From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1758138Ab3CDRCG (ORCPT ); Mon, 4 Mar 2013 12:02:06 -0500 Received: from mail-pb0-f52.google.com ([209.85.160.52]:57382 "EHLO mail-pb0-f52.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757551Ab3CDRCF (ORCPT ); Mon, 4 Mar 2013 12:02:05 -0500 Date: Mon, 4 Mar 2013 09:01:59 -0800 From: Tejun Heo To: Lianwei Wang Cc: linux-kernel@vger.kernel.org, Oleg Nesterov , "Rafael J. Wysocki" Subject: Re: PATCH: freezer: add fake signal clearing back when thaw task Message-ID: <20130304170159.GA30413@htj.dyndns.org> References: <20130225235313.GE2679@htj.dyndns.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Hello, On Mon, Mar 04, 2013 at 11:24:06AM +0800, Lianwei Wang wrote: > Freeze/Thaw is a hot path for the Linux based mobile OS, e.g. Android. > If we don't remove the pending fake signal, then the user space apps Hotness is relative. How often are we talking about? Do you have any numbers backing up that this actually makes any difference? The reason why I'm not convinced is that by the time the control reaches fake_signal_clear() the system already has done most of the heavy lifting and the code path which would be excluded by the extra checking seems pretty short. If you can show otherwise, please do so. > or the related kernel driver has to handle such error. And yes, the > user can handle such case by checking the return value, but it mislead > the user and confuse to them that why wait_event_freezable and friends > return a error on resume path every time? Can we avoid such useless > error return? -ERESTARTSYS may happen without freezing and the syscall path has been capable of handling it for a very long time. Again, do you have anything which shows otherwise? Thanks. -- tejun