diff mbox series

[FFmpeg-devel] avformat/hls: Pass a copy of the URL for probing

Message ID 20200629190013.26657-1-michael@niedermayer.cc
State Accepted
Commit b5e39880fb7269b1b3577cee288e06aa3dc1dfa2
Headers show
Series [FFmpeg-devel] avformat/hls: Pass a copy of the URL for probing | expand

Checks

Context Check Description
andriy/default pending
andriy/make success Make finished
andriy/make_fate success Make fate finished

Commit Message

Michael Niedermayer June 29, 2020, 7 p.m. UTC
The segments / url can be modified by the io read when reloading

This may be an alternative or additional fix for Ticket8673
as a further alternative the reload stuff could be disabled during
probing

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
---
 libavformat/hls.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

Comments

Andreas Rheinhardt June 29, 2020, 7:08 p.m. UTC | #1
Michael Niedermayer:
> The segments / url can be modified by the io read when reloading
> 
> This may be an alternative or additional fix for Ticket8673
> as a further alternative the reload stuff could be disabled during
> probing
> 
> Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
> ---
>  libavformat/hls.c | 6 ++++--
>  1 file changed, 4 insertions(+), 2 deletions(-)
> 
> diff --git a/libavformat/hls.c b/libavformat/hls.c
> index 3798bdd5d1..ba17c4ed96 100644
> --- a/libavformat/hls.c
> +++ b/libavformat/hls.c
> @@ -1946,6 +1946,7 @@ static int hls_read_header(AVFormatContext *s)
>      /* Open the demuxer for each playlist */
>      for (i = 0; i < c->n_playlists; i++) {
>          struct playlist *pls = c->playlists[i];
> +        char *url;
>          ff_const59 AVInputFormat *in_fmt = NULL;
>  
>          if (!(pls->ctx = avformat_alloc_context())) {
> @@ -1983,8 +1984,9 @@ static int hls_read_header(AVFormatContext *s)
>                            read_data, NULL, NULL);
>          pls->ctx->probesize = s->probesize > 0 ? s->probesize : 1024 * 4;
>          pls->ctx->max_analyze_duration = s->max_analyze_duration > 0 ? s->max_analyze_duration : 4 * AV_TIME_BASE;
> -        ret = av_probe_input_buffer(&pls->pb, &in_fmt, pls->segments[0]->url,
> -                                    NULL, 0, 0);
> +        url = av_strdup(pls->segments[0]->url);
> +        ret = av_probe_input_buffer(&pls->pb, &in_fmt, url, NULL, 0, 0);
> +        av_free(url);
>          if (ret < 0) {
>              /* Free the ctx - it isn't initialized properly at this point,
>               * so avformat_close_input shouldn't be called. If
> 
Is the missing allocation check intentional?

- Andreas
Michael Niedermayer June 29, 2020, 7:18 p.m. UTC | #2
On Mon, Jun 29, 2020 at 09:08:07PM +0200, Andreas Rheinhardt wrote:
> Michael Niedermayer:
> > The segments / url can be modified by the io read when reloading
> > 
> > This may be an alternative or additional fix for Ticket8673
> > as a further alternative the reload stuff could be disabled during
> > probing
> > 
> > Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
> > ---
> >  libavformat/hls.c | 6 ++++--
> >  1 file changed, 4 insertions(+), 2 deletions(-)
> > 
> > diff --git a/libavformat/hls.c b/libavformat/hls.c
> > index 3798bdd5d1..ba17c4ed96 100644
> > --- a/libavformat/hls.c
> > +++ b/libavformat/hls.c
> > @@ -1946,6 +1946,7 @@ static int hls_read_header(AVFormatContext *s)
> >      /* Open the demuxer for each playlist */
> >      for (i = 0; i < c->n_playlists; i++) {
> >          struct playlist *pls = c->playlists[i];
> > +        char *url;
> >          ff_const59 AVInputFormat *in_fmt = NULL;
> >  
> >          if (!(pls->ctx = avformat_alloc_context())) {
> > @@ -1983,8 +1984,9 @@ static int hls_read_header(AVFormatContext *s)
> >                            read_data, NULL, NULL);
> >          pls->ctx->probesize = s->probesize > 0 ? s->probesize : 1024 * 4;
> >          pls->ctx->max_analyze_duration = s->max_analyze_duration > 0 ? s->max_analyze_duration : 4 * AV_TIME_BASE;
> > -        ret = av_probe_input_buffer(&pls->pb, &in_fmt, pls->segments[0]->url,
> > -                                    NULL, 0, 0);
> > +        url = av_strdup(pls->segments[0]->url);
> > +        ret = av_probe_input_buffer(&pls->pb, &in_fmt, url, NULL, 0, 0);
> > +        av_free(url);
> >          if (ret < 0) {
> >              /* Free the ctx - it isn't initialized properly at this point,
> >               * so avformat_close_input shouldn't be called. If
> > 
> Is the missing allocation check intentional?

i thought if it fails filename is NULL and the code could still
plausibly continue.
but we can also fail or warn if people prefer

thx

[...]
Michael Niedermayer July 1, 2020, 9:29 a.m. UTC | #3
On Mon, Jun 29, 2020 at 09:18:18PM +0200, Michael Niedermayer wrote:
> On Mon, Jun 29, 2020 at 09:08:07PM +0200, Andreas Rheinhardt wrote:
> > Michael Niedermayer:
> > > The segments / url can be modified by the io read when reloading
> > > 
> > > This may be an alternative or additional fix for Ticket8673
> > > as a further alternative the reload stuff could be disabled during
> > > probing
> > > 
> > > Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
> > > ---
> > >  libavformat/hls.c | 6 ++++--
> > >  1 file changed, 4 insertions(+), 2 deletions(-)
> > > 
> > > diff --git a/libavformat/hls.c b/libavformat/hls.c
> > > index 3798bdd5d1..ba17c4ed96 100644
> > > --- a/libavformat/hls.c
> > > +++ b/libavformat/hls.c
> > > @@ -1946,6 +1946,7 @@ static int hls_read_header(AVFormatContext *s)
> > >      /* Open the demuxer for each playlist */
> > >      for (i = 0; i < c->n_playlists; i++) {
> > >          struct playlist *pls = c->playlists[i];
> > > +        char *url;
> > >          ff_const59 AVInputFormat *in_fmt = NULL;
> > >  
> > >          if (!(pls->ctx = avformat_alloc_context())) {
> > > @@ -1983,8 +1984,9 @@ static int hls_read_header(AVFormatContext *s)
> > >                            read_data, NULL, NULL);
> > >          pls->ctx->probesize = s->probesize > 0 ? s->probesize : 1024 * 4;
> > >          pls->ctx->max_analyze_duration = s->max_analyze_duration > 0 ? s->max_analyze_duration : 4 * AV_TIME_BASE;
> > > -        ret = av_probe_input_buffer(&pls->pb, &in_fmt, pls->segments[0]->url,
> > > -                                    NULL, 0, 0);
> > > +        url = av_strdup(pls->segments[0]->url);
> > > +        ret = av_probe_input_buffer(&pls->pb, &in_fmt, url, NULL, 0, 0);
> > > +        av_free(url);
> > >          if (ret < 0) {
> > >              /* Free the ctx - it isn't initialized properly at this point,
> > >               * so avformat_close_input shouldn't be called. If
> > > 
> > Is the missing allocation check intentional?
> 
> i thought if it fails filename is NULL and the code could still
> plausibly continue.
> but we can also fail or warn if people prefer

will apply this as i think this makes the fix for a public CVE more
robust

[...]
diff mbox series

Patch

diff --git a/libavformat/hls.c b/libavformat/hls.c
index 3798bdd5d1..ba17c4ed96 100644
--- a/libavformat/hls.c
+++ b/libavformat/hls.c
@@ -1946,6 +1946,7 @@  static int hls_read_header(AVFormatContext *s)
     /* Open the demuxer for each playlist */
     for (i = 0; i < c->n_playlists; i++) {
         struct playlist *pls = c->playlists[i];
+        char *url;
         ff_const59 AVInputFormat *in_fmt = NULL;
 
         if (!(pls->ctx = avformat_alloc_context())) {
@@ -1983,8 +1984,9 @@  static int hls_read_header(AVFormatContext *s)
                           read_data, NULL, NULL);
         pls->ctx->probesize = s->probesize > 0 ? s->probesize : 1024 * 4;
         pls->ctx->max_analyze_duration = s->max_analyze_duration > 0 ? s->max_analyze_duration : 4 * AV_TIME_BASE;
-        ret = av_probe_input_buffer(&pls->pb, &in_fmt, pls->segments[0]->url,
-                                    NULL, 0, 0);
+        url = av_strdup(pls->segments[0]->url);
+        ret = av_probe_input_buffer(&pls->pb, &in_fmt, url, NULL, 0, 0);
+        av_free(url);
         if (ret < 0) {
             /* Free the ctx - it isn't initialized properly at this point,
              * so avformat_close_input shouldn't be called. If