linux-kernel - signalfd() question - interaction with threads

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [day] [month] [year] [list]

Message-ID: <BANLkTi=WU6h_0jOVL=_VwJawZd4qVnU-EA@mail.gmail.com>
Date:	Thu, 16 Jun 2011 10:24:44 -0400
From:	Colin Walters <walters@...bum.org>
To:	LKML <linux-kernel@...r.kernel.org>
Subject: signalfd() question - interaction with threads

Hi,

So in a bit of spare time I have been trying to make GLib use some
more modern Linux features, in particular signalfd().  The quick link
for that is here:
https://bugzilla.gnome.org/show_bug.cgi?id=652072

But I ran into a problem - I don't necessarily control all of the
threads in the process, and if any one of them has the signal
unblocked, it will be handled according to its default disposition
there instead of where I want - in the signalfd.

The signalfd example in the man page explicitly documents that you
need to sigprocmask(SIG_BLOCK), which is fine if you control all the
threads in your app, but I don't.  Any chance of a new flag like
SFD_HANDLE or SFD_ALWAYS which was rougly equivalent to using
sigaction()?  Or even better as I say in the bug, *always* deliver the
signal so I can get an event when a child exits even in the case where
some other code uses sigaction(SIGCHLD)?
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/