From owner-freebsd-current Wed Aug 5 02:49:40 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id CAA17539 for freebsd-current-outgoing; Wed, 5 Aug 1998 02:49:40 -0700 (PDT) (envelope-from owner-freebsd-current@FreeBSD.ORG) Received: from cons.org (knight.cons.org [194.233.237.86]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id CAA17534 for ; Wed, 5 Aug 1998 02:49:38 -0700 (PDT) (envelope-from cracauer@cons.org) Received: (from cracauer@localhost) by cons.org (8.8.8/8.7.3) id LAA25974; Wed, 5 Aug 1998 11:48:26 +0200 (CEST) Message-ID: <19980805114825.A25826@cons.org> Date: Wed, 5 Aug 1998 11:48:25 +0200 From: Martin Cracauer To: Bruce Evans , cracauer@cons.org, freebsd-current@FreeBSD.ORG Subject: Re: make/SIGINT (Re: cvs commit: src/bin/sh jobs.c) References: <199805221609.CAA04233@godzilla.zeta.org.au> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.93.1i In-Reply-To: <199805221609.CAA04233@godzilla.zeta.org.au>; from Bruce Evans on Sat, May 23, 1998 at 02:09:25AM +1000 Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG In <199805221609.CAA04233@godzilla.zeta.org.au>, Bruce Evans wrote: > >> I'm not happy with most of the patches. Even the tiny newline fix > >> in the signal handler doesn't work quite right. > > > >I assume you mean the fix to mkdep. Please take in mind that this > >particular fix depends on the latest sh diff I sent, otherwise it will > >not work. > > > >If it's something else, I'm afraid you need to be more precise. > > I mean the fix for calling out2str("\n") (possibly recursively) in sh's > signal handler. Newlines now get printed to often, so the signal handler > must be being called too often. You mean that one that is replaced by write(STDERR_FILENO, "\n", 1); in error.c:onint()? What happens is that nested shells (like in recursive makes or test04.sh of my testsuite) each print their own newline. The right thing would be one newline at all, not one per shell. onint is not being called more than once for each shell, so I consider this cosmetic. Another nitpick is that the newline and the shell prompt are sometimes being printed before the last line(s) put out by the compiler driver in a recursive make. I don't know what this implies, but it doesn't prevent the build from working and from being interruptable. I think some wrong newlines when killing builds should not prevent the fix from being commited, so that other people run their system with it and can point to other application programs that need to exit with proper status before the next release (I already spotted rsync). I put the newest fixes and the newest version of my testsuite on freefall:/home/cracauer/public_html (http://www.freebsd.org/~cracauer/) These include the wait fixes. If your observations were done without them, please repeat them with the newest version. Martin -- %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Martin Cracauer http://www.cons.org/cracauer cracauer@wavehh.hanse.de (batched, preferred for large mails) Tel.: (private) +4940 5221829 Fax.: (private) +4940 5228536 Paper: (private) Waldstrasse 200, 22846 Norderstedt, Germany To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message