Re: [PATCH] filemap_fdatasync & related changes

From: Rik van Riel (riel@conectiva.com.br)
Date: Thu Jan 04 2001 - 11:58:19 EST


On Thu, 4 Jan 2001, Marcelo Tosatti wrote:
> On 4 Jan 2001, Christoph Rohland wrote:
> > Chris Mason <mason@suse.com> writes:

> > > That would loop forever if the writepage func kept returning 1 though
        [snip]

> > return 1 if the swap space is exhausted. So everybody using shared
> > anonymous, SYSV shared or POSIX shared memory can hit this.
>
> In case of swap exhausted I think we should just fail.

No. In case of swap exhausted the page needs to be moved to
the active list so we don't clog up the inactive_dirty list
and have the possibility of a deadlock (already done).

In case of "write ordering not yet satisfied", the writepage
function should leave the page on the inactive_dirty list and
maybe write out something else (reiserfs, ext3, xfs...).

regards,

Rik

--
Hollywood goes for world dumbination,
	Trailer at 11.

http://www.surriel.com/ http://www.conectiva.com/ http://distro.conectiva.com.br/

- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org Please read the FAQ at http://www.tux.org/lkml/



This archive was generated by hypermail 2b29 : Sun Jan 07 2001 - 21:00:18 EST