Re: [PATCH v7] audio/pwaudio.c: Add Pipewire audio backend for QEMU

[Dorinda Bassey]

Hi Volker,

 

To hear this,
start QEMU with qemu-system-x86_64 -machine pcspk-audiodev=audio0
-device ich9-intel-hda -device hda-duplex,audiodev=audio0 -audiodev
pipewire,id=audio0,out.mixing-engine=off ...

I hear the clipped audio stream with these options. IMO, I don't think memset is responsible for that behaviour, I still hear the harsh sound with "-audiodev pa". I also tried using an alternative like:

@@ -117,7 +118,7 @@ playback_on_process(void *data)
     }
 
     if (avail == 0) {
-        memset(p, 0, n_bytes);
+        p = g_malloc0(sizeof(n_bytes));
     } else {

The clipped audio issue is still persistent.

Thanks,

Dorinda.

On Sun, Mar 12, 2023 at 9:01 AM Volker Rümelin <vr_qemu@t-online.de> wrote:

> +/* output data processing function to read stuffs from the buffer */
> +static void
> +playback_on_process(void *data)
> +{
> +    PWVoice *v = (PWVoice *) data;
> +    void *p;
> +    struct pw_buffer *b;
> +    struct spa_buffer *buf;
> +    uint32_t n_frames, req, index, n_bytes;
> +    int32_t avail;
> +
> +    if (!v->stream) {
> +        return;
> +    }
> +
> +    /* obtain a buffer to read from */
> +    b = pw_stream_dequeue_buffer(v->stream);
> +    if (b == NULL) {
> +        error_report("out of buffers: %s", strerror(errno));
> +        return;
> +    }
> +
> +    buf = b->buffer;
> +    p = buf->datas[0].data;
> +    if (p == NULL) {
> +        return;
> +    }
> +    req = b->requested * v->frame_size;
> +    if (req == 0) {
> +        req = 4096 * v->frame_size;
> +    }
> +    n_frames = SPA_MIN(req, buf->datas[0].maxsize);
> +    n_bytes = n_frames * v->frame_size;
> +
> +    /* get no of available bytes to read data from buffer */
> +
> +    avail = spa_ringbuffer_get_read_index(&v->ring, &index);
> +
> +    if (!v->enabled) {
> +        avail = 0;
> +    }
> +
> +    if (avail == 0) {
> +        memset(p, 0, n_bytes);

memset() doesn't work for unsigned samples. For unsigned samples, a
stream of zeros is silence with a DC offset. When Pipewire mixes this
stream with another, the result is a clipped audio stream. To hear this,
start QEMU with qemu-system-x86_64 -machine pcspk-audiodev=audio0
-device ich9-intel-hda -device hda-duplex,audiodev=audio0 -audiodev
pipewire,id=audio0,out.mixing-engine=off ... and start playback with the
hda device.

With best regards,
Volker

> +    } else {
> +        if (avail < (int32_t) n_bytes) {
> +            n_bytes = avail;
> +        }
> +
> +        spa_ringbuffer_read_data(&v->ring,
> +                                    v->buffer, RINGBUFFER_SIZE,
> +                                    index & RINGBUFFER_MASK, p, n_bytes);
> +
> +        index += n_bytes;
> +        spa_ringbuffer_read_update(&v->ring, index);
> +    }
> +
> +    buf->datas[0].chunk->offset = 0;
> +    buf->datas[0].chunk->stride = v->frame_size;
> +    buf->datas[0].chunk->size = n_bytes;
> +
> +    /* queue the buffer for playback */
> +    pw_stream_queue_buffer(v->stream, b);
> +}
> +
>