From mboxrd@z Thu Jan 1 00:00:00 1970 From: Andrea Arcangeli Subject: Re: [RFC][PATCH v3 2/6] mm, directio: fix fork vs direct-io race (read(2) side IOW gup(write) side) Date: Tue, 14 Apr 2009 21:48:25 +0200 Message-ID: <20090414194825.GD9809@random.random> References: <20090414151204.C647.A69D9226@jp.fujitsu.com> <20090414151652.C64D.A69D9226@jp.fujitsu.com> <20090414152500.C65F.A69D9226@jp.fujitsu.com> <20090414175124.GC9809@random.random> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: KOSAKI Motohiro , LKML , Linus Torvalds , Andrew Morton , Nick Piggin , linux-mm@kvack.org, linux-fsdevel@vger.kernel.org, Hugh Dickins , Zach Brown , Andy Grover To: Jeff Moyer Return-path: Content-Disposition: inline In-Reply-To: Sender: owner-linux-mm@kvack.org List-Id: linux-fsdevel.vger.kernel.org On Tue, Apr 14, 2009 at 02:10:08PM -0400, Jeff Moyer wrote: > Really? I don't actually see that in the code, have I missed it? Checking the spinlock version, when any writer is waiting, the sem->wait_list won't empty and down_read will wait too. The wakeup is FIFO with __rwsem_do_wake is doing a wake-one if first one in queue is a down_write. So it looks ok to me. asm version should have an equivalent logic. -- 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