[Fwd: 2.2.14 oopses : is this a clue ?]

From: Hans Reiser (reiser@idiom.com)
Date: Fri Jan 21 2000 - 15:46:01 EST


Thanks for being on top of this willy, it really helps us to get such detailed
error analysis.

Hans

-- 
Get Linux (http://www.kernel.org) plus ReiserFS
 (http://devlinux.org/namesys).  If you sell an OS or
internet appliance, buy a port of ReiserFS!  If you
need customizations and industrial grade support, we sell them.

attached mail follows:


Hi all,

I had an oops yesterday in ext2_check_dir_entry() (fs/ext2/dir.c). Although I couldn't grab the messages, I'm pretty sure it's because the dir_entry pointer (de) was NULL.

Why do I think about this, would you say ?

Simply because of exactly same a problem I reported two weeks ago to the reiserfs developpers, in which case it was really "de" which was NULL. I inserted a test on "de" in this function, and no more crash, but no message either, so no possible conclusion on this... I must say that it occured on a bad disk with lots of hardware errors, and I don't use this disk anymore, but do use reiserfs on my notebook without an error.

In the two last oops reports in LKML, the dir_entry is involved ...

in doubt, I've added a check on "de" at the very beginning of ext2_check_dir_entry, and I'm waiting for the crash to appear again, or at least a message. Is there a reason for which all these functions to be called with a NULL dir_entry ?

I'm wondering about using IKD for this, but I don't know if I could insert break points in the source code, sort of "asserts" which force it to take the control in case of a condition match. Eg:

if (de == NULL) raise_ikd();

Perhaps something such as *(int *)NULL=0 ?

Willy

___________________________________________________________ Do You Yahoo!? Achetez, vendez! À votre prix! Sur http://encheres.yahoo.fr

- 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/

- 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/



This archive was generated by hypermail 2b29 : Sun Jan 23 2000 - 21:00:27 EST