Re: Linux 2.4.34-pre3

[Willy Tarreau <w@1wt.eu>]

On Wed, Sep 20, 2006 at 04:55:25AM +1000, Nick Piggin wrote:
> Willy Tarreau wrote:
> >Hi Nick,
> >
> >On Wed, Sep 20, 2006 at 03:42:06AM +1000, Nick Piggin wrote:
> >
> >[cut -pre3 advertisement]
> >
> >
> >>I wonder if 2.4 doesn't need the memory ordering fix to prevent pagecache
> >>corruption in reclaim? (http://www.gatago.com/linux/kernel/14682626.html)
> >>
> >>What would need to be done is to test page_count before testing PageDirty,
> >>and putting an smp_rmb between the two.
> >
> >
> >I've read the thread, and Linus proposed to add an smp_wmb() in
> >set_page_dirty() too.
> 
> I think that isn't needed because put_page is a RMW, which is defined
> to order memory. And presumably you wouldn't set the page dirty without
> a reference to the page.

OK, thanks for the explanation.

> >I see that an smp_rmb() is already present
> >in shrink_cache() with the adequate comment.
> 
> So there is! My mistake then, I was confused and looking at
> try_to_swap_out, but I see that doesn't actually free the page. Fine,
> I think 2.4 is OK then.

Perfect !

Thanks,
Willy