linux-kernel - Re: [PATCH 5/5] mm/thp: Split huge pmds/puds if they're pinned when fork()

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: <20200923160634.GB79898@xz-x1>
Date:   Wed, 23 Sep 2020 12:06:34 -0400
From:   Peter Xu <peterx@...hat.com>
To:     John Hubbard <jhubbard@...dia.com>
Cc:     linux-mm@...ck.org, linux-kernel@...r.kernel.org,
        Linus Torvalds <torvalds@...ux-foundation.org>,
        Michal Hocko <mhocko@...e.com>,
        Kirill Shutemov <kirill@...temov.name>,
        Jann Horn <jannh@...gle.com>, Oleg Nesterov <oleg@...hat.com>,
        Kirill Tkhai <ktkhai@...tuozzo.com>,
        Hugh Dickins <hughd@...gle.com>,
        Leon Romanovsky <leonro@...dia.com>, Jan Kara <jack@...e.cz>,
        Christoph Hellwig <hch@....de>,
        Andrew Morton <akpm@...ux-foundation.org>,
        Jason Gunthorpe <jgg@...pe.ca>,
        Andrea Arcangeli <aarcange@...hat.com>
Subject: Re: [PATCH 5/5] mm/thp: Split huge pmds/puds if they're pinned when
 fork()

On Mon, Sep 21, 2020 at 11:41:16PM -0700, John Hubbard wrote:
> On 9/21/20 2:20 PM, Peter Xu wrote:
> ...
> > diff --git a/mm/huge_memory.c b/mm/huge_memory.c
> > index 7ff29cc3d55c..c40aac0ad87e 100644
> > --- a/mm/huge_memory.c
> > +++ b/mm/huge_memory.c
> > @@ -1074,6 +1074,23 @@ int copy_huge_pmd(struct mm_struct *dst_mm, struct mm_struct *src_mm,
> >   	src_page = pmd_page(pmd);
> >   	VM_BUG_ON_PAGE(!PageHead(src_page), src_page);
> > +
> > +	/*
> > +	 * If this page is a potentially pinned page, split and retry the fault
> > +	 * with smaller page size.  Normally this should not happen because the
> > +	 * userspace should use MADV_DONTFORK upon pinned regions.  This is a
> > +	 * best effort that the pinned pages won't be replaced by another
> > +	 * random page during the coming copy-on-write.
> > +	 */
> > +	if (unlikely(READ_ONCE(src_mm->has_pinned) &&
> > +		     page_maybe_dma_pinned(src_page))) {

[...]

> > +		pte_free(dst_mm, pgtable);
> > +		spin_unlock(src_ptl);
> > +		spin_unlock(dst_ptl);
> > +		__split_huge_pmd(vma, src_pmd, addr, false, NULL);
> > +		return -EAGAIN;
> > +	}
> 
> 
> Why wait until we are so deep into this routine to detect this and unwind?
> It seems like if you could do a check near the beginning of this routine, and
> handle it there, with less unwinding? In fact, after taking only the src_ptl,
> the check could be made, right?

Because that's where we've fetched the page from the pmd so I can directly
reference src_page.  Also I think at least I need to check against swp entries?
So it seems still easier to keep it here, considering it's an unlikely path.

Thanks,

-- 
Peter Xu