From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S272056AbTG2UTE (ORCPT ); Tue, 29 Jul 2003 16:19:04 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S272058AbTG2UTE (ORCPT ); Tue, 29 Jul 2003 16:19:04 -0400 Received: from hera.cwi.nl ([192.16.191.8]:26272 "EHLO hera.cwi.nl") by vger.kernel.org with ESMTP id S272056AbTG2UTD (ORCPT ); Tue, 29 Jul 2003 16:19:03 -0400 From: Andries.Brouwer@cwi.nl Date: Tue, 29 Jul 2003 22:18:56 +0200 (MEST) Message-Id: To: Valdis.Kletnieks@vt.edu, manfred@colorfullife.com Subject: Re: [PATCH] select fix Cc: Andries.Brouwer@cwi.nl, akpm@osdl.org, linux-kernel@vger.kernel.org, torvalds@osdl.org Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org From: Manfred Spraul Valdis.Kletnieks@vt.edu wrote: > Would including all 3 conditions make sense? I hesitated for a moment. These conditions are not entirely equivalent and neither implies the other. But Manfreds version is better. Many drivers have a write_room() that only counts characters and the corresponding write() will fill the buffer and only check tty->stopped before actually transmitting. So write_room() is a better predictor. Andries