[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: 2.2.19/0.0.7a assertion failure



Hi,

On Thu, Jul 05, 2001 at 10:39:21PM +1000, CaT wrote:

> > Still using ext3?  If 6b works and 7a fails, that's definitely useful
> 
> I thought it might be. :)

I think that the chunk

Index: fs/ext3/file.c
===================================================================
RCS file: /home/rcs/CVS/ext3/fs/ext3/file.c,v
retrieving revision 1.1
retrieving revision 1.2
diff -u -r1.1 -r1.2
--- fs/ext3/file.c	2001/06/04 20:53:25	1.1
+++ fs/ext3/file.c	2001/06/04 20:55:09	1.2
@@ -414,7 +413,7 @@
 		update_vm_cache(inode, pos, bh->b_data + offset, c);
 		pos += c;
 		written += c;
-		written_transaction += c;
+		written_transaction++;
 		buf += c;
 		count -= c;
 		brelse(bh);

is probably responsible for triggering the problem.  (If cdparanoia is
writing data in large chunks greater than 4k, then this change would
affect the size of the write transactions which get generated inside
ext3.)

If that's right, then this change is not _causing_ the problem, merely
allowing it to happen (the old code was accidentally masking the path
on which the error happens.)  You could try reverse-applying that
single chunk to see if that fixes it for you, and I'll hunt around
that area to see what the real problem is.

Cheers,
 Stephen





[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]