linux-kernel - Re: [PATCH 01/17] ptrace: remove silly wait_trap variable from ptrace

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:	Thu, 2 Jun 2011 14:03:40 +0900
From:	Tejun Heo <tj@...nel.org>
To:	Oleg Nesterov <oleg@...hat.com>
Cc:	vda.linux@...glemail.com, jan.kratochvil@...hat.com,
	linux-kernel@...r.kernel.org, torvalds@...ux-foundation.org,
	akpm@...ux-foundation.org, indan@....nu, bdonlan@...il.com,
	pedro@...esourcery.com
Subject: Re: [PATCH 01/17] ptrace: remove silly wait_trap variable from
 ptrace_attach()

Hello,

On Wed, Jun 01, 2011 at 08:47:43PM +0200, Oleg Nesterov wrote:
> On 05/30, Tejun Heo wrote:
> > -	if (wait_trap)
> > -		wait_event(current->signal->wait_chldexit,
> > -			   !(task->group_stop & GROUP_STOP_TRAPPING));
> > +	wait_event(current->signal->wait_chldexit,
> > +		   !(task->group_stop & GROUP_STOP_TRAPPING));
> >  	return retval;
> 
> Well, it doesn't change anything, but only if ptrace_attach() succeeds.
> The caller should not wait if STOP_TRAPPING was already set by another
> tracer and we are going to fail. Afaics, nothing really bad can happen
> but still this doesn't look very clean.
> 
> And. Please note that this patch is buggy until 8/17 "use bit_waitqueue
> for TRAPPING", wait_event(current->signal->wait_chldexit) can hang forever
> in this case since we are not ->parent.

Ah, darn it.  It was part of later patch which got dropped and I just
salavaged this part and put it at the front forgetting about the
chldexit conversion.

> I agree, wait_trap should go away. We can hit STOP_TRAPPING after attach
> if we change detach to set this bit. But perhaps it would be more clean
> to not call wait_event/wait_bit unconditionally anyway, we can check
> retval == 0.

Yeap, that sounds about right.  Will post updated patch.

Thanks.

-- 
tejun
--
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/