Re: 2.6.19-rc2-mm1

[Nick Piggin]

Carsten Otte wrote:
> Andrew Morton wrote:
>
> > The page we're writing into isn't locked, so there's no deadlock afaict.
> >
> > But then, I forget how xip works.  Carsten, is it actually being used for
> > anything?
> >   
>
> The comment may be superfluous. I did not quite understand the deadlock 
> condition refered to by the comment in filemap, therefore I cut&pasted 
> it over. I will send a patch that removes it.

Yes, it wasn't a good comment. Basically we can't hold the page lock and
then enter the page fault handler. mm-fix-pagecache-write-deadlocks.patch
has quite a lot of comments to explain it.

I don't believe filemap_xip holds the page lock, so you don't need to do
the atomic copy, and you don't need to do the fault_in_pages_readable.

--
SUSE Labs, Novell Inc.
Send instant messages to your online friends http://au.messenger.yahoo.com 
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/