From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail137.messagelabs.com (mail137.messagelabs.com [216.82.249.19]) by kanga.kvack.org (Postfix) with SMTP id BBA4E6B01F1 for ; Thu, 26 Aug 2010 14:59:52 -0400 (EDT) Date: Thu, 26 Aug 2010 12:00:02 -0700 (PDT) Message-Id: <20100826.120002.59671031.davem@davemloft.net> Subject: Re: [PATCH] mm: fix hang on anon_vma->root->lock From: David Miller In-Reply-To: References: <20100825.234149.189710316.davem@davemloft.net> Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-linux-mm@kvack.org To: hughd@google.com Cc: torvalds@linux-foundation.org, akpm@linux-foundation.org, riel@redhat.com, aarcange@redhat.com, linux-kernel@vger.kernel.org, linux-mm@kvack.org List-ID: From: Hugh Dickins Date: Thu, 26 Aug 2010 03:54:28 -0700 > On Wed, Aug 25, 2010 at 11:41 PM, David Miller wrote: >> From: Hugh Dickins >> Date: Wed, 25 Aug 2010 23:12:54 -0700 (PDT) >> >>> After several hours, kbuild tests hang with anon_vma_prepare() spinning on >>> a newly allocated anon_vma's lock - on a box with CONFIG_TREE_PREEMPT_RCU=y >>> (which makes this very much more likely, but it could happen without). >>> >>> The ever-subtle page_lock_anon_vma() now needs a further twist: since >>> anon_vma_prepare() and anon_vma_fork() are liable to change the ->root >>> of a reused anon_vma structure at any moment, page_lock_anon_vma() >>> needs to check page_mapped() again before succeeding, otherwise >>> page_unlock_anon_vma() might address a different root->lock. >>> >>> Signed-off-by: Hugh Dickins >> >> Interesting, is the condition which allows this to trigger specific >> to this merge window or was it always possible? > > Just specific to this merge window, which started using > anon_vma->root->lock in place of anon_vma->lock (anon_vma->root is > often anon_vma itself, but not always). Ok, because I've been seeing mysterious hangs recently on one of my sparc64 boxes and I was wondering if this might explain them :-) -- 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: email@kvack.org