From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-ee0-f44.google.com (mail-ee0-f44.google.com [74.125.83.44]) by kanga.kvack.org (Postfix) with ESMTP id A2CD66B0036 for ; Wed, 14 May 2014 12:12:52 -0400 (EDT) Received: by mail-ee0-f44.google.com with SMTP id c41so1531420eek.17 for ; Wed, 14 May 2014 09:12:52 -0700 (PDT) Received: from mx1.redhat.com (mx1.redhat.com. [209.132.183.28]) by mx.google.com with ESMTP id x44si2051504eeo.19.2014.05.14.09.12.49 for ; Wed, 14 May 2014 09:12:51 -0700 (PDT) Date: Wed, 14 May 2014 18:11:52 +0200 From: Oleg Nesterov Subject: Re: [PATCH 19/19] mm: filemap: Avoid unnecessary barries and waitqueue lookups in unlock_page fastpath Message-ID: <20140514161152.GA2615@redhat.com> References: <1399974350-11089-1-git-send-email-mgorman@suse.de> <1399974350-11089-20-git-send-email-mgorman@suse.de> <20140513125313.GR23991@suse.de> <20140513141748.GD2485@laptop.programming.kicks-ass.net> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20140513141748.GD2485@laptop.programming.kicks-ass.net> Sender: owner-linux-mm@kvack.org List-ID: To: Peter Zijlstra Cc: Mel Gorman , Andrew Morton , Johannes Weiner , Vlastimil Babka , Jan Kara , Michal Hocko , Hugh Dickins , Dave Hansen , Linux Kernel , Linux-MM , Linux-FSDevel , Paul McKenney , Linus Torvalds , David Howells The subsequent discussion was "off-topic", and it seems that the patch itself needs a bit more discussion, On 05/13, Peter Zijlstra wrote: > > On Tue, May 13, 2014 at 01:53:13PM +0100, Mel Gorman wrote: > > On Tue, May 13, 2014 at 10:45:50AM +0100, Mel Gorman wrote: > > > void unlock_page(struct page *page) > > > { > > > + wait_queue_head_t *wqh = clear_page_waiters(page); > > > + > > > VM_BUG_ON_PAGE(!PageLocked(page), page); > > > + > > > + /* > > > + * No additional barrier needed due to clear_bit_unlock barriering all updates > > > + * before waking waiters > > > + */ > > > clear_bit_unlock(PG_locked, &page->flags); > > > - smp_mb__after_clear_bit(); > > > - wake_up_page(page, PG_locked); > > > > This is wrong. Yes, > > The smp_mb__after_clear_bit() is still required to ensure > > that the cleared bit is visible before the wakeup on all architectures. But note that "the cleared bit is visible before the wakeup" is confusing. I mean, we do not need mb() before __wake_up(). We need it only because __wake_up_bit() checks waitqueue_active(). And at least fs/cachefiles/namei.c:cachefiles_delete_object() fs/block_dev.c:blkdev_get() kernel/signal.c:task_clear_jobctl_trapping() security/keys/gc.c:key_garbage_collector() look obviously wrong. I would be happy to send the fix, but do I need to split it per-file? Given that it is trivial, perhaps I can send a single patch? Oleg. -- 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