linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Arnaldo Carvalho de Melo <arnaldo.melo@gmail.com>
To: Jiri Olsa <jolsa@redhat.com>
Cc: David Ahern <dsahern@gmail.com>, Jiri Olsa <jolsa@kernel.org>,
	Namhyung Kim <namhyung@kernel.org>,
	Stephane Eranian <eranian@google.com>,
	LKML <linux-kernel@vger.kernel.org>
Subject: Re: [BUG] segfault in perf-top -- thread refcnt
Date: Mon, 30 Mar 2015 10:09:40 -0300	[thread overview]
Message-ID: <20150330130940.GB24063@kernel.org> (raw)
In-Reply-To: <20150330102220.GE1413@krava>

Em Mon, Mar 30, 2015 at 12:22:20PM +0200, Jiri Olsa escreveu:
> On Mon, Mar 30, 2015 at 10:07:37AM +0200, Jiri Olsa wrote:
> 
> SNIP
> 
> > > 
> > > 2 things:
> > > 1. let run for a long time. go about using the server. do lots of builds,
> > > etc. it takes time
> > > 
> > > 2. use a box with a LOT of cpus (1024 in my case)
> > > 
> > > Make sure ulimit is set to get the core.
> > 
> > reproduced under 24 cpu box with kernel build (make -j25)
> > running on background.. will try to look closer
> > 
> > perf: Segmentation fault
> > -------- backtrace --------
> > ./perf[0x4fd79b]
> > /lib64/libc.so.6(+0x358f0)[0x7f9cbff528f0]
> > ./perf(thread__put+0x5b)[0x4b1a7b]
> > ./perf(hists__delete_entries+0x70)[0x4c8670]
> > ./perf[0x436a88]
> > ./perf[0x4fa73d]
> > ./perf(perf_evlist__tui_browse_hists+0x97)[0x4fc437]
> > ./perf[0x4381d0]
> > /lib64/libpthread.so.0(+0x7ee5)[0x7f9cc1ff2ee5]
> > /lib64/libc.so.6(clone+0x6d)[0x7f9cc0011b8d]
> > [0x0]
> 
> looks like race among __machine__findnew_thread and thread__put
> over the machine->threads rb_tree insert/removal
> 
> is there a reason why thread__put does not erase itself from machine->threads?

Thread put should only remove stuff from the rbtree when the refcount
hits zero, and then, it should grab a lock, that is the buggy part, as
this lock isn't there, and remove the entry from the rbtree. Now I see
what Namhyung was talking about...

- Arnaldo
 
> I'm trying attached patch.. so far so gut ;-)
> 
> jirka
> 
> 
> ---
> diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c
> index e335330..7e6abc7 100644
> --- a/tools/perf/util/machine.c
> +++ b/tools/perf/util/machine.c
> @@ -30,6 +30,7 @@ int machine__init(struct machine *machine, const char *root_dir, pid_t pid)
>  	dsos__init(&machine->kernel_dsos);
>  
>  	machine->threads = RB_ROOT;
> +	pthread_mutex_init(&machine->threads_lock, NULL);
>  	INIT_LIST_HEAD(&machine->dead_threads);
>  	machine->last_match = NULL;
>  
> @@ -380,10 +381,13 @@ static struct thread *__machine__findnew_thread(struct machine *machine,
>  	if (!create)
>  		return NULL;
>  
> -	th = thread__new(pid, tid);
> +	th = thread__new(machine, pid, tid);
>  	if (th != NULL) {
> +
> +		pthread_mutex_lock(&machine->threads_lock);
>  		rb_link_node(&th->rb_node, parent, p);
>  		rb_insert_color(&th->rb_node, &machine->threads);
> +		pthread_mutex_unlock(&machine->threads_lock);
>  
>  		/*
>  		 * We have to initialize map_groups separately
> @@ -394,8 +398,10 @@ static struct thread *__machine__findnew_thread(struct machine *machine,
>  		 * leader and that would screwed the rb tree.
>  		 */
>  		if (thread__init_map_groups(th, machine)) {
> +			pthread_mutex_lock(&machine->threads_lock);
>  			rb_erase(&th->rb_node, &machine->threads);
>  			thread__delete(th);
> +			pthread_mutex_unlock(&machine->threads_lock);
>  			return NULL;
>  		}
>  		/*
> diff --git a/tools/perf/util/machine.h b/tools/perf/util/machine.h
> index e2faf3b..e3468d6 100644
> --- a/tools/perf/util/machine.h
> +++ b/tools/perf/util/machine.h
> @@ -30,6 +30,7 @@ struct machine {
>  	bool		  comm_exec;
>  	char		  *root_dir;
>  	struct rb_root	  threads;
> +	pthread_mutex_t	  threads_lock;
>  	struct list_head  dead_threads;
>  	struct thread	  *last_match;
>  	struct vdso_info  *vdso_info;
> diff --git a/tools/perf/util/thread.c b/tools/perf/util/thread.c
> index 1c8fbc9..4592fc4 100644
> --- a/tools/perf/util/thread.c
> +++ b/tools/perf/util/thread.c
> @@ -26,7 +26,7 @@ int thread__init_map_groups(struct thread *thread, struct machine *machine)
>  	return thread->mg ? 0 : -1;
>  }
>  
> -struct thread *thread__new(pid_t pid, pid_t tid)
> +struct thread *thread__new(struct machine *machine, pid_t pid, pid_t tid)
>  {
>  	char *comm_str;
>  	struct comm *comm;
> @@ -38,6 +38,7 @@ struct thread *thread__new(pid_t pid, pid_t tid)
>  		thread->ppid = -1;
>  		thread->cpu = -1;
>  		INIT_LIST_HEAD(&thread->comm_list);
> +		thread->machine = machine;
>  
>  		if (unwind__prepare_access(thread) < 0)
>  			goto err_thread;
> @@ -91,7 +92,14 @@ struct thread *thread__get(struct thread *thread)
>  void thread__put(struct thread *thread)
>  {
>  	if (thread && --thread->refcnt == 0) {
> +		struct machine *machine = thread->machine;
> +
>  		list_del_init(&thread->node);
> +
> +		pthread_mutex_lock(&machine->threads_lock);
> +		rb_erase(&thread->rb_node, &machine->threads);
> +		pthread_mutex_unlock(&machine->threads_lock);
> +
>  		thread__delete(thread);
>  	}
>  }
> diff --git a/tools/perf/util/thread.h b/tools/perf/util/thread.h
> index 9b8a54d..df6fb69 100644
> --- a/tools/perf/util/thread.h
> +++ b/tools/perf/util/thread.h
> @@ -31,12 +31,13 @@ struct thread {
>  
>  	void			*priv;
>  	struct thread_stack	*ts;
> +	struct machine		*machine;
>  };
>  
>  struct machine;
>  struct comm;
>  
> -struct thread *thread__new(pid_t pid, pid_t tid);
> +struct thread *thread__new(struct machine *machine, pid_t pid, pid_t tid);
>  int thread__init_map_groups(struct thread *thread, struct machine *machine);
>  void thread__delete(struct thread *thread);
>  

  parent reply	other threads:[~2015-03-30 13:09 UTC|newest]

Thread overview: 24+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-03-27 17:31 [BUG] segfault in perf-top -- thread refcnt David Ahern
2015-03-27 19:51 ` Arnaldo Carvalho de Melo
2015-03-27 20:11 ` Arnaldo Carvalho de Melo
2015-03-27 20:13   ` David Ahern
2015-03-30  8:07     ` Jiri Olsa
2015-03-30 10:22       ` Jiri Olsa
2015-03-30 11:21         ` Jiri Olsa
2015-03-30 11:49           ` Jiri Olsa
2015-03-30 12:48             ` Namhyung Kim
2015-03-30 12:56               ` Jiri Olsa
2015-03-30 13:06                 ` Namhyung Kim
2015-03-30 14:02                   ` Arnaldo Carvalho de Melo
2015-03-31  0:15                     ` Namhyung Kim
2015-03-30 13:07                 ` Arnaldo Carvalho de Melo
2015-03-30 13:20                   ` Jiri Olsa
2015-03-30 13:59                     ` Arnaldo Carvalho de Melo
2015-03-30 14:58               ` Arnaldo Carvalho de Melo
2015-03-30 15:13                 ` Arnaldo Carvalho de Melo
2015-03-31  0:27                   ` Namhyung Kim
2015-03-31  0:46                     ` Arnaldo Carvalho de Melo
2015-03-31  7:21                       ` Namhyung Kim
2015-03-30 13:22             ` Arnaldo Carvalho de Melo
2015-03-30 13:09         ` Arnaldo Carvalho de Melo [this message]
2015-03-30 13:17         ` Arnaldo Carvalho de Melo

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=20150330130940.GB24063@kernel.org \
    --to=arnaldo.melo@gmail.com \
    --cc=dsahern@gmail.com \
    --cc=eranian@google.com \
    --cc=jolsa@kernel.org \
    --cc=jolsa@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=namhyung@kernel.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;
as well as URLs for NNTP newsgroup(s).