linux-kernel - Re: [PATCH v5 6/9] mm: swap: allow cache reclaim to skip slot cache

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-Id: <20240805015324.45134-1-21cnbao@gmail.com>
Date: Mon,  5 Aug 2024 13:53:24 +1200
From: Barry Song <21cnbao@...il.com>
To: chrisl@...nel.org
Cc: akpm@...ux-foundation.org,
	baohua@...nel.org,
	hughd@...gle.com,
	kaleshsingh@...gle.com,
	linux-kernel@...r.kernel.org,
	linux-mm@...ck.org,
	ryan.roberts@....com,
	ryncsn@...il.com,
	ying.huang@...el.com
Subject: Re: [PATCH v5 6/9] mm: swap: allow cache reclaim to skip slot cache

On Mon, Aug 5, 2024 at 6:07 AM Chris Li <chrisl@...nel.org> wrote:

> > > > +
> > > > +       spin_lock(&si->lock);
> > > > +       /* Only sinple page folio can be backed by zswap */
> > > > +       if (!nr_pages)
> > > > +               zswap_invalidate(entry);
> > >
> > > I am trying to figure out if I am mad :-)  Does nr_pages == 0 means single
> > > page folio?
> > >
> >
> > Hi Barry
> >
> > I'm sorry, this should be nr_pages == 1, I messed up order and nr, as
> > zswap only works for single page folios.
> Ack. Should be  nr_pages == 1.
>

Yes. Andrew, can you please help squash the below fix,

small folios should have nr_pages == 1 but not nr_page == 0

diff --git a/mm/swapfile.c b/mm/swapfile.c
index ea023fc25d08..6de12d712c7e 100644
--- a/mm/swapfile.c
+++ b/mm/swapfile.c
@@ -224,7 +224,7 @@ static int __try_to_reclaim_swap(struct swap_info_struct *si,
 
 	spin_lock(&si->lock);
 	/* Only sinple page folio can be backed by zswap */
-	if (!nr_pages)
+	if (nr_pages == 1)
 		zswap_invalidate(entry);
 	swap_entry_range_free(si, entry, nr_pages);
 	spin_unlock(&si->lock);

> Barry, thanks for catching that.
>
> Chris

Thanks
Barry