Date: Thu, 5 Jul 2012 16:29:52 -0700 (PDT) From: Pedro Giffuni <pfg@freebsd.org> To: Brandon Gooch <jamesbrandongooch@gmail.com>, Peter Jeremy <peter@rulingia.com> Cc: David Shao <davshao@gmail.com>, freebsd-stable@freebsd.org, Jilles Tjoelker <jilles@stack.nl> Subject: Re: sh(1) exiting on SIGWINCH Message-ID: <1341530992.43915.YahooMailClassic@web113511.mail.gq1.yahoo.com> In-Reply-To: <20120705223952.GD85696@server.rulingia.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Hello guys;=0A=0A--- Gio 5/7/12, Peter Jeremy <peter@rulingia.com> ha scrit= to:=0A=0A> Brandon Gooch <jamesbrandongooch@gmail.com> wrote:=0A> >Seems th= at the window resize is somehow causing sh(1) to receive an=0A> >EOF while = the shell is sitting at the prompt, which results in the=0A> >shell exiting= ; haven't dug too deeply into the source yet, but can you=0A> >try to run /= bin/sh with the '-I' (that's capital letter 'i') and it=0A> >the shell shou= ldn't exit (but it will bark at you with a 'Use "exit"=0A> >to leave shell.= ' message on each resize.=0A> =0A> Interesting.=A0 I hadn't tried '-I' but = now also see=0A> that.=0A> =0A> >I'm CC'ing jilles@ for any potential insig= ht into the behavior of=0A> >sh(1) (and perhaps this updated libedit snapsh= ot).=0A> =0A> I would also welcome any insights jilles@ can offer.=0A> =0A>= pfg@ (who shepherded the libedit update into the tree), David Shao=0A> (or= iginator of kern/169603) and I have been investigating=0A> fixes to libedit= but do not have a solution yet.=A0 There is a=0A> possibility that sh(1) i= s relying on bugs in the old libedit.=0A> =0A> At this stage, it seems like= ly that the libedit update=0A> (r237738) will be reverted for 9.1-RELEASE.= =0A> =0A=0AI am trying hard to identify the specific changes from=0ANetBSD = that caused the issue. The problem is known to=0Ahappen in the latest versi= on of the upstream sources=0Atoo but I don't want to revert all the patch i= f I can=0Aavoid it because it is becoming increasingly difficult=0Aupdate l= ibedit and we need some of the new features.=0A=0AIt seems like David Shao'= s issue is different but so=0Afar I am aware that the issue was introduced = in the=0Achanges associated with read.c=0A=0Ahttp://cvsweb.netbsd.org/bsdwe= b.cgi/src/lib/libedit/read.c=0A=0ANetBSD fixed the SIGWINCH issue in revisi= ons 1.55, 1.56=0Aand 1.57 of that file but it was caused by revision 1.43= =0A=0Athe EOF issue seems to be caused by Revision 1.49 =0A=0AI still have = to play more with this.=0A=0APedro.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?1341530992.43915.YahooMailClassic>