>
>
> On Tue, 12 Jun 2001, Steve Lord wrote:
>
> > Hmm, are you sure this will fix it, b_pprev is only cleared by
> > __hash_unlink via __remove_from_queues and __remove_from_queues
> > is only called by try_to_free_buffers in the case where it
> > decides it is going to actually free the buffer_head. In the
> > case where it leaves it on the inode, b_pprev is still set.
>
> Page buffers are never included into the hash. If they are - you've
> got much uglier problems.
Hmm, good point, I need to take another look at the buffer and
see what is set in there, I have a trap in put_last_free and I
am still hitting it with the fix, something else fishy must be
going on here, or the trap code I am running has a hole in it.
We do not hash the buffers, just put them on the inode queue.
Steve
-
To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in
the body of a message to [EMAIL PROTECTED]