From mboxrd@z Thu Jan 1 00:00:00 1970 From: Tejun Heo Subject: Re: poll: allow f_op->poll to sleep, take #3 Date: Mon, 24 Nov 2008 13:29:51 +0900 Message-ID: <492A2DBF.3030208@gmail.com> References: <20081122123942.GF5707@parisc-linux.org> <4927FE87.6050005@gmail.com> <20081122105356.87856d04.akpm@linux-foundation.org> <4928B162.9030404@gmail.com> <20081123085902.GI30453@elte.hu> <49291F06.5040907@gmail.com> <20081123093420.GN30453@elte.hu> <492925B0.9030404@gmail.com> <20081123094551.GQ30453@elte.hu> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Cc: Davide Libenzi , Andrew Morton , Matthew Wilcox , Miklos Szeredi , arjan@linux.intel.com, Linus Torvalds , hch@infradead.org, rminnich@sandia.gov, ericvh@gmail.com, Linux Kernel Mailing List , linux-fsdevel@vger.kernel.org, Peter Zijlstra To: Ingo Molnar Return-path: Received: from mail-gx0-f29.google.com ([209.85.217.29]:60509 "EHLO mail-gx0-f29.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750837AbYKXE37 (ORCPT ); Sun, 23 Nov 2008 23:29:59 -0500 Received: by gxk10 with SMTP id 10so1147554gxk.13 for ; Sun, 23 Nov 2008 20:29:58 -0800 (PST) In-Reply-To: <20081123094551.GQ30453@elte.hu> Sender: linux-fsdevel-owner@vger.kernel.org List-ID: Ingo Molnar wrote: >> And @sync might be useful depending on who's waking it up, so we >> either need to change the wake interface or give it an easier way to >> pass those parameters as received. The callback function isn't the >> right place to ignore those parameters. It simply doesn't know why >> the caller is passing them in or what they mean under the >> circumstances. > > We'll likely eliminate the 'sync' parameter from the scheduler. It's > not a flag that should be proliferated. But it's still being used in quite hot paths (pipe, splice, socket) and I don't really wanna mix up a change which can cause subtle scheduling related performance regression into this patch. How about using the dummy waitqueue hack for now and when removing the @sync param, switch it to one of wakeup APIs? I'll be happy to add big /* TODO */ comment in the function. Thanks. -- tejun