[PATCH] ALSA: usb-audio: fix set_format altsetting sanity check

[PATCH] ALSA: usb-audio: fix set_format altsetting sanity check

[Johan Hovold]

Make sure to check the return value of usb_altnum_to_altsetting() to
avoid dereferencing a NULL pointer when the requested alternate settings
is missing.

The format altsetting number may come from a quirk table and there does
not seem to be any other validation of it (the corresponding index is
checked however).

Fixes: b099b9693d23 ("ALSA: usb-audio: Avoid superfluous usb_set_interface() calls")
Cc: stable <stable@vger.kernel.org>     # 4.18
Signed-off-by: Johan Hovold <johan@kernel.org>
---
 sound/usb/pcm.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/sound/usb/pcm.c b/sound/usb/pcm.c
index 9c8930bb00c8..73dd9d21bb42 100644
--- a/sound/usb/pcm.c
+++ b/sound/usb/pcm.c
@@ -506,9 +506,9 @@ static int set_format(struct snd_usb_substream *subs, struct audioformat *fmt)
 	if (WARN_ON(!iface))
 		return -EINVAL;
 	alts = usb_altnum_to_altsetting(iface, fmt->altsetting);
-	altsd = get_iface_desc(alts);
-	if (WARN_ON(altsd->bAlternateSetting != fmt->altsetting))
+	if (WARN_ON(!alts))
 		return -EINVAL;
+	altsd = get_iface_desc(alts);
 
 	if (fmt == subs->cur_audiofmt)
 		return 0;
-- 
2.24.1

Re: [PATCH] ALSA: usb-audio: fix set_format altsetting sanity check

[Takashi Iwai]

On Fri, 20 Dec 2019 10:31:34 +0100,
Johan Hovold wrote:
> 
> Make sure to check the return value of usb_altnum_to_altsetting() to
> avoid dereferencing a NULL pointer when the requested alternate settings
> is missing.
> 
> The format altsetting number may come from a quirk table and there does
> not seem to be any other validation of it (the corresponding index is
> checked however).
> 
> Fixes: b099b9693d23 ("ALSA: usb-audio: Avoid superfluous usb_set_interface() calls")
> Cc: stable <stable@vger.kernel.org>     # 4.18
> Signed-off-by: Johan Hovold <johan@kernel.org>
> ---
>  sound/usb/pcm.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/sound/usb/pcm.c b/sound/usb/pcm.c
> index 9c8930bb00c8..73dd9d21bb42 100644
> --- a/sound/usb/pcm.c
> +++ b/sound/usb/pcm.c
> @@ -506,9 +506,9 @@ static int set_format(struct snd_usb_substream *subs, struct audioformat *fmt)
>  	if (WARN_ON(!iface))
>  		return -EINVAL;
>  	alts = usb_altnum_to_altsetting(iface, fmt->altsetting);
> -	altsd = get_iface_desc(alts);
> -	if (WARN_ON(altsd->bAlternateSetting != fmt->altsetting))
> +	if (WARN_ON(!alts))
>  		return -EINVAL;

Do we need WARN_ON() here?  If this may hit on syzbot, it'll stop at
this point because of panic_on_warn.


thanks,

Takashi

> +	altsd = get_iface_desc(alts);
>  
>  	if (fmt == subs->cur_audiofmt)
>  		return 0;
> -- 
> 2.24.1
> 

Re: [PATCH] ALSA: usb-audio: fix set_format altsetting sanity check

[Johan Hovold]

On Fri, Dec 20, 2019 at 10:46:50AM +0100, Takashi Iwai wrote:
> On Fri, 20 Dec 2019 10:31:34 +0100,
> Johan Hovold wrote:
> > 
> > Make sure to check the return value of usb_altnum_to_altsetting() to
> > avoid dereferencing a NULL pointer when the requested alternate settings
> > is missing.
> > 
> > The format altsetting number may come from a quirk table and there does
> > not seem to be any other validation of it (the corresponding index is
> > checked however).
> > 
> > Fixes: b099b9693d23 ("ALSA: usb-audio: Avoid superfluous usb_set_interface() calls")
> > Cc: stable <stable@vger.kernel.org>     # 4.18
> > Signed-off-by: Johan Hovold <johan@kernel.org>
> > ---
> >  sound/usb/pcm.c | 4 ++--
> >  1 file changed, 2 insertions(+), 2 deletions(-)
> > 
> > diff --git a/sound/usb/pcm.c b/sound/usb/pcm.c
> > index 9c8930bb00c8..73dd9d21bb42 100644
> > --- a/sound/usb/pcm.c
> > +++ b/sound/usb/pcm.c
> > @@ -506,9 +506,9 @@ static int set_format(struct snd_usb_substream *subs, struct audioformat *fmt)
> >  	if (WARN_ON(!iface))
> >  		return -EINVAL;
> >  	alts = usb_altnum_to_altsetting(iface, fmt->altsetting);
> > -	altsd = get_iface_desc(alts);
> > -	if (WARN_ON(altsd->bAlternateSetting != fmt->altsetting))
> > +	if (WARN_ON(!alts))
> >  		return -EINVAL;
> 
> Do we need WARN_ON() here?  If this may hit on syzbot, it'll stop at
> this point because of panic_on_warn.

Yeah, I considered that too and decided to leave it in. Just like for
the WARN_ON(iface), those numbers should be verified at probe.

I tried tracking where fmt->altsetting comes from, and it seems like
a sanity check needs to be added at least to create_fixed_stream_quirk()
where, for example, fmt->iface, fmt->altset_idx and the number of
endpoints are verified.

If there are other paths that can end up setting these fields to invalid
values, we want that WARN_ON() in there so we can fix those.

Johan

Re: [PATCH] ALSA: usb-audio: fix set_format altsetting sanity check

[Takashi Iwai]

On Fri, 20 Dec 2019 11:23:15 +0100,
Johan Hovold wrote:
> 
> On Fri, Dec 20, 2019 at 10:46:50AM +0100, Takashi Iwai wrote:
> > On Fri, 20 Dec 2019 10:31:34 +0100,
> > Johan Hovold wrote:
> > > 
> > > Make sure to check the return value of usb_altnum_to_altsetting() to
> > > avoid dereferencing a NULL pointer when the requested alternate settings
> > > is missing.
> > > 
> > > The format altsetting number may come from a quirk table and there does
> > > not seem to be any other validation of it (the corresponding index is
> > > checked however).
> > > 
> > > Fixes: b099b9693d23 ("ALSA: usb-audio: Avoid superfluous usb_set_interface() calls")
> > > Cc: stable <stable@vger.kernel.org>     # 4.18
> > > Signed-off-by: Johan Hovold <johan@kernel.org>
> > > ---
> > >  sound/usb/pcm.c | 4 ++--
> > >  1 file changed, 2 insertions(+), 2 deletions(-)
> > > 
> > > diff --git a/sound/usb/pcm.c b/sound/usb/pcm.c
> > > index 9c8930bb00c8..73dd9d21bb42 100644
> > > --- a/sound/usb/pcm.c
> > > +++ b/sound/usb/pcm.c
> > > @@ -506,9 +506,9 @@ static int set_format(struct snd_usb_substream *subs, struct audioformat *fmt)
> > >  	if (WARN_ON(!iface))
> > >  		return -EINVAL;
> > >  	alts = usb_altnum_to_altsetting(iface, fmt->altsetting);
> > > -	altsd = get_iface_desc(alts);
> > > -	if (WARN_ON(altsd->bAlternateSetting != fmt->altsetting))
> > > +	if (WARN_ON(!alts))
> > >  		return -EINVAL;
> > 
> > Do we need WARN_ON() here?  If this may hit on syzbot, it'll stop at
> > this point because of panic_on_warn.
> 
> Yeah, I considered that too and decided to leave it in. Just like for
> the WARN_ON(iface), those numbers should be verified at probe.
> 
> I tried tracking where fmt->altsetting comes from, and it seems like
> a sanity check needs to be added at least to create_fixed_stream_quirk()
> where, for example, fmt->iface, fmt->altset_idx and the number of
> endpoints are verified.
> 
> If there are other paths that can end up setting these fields to invalid
> values, we want that WARN_ON() in there so we can fix those.

Fair enough.

I applied now as-is.  Thanks!


Takashi