linux-kernel - Re: [PATCH] net: xen-netbank: hash.c: Use built-in RCU list checking

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] [day] [month] [year] [list]

Date:   Wed, 15 Jan 2020 13:56:31 +0000
From:   Wei Liu <wei.liu@...nel.org>
To:     madhuparnabhowmik04@...il.com
Cc:     wei.liu@...nel.org, paul@....org, davem@...emloft.net,
        xen-devel@...ts.xenproject.org, netdev@...r.kernel.org,
        linux-kernel@...r.kernel.org, paulmck@...nel.org,
        joel@...lfernandes.org, frextrite@...il.com,
        linux-kernel-mentees@...ts.linuxfoundation.org
Subject: Re: [PATCH] net: xen-netbank: hash.c: Use built-in RCU list checking

Thanks for the patch.

There is a typo in the subject line. It should say xen-netback, not
xen-netbank.

On Wed, Jan 15, 2020 at 06:11:28PM +0530, madhuparnabhowmik04@...il.com wrote:
> From: Madhuparna Bhowmik <madhuparnabhowmik04@...il.com>
> 
> list_for_each_entry_rcu has built-in RCU and lock checking.
> Pass cond argument to list_for_each_entry_rcu.
> 
> Signed-off-by: Madhuparna Bhowmik <madhuparnabhowmik04@...il.com>
> ---
>  drivers/net/xen-netback/hash.c | 6 ++++--
>  1 file changed, 4 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/net/xen-netback/hash.c b/drivers/net/xen-netback/hash.c
> index 10d580c3dea3..30709bc9d170 100644
> --- a/drivers/net/xen-netback/hash.c
> +++ b/drivers/net/xen-netback/hash.c
> @@ -51,7 +51,8 @@ static void xenvif_add_hash(struct xenvif *vif, const u8 *tag,
>  
>  	found = false;
>  	oldest = NULL;
> -	list_for_each_entry_rcu(entry, &vif->hash.cache.list, link) {
> +	list_for_each_entry_rcu(entry, &vif->hash.cache.list, link,
> +							lockdep_is_held(&vif->hash.cache.lock)) {

There are probably too many tabs here. Indentation looks wrong.

The surrounding code makes it pretty clear that the lock is already held
by the time list_for_each_entry_rcu is called, yet the checking involved
in lockdep_is_held is not trivial, so I'm afraid I don't consider this a
strict improvement over the existing code.

If there is something I misunderstood, let me know.

Wei.

>  		/* Make sure we don't add duplicate entries */
>  		if (entry->len == len &&
>  		    memcmp(entry->tag, tag, len) == 0)
> @@ -102,7 +103,8 @@ static void xenvif_flush_hash(struct xenvif *vif)
>  
>  	spin_lock_irqsave(&vif->hash.cache.lock, flags);
>  
> -	list_for_each_entry_rcu(entry, &vif->hash.cache.list, link) {
> +	list_for_each_entry_rcu(entry, &vif->hash.cache.list, link,
> +							lockdep_is_held(&vif->hash.cache.lock)) {
>  		list_del_rcu(&entry->link);
>  		vif->hash.cache.count--;
>  		kfree_rcu(entry, rcu);
> -- 
> 2.17.1
>