From mboxrd@z Thu Jan 1 00:00:00 1970 Date: Mon, 9 Oct 2006 20:37:18 -0700 From: Andrew Morton Subject: Re: [patch] mm: bug in set_page_dirty_buffers Message-Id: <20061009203718.d6a8f803.akpm@osdl.org> In-Reply-To: <20061009202039.b6948a93.akpm@osdl.org> References: <20061010023654.GD15822@wotan.suse.de> <20061009202039.b6948a93.akpm@osdl.org> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-linux-mm@kvack.org Return-Path: To: Linus Torvalds , Nick Piggin , Peter Zijlstra , Linux Memory Management List , Greg KH List-ID: On Mon, 9 Oct 2006 20:20:39 -0700 Andrew Morton wrote: > On Mon, 9 Oct 2006 20:06:05 -0700 (PDT) > Linus Torvalds wrote: > > > On Tue, 10 Oct 2006, Nick Piggin wrote: > > > > > > This was triggered, but not the fault of, the dirty page accounting > > > patches. Suitable for -stable as well, after it goes upstream. > > > > Applied. However, I wonder what protects "page_mapping()" here? > > Nothing. And I don't understand the (unchangelogged) switch from > page->mapping to page_mapping(). > > > I don't > > think we hold the page lock anywhere, so "page->mapping" can change at any > > time, no? > > Yes. The patch makes the race window a bit smaller. OK, the address_space is protected from reclaim here by virtue of the caller's ref on vma->vm_file (needs a comment). The page_mapping() change remains a mystery. -- 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