserfs, only inode->i_sem
and reiserfs_end_io_sem are accessed. reiserfs_end_io_sem looks like the
culprit. However the related code in fs/reiserfs/journal.c looks perfectly
sane, DECLARE_MUTEX() in the global section, only up() and down() calls
otherwise. Module loading is a black box to me, but from the magics given
I assume that some relocation of global data after initialization is going
on here. Which would mean that DECLARE_MUTEX()+wq debugging is broken in a
module's global data. Maybe someone more knowledgeable than me could
confirm this (or explain what's wrong otherwise)?

Regards,

Daniel.

-- GNU/Linux Audio Mechanics - http://www.glame.de GPG Key ID 89BF7E2B - http://www.keyserver.net

- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/

Next message: Chris Wedgwood: "Re: Multithreaded TCP/IP stack"
Previous message: Kai Vehmanen: "Re: [linux-audio-dev] low-latency patch"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

This archive was generated by hypermail 2b29 : Sat Jul 15 2000 - 21:00:17 EST