From mats@snowbee.dyns.cx Tue Sep 2 10:17:54 2003 Return-Path: Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 6836916A4BF for ; Tue, 2 Sep 2003 10:17:54 -0700 (PDT) Received: from snowbee.dyns.cx (as13-5-6.n.n.bonet.se [217.215.6.225]) by mx1.FreeBSD.org (Postfix) with ESMTP id 03BCF43FBF for ; Tue, 2 Sep 2003 10:17:53 -0700 (PDT) (envelope-from mats@snowbee.dyns.cx) Received: from pc10.snowbee.foo (localhost [127.0.0.1]) by snowbee.dyns.cx (8.12.3/8.12.3) with ESMTP id h82HHp5H019981; Tue, 2 Sep 2003 19:17:51 +0200 (CEST) (envelope-from mats@pc10.snowbee.foo) Received: (from mats@localhost) by pc10.snowbee.foo (8.12.3/8.12.3/Submit) id h82HHold019980; Tue, 2 Sep 2003 19:17:50 +0200 (CEST) (envelope-from mats) Message-Id: <200309021717.h82HHold019980@pc10.snowbee.foo> Date: Tue, 2 Sep 2003 19:17:50 +0200 (CEST) From: Mats Peterson Reply-To: Mats Peterson To: FreeBSD-gnats-submit@freebsd.org Cc: mats@snowbee.dyns.cx Subject: Incorrect information in /etc/gettytab X-Send-Pr-Version: 3.113 X-GNATS-Notify: >Number: 56325 >Category: bin >Synopsis: Incorrect information in /etc/gettytab >Confidential: no >Severity: non-critical >Priority: medium >Responsible: yar >State: closed >Quarter: >Keywords: >Date-Required: >Class: sw-bug >Submitter-Id: current-users >Arrival-Date: Tue Sep 02 10:20:15 PDT 2003 >Closed-Date: Tue Sep 16 02:36:37 PDT 2003 >Last-Modified: Tue Sep 16 02:36:37 PDT 2003 >Originator: Mats Peterson >Release: FreeBSD 4.6.2-RELEASE-p12 i386 >Organization: None >Environment: System: FreeBSD pc10.snowbee.foo 4.6.2-RELEASE-p12 FreeBSD 4.6.2-RELEASE-p12 #0: Sat Aug 30 10:57:24 CEST 2003 mats@pc10.snowbee.foo:/usr/obj/usr/src/sys/MYKERNEL i386 >Description: In /etc/gettytab, the three lines below are not valid. Using the 'np' flag will override any 'ep', 'op' or 'ap' flags (have a look at /usr/src/libexec/getty/subr.c). I suppose these lines are from some older incarnation of getty. # np:ep: same as np except inpck. # np:op: same as np:ep except for parodd (but parodd is overridden). # np:ap: same as np except istrip. >How-To-Repeat: Use ':np:ep:', ':np:op:' or ':np:ap:' in a gettytab entry. >Fix: Remove the three lines from /etc/gettytab. >Release-Note: >Audit-Trail: From: Yar Tikhiy To: Bruce Evans Cc: mats@snowbee.dyns.cx, FreeBSD-gnats-submit@freebsd.org Subject: Re: bin/56325: Incorrect information in /etc/gettytab Date: Mon, 8 Sep 2003 19:47:54 +0400 On Sun, Sep 07, 2003 at 08:04:01PM +1000, Bruce Evans wrote: > I looked at what the various combinations of parity did when I wrote > thos lines in gettytab. I haven't looked at them recently (i never > use anything except np). Perhaps changing getty to termios canonicalized > the parity support. Indeed, it was rev. 1.6 of subr.c that introduced the new behaviour of the parity flags along with using termios instead of the old BSD interface to terminals. I believe that people use just np nowadays, so the question is whether the old behaviour of parity flag combinations is worth restoring. How do you think? -- Yar From: Bruce Evans To: Yar Tikhiy Cc: mats@snowbee.dyns.cx, FreeBSD-gnats-submit@freebsd.org Subject: Re: bin/56325: Incorrect information in /etc/gettytab Date: Tue, 9 Sep 2003 03:14:29 +1000 (EST) On Mon, 8 Sep 2003, Yar Tikhiy wrote: > On Sun, Sep 07, 2003 at 08:04:01PM +1000, Bruce Evans wrote: > > I looked at what the various combinations of parity did when I wrote > > thos lines in gettytab. I haven't looked at them recently (i never > > use anything except np). Perhaps changing getty to termios canonicalized > > the parity support. > > Indeed, it was rev. 1.6 of subr.c that introduced the new behaviour > of the parity flags along with using termios instead of the old BSD > interface to terminals. > > I believe that people use just np nowadays, so the question is > whether the old behaviour of parity flag combinations is worth > restoring. How do you think? Not having it is OK with me. I used just np back when I wrote those lines in gettytab :-). Bruce From: Yar Tikhiy To: Bruce Evans Cc: mats@snowbee.dyns.cx, FreeBSD-gnats-submit@freebsd.org Subject: Re: bin/56325: Incorrect information in /etc/gettytab Date: Tue, 9 Sep 2003 14:27:48 +0400 On Tue, Sep 09, 2003 at 03:14:29AM +1000, Bruce Evans wrote: > > > > I believe that people use just np nowadays, so the question is > > whether the old behaviour of parity flag combinations is worth > > restoring. How do you think? > > Not having it is OK with me. I used just np back when I wrote those > lines in gettytab :-). As long as nobody has actually needed the old functionality in the last 9 years, I'll just match the comment at the top of gettytab with the harsh reality of the XXI century :-) -- Yar From: Mats Peterson To: bde@zeta.org.au, yar@freebsd.org Cc: FreeBSD-gnats-submit@freebsd.org, mats@snowbee.dyns.cx Subject: Re: bin/56325: Incorrect information in /etc/gettytab Date: Tue, 9 Sep 2003 13:12:56 +0200 (CEST) > On Tue, Sep 09, 2003 at 03:14:29AM +1000, Bruce Evans wrote: > > > > > > I believe that people use just np nowadays, so the question is > > > whether the old behaviour of parity flag combinations is worth > > > restoring. How do you think? > > > > Not having it is OK with me. I used just np back when I wrote those > > lines in gettytab :-). > > As long as nobody has actually needed the old functionality in the > last 9 years, I'll just match the comment at the top of gettytab > with the harsh reality of the XXI century :-) > > -- > Yar > It's still nice if it's there for a reason, though. EVEN if most people use no parity nowadays. I had to look at the source to find out why it didn't work as expected. I guess not everyone is too keen on doing that... Mats From: Yar Tikhiy To: Mats Peterson Cc: bde@zeta.org.au, FreeBSD-gnats-submit@freebsd.org Subject: Re: bin/56325: Incorrect information in /etc/gettytab Date: Tue, 9 Sep 2003 18:34:32 +0400 On Tue, Sep 09, 2003 at 01:12:56PM +0200, Mats Peterson wrote: > > It's still nice if it's there for a reason, though. EVEN if most people > use no parity nowadays. I had to look at the source to find out why it > didn't work as expected. I guess not everyone is too keen on doing that... I'm not speaking of removing the parity capabilities completely. However, I'd rather drop most double combinations of them because I can hardly see how serial/terminal hardware would do "-parenb inpck" (np:ep:), leave alone "-parenb inpck parodd" (np:op:). Bruce simply documented the present state of getty code when he wrote the comment in gettytab 9 years ago. I believe it's high time to revise it. Perhaps the only combination that makes sense is np:ap: which used to set istrip without turning on parity stuff. -- Yar State-Changed-From-To: open->patched State-Changed-By: yar State-Changed-When: Tue Sep 9 08:37:47 PDT 2003 State-Changed-Why: The comment in gettytab(5) has been revised. Responsible-Changed-From-To: freebsd-bugs->yar Responsible-Changed-By: yar Responsible-Changed-When: Tue Sep 9 08:37:47 PDT 2003 Responsible-Changed-Why: I'm taking this one. http://www.freebsd.org/cgi/query-pr.cgi?pr=56325 From: mats@snowbee.dyns.cx To: freebsd-gnats-submit@FreeBSD.org, mats@snowbee.dyns.cx Cc: mats@snowbee.dyns.cx Subject: Re: bin/56325: Incorrect information in /etc/gettytab Date: Tue, 9 Sep 2003 17:55:38 +0200 (CEST) > > It's still nice if it's there for a reason, though. EVEN if most people > > use no parity nowadays. I had to look at the source to find out why it > > didn't work as expected. I guess not everyone is too keen on doing that... > > I'm not speaking of removing the parity capabilities completely. > However, I'd rather drop most double combinations of them because > I can hardly see how serial/terminal hardware would do "-parenb > inpck" (np:ep:), leave alone "-parenb inpck parodd" (np:op:). Bruce > simply documented the present state of getty code when he wrote the > comment in gettytab 9 years ago. I believe it's high time to revise > it. Perhaps the only combination that makes sense is np:ap: which > used to set istrip without turning on parity stuff. > I wasn't referring to the parity when I said "It's still nice if it's there for a reason", but to the lines in gettytab, and that it's nice if they somehow reflect the functionality in getty. Sorry for being unclear. They were surely valid in the past, but apparently were overlooked when getty was modified later on. Regards, Mats From: Yar Tikhiy To: mats@snowbee.dyns.cx Cc: freebsd-gnats-submit@FreeBSD.org Subject: Re: bin/56325: Incorrect information in /etc/gettytab Date: Tue, 9 Sep 2003 20:10:50 +0400 On Tue, Sep 09, 2003 at 09:00:34AM -0700, mats@snowbee.dyns.cx wrote: > > I wasn't referring to the parity when I said "It's still nice if it's > there for a reason", but to the lines in gettytab, and that it's nice > if they somehow reflect the functionality in getty. Sorry for being > unclear. They were surely valid in the past, but apparently were > overlooked when getty was modified later on. I've just revised the lines so they match getty(8). Would you mind taking a glance at the new revision? Thanks! -- Yar From: Mats Peterson To: mats@snowbee.dyns.cx, yar@comp.chem.msu.su Cc: freebsd-gnats-submit@FreeBSD.org Subject: Re: bin/56325: Incorrect information in /etc/gettytab Date: Tue, 9 Sep 2003 18:23:35 +0200 (CEST) > On Tue, Sep 09, 2003 at 09:00:34AM -0700, mats@snowbee.dyns.cx wrote: > > > > I wasn't referring to the parity when I said "It's still nice if it's > > there for a reason", but to the lines in gettytab, and that it's nice > > if they somehow reflect the functionality in getty. Sorry for being > > unclear. They were surely valid in the past, but apparently were > > overlooked when getty was modified later on. > > I've just revised the lines so they match getty(8). Would you mind > taking a glance at the new revision? Thanks! > > -- > Yar > Sorry for being ignorant, but where is it? Mats From: Yar Tikhiy To: Mats Peterson Cc: freebsd-gnats-submit@FreeBSD.org Subject: Re: bin/56325: Incorrect information in /etc/gettytab Date: Tue, 9 Sep 2003 20:28:51 +0400 On Tue, Sep 09, 2003 at 06:23:35PM +0200, Mats Peterson wrote: > > > > I've just revised the lines so they match getty(8). Would you mind > > taking a glance at the new revision? Thanks! > > Sorry for being ignorant, but where is it? For instance, there: http://www.freebsd.org/cgi/cvsweb.cgi/src/etc/gettytab?rev=HEAD -- Yar From: Mats Peterson To: mats@snowbee.dyns.cx, yar@FreeBSD.org Cc: freebsd-gnats-submit@FreeBSD.org Subject: Re: bin/56325: Incorrect information in /etc/gettytab Date: Tue, 9 Sep 2003 18:37:26 +0200 (CEST) > On Tue, Sep 09, 2003 at 06:23:35PM +0200, Mats Peterson wrote: > > > > > > I've just revised the lines so they match getty(8). Would you mind > > > taking a glance at the new revision? Thanks! > > > > Sorry for being ignorant, but where is it? > > For instance, there: > > http://www.freebsd.org/cgi/cvsweb.cgi/src/etc/gettytab?rev=HEAD > > -- > Yar > Looks fine to me:) I'm no authority really, but it matches the internal behaviour of getty much better now. The most important thing was to get rid of those np:ep, np:op and np:ap lines which were totally bogus... Mats From: Yar Tikhiy To: Mats Peterson Cc: freebsd-gnats-submit@FreeBSD.org Subject: Re: bin/56325: Incorrect information in /etc/gettytab Date: Wed, 10 Sep 2003 09:00:16 +0400 On Tue, Sep 09, 2003 at 06:37:26PM +0200, Mats Peterson wrote: > > Looks fine to me:) I'm no authority really, but it matches the internal > behaviour of getty much better now. Your review is much appreciated. Thank you! In a couple of days I'll submit my change for approval by a Release Engineer so it will appear in 4.9-RELEASE, too. > The most important thing was to get rid of those np:ep, np:op and np:ap > lines which were totally bogus... Yeah, it's exactly what I thought about them :-) -- Yar From: Mats Peterson To: mats@snowbee.dyns.cx, yar@FreeBSD.org Cc: freebsd-gnats-submit@FreeBSD.org Subject: Re: bin/56325: Incorrect information in /etc/gettytab Date: Wed, 10 Sep 2003 09:01:33 +0200 (CEST) > > On Tue, Sep 09, 2003 at 06:37:26PM +0200, Mats Peterson wrote: > > > > Looks fine to me:) I'm no authority really, but it matches the internal > > behaviour of getty much better now. > > Your review is much appreciated. Thank you! > In a couple of days I'll submit my change for approval by a Release > Engineer so it will appear in 4.9-RELEASE, too. > Thank you too. > > The most important thing was to get rid of those np:ep, np:op and np:ap > > lines which were totally bogus... > > Yeah, it's exactly what I thought about them :-) > > -- > Yar > :) State-Changed-From-To: patched->closed State-Changed-By: yar State-Changed-When: Tue Sep 16 02:34:45 PDT 2003 State-Changed-Why: The comment paragraph on parity capabilities in /etc/gettytab has been corrected in both active branches. http://www.freebsd.org/cgi/query-pr.cgi?pr=56325 >Unformatted: