Zsh Mailing List Archive
Messages sorted by:
Reverse Date,
Date,
Thread,
Author
Re: segfault in completion for configure
- X-seq: zsh-workers 38136
- From: Bart Schaefer <schaefer@xxxxxxxxxxxxxxxx>
- To: zsh-workers@xxxxxxx
- Subject: Re: segfault in completion for configure
- Date: Fri, 11 Mar 2016 15:00:56 -0800
- Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=brasslantern-com.20150623.gappssmtp.com; s=20150623; h=from:message-id:date:in-reply-to:comments:references:to:subject :mime-version; bh=GmmLP6e8ZylGoBUUq1m+Qy99KjYYEarsgb1riyg7N3E=; b=RnsioQpGz2goUB3mrURugwBlSPwbn7uEO1Zn0VEtd8CdvxRcacNuHQ8IckdtPVeLE3 p3TwPVB6xHiI8zkeXsp6pBZe3e2naBi8Sz3OmSVapTIBvrE0B0BMhwivGl8mO/k97BP3 hNu2vns61bmIlqGiHs2hc3v69ebW76U3aQzwV8ZaA5EQ+0YsysoOZS2sKQrb/7vz9/vr hHd6Rf+Ka9y04vgLFTpYtQAM/b4xIm25LT7kVgRFNZ5OjaJYi0WZ+aeO1s60FWtMqq6t dLwQ6xWG5xy7eesHQ3hbx6Lmopd3FaXM29jeP/24ABDffj3a1squJ8mS8lfPgVP5xJPh nIxg==
- In-reply-to: <20160311143202.4008e29b@pwslap01u.europe.root.pri>
- List-help: <mailto:zsh-workers-help@zsh.org>
- List-id: Zsh Workers List <zsh-workers.zsh.org>
- List-post: <mailto:zsh-workers@zsh.org>
- Mailing-list: contact zsh-workers-help@xxxxxxx; run by ezmlm
- References: <20160311134729.GA32476@cventin.lip.ens-lyon.fr> <20160311143202.4008e29b@pwslap01u.europe.root.pri>
On Mar 11, 2:32pm, Peter Stephenson wrote:
}
} On Fri, 11 Mar 2016 14:47:29 +0100
} Vincent Lefevre <vincent@xxxxxxxxxx> wrote:
} > With zsh 5.2 under Debian/unstable, zsh sometimes crashes when
} > I do:
} >
} > cventin:~/software/mutt/mutt> ./configure --[TAB]
}
} Looks like something's already gone wrong by this point
Yes, this is the kind of crash where the stack trace is probably not
very helpful, because frames have returned and/or new frames entered
an unknown number of times after the pointers went awry but before
the bad dereference that trips the fault.
Shot in the [mostly] dark: I see TRAPCHLD embedded in one of the strings
in your backtrace. Completion after "configure --" forks at least one
child process (says valgrind) so if you have a TRAPCHLD function it's
possible that it's corrupting something. We do attempt signal handling
during pattern matching in order that deep recursive globs can be stopped
with SIGINT; it could be that needs a bit more regulation.
Messages sorted by:
Reverse Date,
Date,
Thread,
Author