On 16/10/15 19:07, wm4 wrote: > On Thu, 15 Oct 2015 02:26:42 +0200 > Luca Barbato <[email protected]> wrote: > >> --- >> avconv.c | 2 +- >> 1 file changed, 1 insertion(+), 1 deletion(-) >> >> diff --git a/avconv.c b/avconv.c >> index 79569f6..34f3d42 100644 >> --- a/avconv.c >> +++ b/avconv.c >> @@ -2125,7 +2125,7 @@ static void *input_thread(void *arg) >> while (!av_fifo_space(f->fifo)) >> pthread_cond_wait(&f->fifo_cond, &f->fifo_lock); >> >> - av_dup_packet(&pkt); >> + av_packet_ref(&pkt, &pkt); >> av_fifo_generic_write(f->fifo, &pkt, sizeof(pkt), NULL); >> >> pthread_mutex_unlock(&f->fifo_lock); > > Is this really valid? I don't think it is. av_packet_copy_props() will > cause a mess, and av_packet_ref() itself will leave behind trash on > memory alloc failure if the packet is already refcounted.
asan seems happy with it, I can doublecheck the error path. lu _______________________________________________ libav-devel mailing list [email protected] https://lists.libav.org/mailman/listinfo/libav-devel
