linux-kernel - Re: [PATCH] x86_64: fix delayed signals

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

Date:	Sat, 12 Jul 2008 10:41:39 -0700 (PDT)
From:	Linus Torvalds <torvalds@...ux-foundation.org>
To:	Andi Kleen <andi@...stfloor.org>
cc:	Ingo Molnar <mingo@...e.hu>, Roland McGrath <roland@...hat.com>,
	Thomas Gleixner <tglx@...utronix.de>,
	Andrew Morton <akpm@...ux-foundation.org>,
	linux-kernel@...r.kernel.org, Elias Oltmanns <eo@...ensachen.de>,
	Török Edwin <edwintorok@...il.com>,
	Arjan van de Ven <arjan@...radead.org>
Subject: Re: [PATCH] x86_64: fix delayed signals

On Sat, 12 Jul 2008, Andi Kleen wrote:
> 
> Linus, the fact that is not explained by your theory is why
> Ctrl-Z+kill works but Ctrl-C doesn't.

Umm. Read the reports more carefully. Many of the complaints are about ^Z 
too.

More-over, ^Z and ^C are very different from an app standpoint. A _lot_ of 
applications block or catch ^C (SIGINT), while doing the same for ^Z 
(SIGTSTP) and SIGTERM is much less common (but possible: a tty sends 
SIGTSTP, which is blockable, not SIGSTOP, which is not).

The reason? It's quite common to catch ^C in order to do cleanup. The 
common handler case is that you catch SIGINT, do cleanup, and then kill 
yourself with SIGINT in order to make it _look_ like you reacted directly 
to the ^C.

So there are a _lot_ of loads where ^C will not kill things immediately, 
but ^Z + SIGKILL (or SIGTERM, the default for "sig") will.

Try doing a "yum update", for example.

Or use 'git', for that matter. git does exactly that kind of thing: 
catches SIGINT and removes temp-files etc. But if you kill it manually 
using some "stronger" signal (and using ^Z to get the shell back before 
you do that is obviously one way), that won't happen.

And trust me, git is not at all unusual in this respect.

			Linus
--
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/