[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <Y6lY+RB/LXY6V5tH@lucifer>
Date: Mon, 26 Dec 2022 08:19:05 +0000
From: Lorenzo Stoakes <lstoakes@...il.com>
To: Matthew Wilcox <willy@...radead.org>
Cc: linux-mm@...ck.org, Andrew Morton <akpm@...ux-foundation.org>,
linux-kernel@...r.kernel.org, Hugh Dickins <hughd@...gle.com>,
Vlastimil Babka <vbabka@...e.cz>,
Liam Howlett <liam.howlett@...cle.com>,
William Kucharski <william.kucharski@...cle.com>,
Christian Brauner <brauner@...nel.org>,
Jonathan Corbet <corbet@....net>,
Mike Rapoport <rppt@...nel.org>,
Joel Fernandes <joel@...lfernandes.org>
Subject: Re: [PATCH v2 3/4] mm: mlock: update the interface to use folios
On Mon, Dec 26, 2022 at 08:15:29AM +0000, Matthew Wilcox wrote:
> On Mon, Dec 26, 2022 at 07:08:46AM +0000, Lorenzo Stoakes wrote:
> > goto out;
> > - page = pmd_page(*pmd);
> > + folio = page_folio((struct page *)pmd_page(*pmd));
>
> I do not like this fix. Better to fix m68k to:
>
> #define pmd_page(pmd) (struct page *)NULL
>
I actually completely agree with this, and felt it was essentially an issue with
the fact m68k aren't doing this cast, but I thought perhaps it'd be out of scope
for this series to make the change there.
However now you're suggesting the same thing it seems like it isn't out of scope
at all :)
Will spin a v3 moving the fix there.
Powered by blists - more mailing lists