diff mbox

[FFmpeg-devel,6/6] pvfdec: prevent overflow during block alignment, calculation

Message ID 1cd2cebb-86ff-37f5-cdda-bf72cb58fe0f@googlemail.com
State New
Headers show

Commit Message

Andreas Cadhalpun Dec. 15, 2016, 1:19 a.m. UTC
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
---
 libavformat/pvfdec.c | 5 +++++
 1 file changed, 5 insertions(+)

Comments

Ronald S. Bultje Dec. 15, 2016, 1:04 p.m. UTC | #1
Hi,

On Wed, Dec 14, 2016 at 8:19 PM, Andreas Cadhalpun <
andreas.cadhalpun@googlemail.com> wrote:

> Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
> ---
>  libavformat/pvfdec.c | 5 +++++
>  1 file changed, 5 insertions(+)
>
> diff --git a/libavformat/pvfdec.c b/libavformat/pvfdec.c
> index b9f6d4f..5eecc22 100644
> --- a/libavformat/pvfdec.c
> +++ b/libavformat/pvfdec.c
> @@ -56,6 +56,11 @@ static int pvf_read_header(AVFormatContext *s)
>      st->codecpar->sample_rate = sample_rate;
>      st->codecpar->codec_id    = ff_get_pcm_codec_id(bps, 0, 1, 0xFFFF);
>      st->codecpar->bits_per_coded_sample = bps;
> +    if (bps > INT_MAX / st->codecpar->channels) {
> +        av_log(s, AV_LOG_ERROR, "Overflow during block alignment
> calculation %d * %d\n",
> +               bps, st->codecpar->channels);
> +        return AVERROR_INVALIDDATA;
> +    }


And this is what I meant.

Please stop. No. No. No. No. No. Not in codec code. Add these checks in
generic code if you care about the outcome, but please don't make each
codec a crapshoot like this.

Please. From a maintenance point of view, that's a much better approach.
Please stop for a second and think about my point of view here. I beg you.
Please.

Ronald
diff mbox

Patch

diff --git a/libavformat/pvfdec.c b/libavformat/pvfdec.c
index b9f6d4f..5eecc22 100644
--- a/libavformat/pvfdec.c
+++ b/libavformat/pvfdec.c
@@ -56,6 +56,11 @@  static int pvf_read_header(AVFormatContext *s)
     st->codecpar->sample_rate = sample_rate;
     st->codecpar->codec_id    = ff_get_pcm_codec_id(bps, 0, 1, 0xFFFF);
     st->codecpar->bits_per_coded_sample = bps;
+    if (bps > INT_MAX / st->codecpar->channels) {
+        av_log(s, AV_LOG_ERROR, "Overflow during block alignment calculation %d * %d\n",
+               bps, st->codecpar->channels);
+        return AVERROR_INVALIDDATA;
+    }
     st->codecpar->block_align = bps * st->codecpar->channels / 8;
 
     avpriv_set_pts_info(st, 64, 1, st->codecpar->sample_rate);