Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.36 as permitted sender) client-ip=23.128.96.36;
Date:   Tue, 12 Sep 2023 10:50:22 +0200
Message-ID: <87o7i77p2p.wl-tiwai@suse.de>
From:   Takashi Iwai <tiwai@suse.de>
To:     Ma Ke <make_ruc2021@163.com>
Cc:     perex@perex.cz, tiwai@suse.com, cujomalainey@chromium.org,
        maciej.szmigiero@oracle.com, clecigne@google.com,
        alsa-devel@alsa-project.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH] ALSA: control: do not access controls without possession of r_w lock
In-Reply-To: <20230912084530.3307329-1-make_ruc2021@163.com>
References: <20230912084530.3307329-1-make_ruc2021@163.com>
User-Agent: Wanderlust/2.15.9 (Almost Unreal) Emacs/27.2 Mule/6.0
MIME-Version: 1.0 (generated by SEMI-EPG 1.14.7 - "Harue")
Content-Type: text/plain; charset=US-ASCII
Precedence: bulk

On Tue, 12 Sep 2023 10:45:30 +0200,
Ma Ke wrote:
> 
> we should not access any of its memory when we don't ensure possession of
> a read/write lock. Otherwise we risk a use after free access, which allows
> local users to cause a denial of service and obtain sensitive information
> from kernel memory.

This function is called always under card->controls_rwsem.
So it should be OK as is.


thanks,

Takashi

> 
> Signed-off-by: Ma Ke <make_ruc2021@163.com>
> ---
>  sound/core/control.c | 3 ++-
>  1 file changed, 2 insertions(+), 1 deletion(-)
> 
> diff --git a/sound/core/control.c b/sound/core/control.c
> index 59c8658966d4..98782cc68ee1 100644
> --- a/sound/core/control.c
> +++ b/sound/core/control.c
> @@ -590,7 +590,8 @@ static int __snd_ctl_remove(struct snd_card *card,
>  		remove_hash_entries(card, kcontrol);
>  
>  	card->controls_count -= kcontrol->count;
> -	for (idx = 0; idx < kcontrol->count; idx++)
> +	count = kcontrol->count;
> +	for (idx = 0; idx < count; idx++)
>  		snd_ctl_notify_one(card, SNDRV_CTL_EVENT_MASK_REMOVE, kcontrol, idx);
>  	snd_ctl_free_one(kcontrol);
>  	return 0;
> -- 
> 2.37.2
>