mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Steven Rostedt <rostedt@goodmis.org>
To: Peter Zijlstra <peterz@infradead.org>
Cc: linux-kernel@vger.kernel.org, Ingo Molnar <mingo@elte.hu>,
	Andrew Morton <akpm@linux-foundation.org>,
	Frederic Weisbecker <fweisbec@gmail.com>,
	Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>,
	Jason Baron <jbaron@redhat.com>
Subject: Re: [PATCH 4/9] jump label: Fix deadlock b/w jump_label_mutex vs. text_mutex
Date: Sat, 16 Oct 2010 00:21:06 -0400	[thread overview]
Message-ID: <1287202866.16971.47.camel@gandalf.stny.rr.com> (raw)
In-Reply-To: <1287195949.16971.45.camel@gandalf.stny.rr.com>

On Fri, 2010-10-15 at 22:25 -0400, Steven Rostedt wrote:
> On Fri, 2010-10-15 at 22:55 +0200, Peter Zijlstra wrote:
> > On Fri, 2010-10-15 at 16:09 -0400, Steven Rostedt wrote:
> > > +void jump_label_lock(void)
> > > +{
> > > +       mutex_lock(&jump_label_mutex);
> > > +}
> > > +
> > 
> > > +++ b/kernel/kprobes.c
> > > @@ -1145,13 +1145,16 @@ int __kprobes register_kprobe(struct kprobe *p)
> > >                 return ret;
> > >  
> > >         preempt_disable();
> > > +       jump_label_lock();
> > 
> > How exactly does that work?
> > 
> 
> This patch should fix it (compiled tested only). If all agree, I'll add
> it to the series.

Well, I just did more than compile test it. I booted first without this
patch, and added a kprobe and lockdep spewed its guts about this
condition. Then I applied the patch, compiled and rebooted, tried the
probe again, and lockdep stayed quiet :-)

-- Steve


> diff --git a/kernel/kprobes.c b/kernel/kprobes.c
> index d45d72e..0dbd328 100644
> --- a/kernel/kprobes.c
> +++ b/kernel/kprobes.c
> @@ -1144,17 +1144,13 @@ int __kprobes register_kprobe(struct kprobe *p)
>  	if (ret)
>  		return ret;
>  
> -	preempt_disable();
>  	jump_label_lock();
> +	preempt_disable();
>  	if (!kernel_text_address((unsigned long) p->addr) ||
>  	    in_kprobes_functions((unsigned long) p->addr) ||
>  	    ftrace_text_reserved(p->addr, p->addr) ||
> -	    jump_label_text_reserved(p->addr, p->addr)) {
> -		preempt_enable();
> -		jump_label_unlock();
> -		return -EINVAL;
> -	}
> -	jump_label_unlock();
> +	    jump_label_text_reserved(p->addr, p->addr))
> +		goto fail_with_jump_label;
>  
>  	/* User can pass only KPROBE_FLAG_DISABLED to register_kprobe */
>  	p->flags &= KPROBE_FLAG_DISABLED;
> @@ -1168,10 +1164,9 @@ int __kprobes register_kprobe(struct kprobe *p)
>  		 * We must hold a refcount of the probed module while updating
>  		 * its code to prohibit unexpected unloading.
>  		 */
> -		if (unlikely(!try_module_get(probed_mod))) {
> -			preempt_enable();
> -			return -EINVAL;
> -		}
> +		if (unlikely(!try_module_get(probed_mod)))
> +			goto fail_with_jump_label;
> +
>  		/*
>  		 * If the module freed .init.text, we couldn't insert
>  		 * kprobes in there.
> @@ -1179,11 +1174,11 @@ int __kprobes register_kprobe(struct kprobe *p)
>  		if (within_module_init((unsigned long)p->addr, probed_mod) &&
>  		    probed_mod->state != MODULE_STATE_COMING) {
>  			module_put(probed_mod);
> -			preempt_enable();
> -			return -EINVAL;
> +			goto fail_with_jump_label;
>  		}
>  	}
>  	preempt_enable();
> +	jump_label_unlock();
>  
>  	p->nmissed = 0;
>  	INIT_LIST_HEAD(&p->list);
> @@ -1225,6 +1220,11 @@ out:
>  		module_put(probed_mod);
>  
>  	return ret;
> +
> +fail_with_jump_label:
> +	preempt_enable();
> +	jump_label_unlock();
> +	return -EINVAL;
>  }
>  EXPORT_SYMBOL_GPL(register_kprobe);
>  
> 



  reply	other threads:[~2010-10-16  4:21 UTC|newest]

Thread overview: 40+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-10-15 20:09 [PATCH 0/9] [GIT PULL] jump label: various updates Steven Rostedt
2010-10-15 20:09 ` [PATCH 1/9] jump label/x86: Move arch_init_ideal_nop5 later Steven Rostedt
2010-10-15 20:09 ` [PATCH 2/9] tracing/x86: No need to disable interrupts when calling arch_init_ideal_nop5 Steven Rostedt
2010-10-15 20:09 ` [PATCH 3/9] jump label: Fix module __init section race Steven Rostedt
2010-10-16  2:09   ` Steven Rostedt
2010-10-16  6:23     ` Ingo Molnar
2010-10-16 16:23       ` Steven Rostedt
2010-10-18 14:14         ` Jason Baron
2010-10-15 20:09 ` [PATCH 4/9] jump label: Fix deadlock b/w jump_label_mutex vs. text_mutex Steven Rostedt
2010-10-15 20:55   ` Peter Zijlstra
2010-10-16  2:16     ` Steven Rostedt
2010-10-16  2:25     ` Steven Rostedt
2010-10-16  4:21       ` Steven Rostedt [this message]
2010-10-15 20:09 ` [PATCH 5/9] jump label: Add register_jump_label_key/unregister_jump_label_key Steven Rostedt
2010-10-15 20:58   ` Peter Zijlstra
2010-10-16  2:11     ` Steven Rostedt
2010-10-15 21:03   ` Peter Zijlstra
2010-10-15 21:09     ` Steven Rostedt
2010-10-15 21:13       ` Peter Zijlstra
2010-10-18 12:05         ` Peter Zijlstra
2010-10-18 14:03           ` Jason Baron
2010-10-18 14:07             ` Peter Zijlstra
2010-10-15 20:09 ` [PATCH 6/9] jump label: Move jump table to r/w section Steven Rostedt
2010-10-15 20:09 ` [PATCH 7/9] jump label: Add docs Steven Rostedt
2010-10-15 20:09 ` [PATCH 8/9] jump label: Make arch_jump_label_text_poke_early() optional Steven Rostedt
2010-10-15 21:06   ` Peter Zijlstra
2010-10-15 21:08     ` Steven Rostedt
2010-10-15 21:22       ` David Daney
2010-10-15 21:35         ` Steven Rostedt
2010-10-15 21:38           ` David Daney
2010-10-16  1:08             ` Steven Rostedt
2010-10-30 10:40   ` [tip:perf/urgent] " tip-bot for Steven Rostedt
2010-10-15 20:09 ` [PATCH 9/9] jump label: Add MIPS support Steven Rostedt
2010-10-15 20:17 ` [PATCH 0/9] [GIT PULL] jump label: various updates Steven Rostedt
2010-10-16  1:10 ` Steven Rostedt
  -- strict thread matches above, loose matches on Subject: below --
2010-10-04 18:56 [PATCH v2 0/2] jump label: Add MIPS architecture support David Daney
2010-10-04 18:56 ` [PATCH v2 1/2] jump label: Make arch_jump_label_text_poke_early() optional David Daney
2010-10-04 18:56 ` [PATCH v2 2/2] jump label: Add MIPS support David Daney
2010-10-06 23:00   ` Ralf Baechle
2010-10-07  1:26     ` David Daney

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1287202866.16971.47.camel@gandalf.stny.rr.com \
    --to=rostedt@goodmis.org \
    --cc=akpm@linux-foundation.org \
    --cc=fweisbec@gmail.com \
    --cc=jbaron@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=masami.hiramatsu.pt@hitachi.com \
    --cc=mingo@elte.hu \
    --cc=peterz@infradead.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®