[FFmpeg-devel] [PATCH] avutil/hwcontext: Don't assume frames_uninit is reentrant
Zhao Zhili
quinkblack at foxmail.com
Mon Oct 28 14:46:34 EET 2024
> On Oct 24, 2024, at 02:14, 徐浩宇 <dawndreaming97 at gmail.com> wrote:
>
> I have discovered that the vulnerability addressed by commit 3bb00c0
> <https://github.com/ffmpeg/ffmpeg/commit/3bb00c0a420c3ce83c6fafee30270d69622ccad7>
> also
> affects the release/6.1 branch,but the fix has not yet been applied. Given
> that there were relevant commits to the release/6.1 branch just two weeks
> ago, I believe it is an active branch and should be properly patched.To
> address this, I have created this patch for release/6.1 based on 3bb00c0
> <https://github.com/ffmpeg/ffmpeg/commit/3bb00c0a420c3ce83c6fafee30270d69622ccad7>
> .
Thanks for the report. I have cherry-picked commit 3bb00c0 to release/6.1 branch.
>
> Signed-off-by: sprinkle <dawndreaming97 at gmail.com>
> ---
> libavutil/hwcontext.c | 8 ++------
> 1 file changed, 2 insertions(+), 6 deletions(-)
>
> diff --git a/libavutil/hwcontext.c b/libavutil/hwcontext.c
> index 3650d46..0ef3479 100644
> --- a/libavutil/hwcontext.c
> +++ b/libavutil/hwcontext.c
> @@ -363,7 +363,7 @@ int av_hwframe_ctx_init(AVBufferRef *ref)
> if (ctx->internal->hw_type->frames_init) {
> ret = ctx->internal->hw_type->frames_init(ctx);
> if (ret < 0)
> - goto fail;
> + return ret;
> }
>
> if (ctx->internal->pool_internal && !ctx->pool)
> @@ -373,14 +373,10 @@ int av_hwframe_ctx_init(AVBufferRef *ref)
> if (ctx->initial_pool_size > 0) {
> ret = hwframe_pool_prealloc(ref);
> if (ret < 0)
> - goto fail;
> + return ret;
> }
>
> return 0;
> -fail:
> - if (ctx->internal->hw_type->frames_uninit)
> - ctx->internal->hw_type->frames_uninit(ctx);
> - return ret;
> }
>
> int av_hwframe_transfer_get_formats(AVBufferRef *hwframe_ref,
> --
> 2.7.4
> _______________________________________________
> ffmpeg-devel mailing list
> ffmpeg-devel at ffmpeg.org
> https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
>
> To unsubscribe, visit link above, or email
> ffmpeg-devel-request at ffmpeg.org with subject "unsubscribe".
More information about the ffmpeg-devel
mailing list