linux-kernel - Re: [PATCH] ASoC: Intel: sof_rt5682: fix out-of-bounds array access

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]

Message-ID: <65a5e374-ff3a-ec2d-af70-c042c22decdb@linux.intel.com>
Date:   Fri, 1 Jul 2022 08:53:35 -0500
From:   Pierre-Louis Bossart <pierre-louis.bossart@...ux.intel.com>
To:     Brent Lu <brent.lu@...el.com>, alsa-devel@...a-project.org
Cc:     Cezary Rojewski <cezary.rojewski@...el.com>,
        Liam Girdwood <liam.r.girdwood@...ux.intel.com>,
        Peter Ujfalusi <peter.ujfalusi@...ux.intel.com>,
        Bard Liao <yung-chuan.liao@...ux.intel.com>,
        Ranjani Sridharan <ranjani.sridharan@...ux.intel.com>,
        Kai Vehmanen <kai.vehmanen@...ux.intel.com>,
        Mark Brown <broonie@...nel.org>,
        Jaroslav Kysela <perex@...ex.cz>,
        Takashi Iwai <tiwai@...e.com>, Yong Zhi <yong.zhi@...el.com>,
        Mac Chiang <mac.chiang@...el.com>,
        Ajye Huang <ajye.huang@...il.com>, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] ASoC: Intel: sof_rt5682: fix out-of-bounds array access



On 7/1/22 03:19, Brent Lu wrote:
> Starting from ADL platform we have four HDMI PCM devices which exceeds
> the size of sof_hdmi array. Since each sof_hdmi_pcm structure
> represents one HDMI PCM device, we remove the sof_hdmi array and add a
> new member hdmi_jack to the snd_soc_jack structure to fix the
> out-of-bounds problem.

Valid fix aligned with other machine drivers, but the commit message is
incorrect: the 4th HDMI link was added in TGL, not ADL.

See e.g. 'Google Volteer', this quirk is already enabled:
SOF_RT5682_NUM_HDMIDEV(4)),

I guess existing topologies for Volteer never used the 4th link?

Wondering how we didn't this problem sooner - and if this needs to be
added to -stable, or if this is only needed for newer platforms.

> Signed-off-by: Brent Lu <brent.lu@...el.com>
> ---
>  sound/soc/intel/boards/sof_rt5682.c | 10 +++-------
>  1 file changed, 3 insertions(+), 7 deletions(-)
> 
> diff --git a/sound/soc/intel/boards/sof_rt5682.c b/sound/soc/intel/boards/sof_rt5682.c
> index a24fb71d5ff3..1384716c6360 100644
> --- a/sound/soc/intel/boards/sof_rt5682.c
> +++ b/sound/soc/intel/boards/sof_rt5682.c
> @@ -69,11 +69,10 @@ static unsigned long sof_rt5682_quirk = SOF_RT5682_MCLK_EN |
>  
>  static int is_legacy_cpu;
>  
> -static struct snd_soc_jack sof_hdmi[3];
> -
>  struct sof_hdmi_pcm {
>  	struct list_head head;
>  	struct snd_soc_dai *codec_dai;
> +	struct snd_soc_jack hdmi_jack;
>  	int device;
>  };
>  
> @@ -447,7 +446,6 @@ static int sof_card_late_probe(struct snd_soc_card *card)
>  	char jack_name[NAME_SIZE];
>  	struct sof_hdmi_pcm *pcm;
>  	int err;
> -	int i = 0;
>  
>  	/* HDMI is not supported by SOF on Baytrail/CherryTrail */
>  	if (is_legacy_cpu || !ctx->idisp_codec)
> @@ -468,17 +466,15 @@ static int sof_card_late_probe(struct snd_soc_card *card)
>  		snprintf(jack_name, sizeof(jack_name),
>  			 "HDMI/DP, pcm=%d Jack", pcm->device);
>  		err = snd_soc_card_jack_new(card, jack_name,
> -					    SND_JACK_AVOUT, &sof_hdmi[i]);
> +					    SND_JACK_AVOUT, &pcm->hdmi_jack);
>  
>  		if (err)
>  			return err;
>  
>  		err = hdac_hdmi_jack_init(pcm->codec_dai, pcm->device,
> -					  &sof_hdmi[i]);
> +					  &pcm->hdmi_jack);
>  		if (err < 0)
>  			return err;
> -
> -		i++;
>  	}
>  
>  	if (sof_rt5682_quirk & SOF_MAX98373_SPEAKER_AMP_PRESENT) {