<html><head><style type="text/css"></style></head><body><div style="font-family:Courier New,courier,monaco,monospace,sans-serif;font-size:8pt">Cool - I'm glad you got rid of the a52 string check hack. I didn't have the line pa_sink_set_max_rewind_within_thread(u->sink, u->hwbuf_size); I guess the version of Pulseaudio source code in Lucid is much older than yours? Anyway I did have the line with pa_sink_set_max_request_within_thread so I added your lines below and didn't get any compile errors, so the function exists in my version, just wasn't used. It still all works for me. Cheers, Mike <div> </div><div style="font-family: Courier New,courier,monaco,monospace,sans-serif; font-size: 8pt;"> <div style="font-family: arial,helvetica,sans-serif; font-size: 13px;"><hr size="1">From: pl bossart <bossart.nospam@gmail.com> To: General PulseAudio Discussion <pulseaudio-discuss@mail.0pointer.de> Sent: Mon, 12 July, 2010 22:08:44 Subject: Re: [pulseaudio-discuss] {PATCH][RFC] AC3 passthrough support > 2. Replace: pa_sink_set_max_rewind(u->sink, u->hwbuf_size);      with: >     if(strcmp(u->device_name, "a52") == 0) { >         pa_sink_set_max_rewind(u->sink, 0); >     } else { >         pa_sink_set_max_rewind(u->sink, u->hwbuf_size); >     } NAK. This does not work for me. You need to rewind also in the thread context: diff --git a/src/modules/alsa/alsa-sink.c b/src/modules/alsa/alsa-sink.c index 2995c3c..4059a9b 100644 --- a/src/modules/alsa/alsa-sink.c +++ b/src/modules/alsa/alsa-sink.c @@ -953,7 +953,12 @@ static int update_sw_params(struct userdata *u) {     }     pa_sink_set_max_request_within_thread(u->sink, u->hwbuf_size - u->hwbuf_unused); -    pa_sink_set_max_rewind_within_thread(u->sink, u->hwbuf_size); +    if (pa_alsa_pcm_is_hw(u->pcm_handle)) +        pa_sink_set_max_rewind_within_thread(u->sink, u->hwbuf_size); +    else { +        pa_log("Disabling rewind_within_thread for device %s", u->device_name); +        pa_sink_set_max_rewind_within_thread(u->sink, 0); +    }     return 0; } @@ -1906,7 +1911,12 @@ pa_sink *pa_alsa_sink_new(pa_module *m, pa_modargs *ma, const char*driver, pa_ca                 (double) pa_bytes_to_usec(u->hwbuf_size, &ss) / PA_USEC_PER_MSEC);     pa_sink_set_max_request(u->sink, u->hwbuf_size); -    pa_sink_set_max_rewind(u->sink, u->hwbuf_size); +    if (pa_alsa_pcm_is_hw(u->pcm_handle)) +        pa_sink_set_max_rewind(u->sink, u->hwbuf_size); +    else { +        pa_log("Disabling rewind for device %s", u->device_name); +        pa_sink_set_max_rewind(u->sink, 0); +    }     if (u->use_tsched) {         u->tsched_watermark = pa_usec_to_bytes_round_up(pa_bytes_to_usec_round_up(tsched_watermark, &requested_ss), &u->sink->sample_spec); pa_pcm_is_hw detects a52 but allows rewind for iec958. Looks like a better fix than a hard-coded string compare. Even with this fix, the sink doesn't seem to do anything: D: alsa-sink.c: Requested to rewind 0 bytes. D: alsa-sink.c: Mhmm, actually there is nothing to rewind. D: protocol-native.c: Requesting rewind due to rewrite. D: sink-input.c: Requesting rewind due to corking D: alsa-sink.c: Requested to rewind 0 bytes. D: alsa-sink.c: Mhmm, actually there is nothing to rewind. D: module-suspend-on-idle.c: Sink alsa_output.pci-0000_00_1b.0.iec958-ac3-surround-51 becomes idle, timeout in 5 seconds. D: alsa-sink.c: Requested to rewind 0 bytes. D: module-suspend-on-idle.c: Sink alsa_output.pci-0000_00_1b.0.iec958-ac3-surround-51 becomes idle, timeout in 5 seconds. D: alsa-sink.c: Mhmm, actually there is nothing to rewind. D: protocol-dbus.c: Interface org.PulseAudio.Core1.Stream removed from object /org/pulseaudio/core1/playback_stream0 I: sink-input.c: Freeing input 0 "Playback Stream" I: client.c: Freed 5 "gst-launch-0.10" I: protocol-native.c: Connection died. You'll find my diffs against git master below: - Pierre </div></div> </div> </body></html>