From owner-freebsd-commit Sun Jun 4 15:03:38 1995 Return-Path: commit-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id PAA09241 for commit-outgoing; Sun, 4 Jun 1995 15:03:38 -0700 Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id PAA09161 for cvs-sys-outgoing; Sun, 4 Jun 1995 15:02:14 -0700 Received: from sequent.kiae.su (sequent.kiae.su [144.206.136.6]) by freefall.cdrom.com (8.6.10/8.6.6) with SMTP id PAA09147 ; Sun, 4 Jun 1995 15:02:07 -0700 Received: by sequent.kiae.su id AA06014 (5.65.kiae-2 ); Mon, 5 Jun 1995 02:01:58 +0400 Received: by sequent.KIAE.su (UUMAIL/2.0); Mon, 5 Jun 95 02:01:57 +0400 Received: (from ache@localhost) by astral.msk.su (8.6.8/8.6.6) id BAA06688; Mon, 5 Jun 1995 01:46:06 +0400 To: CVS-commiters@freefall.cdrom.com, cvs-sys@freefall.cdrom.com, David Greenman Cc: Bruce Evans References: <199506041628.JAA15415@freefall.cdrom.com> In-Reply-To: <199506041628.JAA15415@freefall.cdrom.com>; from David Greenman at Sun, 4 Jun 1995 09:28:44 -0700 Message-Id: Organization: Olahm Ha-Yetzirah Date: Mon, 5 Jun 1995 01:46:05 +0400 (MSD) X-Mailer: Mail/@ [v2.37 FreeBSD] From: =?KOI8-R?Q?=E1=CE=C4=D2=C5=CA=?= =?KOI8-R?Q?=FE=C5=D2=CE=CF=D7=?= aka "Andrey A. Chernov, Black Mage" X-Class: Fast Subject: Re: cvs commit: src/sys/kern tty.c Lines: 53 Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Length: 1845 Sender: commit-owner@FreeBSD.org Precedence: bulk -----BEGIN PGP SIGNED MESSAGE----- In message <199506041628.JAA15415@freefall.cdrom.com> David Greenman writes: >davidg 95/06/04 09:28:44 > Modified: sys/kern tty.c > Log: > Remove t_timeout from ttysleep() call in ttwrite() (undo part of revision > 1.44). Nonblocking writes can't return EWOULDBLOCK so there is no point > in timing out. > Fixes PR 445. > Submitted by: Bruce Evans I disagree. 1) Variant before "fix" can't affect default case: if you don't set t_timeout directly, it is equal to zero (acts completely like in old case). I don't see PR 445 right now, but can say that this fix can't fix PR unless t_timeout sets directly by user. If user sets t_timeout directly to non-zero, he must know what he do. 2) This "fix" break non-default case (purpose of t_timeout as introduced), i.e. if process issue write operation on modem with hanging CTS, process hangs forever. From very practical reasons I prefer return of EWOULDBLOCK than hanging forever... If you try to run modem spool site with about 200 numbers, hanged modems becomes real pain. 3) t_timeout is very special case and already was non-standard as invented, so I consider this fix like attepmt of void standartization with big blood. Please, back out it. - -- Andrey A. Chernov : And I rest so composedly, /Now, in my bed, ache@astral.msk.su : That any beholder /Might fancy me dead - FidoNet: 2:5020/230.3 : Might start at beholding me, /Thinking me dead. RELCOM Team,FreeBSD Team : E.A.Poe From "For Annie" 1849 -----BEGIN PGP SIGNATURE----- Version: 2.6.2i iQCVAwUBL9IpmuJgpPLZnQjrAQH/6QQA6OS+LPDpLjElqpW+qXnAsE5WAA3HDW1I XdTTMIW+USPsAtJS+wRKEiY/fw4Jmd2+TaXapmHVt9qTmJcGU2PzB3r1+0wLXLIg Df9pL41+2Yf8/C60z9yqx/uHYz7imI8if3n02KkPuEnrtT8SRt6T8ycP4wpAPqNE tBa4A/W5Q9k= =xB2Q -----END PGP SIGNATURE-----