linux-kernel - Re: [RFC PATCH V2 0/9] Introduce attach/clear_page

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: Windows password security audit tool. GUI, reports in PDF.

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

Message-ID: <20200502004158.GD29705@bombadil.infradead.org>
Date:   Fri, 1 May 2020 17:41:58 -0700
From:   Matthew Wilcox <willy@...radead.org>
To:     Guoqing Jiang <guoqing.jiang@...ud.ionos.com>
Cc:     linux-fsdevel@...r.kernel.org, linux-kernel@...r.kernel.org,
        hch@...radead.org, david@...morbit.com
Subject: Re: [RFC PATCH V2 0/9] Introduce attach/clear_page_private to
 cleanup code

On Sat, May 02, 2020 at 12:42:15AM +0200, Guoqing Jiang wrote:
> On 5/2/20 12:16 AM, Matthew Wilcox wrote:
> > On Thu, Apr 30, 2020 at 11:44:41PM +0200, Guoqing Jiang wrote:
> > >    include/linux/pagemap.h: introduce attach/clear_page_private
> > >    md: remove __clear_page_buffers and use attach/clear_page_private
> > >    btrfs: use attach/clear_page_private
> > >    fs/buffer.c: use attach/clear_page_private
> > >    f2fs: use attach/clear_page_private
> > >    iomap: use attach/clear_page_private
> > >    ntfs: replace attach_page_buffers with attach_page_private
> > >    orangefs: use attach/clear_page_private
> > >    buffer_head.h: remove attach_page_buffers
> > I think mm/migrate.c could also use this:
> > 
> >          ClearPagePrivate(page);
> >          set_page_private(newpage, page_private(page));
> >          set_page_private(page, 0);
> >          put_page(page);
> >          get_page(newpage);
> > 
> 
> Thanks for checking!  Assume the below change is appropriate.
> 
> diff --git a/mm/migrate.c b/mm/migrate.c
> index 7160c1556f79..f214adfb3fa4 100644
> --- a/mm/migrate.c
> +++ b/mm/migrate.c
> @@ -797,10 +797,7 @@ static int __buffer_migrate_page(struct address_space
> *mapping,
>         if (rc != MIGRATEPAGE_SUCCESS)
>                 goto unlock_buffers;
> 
> -       ClearPagePrivate(page);
> -       set_page_private(newpage, page_private(page));
> -       set_page_private(page, 0);
> -       put_page(page);
> +       set_page_private(newpage, detach_page_private(page));
>         get_page(newpage);

I think you can do:

@@ -797,11 +797,7 @@ static int __buffer_migrate_page(struct address_space *mapping,
        if (rc != MIGRATEPAGE_SUCCESS)
                goto unlock_buffers;
 
-       ClearPagePrivate(page);
-       set_page_private(newpage, page_private(page));
-       set_page_private(page, 0);
-       put_page(page);
-       get_page(newpage);
+       attach_page_private(newpage, detach_page_private(page));
 
        bh = head;
        do {
@@ -810,8 +806,6 @@ static int __buffer_migrate_page(struct address_space *mapping,
 
        } while (bh != head);
 
-       SetPagePrivate(newpage);
-
        if (mode != MIGRATE_SYNC_NO_COPY)

... but maybe there's a subtlety to the ordering of the setup of the bh
and setting PagePrivate that means what you have there is a better patch.
Anybody know?