From owner-freebsd-threads@FreeBSD.ORG  Wed May 15 09:06:15 2013
Return-Path: <owner-freebsd-threads@FreeBSD.ORG>
Delivered-To: threads@freebsd.org
Received: from mx1.freebsd.org (mx1.freebsd.org
 [IPv6:2001:1900:2254:206a::19:1])
 by hub.freebsd.org (Postfix) with ESMTP id 0EF87470;
 Wed, 15 May 2013 09:06:15 +0000 (UTC)
 (envelope-from davidxu@freebsd.org)
Received: from freefall.freebsd.org (freefall.freebsd.org
 [IPv6:2001:1900:2254:206c::16:87])
 by mx1.freebsd.org (Postfix) with ESMTP id 01579A33;
 Wed, 15 May 2013 09:06:15 +0000 (UTC)
Received: from xyf.my.dom (localhost [127.0.0.1])
 by freefall.freebsd.org (8.14.7/8.14.7) with ESMTP id r4F96DYK084352;
 Wed, 15 May 2013 09:06:14 GMT (envelope-from davidxu@freebsd.org)
Message-ID: <51935024.3080107@freebsd.org>
Date: Wed, 15 May 2013 17:06:44 +0800
From: David Xu <davidxu@freebsd.org>
User-Agent: Mozilla/5.0 (X11; FreeBSD i386;
 rv:17.0) Gecko/20130416 Thunderbird/17.0.5
MIME-Version: 1.0
To: Poul-Henning Kamp <phk@freebsd.org>
Subject: Re: pthread_key_create() should avoid key#0
References: <79159.1368525129@critter.freebsd.dk>
In-Reply-To: <79159.1368525129@critter.freebsd.dk>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Cc: threads@freebsd.org
X-BeenThere: freebsd-threads@freebsd.org
X-Mailman-Version: 2.1.14
Precedence: list
List-Id: Threading on FreeBSD <freebsd-threads.freebsd.org>
List-Unsubscribe: <http://lists.freebsd.org/mailman/options/freebsd-threads>, 
 <mailto:freebsd-threads-request@freebsd.org?subject=unsubscribe>
List-Archive: <http://lists.freebsd.org/pipermail/freebsd-threads>
List-Post: <mailto:freebsd-threads@freebsd.org>
List-Help: <mailto:freebsd-threads-request@freebsd.org?subject=help>
List-Subscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-threads>,
 <mailto:freebsd-threads-request@freebsd.org?subject=subscribe>
X-List-Received-Date: Wed, 15 May 2013 09:06:15 -0000

On 2013/05/14 17:52, Poul-Henning Kamp wrote:
>
> I think this patch would be a good idea, it avoids allocing
> a thread specific key with numeric value zero, which helps
> expose cases where a key hasn't been allocated in the first
> place.
>
>
> Index: libkse/thread/thr_spec.c
> ===================================================================
> --- libkse/thread/thr_spec.c	(revision 248293)
> +++ libkse/thread/thr_spec.c	(working copy)
> @@ -59,7 +59,7 @@
>
>   	/* Lock the key table: */
>   	THR_LOCK_ACQUIRE(curthread, &_keytable_lock);
> -	for (i = 0; i < PTHREAD_KEYS_MAX; i++) {
> +	for (i = 1; i < PTHREAD_KEYS_MAX; i++) {
>
>   		if (_thread_keytable[i].allocated == 0) {
>   			_thread_keytable[i].allocated = 1;
> @@ -84,7 +84,7 @@
>   	struct pthread *curthread = _get_curthread();
>   	int ret = 0;
>
> -	if ((unsigned int)key < PTHREAD_KEYS_MAX) {
> +	if (key > 0 && (unsigned int)key < PTHREAD_KEYS_MAX) {
>   		/* Lock the key table: */
>   		THR_LOCK_ACQUIRE(curthread, &_keytable_lock);
>
> Index: libthr/thread/thr_spec.c
> ===================================================================
> --- libthr/thread/thr_spec.c	(revision 248293)
> +++ libthr/thread/thr_spec.c	(working copy)
> @@ -61,7 +61,7 @@
>
>   	/* Lock the key table: */
>   	THR_LOCK_ACQUIRE(curthread, &_keytable_lock);
> -	for (i = 0; i < PTHREAD_KEYS_MAX; i++) {
> +	for (i = 1; i < PTHREAD_KEYS_MAX; i++) {
>
>   		if (_thread_keytable[i].allocated == 0) {
>   			_thread_keytable[i].allocated = 1;
> @@ -86,7 +86,7 @@
>   	struct pthread *curthread = _get_curthread();
>   	int ret = 0;
>
> -	if ((unsigned int)key < PTHREAD_KEYS_MAX) {
> +	if (key > 0 && (unsigned int)key < PTHREAD_KEYS_MAX) {
>   		/* Lock the key table: */
>   		THR_LOCK_ACQUIRE(curthread, &_keytable_lock);
>
>

while key#0 is avoided, I would like to see we still support
PTHREAD_KEYS_MAX keys. I am working on it.