Hello,
> On Mon, 2005-07-11 at 17:10, Jan Kara wrote:
>
> > attached patch should fix the following race:
> ...
> > and we have sent wrong data to disk... We now clean the dirty buffer
> > flag under buffer lock in all cases and hence we know that whenever a buffer
> > is starting to be journaled we either finish the pending write-out
> > before attaching a buffer to a transaction or we won't write the buffer
> > until the transaction is going to be committed... Please apply.
>
> Looks good to me.
>
> Btw, how did you find this? Were you able to reproduce this in
> practice?
I was not able to reproduce the problem in practice (not that I'd try
very much). When I was fixing the checkpointing code I noted that
ll_rw_block() need not actually send data to disk because of buffer
being locked. So I started inspecting how do we actually use buffer
lock and found this problem...
Honza
--
Jan Kara <[EMAIL PROTECTED]>
SuSE CR Labs
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
