[PATCH] ALSA: pcm: fix a memory leak bug

[PATCH] ALSA: pcm: fix a memory leak bug

[Wenwen Wang]

In hiface_pcm_init(), 'rt' is firstly allocated through kzalloc(). Later
on, hiface_pcm_init_urb() is invoked to initialize 'rt->out_urbs[i]'.
However, if the initialization fails, 'rt' is not deallocated, leading to a
memory leak bug.

To fix the above issue, free 'rt' before returning the error.

Signed-off-by: Wenwen Wang <wenwen@cs.uga.edu>
---
 sound/usb/hiface/pcm.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/sound/usb/hiface/pcm.c b/sound/usb/hiface/pcm.c
index 14fc1e1..5dbcd0d 100644
--- a/sound/usb/hiface/pcm.c
+++ b/sound/usb/hiface/pcm.c
@@ -599,8 +599,10 @@ int hiface_pcm_init(struct hiface_chip *chip, u8
extra_freq)
        for (i = 0; i < PCM_N_URBS; i++) {
                ret = hiface_pcm_init_urb(&rt->out_urbs[i], chip, OUT_EP,
                                    hiface_pcm_out_urb_handler);
-               if (ret < 0)
+               if (ret < 0) {
+                       kfree(rt);
                        return ret;
+               }
        }

        ret = snd_pcm_new(chip->card, "USB-SPDIF Audio", 0, 1, 0, &pcm);
-- 
2.7.4

Re: [PATCH] ALSA: pcm: fix a memory leak bug

[Takashi Iwai]

On Wed, 07 Aug 2019 08:15:17 +0200,
Wenwen Wang wrote:
> 
> In hiface_pcm_init(), 'rt' is firstly allocated through kzalloc(). Later
> on, hiface_pcm_init_urb() is invoked to initialize 'rt->out_urbs[i]'.
> However, if the initialization fails, 'rt' is not deallocated, leading to a
> memory leak bug.
> 
> To fix the above issue, free 'rt' before returning the error.
> 
> Signed-off-by: Wenwen Wang <wenwen@cs.uga.edu>
> ---
>  sound/usb/hiface/pcm.c | 4 +++-
>  1 file changed, 3 insertions(+), 1 deletion(-)
> 
> diff --git a/sound/usb/hiface/pcm.c b/sound/usb/hiface/pcm.c
> index 14fc1e1..5dbcd0d 100644
> --- a/sound/usb/hiface/pcm.c
> +++ b/sound/usb/hiface/pcm.c
> @@ -599,8 +599,10 @@ int hiface_pcm_init(struct hiface_chip *chip, u8
> extra_freq)
>         for (i = 0; i < PCM_N_URBS; i++) {
>                 ret = hiface_pcm_init_urb(&rt->out_urbs[i], chip, OUT_EP,
>                                     hiface_pcm_out_urb_handler);
> -               if (ret < 0)
> +               if (ret < 0) {
> +                       kfree(rt);
>                         return ret;
> +               }

Unfortunately this still leaves some memory.  We need to release
rt->out_urbs[], too.  The relevant code is already in
hiface_pcm_destroy(), so factor out the looped kfree() there and call
it from both places.

Care to resubmit with more fixes?


thanks,

Takashi

Re: [PATCH] ALSA: pcm: fix a memory leak bug

[Wenwen Wang]

On Wed, Aug 7, 2019 at 2:33 AM Takashi Iwai <tiwai@suse.de> wrote:
>
> On Wed, 07 Aug 2019 08:15:17 +0200,
> Wenwen Wang wrote:
> >
> > In hiface_pcm_init(), 'rt' is firstly allocated through kzalloc(). Later
> > on, hiface_pcm_init_urb() is invoked to initialize 'rt->out_urbs[i]'.
> > However, if the initialization fails, 'rt' is not deallocated, leading to a
> > memory leak bug.
> >
> > To fix the above issue, free 'rt' before returning the error.
> >
> > Signed-off-by: Wenwen Wang <wenwen@cs.uga.edu>
> > ---
> >  sound/usb/hiface/pcm.c | 4 +++-
> >  1 file changed, 3 insertions(+), 1 deletion(-)
> >
> > diff --git a/sound/usb/hiface/pcm.c b/sound/usb/hiface/pcm.c
> > index 14fc1e1..5dbcd0d 100644
> > --- a/sound/usb/hiface/pcm.c
> > +++ b/sound/usb/hiface/pcm.c
> > @@ -599,8 +599,10 @@ int hiface_pcm_init(struct hiface_chip *chip, u8
> > extra_freq)
> >         for (i = 0; i < PCM_N_URBS; i++) {
> >                 ret = hiface_pcm_init_urb(&rt->out_urbs[i], chip, OUT_EP,
> >                                     hiface_pcm_out_urb_handler);
> > -               if (ret < 0)
> > +               if (ret < 0) {
> > +                       kfree(rt);
> >                         return ret;
> > +               }
>
> Unfortunately this still leaves some memory.  We need to release
> rt->out_urbs[], too.  The relevant code is already in
> hiface_pcm_destroy(), so factor out the looped kfree() there and call
> it from both places.
>
> Care to resubmit with more fixes?

Thanks for your comments! I also found this issue, and am working on
another patch to fix it.

Wenwen