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

Re: Max Mount Count

On Fri, 8 Jun 2001, Stephen C. Tweedie wrote:

> > Yeah, I'm using the following which AFAIK seems to cover the occurances
> > that LVM surfaced.  I can run multiple pvscans, etc for hours without
> > generating the refile buffers at least.
> I'll apply to 2.2.  For 2.4, we need to take appropriate locking
> before we can make such a test.
> Andrew, what do you think?  I'm tempted to solve this by taking the
> lru_list_lock and bumping b_count when adding the journal_head to an
> existing buffer_head.  That would certainly protect against buffer
> invalidation and try_to_free_buffers, without requiring us to
> hard-code the protection of journaled buffers into the rest of the VM.
> However, I'm still nervous: journaling can keep buffers around for
> quite a while for various reasons.  WHY does LVM want to invalidate
> buffers?  I'm scared that by "fixing" this we actually mask a deeper
> problem.  By causing journaled buffers to remain in the system after
> an invalidate, are we going to violate some assumption that LVM is
> making about its ability to flush things from the kernel?

>From my experience, I'm not sure LVM should be calling
invalidate_buffers() on all the filesystems it is.  I could understand it
calling it upon it's own LVM devices, but in this case it's actually just
enumerating the /proc/mounts list and calling it on every device in that
list according to following the printk's I added prior to it's
invalidate_buffers() call.  Generally this would cause refile buffer
attempts on devices which weren't even LVM devices on my system (ie. my
root fs).  Hence, that seems wrong to me.

Anyway, thats my knowledge. :)

> Cheers,
>  Stephen
> _______________________________________________
> Ext3-users mailing list
> Ext3-users redhat com
> https://listman.redhat.com/mailman/listinfo/ext3-users

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