linux-kernel - Re: kernel BUG at mm/page-writeback.c:LINE!

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]

Date:   Tue, 5 Jan 2021 21:34:33 +0000
From:   Matthew Wilcox <willy@...radead.org>
To:     Linus Torvalds <torvalds@...ux-foundation.org>
Cc:     Hugh Dickins <hughd@...gle.com>,
        Andrew Morton <akpm@...ux-foundation.org>,
        syzbot <syzbot+2fc0712f8f8b8b8fa0ef@...kaller.appspotmail.com>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        Linux-MM <linux-mm@...ck.org>,
        syzkaller-bugs <syzkaller-bugs@...glegroups.com>
Subject: Re: kernel BUG at mm/page-writeback.c:LINE!

On Tue, Jan 05, 2021 at 01:22:49PM -0800, Linus Torvalds wrote:
> On Tue, Jan 5, 2021 at 1:13 PM Hugh Dickins <hughd@...gle.com> wrote:
> >
> > I was going to raise a question, whether you should now revert
> > 073861ed77b6 ("mm: fix VM_BUG_ON(PageTail) and BUG_ON(PageWriteback)"):
> > which would not have gone in like that if c2407cf7d22d were already in.
> 
> Honestly, even if it wasn't for that PageTail issue, I think
> 073861ed77b6 is just the right thing to do anyway. It just feels _so_
> much safer to not have the possibility of that page wait thing
> following while the page is possibly then being free'd and re-used at
> the same time.
> 
> So I think the only reason to revert that commit would be if we were
> to find that it's a huge performance problem to raise the page
> refcount temporarily. Which I think is very unlikely (since we already
> dirty the page structure due to the page flags modification - although
> they are far enough apart that it might be a different cacheline).

struct pages _tend_ to be 64 bytes on 64-bit platforms (and i suspect
you're long past caring about performance on 32-bit platforms).