All of lore.kernel.org
 help / color / mirror / Atom feed
From: Catalin Marinas <catalin.marinas@arm.com>
To: majianpeng <majianpeng@gmail.com>
Cc: linux-mm <linux-mm@kvack.org>,
	linux-kernel <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH 2/3] mm/kmemleak.c: Use list_for_each_entry_safe to reconstruct function scan_gray_list
Date: Thu, 30 May 2013 15:40:28 +0100	[thread overview]
Message-ID: <20130530144028.GF23631@arm.com> (raw)
In-Reply-To: <519224D8.5090704@gmail.com>

On Tue, May 14, 2013 at 12:49:44PM +0100, majianpeng wrote:
> Signed-off-by: Jianpeng Ma <majianpeng@gmail.com>
> ---
>  mm/kmemleak.c | 8 +-------
>  1 file changed, 1 insertion(+), 7 deletions(-)
> 
> diff --git a/mm/kmemleak.c b/mm/kmemleak.c
> index b1525db..f0ece93 100644
> --- a/mm/kmemleak.c
> +++ b/mm/kmemleak.c
> @@ -1225,22 +1225,16 @@ static void scan_gray_list(void)
>  	 * from inside the loop. The kmemleak objects cannot be freed from
>  	 * outside the loop because their use_count was incremented.
>  	 */
> -	object = list_entry(gray_list.next, typeof(*object), gray_list);
> -	while (&object->gray_list != &gray_list) {
> +	list_for_each_entry_safe(object, tmp, &gray_list, gray_list) {
>  		cond_resched();
>  
>  		/* may add new objects to the list */
>  		if (!scan_should_stop())
>  			scan_object(object);
>  
> -		tmp = list_entry(object->gray_list.next, typeof(*object),
> -				 gray_list);
> -
>  		/* remove the object from the list and release it */
>  		list_del(&object->gray_list);
>  		put_object(object);
> -
> -		object = tmp;
>  	}
>  	WARN_ON(!list_empty(&gray_list));

I tried this patch for a few days and I hit the WARN_ON after the loop.
During scanning, new entries may be added at the end of the loop but we
need to loop until all the entries have been removed. I probably had a
reason why I had the 'while' loop.

The key difference is that list_for_each_entry_safe() gets the next
entry (n or tmp above) before scan_object() and it may hit the end of
the list. However, scan_object() may do a list_add_tail(&gray_list)
hence we need to get the next entry after this function.

Basically list_for_each_entry_safe() is not safe with tail additions.
I'll revert this patch (hasn't reached mainline anyway).

Thanks.

-- 
Catalin

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

WARNING: multiple messages have this Message-ID (diff)
From: Catalin Marinas <catalin.marinas@arm.com>
To: majianpeng <majianpeng@gmail.com>
Cc: linux-mm <linux-mm@kvack.org>,
	linux-kernel <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH 2/3] mm/kmemleak.c: Use list_for_each_entry_safe to reconstruct function scan_gray_list
Date: Thu, 30 May 2013 15:40:28 +0100	[thread overview]
Message-ID: <20130530144028.GF23631@arm.com> (raw)
In-Reply-To: <519224D8.5090704@gmail.com>

On Tue, May 14, 2013 at 12:49:44PM +0100, majianpeng wrote:
> Signed-off-by: Jianpeng Ma <majianpeng@gmail.com>
> ---
>  mm/kmemleak.c | 8 +-------
>  1 file changed, 1 insertion(+), 7 deletions(-)
> 
> diff --git a/mm/kmemleak.c b/mm/kmemleak.c
> index b1525db..f0ece93 100644
> --- a/mm/kmemleak.c
> +++ b/mm/kmemleak.c
> @@ -1225,22 +1225,16 @@ static void scan_gray_list(void)
>  	 * from inside the loop. The kmemleak objects cannot be freed from
>  	 * outside the loop because their use_count was incremented.
>  	 */
> -	object = list_entry(gray_list.next, typeof(*object), gray_list);
> -	while (&object->gray_list != &gray_list) {
> +	list_for_each_entry_safe(object, tmp, &gray_list, gray_list) {
>  		cond_resched();
>  
>  		/* may add new objects to the list */
>  		if (!scan_should_stop())
>  			scan_object(object);
>  
> -		tmp = list_entry(object->gray_list.next, typeof(*object),
> -				 gray_list);
> -
>  		/* remove the object from the list and release it */
>  		list_del(&object->gray_list);
>  		put_object(object);
> -
> -		object = tmp;
>  	}
>  	WARN_ON(!list_empty(&gray_list));

I tried this patch for a few days and I hit the WARN_ON after the loop.
During scanning, new entries may be added at the end of the loop but we
need to loop until all the entries have been removed. I probably had a
reason why I had the 'while' loop.

The key difference is that list_for_each_entry_safe() gets the next
entry (n or tmp above) before scan_object() and it may hit the end of
the list. However, scan_object() may do a list_add_tail(&gray_list)
hence we need to get the next entry after this function.

Basically list_for_each_entry_safe() is not safe with tail additions.
I'll revert this patch (hasn't reached mainline anyway).

Thanks.

-- 
Catalin

  reply	other threads:[~2013-05-30 14:40 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-05-14 11:49 [PATCH 2/3] mm/kmemleak.c: Use list_for_each_entry_safe to reconstruct function scan_gray_list majianpeng
2013-05-14 11:49 ` majianpeng
2013-05-30 14:40 ` Catalin Marinas [this message]
2013-05-30 14:40   ` Catalin Marinas
2013-06-03  0:58   ` majianpeng
2013-06-03  0:58     ` majianpeng

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=20130530144028.GF23631@arm.com \
    --to=catalin.marinas@arm.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=majianpeng@gmail.com \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.