All of lore.kernel.org
 help / color / mirror / Atom feed
From: Linus Torvalds <torvalds@linux-foundation.org>
To: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: LKML <linux-kernel@vger.kernel.org>,
	Andrea Arcangeli <aarcange@redhat.com>,
	Hugh Dickins <hughd@google.com>, Christoph Lameter <cl@linux.com>,
	Linus Torvalds <torvalds@linux-foundation.org>,
	Andrew Morton <akpm@linux-foundation.org>,
	Rik van Riel <riel@redhat.com>,
	Peter Zijlstra <peterz@infradead.org>,
	Andi Kleen <andi@firstfloor.org>, linux-mm <linux-mm@kvack.org>
Subject: Re: [PATCH 3/4] hugetlb, rmap: fix confusing page locking in hugetlb_cow()
Date: Fri, 10 Sep 2010 10:15:46 -0700 (PDT)	[thread overview]
Message-ID: <alpine.LFD.2.00.1009101011140.9670@i5.linux-foundation.org> (raw)
In-Reply-To: <1284092586-1179-4-git-send-email-n-horiguchi@ah.jp.nec.com>



On Fri, 10 Sep 2010, Naoya Horiguchi wrote:
>  
> -	if (!pagecache_page) {
> -		page = pte_page(entry);
> +	/*
> +	 * hugetlb_cow() requires page locks of pte_page(entry) and
> +	 * pagecache_page, so here we need take the former one
> +	 * when page != pagecache_page or !pagecache_page.
> +	 */
> +	page = pte_page(entry);
> +	if (page != pagecache_page)
>  		lock_page(page);

Why isn't this a potential deadlock? You have two pages, and lock them 
both. Is there some ordering guarantee that says that 'pagecache_page' and 
'page' will always be in a certain relationship so that you cannot get 
A->B and B->A lock ordering?

Please document that ordering rule if so.

			Linus

WARNING: multiple messages have this Message-ID (diff)
From: Linus Torvalds <torvalds@linux-foundation.org>
To: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: LKML <linux-kernel@vger.kernel.org>,
	Andrea Arcangeli <aarcange@redhat.com>,
	Hugh Dickins <hughd@google.com>, Christoph Lameter <cl@linux.com>,
	Linus Torvalds <torvalds@linux-foundation.org>,
	Andrew Morton <akpm@linux-foundation.org>,
	Rik van Riel <riel@redhat.com>,
	Peter Zijlstra <peterz@infradead.org>,
	Andi Kleen <andi@firstfloor.org>, linux-mm <linux-mm@kvack.org>
Subject: Re: [PATCH 3/4] hugetlb, rmap: fix confusing page locking in hugetlb_cow()
Date: Fri, 10 Sep 2010 10:15:46 -0700 (PDT)	[thread overview]
Message-ID: <alpine.LFD.2.00.1009101011140.9670@i5.linux-foundation.org> (raw)
In-Reply-To: <1284092586-1179-4-git-send-email-n-horiguchi@ah.jp.nec.com>



On Fri, 10 Sep 2010, Naoya Horiguchi wrote:
>  
> -	if (!pagecache_page) {
> -		page = pte_page(entry);
> +	/*
> +	 * hugetlb_cow() requires page locks of pte_page(entry) and
> +	 * pagecache_page, so here we need take the former one
> +	 * when page != pagecache_page or !pagecache_page.
> +	 */
> +	page = pte_page(entry);
> +	if (page != pagecache_page)
>  		lock_page(page);

Why isn't this a potential deadlock? You have two pages, and lock them 
both. Is there some ordering guarantee that says that 'pagecache_page' and 
'page' will always be in a certain relationship so that you cannot get 
A->B and B->A lock ordering?

Please document that ordering rule if so.

			Linus

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

  parent reply	other threads:[~2010-09-10 17:15 UTC|newest]

Thread overview: 42+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-09-10  4:23 [PATCH 0/4] hugetlb, rmap: fixes and cleanups Naoya Horiguchi
2010-09-10  4:23 ` Naoya Horiguchi
2010-09-10  4:23 ` [PATCH 1/4] hugetlb, rmap: always use anon_vma root pointer Naoya Horiguchi
2010-09-10  4:23   ` Naoya Horiguchi
2010-09-10 14:37   ` Andrea Arcangeli
2010-09-10 14:37     ` Andrea Arcangeli
2010-09-10 14:40   ` Rik van Riel
2010-09-10 14:40     ` Rik van Riel
2010-09-10 17:19   ` Linus Torvalds
2010-09-10 17:19     ` Linus Torvalds
2010-09-10 21:50     ` Andi Kleen
2010-09-10 21:50       ` Andi Kleen
2010-09-10 22:07       ` Linus Torvalds
2010-09-10 22:07         ` Linus Torvalds
2010-09-11  2:09       ` Rik van Riel
2010-09-11  2:09         ` Rik van Riel
2010-09-10  4:23 ` [PATCH 2/4] hugetlb, rmap: use hugepage_add_new_anon_rmap() in hugetlb_cow() Naoya Horiguchi
2010-09-10  4:23   ` Naoya Horiguchi
2010-09-10 14:39   ` Andrea Arcangeli
2010-09-10 14:39     ` Andrea Arcangeli
2010-09-10 14:40   ` Rik van Riel
2010-09-10 14:40     ` Rik van Riel
2010-09-10  4:23 ` [PATCH 3/4] hugetlb, rmap: fix confusing page locking " Naoya Horiguchi
2010-09-10  4:23   ` Naoya Horiguchi
2010-09-10 14:41   ` Rik van Riel
2010-09-10 14:41     ` Rik van Riel
2010-09-10 17:15   ` Linus Torvalds [this message]
2010-09-10 17:15     ` Linus Torvalds
2010-09-23 23:50     ` Naoya Horiguchi
2010-09-23 23:50       ` Naoya Horiguchi
2010-09-10  4:23 ` [PATCH 4/4] hugetlb, rmap: add BUG_ON(!PageLocked) in hugetlb_add_anon_rmap() Naoya Horiguchi
2010-09-10  4:23   ` Naoya Horiguchi
2010-09-10 14:42   ` Rik van Riel
2010-09-10 14:42     ` Rik van Riel
2010-09-10 14:44   ` Andrea Arcangeli
2010-09-10 14:44     ` Andrea Arcangeli
2010-09-10  9:04 ` [PATCH 0/4] hugetlb, rmap: fixes and cleanups Andi Kleen
2010-09-10  9:04   ` Andi Kleen
2010-09-10 11:47   ` Naoya Horiguchi
2010-09-10 11:47     ` Naoya Horiguchi
2010-09-10 14:30     ` Andi Kleen
2010-09-10 14:30       ` Andi Kleen

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=alpine.LFD.2.00.1009101011140.9670@i5.linux-foundation.org \
    --to=torvalds@linux-foundation.org \
    --cc=aarcange@redhat.com \
    --cc=akpm@linux-foundation.org \
    --cc=andi@firstfloor.org \
    --cc=cl@linux.com \
    --cc=hughd@google.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=n-horiguchi@ah.jp.nec.com \
    --cc=peterz@infradead.org \
    --cc=riel@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body. 
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.