linux-kernel - Re: [PATCH] media: vidtv: fix to initialize local pointers upon transfer of memory ownership

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 for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <CAO9qdTH=7qdR2KGQiAeKu11g8T9pPjYN0zjarLGGW_EPQ_nZ3w@mail.gmail.com>
Date: Sat, 23 Aug 2025 00:19:57 +0900
From: Jeongjun Park <aha310510@...il.com>
To: Daniel Almeida <dwlsalmeida@...il.com>
Cc: mchehab@...nel.org, linux-media@...r.kernel.org, 
	linux-kernel@...r.kernel.org, 
	syzbot+1d9c0edea5907af239e0@...kaller.appspotmail.com
Subject: Re: [PATCH] media: vidtv: fix to initialize local pointers upon
 transfer of memory ownership

Hello Daniel,


Daniel Almeida <dwlsalmeida@...il.com> wrote:
>
> Hi Jeongjun,
>
> > On 22 Aug 2025, at 03:58, Jeongjun Park <aha310510@...il.com> wrote:
> >
> > vidtv_channel_si_init() transfers ownership of each object to the correct
> > table through functions such as vidtv_psi_*_assign().
> >
> > However, since it does not set the local pointer to NULL afterwards, if it
> > fails for various reasons and jumps to a place such as free_*it, memory
> > that was freed in vidtv_psi_*_table_destroy() will be accessed and freed
> > again, resulting in use-after-free and double-free vuln.
>
> I get what you’re trying to say, but what you actually wrote is hard to
> parse, can you improve the wording here?

Oops, I think I made the explanation difficult to understand and confusing.
I'll try to rewrite it so it's as easy to understand as possible.

>
> >
> > Therefore, local pointers that have completed ownership transfer must be
> > initialized to NULL to prevent re-access to already freed memory.
> >
> > Reported-by: syzbot+1d9c0edea5907af239e0@...kaller.appspotmail.com
> > Closes: https://syzkaller.appspot.com/bug?extid=1d9c0edea5907af239e0
> > Fixes: 3be8037960bc ("media: vidtv: add error checks")
> > Signed-off-by: Jeongjun Park <aha310510@...il.com>
> > ---
> > drivers/media/test-drivers/vidtv/vidtv_channel.c | 3 +++
> > 1 file changed, 3 insertions(+)
> >
> > diff --git a/drivers/media/test-drivers/vidtv/vidtv_channel.c b/drivers/media/test-drivers/vidtv/vidtv_channel.c
> > index f3023e91b3eb..3541155c6fc6 100644
> > --- a/drivers/media/test-drivers/vidtv/vidtv_channel.c
> > +++ b/drivers/media/test-drivers/vidtv/vidtv_channel.c
> > @@ -461,12 +461,15 @@ int vidtv_channel_si_init(struct vidtv_mux *m)
> >
> > /* assemble all programs and assign to PAT */
> > vidtv_psi_pat_program_assign(m->si.pat, programs);
> > + programs = NULL;
> >
> > /* assemble all services and assign to SDT */
> > vidtv_psi_sdt_service_assign(m->si.sdt, services);
> > + services = NULL;
> >
> > /* assemble all events and assign to EIT */
> > vidtv_psi_eit_event_assign(m->si.eit, events);
> > + events = NULL;
> >
> > m->si.pmt_secs = vidtv_psi_pmt_create_sec_for_each_pat_entry(m->si.pat,
> >     m->pcr_pid);
> > --
>
> The patch itself is ok, thanks a lot, however:
>
> > media: vidtv: fix to initialize local pointers upon transfer of memory ownership
>
> Can you please use imperative voice here? i.e.:
>
> "media: vidtv: initialize local pointers upon transfer of memory ownership”

If I do that, the patch subject will change. Should I just change the
subject and send it to you, or should I also send it with the v2 tag
included?

>
> — Daniel

Regards,
Jeongjun Park