linux-kernel - core_uses_pid==0 still produces core.PID

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]

Date:	Tue, 15 Jul 2008 12:55:04 +0200
From:	Philippe De Muyter <phdm@...qel.be>
To:	linux-kernel@...r.kernel.org
Subject: core_uses_pid==0 still produces core.PID

Hello everybody

I have a process that produces core files of the form "core.PID".

I read in the docs the core file name can be controlled by
/proc/sys/kernel/core_pattern and /proc/sys/kernel/core_uses_pid.

On my machine, they have the value "core" and 0, so I would expect the
core files named "core", not "core.PID".

If I start "sleep 100" and kill it with CTRL-C, I indeed get a file
named "core", but with my interesting process the files are always named
"core.PID".

Looking at the fs/exec.c sources, I see that core_uses_pid is superseded
by
	atomic_read(&current->mm->mm_users) != 1

Why is that needed there ? When my process crashes or is killed, I always
get only one core file, not many core files, although my process is
multi-threaded.

Is that a rest of a previous thread implementation that could now be removed ?

Best regards

Philippe
--
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/