On Fri, 2011-11-11 at 16:22 +0200, Tanu Kaskinen wrote:
> From: Antti-Ville Jansson <antti-ville.jansson at digia.com>
>
> In pa_create_stream_callback, a stream is inserted into
> s->context->record_streams only if it's a record stream. Otherwise it's
> inserted into s->context->playback_streams. However, in stream_unlink the
> stream is removed from s->context->playback_streams only if it's a playback
> stream and otherwise it's removed from s->context->record_streams.
>
> Thus, if the stream is an upload stream, we first insert it into
> s->context->playback_streams in pa_create_stream_callback and then try to
> remove it unsuccessfully from s->context->record_streams in stream_unlink. This
> means that we are leaking hashmap entries until the context is freed,
> constantly consuming more memory with applications that upload and unload a
> large number of samples through one context.
>
> Of course, this begs the question whether upload streams even belong in either
> of those hashmaps. I don't want to mess around with the code too much at this
> point though, so this patch should be a sufficient improvement.
> ---
> src/pulse/stream.c | 2 +-
> 1 files changed, 1 insertions(+), 1 deletions(-)
>
> diff --git a/src/pulse/stream.c b/src/pulse/stream.c
> index 3bf2d96..5433abf 100644
> --- a/src/pulse/stream.c
> +++ b/src/pulse/stream.c
> @@ -258,7 +258,7 @@ static void stream_unlink(pa_stream *s) {
> pa_pdispatch_unregister_reply(s->context->pdispatch, s);
>
> if (s->channel_valid) {
> - pa_hashmap_remove((s->direction == PA_STREAM_PLAYBACK) ? s->context->playback_streams : s->context->record_streams, PA_UINT32_TO_PTR(s->channel));
> + pa_hashmap_remove((s->direction == PA_STREAM_RECORD) ? s->context->record_streams : s->context->playback_streams, PA_UINT32_TO_PTR(s->channel));
> s->channel = 0;
> s->channel_valid = FALSE;
> }
Pulled to my tree with a shortened summary for easier reading in
shortlogs and gitweb.
Thanks,
Arun
