gstreamer/sys/wasapi2/gstwasapi2src.c

482 lines
14 KiB
C
Raw Normal View History

/*
* Copyright (C) 2008 Ole André Vadla Ravnås <ole.andre.ravnas@tandberg.com>
* Copyright (C) 2018 Centricular Ltd.
* Author: Nirbheek Chauhan <nirbheek@centricular.com>
* Copyright (C) 2020 Seungha Yang <seungha@centricular.com>
*
* This library is free software; you can redistribute it and/or
* modify it under the terms of the GNU Library General Public
* License as published by the Free Software Foundation; either
* version 2 of the License, or (at your option) any later version.
*
* This library is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Library General Public License for more details.
*
* You should have received a copy of the GNU Library General Public
* License along with this library; if not, write to the
* Free Software Foundation, Inc., 51 Franklin St, Fifth Floor,
* Boston, MA 02110-1301, USA.
*/
/**
* SECTION:element-wasapi2src
* @title: wasapi2src
*
* Provides audio capture from the Windows Audio Session API available with
* Windows 10.
*
* ## Example pipelines
* |[
* gst-launch-1.0 -v wasapi2src ! fakesink
* ]| Capture from the default audio device and render to fakesink.
*
* |[
* gst-launch-1.0 -v wasapi2src low-latency=true ! fakesink
* ]| Capture from the default audio device with the minimum possible latency and render to fakesink.
*
*/
#ifdef HAVE_CONFIG_H
#include <config.h>
#endif
#include "gstwasapi2src.h"
#include "gstwasapi2util.h"
wasapi2: Rewrite plugin and implement audioringbuffer subclass ... based on MediaFoundation work queue API. By this commit, wasapi2 plugin will make use of pull mode scheduling with audioringbuffer subclass. There are several drawbacks of audiosrc/audiosink subclassing (not audiobasesrc/audiobasesink) for WASAPI API, which are: * audiosrc/audiosink classes try to set high priority to read/write thread via MMCSS (Multimedia Class Scheduler Service) but it's not allowed in case of UWP application. In order to use MMCSS in UWP, application should use MediaFoundation work queue indirectly. Since audiosrc/audiosink scheduling model is not compatible with MediaFoundation's work queue model, audioringbuffer subclassing is required. * WASAPI capture device might report larger packet size than expected (i.e., larger frames we can read than expected frame size per period). Meanwhile, in any case, application should drain all packets at that moment. In order to handle the case, wasapi/wasapi2 plugins were making use of GstAdapter which is obviously sub-optimal because it requires additional memory allocation and copy. By implementing audioringbuffer subclassing, we can avoid such inefficiency. In this commit, all the device read/write operations will be moved to newly implemented wasapi2ringbuffer class and existing wasapi2client class will take care of device enumeration and activation parts only. Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2306>
2021-05-10 11:45:28 +00:00
#include "gstwasapi2ringbuffer.h"
GST_DEBUG_CATEGORY_STATIC (gst_wasapi2_src_debug);
#define GST_CAT_DEFAULT gst_wasapi2_src_debug
static GstStaticPadTemplate src_template = GST_STATIC_PAD_TEMPLATE ("src",
GST_PAD_SRC,
GST_PAD_ALWAYS,
GST_STATIC_CAPS (GST_WASAPI2_STATIC_CAPS));
#define DEFAULT_LOW_LATENCY FALSE
#define DEFAULT_MUTE FALSE
#define DEFAULT_VOLUME 1.0
#define DEFAULT_LOOPBACK FALSE
enum
{
PROP_0,
PROP_DEVICE,
PROP_LOW_LATENCY,
PROP_MUTE,
PROP_VOLUME,
PROP_DISPATCHER,
PROP_LOOPBACK,
};
struct _GstWasapi2Src
{
wasapi2: Rewrite plugin and implement audioringbuffer subclass ... based on MediaFoundation work queue API. By this commit, wasapi2 plugin will make use of pull mode scheduling with audioringbuffer subclass. There are several drawbacks of audiosrc/audiosink subclassing (not audiobasesrc/audiobasesink) for WASAPI API, which are: * audiosrc/audiosink classes try to set high priority to read/write thread via MMCSS (Multimedia Class Scheduler Service) but it's not allowed in case of UWP application. In order to use MMCSS in UWP, application should use MediaFoundation work queue indirectly. Since audiosrc/audiosink scheduling model is not compatible with MediaFoundation's work queue model, audioringbuffer subclassing is required. * WASAPI capture device might report larger packet size than expected (i.e., larger frames we can read than expected frame size per period). Meanwhile, in any case, application should drain all packets at that moment. In order to handle the case, wasapi/wasapi2 plugins were making use of GstAdapter which is obviously sub-optimal because it requires additional memory allocation and copy. By implementing audioringbuffer subclassing, we can avoid such inefficiency. In this commit, all the device read/write operations will be moved to newly implemented wasapi2ringbuffer class and existing wasapi2client class will take care of device enumeration and activation parts only. Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2306>
2021-05-10 11:45:28 +00:00
GstAudioBaseSrc parent;
/* properties */
gchar *device_id;
gboolean low_latency;
gboolean mute;
gdouble volume;
gpointer dispatcher;
gboolean loopback;
gboolean mute_changed;
gboolean volume_changed;
};
static void gst_wasapi2_src_finalize (GObject * object);
static void gst_wasapi2_src_set_property (GObject * object, guint prop_id,
const GValue * value, GParamSpec * pspec);
static void gst_wasapi2_src_get_property (GObject * object, guint prop_id,
GValue * value, GParamSpec * pspec);
wasapi2: Rewrite plugin and implement audioringbuffer subclass ... based on MediaFoundation work queue API. By this commit, wasapi2 plugin will make use of pull mode scheduling with audioringbuffer subclass. There are several drawbacks of audiosrc/audiosink subclassing (not audiobasesrc/audiobasesink) for WASAPI API, which are: * audiosrc/audiosink classes try to set high priority to read/write thread via MMCSS (Multimedia Class Scheduler Service) but it's not allowed in case of UWP application. In order to use MMCSS in UWP, application should use MediaFoundation work queue indirectly. Since audiosrc/audiosink scheduling model is not compatible with MediaFoundation's work queue model, audioringbuffer subclassing is required. * WASAPI capture device might report larger packet size than expected (i.e., larger frames we can read than expected frame size per period). Meanwhile, in any case, application should drain all packets at that moment. In order to handle the case, wasapi/wasapi2 plugins were making use of GstAdapter which is obviously sub-optimal because it requires additional memory allocation and copy. By implementing audioringbuffer subclassing, we can avoid such inefficiency. In this commit, all the device read/write operations will be moved to newly implemented wasapi2ringbuffer class and existing wasapi2client class will take care of device enumeration and activation parts only. Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2306>
2021-05-10 11:45:28 +00:00
static GstStateChangeReturn gst_wasapi2_src_change_state (GstElement *
element, GstStateChange transition);
wasapi2: Rewrite plugin and implement audioringbuffer subclass ... based on MediaFoundation work queue API. By this commit, wasapi2 plugin will make use of pull mode scheduling with audioringbuffer subclass. There are several drawbacks of audiosrc/audiosink subclassing (not audiobasesrc/audiobasesink) for WASAPI API, which are: * audiosrc/audiosink classes try to set high priority to read/write thread via MMCSS (Multimedia Class Scheduler Service) but it's not allowed in case of UWP application. In order to use MMCSS in UWP, application should use MediaFoundation work queue indirectly. Since audiosrc/audiosink scheduling model is not compatible with MediaFoundation's work queue model, audioringbuffer subclassing is required. * WASAPI capture device might report larger packet size than expected (i.e., larger frames we can read than expected frame size per period). Meanwhile, in any case, application should drain all packets at that moment. In order to handle the case, wasapi/wasapi2 plugins were making use of GstAdapter which is obviously sub-optimal because it requires additional memory allocation and copy. By implementing audioringbuffer subclassing, we can avoid such inefficiency. In this commit, all the device read/write operations will be moved to newly implemented wasapi2ringbuffer class and existing wasapi2client class will take care of device enumeration and activation parts only. Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2306>
2021-05-10 11:45:28 +00:00
static GstCaps *gst_wasapi2_src_get_caps (GstBaseSrc * bsrc, GstCaps * filter);
static GstAudioRingBuffer *gst_wasapi2_src_create_ringbuffer (GstAudioBaseSrc *
src);
static void gst_wasapi2_src_set_mute (GstWasapi2Src * self, gboolean mute);
static gboolean gst_wasapi2_src_get_mute (GstWasapi2Src * self);
static void gst_wasapi2_src_set_volume (GstWasapi2Src * self, gdouble volume);
static gdouble gst_wasapi2_src_get_volume (GstWasapi2Src * self);
#define gst_wasapi2_src_parent_class parent_class
wasapi2: Rewrite plugin and implement audioringbuffer subclass ... based on MediaFoundation work queue API. By this commit, wasapi2 plugin will make use of pull mode scheduling with audioringbuffer subclass. There are several drawbacks of audiosrc/audiosink subclassing (not audiobasesrc/audiobasesink) for WASAPI API, which are: * audiosrc/audiosink classes try to set high priority to read/write thread via MMCSS (Multimedia Class Scheduler Service) but it's not allowed in case of UWP application. In order to use MMCSS in UWP, application should use MediaFoundation work queue indirectly. Since audiosrc/audiosink scheduling model is not compatible with MediaFoundation's work queue model, audioringbuffer subclassing is required. * WASAPI capture device might report larger packet size than expected (i.e., larger frames we can read than expected frame size per period). Meanwhile, in any case, application should drain all packets at that moment. In order to handle the case, wasapi/wasapi2 plugins were making use of GstAdapter which is obviously sub-optimal because it requires additional memory allocation and copy. By implementing audioringbuffer subclassing, we can avoid such inefficiency. In this commit, all the device read/write operations will be moved to newly implemented wasapi2ringbuffer class and existing wasapi2client class will take care of device enumeration and activation parts only. Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2306>
2021-05-10 11:45:28 +00:00
G_DEFINE_TYPE_WITH_CODE (GstWasapi2Src, gst_wasapi2_src,
GST_TYPE_AUDIO_BASE_SRC,
G_IMPLEMENT_INTERFACE (GST_TYPE_STREAM_VOLUME, NULL));
static void
gst_wasapi2_src_class_init (GstWasapi2SrcClass * klass)
{
GObjectClass *gobject_class = G_OBJECT_CLASS (klass);
GstElementClass *element_class = GST_ELEMENT_CLASS (klass);
GstBaseSrcClass *basesrc_class = GST_BASE_SRC_CLASS (klass);
wasapi2: Rewrite plugin and implement audioringbuffer subclass ... based on MediaFoundation work queue API. By this commit, wasapi2 plugin will make use of pull mode scheduling with audioringbuffer subclass. There are several drawbacks of audiosrc/audiosink subclassing (not audiobasesrc/audiobasesink) for WASAPI API, which are: * audiosrc/audiosink classes try to set high priority to read/write thread via MMCSS (Multimedia Class Scheduler Service) but it's not allowed in case of UWP application. In order to use MMCSS in UWP, application should use MediaFoundation work queue indirectly. Since audiosrc/audiosink scheduling model is not compatible with MediaFoundation's work queue model, audioringbuffer subclassing is required. * WASAPI capture device might report larger packet size than expected (i.e., larger frames we can read than expected frame size per period). Meanwhile, in any case, application should drain all packets at that moment. In order to handle the case, wasapi/wasapi2 plugins were making use of GstAdapter which is obviously sub-optimal because it requires additional memory allocation and copy. By implementing audioringbuffer subclassing, we can avoid such inefficiency. In this commit, all the device read/write operations will be moved to newly implemented wasapi2ringbuffer class and existing wasapi2client class will take care of device enumeration and activation parts only. Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2306>
2021-05-10 11:45:28 +00:00
GstAudioBaseSrcClass *audiobasesrc_class = GST_AUDIO_BASE_SRC_CLASS (klass);
gobject_class->finalize = gst_wasapi2_src_finalize;
gobject_class->set_property = gst_wasapi2_src_set_property;
gobject_class->get_property = gst_wasapi2_src_get_property;
g_object_class_install_property (gobject_class, PROP_DEVICE,
g_param_spec_string ("device", "Device",
"WASAPI playback device as a GUID string",
NULL, GST_PARAM_MUTABLE_READY | G_PARAM_READWRITE |
G_PARAM_STATIC_STRINGS));
g_object_class_install_property (gobject_class, PROP_LOW_LATENCY,
g_param_spec_boolean ("low-latency", "Low latency",
"Optimize all settings for lowest latency. Always safe to enable.",
DEFAULT_LOW_LATENCY, GST_PARAM_MUTABLE_READY | G_PARAM_READWRITE |
G_PARAM_STATIC_STRINGS));
g_object_class_install_property (gobject_class, PROP_MUTE,
g_param_spec_boolean ("mute", "Mute", "Mute state of this stream",
DEFAULT_MUTE, GST_PARAM_MUTABLE_PLAYING | G_PARAM_READWRITE |
G_PARAM_STATIC_STRINGS));
g_object_class_install_property (gobject_class, PROP_VOLUME,
g_param_spec_double ("volume", "Volume", "Volume of this stream",
0.0, 1.0, DEFAULT_VOLUME,
GST_PARAM_MUTABLE_PLAYING | G_PARAM_READWRITE |
G_PARAM_STATIC_STRINGS));
/**
* GstWasapi2Src:dispatcher:
*
* ICoreDispatcher COM object used for activating device from UI thread.
*
* Since: 1.18
*/
g_object_class_install_property (gobject_class, PROP_DISPATCHER,
g_param_spec_pointer ("dispatcher", "Dispatcher",
"ICoreDispatcher COM object to use. In order for application to ask "
"permission of audio device, device activation should be running "
"on UI thread via ICoreDispatcher. This element will increase "
"the reference count of given ICoreDispatcher and release it after "
"use. Therefore, caller does not need to consider additional "
"reference count management",
GST_PARAM_MUTABLE_READY | G_PARAM_WRITABLE | G_PARAM_STATIC_STRINGS));
/**
* GstWasapi2Src:loopback:
*
* Open render device for loopback recording
*
* Since: 1.20
*/
g_object_class_install_property (gobject_class, PROP_LOOPBACK,
g_param_spec_boolean ("loopback", "Loopback recording",
"Open render device for loopback recording", DEFAULT_LOOPBACK,
GST_PARAM_MUTABLE_READY | G_PARAM_READWRITE |
G_PARAM_STATIC_STRINGS));
gst_element_class_add_static_pad_template (element_class, &src_template);
gst_element_class_set_static_metadata (element_class, "Wasapi2Src",
"Source/Audio/Hardware",
"Stream audio from an audio capture device through WASAPI",
"Nirbheek Chauhan <nirbheek@centricular.com>, "
"Ole André Vadla Ravnås <ole.andre.ravnas@tandberg.com>, "
"Seungha Yang <seungha@centricular.com>");
wasapi2: Rewrite plugin and implement audioringbuffer subclass ... based on MediaFoundation work queue API. By this commit, wasapi2 plugin will make use of pull mode scheduling with audioringbuffer subclass. There are several drawbacks of audiosrc/audiosink subclassing (not audiobasesrc/audiobasesink) for WASAPI API, which are: * audiosrc/audiosink classes try to set high priority to read/write thread via MMCSS (Multimedia Class Scheduler Service) but it's not allowed in case of UWP application. In order to use MMCSS in UWP, application should use MediaFoundation work queue indirectly. Since audiosrc/audiosink scheduling model is not compatible with MediaFoundation's work queue model, audioringbuffer subclassing is required. * WASAPI capture device might report larger packet size than expected (i.e., larger frames we can read than expected frame size per period). Meanwhile, in any case, application should drain all packets at that moment. In order to handle the case, wasapi/wasapi2 plugins were making use of GstAdapter which is obviously sub-optimal because it requires additional memory allocation and copy. By implementing audioringbuffer subclassing, we can avoid such inefficiency. In this commit, all the device read/write operations will be moved to newly implemented wasapi2ringbuffer class and existing wasapi2client class will take care of device enumeration and activation parts only. Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2306>
2021-05-10 11:45:28 +00:00
element_class->change_state =
GST_DEBUG_FUNCPTR (gst_wasapi2_src_change_state);
basesrc_class->get_caps = GST_DEBUG_FUNCPTR (gst_wasapi2_src_get_caps);
wasapi2: Rewrite plugin and implement audioringbuffer subclass ... based on MediaFoundation work queue API. By this commit, wasapi2 plugin will make use of pull mode scheduling with audioringbuffer subclass. There are several drawbacks of audiosrc/audiosink subclassing (not audiobasesrc/audiobasesink) for WASAPI API, which are: * audiosrc/audiosink classes try to set high priority to read/write thread via MMCSS (Multimedia Class Scheduler Service) but it's not allowed in case of UWP application. In order to use MMCSS in UWP, application should use MediaFoundation work queue indirectly. Since audiosrc/audiosink scheduling model is not compatible with MediaFoundation's work queue model, audioringbuffer subclassing is required. * WASAPI capture device might report larger packet size than expected (i.e., larger frames we can read than expected frame size per period). Meanwhile, in any case, application should drain all packets at that moment. In order to handle the case, wasapi/wasapi2 plugins were making use of GstAdapter which is obviously sub-optimal because it requires additional memory allocation and copy. By implementing audioringbuffer subclassing, we can avoid such inefficiency. In this commit, all the device read/write operations will be moved to newly implemented wasapi2ringbuffer class and existing wasapi2client class will take care of device enumeration and activation parts only. Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2306>
2021-05-10 11:45:28 +00:00
audiobasesrc_class->create_ringbuffer =
GST_DEBUG_FUNCPTR (gst_wasapi2_src_create_ringbuffer);
GST_DEBUG_CATEGORY_INIT (gst_wasapi2_src_debug, "wasapi2src",
0, "Windows audio session API source");
}
static void
gst_wasapi2_src_init (GstWasapi2Src * self)
{
self->mute = DEFAULT_MUTE;
self->volume = DEFAULT_VOLUME;
self->low_latency = DEFAULT_LOW_LATENCY;
self->loopback = DEFAULT_LOOPBACK;
}
static void
gst_wasapi2_src_finalize (GObject * object)
{
GstWasapi2Src *self = GST_WASAPI2_SRC (object);
g_free (self->device_id);
G_OBJECT_CLASS (parent_class)->finalize (object);
}
static void
gst_wasapi2_src_set_property (GObject * object, guint prop_id,
const GValue * value, GParamSpec * pspec)
{
GstWasapi2Src *self = GST_WASAPI2_SRC (object);
switch (prop_id) {
case PROP_DEVICE:
g_free (self->device_id);
self->device_id = g_value_dup_string (value);
break;
case PROP_LOW_LATENCY:
self->low_latency = g_value_get_boolean (value);
break;
case PROP_MUTE:
gst_wasapi2_src_set_mute (self, g_value_get_boolean (value));
break;
case PROP_VOLUME:
gst_wasapi2_src_set_volume (self, g_value_get_double (value));
break;
case PROP_DISPATCHER:
self->dispatcher = g_value_get_pointer (value);
break;
case PROP_LOOPBACK:
self->loopback = g_value_get_boolean (value);
break;
default:
G_OBJECT_WARN_INVALID_PROPERTY_ID (object, prop_id, pspec);
break;
}
}
static void
gst_wasapi2_src_get_property (GObject * object, guint prop_id,
GValue * value, GParamSpec * pspec)
{
GstWasapi2Src *self = GST_WASAPI2_SRC (object);
switch (prop_id) {
case PROP_DEVICE:
g_value_set_string (value, self->device_id);
break;
case PROP_LOW_LATENCY:
g_value_set_boolean (value, self->low_latency);
break;
case PROP_MUTE:
g_value_set_boolean (value, gst_wasapi2_src_get_mute (self));
break;
case PROP_VOLUME:
g_value_set_double (value, gst_wasapi2_src_get_volume (self));
break;
case PROP_LOOPBACK:
g_value_set_boolean (value, self->loopback);
break;
default:
G_OBJECT_WARN_INVALID_PROPERTY_ID (object, prop_id, pspec);
break;
}
}
wasapi2: Rewrite plugin and implement audioringbuffer subclass ... based on MediaFoundation work queue API. By this commit, wasapi2 plugin will make use of pull mode scheduling with audioringbuffer subclass. There are several drawbacks of audiosrc/audiosink subclassing (not audiobasesrc/audiobasesink) for WASAPI API, which are: * audiosrc/audiosink classes try to set high priority to read/write thread via MMCSS (Multimedia Class Scheduler Service) but it's not allowed in case of UWP application. In order to use MMCSS in UWP, application should use MediaFoundation work queue indirectly. Since audiosrc/audiosink scheduling model is not compatible with MediaFoundation's work queue model, audioringbuffer subclassing is required. * WASAPI capture device might report larger packet size than expected (i.e., larger frames we can read than expected frame size per period). Meanwhile, in any case, application should drain all packets at that moment. In order to handle the case, wasapi/wasapi2 plugins were making use of GstAdapter which is obviously sub-optimal because it requires additional memory allocation and copy. By implementing audioringbuffer subclassing, we can avoid such inefficiency. In this commit, all the device read/write operations will be moved to newly implemented wasapi2ringbuffer class and existing wasapi2client class will take care of device enumeration and activation parts only. Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2306>
2021-05-10 11:45:28 +00:00
static GstStateChangeReturn
gst_wasapi2_src_change_state (GstElement * element, GstStateChange transition)
{
GstWasapi2Src *self = GST_WASAPI2_SRC (element);
GstAudioBaseSrc *asrc = GST_AUDIO_BASE_SRC_CAST (element);
switch (transition) {
case GST_STATE_CHANGE_READY_TO_PAUSED:
/* If we have pending volume/mute values to set, do here */
GST_OBJECT_LOCK (self);
if (asrc->ringbuffer) {
GstWasapi2RingBuffer *ringbuffer =
GST_WASAPI2_RING_BUFFER (asrc->ringbuffer);
if (self->volume_changed) {
gst_wasapi2_ring_buffer_set_volume (ringbuffer, self->volume);
self->volume_changed = FALSE;
}
if (self->mute_changed) {
gst_wasapi2_ring_buffer_set_mute (ringbuffer, self->mute);
self->mute_changed = FALSE;
}
}
GST_OBJECT_UNLOCK (self);
break;
default:
break;
}
return GST_ELEMENT_CLASS (parent_class)->change_state (element, transition);
}
static GstCaps *
gst_wasapi2_src_get_caps (GstBaseSrc * bsrc, GstCaps * filter)
{
wasapi2: Rewrite plugin and implement audioringbuffer subclass ... based on MediaFoundation work queue API. By this commit, wasapi2 plugin will make use of pull mode scheduling with audioringbuffer subclass. There are several drawbacks of audiosrc/audiosink subclassing (not audiobasesrc/audiobasesink) for WASAPI API, which are: * audiosrc/audiosink classes try to set high priority to read/write thread via MMCSS (Multimedia Class Scheduler Service) but it's not allowed in case of UWP application. In order to use MMCSS in UWP, application should use MediaFoundation work queue indirectly. Since audiosrc/audiosink scheduling model is not compatible with MediaFoundation's work queue model, audioringbuffer subclassing is required. * WASAPI capture device might report larger packet size than expected (i.e., larger frames we can read than expected frame size per period). Meanwhile, in any case, application should drain all packets at that moment. In order to handle the case, wasapi/wasapi2 plugins were making use of GstAdapter which is obviously sub-optimal because it requires additional memory allocation and copy. By implementing audioringbuffer subclassing, we can avoid such inefficiency. In this commit, all the device read/write operations will be moved to newly implemented wasapi2ringbuffer class and existing wasapi2client class will take care of device enumeration and activation parts only. Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2306>
2021-05-10 11:45:28 +00:00
GstAudioBaseSrc *asrc = GST_AUDIO_BASE_SRC_CAST (bsrc);
GstCaps *caps = NULL;
wasapi2: Rewrite plugin and implement audioringbuffer subclass ... based on MediaFoundation work queue API. By this commit, wasapi2 plugin will make use of pull mode scheduling with audioringbuffer subclass. There are several drawbacks of audiosrc/audiosink subclassing (not audiobasesrc/audiobasesink) for WASAPI API, which are: * audiosrc/audiosink classes try to set high priority to read/write thread via MMCSS (Multimedia Class Scheduler Service) but it's not allowed in case of UWP application. In order to use MMCSS in UWP, application should use MediaFoundation work queue indirectly. Since audiosrc/audiosink scheduling model is not compatible with MediaFoundation's work queue model, audioringbuffer subclassing is required. * WASAPI capture device might report larger packet size than expected (i.e., larger frames we can read than expected frame size per period). Meanwhile, in any case, application should drain all packets at that moment. In order to handle the case, wasapi/wasapi2 plugins were making use of GstAdapter which is obviously sub-optimal because it requires additional memory allocation and copy. By implementing audioringbuffer subclassing, we can avoid such inefficiency. In this commit, all the device read/write operations will be moved to newly implemented wasapi2ringbuffer class and existing wasapi2client class will take care of device enumeration and activation parts only. Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2306>
2021-05-10 11:45:28 +00:00
GST_OBJECT_LOCK (bsrc);
if (asrc->ringbuffer) {
GstWasapi2RingBuffer *ringbuffer =
GST_WASAPI2_RING_BUFFER (asrc->ringbuffer);
wasapi2: Rewrite plugin and implement audioringbuffer subclass ... based on MediaFoundation work queue API. By this commit, wasapi2 plugin will make use of pull mode scheduling with audioringbuffer subclass. There are several drawbacks of audiosrc/audiosink subclassing (not audiobasesrc/audiobasesink) for WASAPI API, which are: * audiosrc/audiosink classes try to set high priority to read/write thread via MMCSS (Multimedia Class Scheduler Service) but it's not allowed in case of UWP application. In order to use MMCSS in UWP, application should use MediaFoundation work queue indirectly. Since audiosrc/audiosink scheduling model is not compatible with MediaFoundation's work queue model, audioringbuffer subclassing is required. * WASAPI capture device might report larger packet size than expected (i.e., larger frames we can read than expected frame size per period). Meanwhile, in any case, application should drain all packets at that moment. In order to handle the case, wasapi/wasapi2 plugins were making use of GstAdapter which is obviously sub-optimal because it requires additional memory allocation and copy. By implementing audioringbuffer subclassing, we can avoid such inefficiency. In this commit, all the device read/write operations will be moved to newly implemented wasapi2ringbuffer class and existing wasapi2client class will take care of device enumeration and activation parts only. Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2306>
2021-05-10 11:45:28 +00:00
gst_object_ref (ringbuffer);
GST_OBJECT_UNLOCK (bsrc);
wasapi2: Rewrite plugin and implement audioringbuffer subclass ... based on MediaFoundation work queue API. By this commit, wasapi2 plugin will make use of pull mode scheduling with audioringbuffer subclass. There are several drawbacks of audiosrc/audiosink subclassing (not audiobasesrc/audiobasesink) for WASAPI API, which are: * audiosrc/audiosink classes try to set high priority to read/write thread via MMCSS (Multimedia Class Scheduler Service) but it's not allowed in case of UWP application. In order to use MMCSS in UWP, application should use MediaFoundation work queue indirectly. Since audiosrc/audiosink scheduling model is not compatible with MediaFoundation's work queue model, audioringbuffer subclassing is required. * WASAPI capture device might report larger packet size than expected (i.e., larger frames we can read than expected frame size per period). Meanwhile, in any case, application should drain all packets at that moment. In order to handle the case, wasapi/wasapi2 plugins were making use of GstAdapter which is obviously sub-optimal because it requires additional memory allocation and copy. By implementing audioringbuffer subclassing, we can avoid such inefficiency. In this commit, all the device read/write operations will be moved to newly implemented wasapi2ringbuffer class and existing wasapi2client class will take care of device enumeration and activation parts only. Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2306>
2021-05-10 11:45:28 +00:00
/* Get caps might be able to block if device is not activated yet */
caps = gst_wasapi2_ring_buffer_get_caps (ringbuffer);
gst_object_unref (ringbuffer);
} else {
GST_OBJECT_UNLOCK (bsrc);
}
if (!caps)
caps = gst_pad_get_pad_template_caps (bsrc->srcpad);
if (filter) {
GstCaps *filtered =
gst_caps_intersect_full (filter, caps, GST_CAPS_INTERSECT_FIRST);
gst_caps_unref (caps);
caps = filtered;
}
wasapi2: Rewrite plugin and implement audioringbuffer subclass ... based on MediaFoundation work queue API. By this commit, wasapi2 plugin will make use of pull mode scheduling with audioringbuffer subclass. There are several drawbacks of audiosrc/audiosink subclassing (not audiobasesrc/audiobasesink) for WASAPI API, which are: * audiosrc/audiosink classes try to set high priority to read/write thread via MMCSS (Multimedia Class Scheduler Service) but it's not allowed in case of UWP application. In order to use MMCSS in UWP, application should use MediaFoundation work queue indirectly. Since audiosrc/audiosink scheduling model is not compatible with MediaFoundation's work queue model, audioringbuffer subclassing is required. * WASAPI capture device might report larger packet size than expected (i.e., larger frames we can read than expected frame size per period). Meanwhile, in any case, application should drain all packets at that moment. In order to handle the case, wasapi/wasapi2 plugins were making use of GstAdapter which is obviously sub-optimal because it requires additional memory allocation and copy. By implementing audioringbuffer subclassing, we can avoid such inefficiency. In this commit, all the device read/write operations will be moved to newly implemented wasapi2ringbuffer class and existing wasapi2client class will take care of device enumeration and activation parts only. Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2306>
2021-05-10 11:45:28 +00:00
GST_DEBUG_OBJECT (bsrc, "returning caps %" GST_PTR_FORMAT, caps);
return caps;
}
wasapi2: Rewrite plugin and implement audioringbuffer subclass ... based on MediaFoundation work queue API. By this commit, wasapi2 plugin will make use of pull mode scheduling with audioringbuffer subclass. There are several drawbacks of audiosrc/audiosink subclassing (not audiobasesrc/audiobasesink) for WASAPI API, which are: * audiosrc/audiosink classes try to set high priority to read/write thread via MMCSS (Multimedia Class Scheduler Service) but it's not allowed in case of UWP application. In order to use MMCSS in UWP, application should use MediaFoundation work queue indirectly. Since audiosrc/audiosink scheduling model is not compatible with MediaFoundation's work queue model, audioringbuffer subclassing is required. * WASAPI capture device might report larger packet size than expected (i.e., larger frames we can read than expected frame size per period). Meanwhile, in any case, application should drain all packets at that moment. In order to handle the case, wasapi/wasapi2 plugins were making use of GstAdapter which is obviously sub-optimal because it requires additional memory allocation and copy. By implementing audioringbuffer subclassing, we can avoid such inefficiency. In this commit, all the device read/write operations will be moved to newly implemented wasapi2ringbuffer class and existing wasapi2client class will take care of device enumeration and activation parts only. Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2306>
2021-05-10 11:45:28 +00:00
static GstAudioRingBuffer *
gst_wasapi2_src_create_ringbuffer (GstAudioBaseSrc * src)
{
wasapi2: Rewrite plugin and implement audioringbuffer subclass ... based on MediaFoundation work queue API. By this commit, wasapi2 plugin will make use of pull mode scheduling with audioringbuffer subclass. There are several drawbacks of audiosrc/audiosink subclassing (not audiobasesrc/audiobasesink) for WASAPI API, which are: * audiosrc/audiosink classes try to set high priority to read/write thread via MMCSS (Multimedia Class Scheduler Service) but it's not allowed in case of UWP application. In order to use MMCSS in UWP, application should use MediaFoundation work queue indirectly. Since audiosrc/audiosink scheduling model is not compatible with MediaFoundation's work queue model, audioringbuffer subclassing is required. * WASAPI capture device might report larger packet size than expected (i.e., larger frames we can read than expected frame size per period). Meanwhile, in any case, application should drain all packets at that moment. In order to handle the case, wasapi/wasapi2 plugins were making use of GstAdapter which is obviously sub-optimal because it requires additional memory allocation and copy. By implementing audioringbuffer subclassing, we can avoid such inefficiency. In this commit, all the device read/write operations will be moved to newly implemented wasapi2ringbuffer class and existing wasapi2client class will take care of device enumeration and activation parts only. Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2306>
2021-05-10 11:45:28 +00:00
GstWasapi2Src *self = GST_WASAPI2_SRC (src);
GstAudioRingBuffer *ringbuffer;
gchar *name;
GstWasapi2ClientDeviceClass device_class =
GST_WASAPI2_CLIENT_DEVICE_CLASS_CAPTURE;
if (self->loopback)
device_class = GST_WASAPI2_CLIENT_DEVICE_CLASS_LOOPBACK_CAPTURE;
wasapi2: Rewrite plugin and implement audioringbuffer subclass ... based on MediaFoundation work queue API. By this commit, wasapi2 plugin will make use of pull mode scheduling with audioringbuffer subclass. There are several drawbacks of audiosrc/audiosink subclassing (not audiobasesrc/audiobasesink) for WASAPI API, which are: * audiosrc/audiosink classes try to set high priority to read/write thread via MMCSS (Multimedia Class Scheduler Service) but it's not allowed in case of UWP application. In order to use MMCSS in UWP, application should use MediaFoundation work queue indirectly. Since audiosrc/audiosink scheduling model is not compatible with MediaFoundation's work queue model, audioringbuffer subclassing is required. * WASAPI capture device might report larger packet size than expected (i.e., larger frames we can read than expected frame size per period). Meanwhile, in any case, application should drain all packets at that moment. In order to handle the case, wasapi/wasapi2 plugins were making use of GstAdapter which is obviously sub-optimal because it requires additional memory allocation and copy. By implementing audioringbuffer subclassing, we can avoid such inefficiency. In this commit, all the device read/write operations will be moved to newly implemented wasapi2ringbuffer class and existing wasapi2client class will take care of device enumeration and activation parts only. Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2306>
2021-05-10 11:45:28 +00:00
name = g_strdup_printf ("%s-ringbuffer", GST_OBJECT_NAME (src));
wasapi2: Rewrite plugin and implement audioringbuffer subclass ... based on MediaFoundation work queue API. By this commit, wasapi2 plugin will make use of pull mode scheduling with audioringbuffer subclass. There are several drawbacks of audiosrc/audiosink subclassing (not audiobasesrc/audiobasesink) for WASAPI API, which are: * audiosrc/audiosink classes try to set high priority to read/write thread via MMCSS (Multimedia Class Scheduler Service) but it's not allowed in case of UWP application. In order to use MMCSS in UWP, application should use MediaFoundation work queue indirectly. Since audiosrc/audiosink scheduling model is not compatible with MediaFoundation's work queue model, audioringbuffer subclassing is required. * WASAPI capture device might report larger packet size than expected (i.e., larger frames we can read than expected frame size per period). Meanwhile, in any case, application should drain all packets at that moment. In order to handle the case, wasapi/wasapi2 plugins were making use of GstAdapter which is obviously sub-optimal because it requires additional memory allocation and copy. By implementing audioringbuffer subclassing, we can avoid such inefficiency. In this commit, all the device read/write operations will be moved to newly implemented wasapi2ringbuffer class and existing wasapi2client class will take care of device enumeration and activation parts only. Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2306>
2021-05-10 11:45:28 +00:00
ringbuffer =
gst_wasapi2_ring_buffer_new (device_class,
wasapi2: Rewrite plugin and implement audioringbuffer subclass ... based on MediaFoundation work queue API. By this commit, wasapi2 plugin will make use of pull mode scheduling with audioringbuffer subclass. There are several drawbacks of audiosrc/audiosink subclassing (not audiobasesrc/audiobasesink) for WASAPI API, which are: * audiosrc/audiosink classes try to set high priority to read/write thread via MMCSS (Multimedia Class Scheduler Service) but it's not allowed in case of UWP application. In order to use MMCSS in UWP, application should use MediaFoundation work queue indirectly. Since audiosrc/audiosink scheduling model is not compatible with MediaFoundation's work queue model, audioringbuffer subclassing is required. * WASAPI capture device might report larger packet size than expected (i.e., larger frames we can read than expected frame size per period). Meanwhile, in any case, application should drain all packets at that moment. In order to handle the case, wasapi/wasapi2 plugins were making use of GstAdapter which is obviously sub-optimal because it requires additional memory allocation and copy. By implementing audioringbuffer subclassing, we can avoid such inefficiency. In this commit, all the device read/write operations will be moved to newly implemented wasapi2ringbuffer class and existing wasapi2client class will take care of device enumeration and activation parts only. Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2306>
2021-05-10 11:45:28 +00:00
self->low_latency, self->device_id, self->dispatcher, name);
g_free (name);
wasapi2: Rewrite plugin and implement audioringbuffer subclass ... based on MediaFoundation work queue API. By this commit, wasapi2 plugin will make use of pull mode scheduling with audioringbuffer subclass. There are several drawbacks of audiosrc/audiosink subclassing (not audiobasesrc/audiobasesink) for WASAPI API, which are: * audiosrc/audiosink classes try to set high priority to read/write thread via MMCSS (Multimedia Class Scheduler Service) but it's not allowed in case of UWP application. In order to use MMCSS in UWP, application should use MediaFoundation work queue indirectly. Since audiosrc/audiosink scheduling model is not compatible with MediaFoundation's work queue model, audioringbuffer subclassing is required. * WASAPI capture device might report larger packet size than expected (i.e., larger frames we can read than expected frame size per period). Meanwhile, in any case, application should drain all packets at that moment. In order to handle the case, wasapi/wasapi2 plugins were making use of GstAdapter which is obviously sub-optimal because it requires additional memory allocation and copy. By implementing audioringbuffer subclassing, we can avoid such inefficiency. In this commit, all the device read/write operations will be moved to newly implemented wasapi2ringbuffer class and existing wasapi2client class will take care of device enumeration and activation parts only. Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2306>
2021-05-10 11:45:28 +00:00
return ringbuffer;
}
static void
gst_wasapi2_src_set_mute (GstWasapi2Src * self, gboolean mute)
{
wasapi2: Rewrite plugin and implement audioringbuffer subclass ... based on MediaFoundation work queue API. By this commit, wasapi2 plugin will make use of pull mode scheduling with audioringbuffer subclass. There are several drawbacks of audiosrc/audiosink subclassing (not audiobasesrc/audiobasesink) for WASAPI API, which are: * audiosrc/audiosink classes try to set high priority to read/write thread via MMCSS (Multimedia Class Scheduler Service) but it's not allowed in case of UWP application. In order to use MMCSS in UWP, application should use MediaFoundation work queue indirectly. Since audiosrc/audiosink scheduling model is not compatible with MediaFoundation's work queue model, audioringbuffer subclassing is required. * WASAPI capture device might report larger packet size than expected (i.e., larger frames we can read than expected frame size per period). Meanwhile, in any case, application should drain all packets at that moment. In order to handle the case, wasapi/wasapi2 plugins were making use of GstAdapter which is obviously sub-optimal because it requires additional memory allocation and copy. By implementing audioringbuffer subclassing, we can avoid such inefficiency. In this commit, all the device read/write operations will be moved to newly implemented wasapi2ringbuffer class and existing wasapi2client class will take care of device enumeration and activation parts only. Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2306>
2021-05-10 11:45:28 +00:00
GstAudioBaseSrc *bsrc = GST_AUDIO_BASE_SRC_CAST (self);
HRESULT hr;
GST_OBJECT_LOCK (self);
self->mute = mute;
self->mute_changed = TRUE;
wasapi2: Rewrite plugin and implement audioringbuffer subclass ... based on MediaFoundation work queue API. By this commit, wasapi2 plugin will make use of pull mode scheduling with audioringbuffer subclass. There are several drawbacks of audiosrc/audiosink subclassing (not audiobasesrc/audiobasesink) for WASAPI API, which are: * audiosrc/audiosink classes try to set high priority to read/write thread via MMCSS (Multimedia Class Scheduler Service) but it's not allowed in case of UWP application. In order to use MMCSS in UWP, application should use MediaFoundation work queue indirectly. Since audiosrc/audiosink scheduling model is not compatible with MediaFoundation's work queue model, audioringbuffer subclassing is required. * WASAPI capture device might report larger packet size than expected (i.e., larger frames we can read than expected frame size per period). Meanwhile, in any case, application should drain all packets at that moment. In order to handle the case, wasapi/wasapi2 plugins were making use of GstAdapter which is obviously sub-optimal because it requires additional memory allocation and copy. By implementing audioringbuffer subclassing, we can avoid such inefficiency. In this commit, all the device read/write operations will be moved to newly implemented wasapi2ringbuffer class and existing wasapi2client class will take care of device enumeration and activation parts only. Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2306>
2021-05-10 11:45:28 +00:00
if (bsrc->ringbuffer) {
GstWasapi2RingBuffer *ringbuffer =
GST_WASAPI2_RING_BUFFER (bsrc->ringbuffer);
hr = gst_wasapi2_ring_buffer_set_mute (ringbuffer, mute);
if (FAILED (hr)) {
GST_INFO_OBJECT (self, "Couldn't set mute");
} else {
self->mute_changed = FALSE;
}
}
wasapi2: Rewrite plugin and implement audioringbuffer subclass ... based on MediaFoundation work queue API. By this commit, wasapi2 plugin will make use of pull mode scheduling with audioringbuffer subclass. There are several drawbacks of audiosrc/audiosink subclassing (not audiobasesrc/audiobasesink) for WASAPI API, which are: * audiosrc/audiosink classes try to set high priority to read/write thread via MMCSS (Multimedia Class Scheduler Service) but it's not allowed in case of UWP application. In order to use MMCSS in UWP, application should use MediaFoundation work queue indirectly. Since audiosrc/audiosink scheduling model is not compatible with MediaFoundation's work queue model, audioringbuffer subclassing is required. * WASAPI capture device might report larger packet size than expected (i.e., larger frames we can read than expected frame size per period). Meanwhile, in any case, application should drain all packets at that moment. In order to handle the case, wasapi/wasapi2 plugins were making use of GstAdapter which is obviously sub-optimal because it requires additional memory allocation and copy. By implementing audioringbuffer subclassing, we can avoid such inefficiency. In this commit, all the device read/write operations will be moved to newly implemented wasapi2ringbuffer class and existing wasapi2client class will take care of device enumeration and activation parts only. Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2306>
2021-05-10 11:45:28 +00:00
GST_OBJECT_UNLOCK (self);
}
static gboolean
gst_wasapi2_src_get_mute (GstWasapi2Src * self)
{
wasapi2: Rewrite plugin and implement audioringbuffer subclass ... based on MediaFoundation work queue API. By this commit, wasapi2 plugin will make use of pull mode scheduling with audioringbuffer subclass. There are several drawbacks of audiosrc/audiosink subclassing (not audiobasesrc/audiobasesink) for WASAPI API, which are: * audiosrc/audiosink classes try to set high priority to read/write thread via MMCSS (Multimedia Class Scheduler Service) but it's not allowed in case of UWP application. In order to use MMCSS in UWP, application should use MediaFoundation work queue indirectly. Since audiosrc/audiosink scheduling model is not compatible with MediaFoundation's work queue model, audioringbuffer subclassing is required. * WASAPI capture device might report larger packet size than expected (i.e., larger frames we can read than expected frame size per period). Meanwhile, in any case, application should drain all packets at that moment. In order to handle the case, wasapi/wasapi2 plugins were making use of GstAdapter which is obviously sub-optimal because it requires additional memory allocation and copy. By implementing audioringbuffer subclassing, we can avoid such inefficiency. In this commit, all the device read/write operations will be moved to newly implemented wasapi2ringbuffer class and existing wasapi2client class will take care of device enumeration and activation parts only. Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2306>
2021-05-10 11:45:28 +00:00
GstAudioBaseSrc *bsrc = GST_AUDIO_BASE_SRC_CAST (self);
gboolean mute;
wasapi2: Rewrite plugin and implement audioringbuffer subclass ... based on MediaFoundation work queue API. By this commit, wasapi2 plugin will make use of pull mode scheduling with audioringbuffer subclass. There are several drawbacks of audiosrc/audiosink subclassing (not audiobasesrc/audiobasesink) for WASAPI API, which are: * audiosrc/audiosink classes try to set high priority to read/write thread via MMCSS (Multimedia Class Scheduler Service) but it's not allowed in case of UWP application. In order to use MMCSS in UWP, application should use MediaFoundation work queue indirectly. Since audiosrc/audiosink scheduling model is not compatible with MediaFoundation's work queue model, audioringbuffer subclassing is required. * WASAPI capture device might report larger packet size than expected (i.e., larger frames we can read than expected frame size per period). Meanwhile, in any case, application should drain all packets at that moment. In order to handle the case, wasapi/wasapi2 plugins were making use of GstAdapter which is obviously sub-optimal because it requires additional memory allocation and copy. By implementing audioringbuffer subclassing, we can avoid such inefficiency. In this commit, all the device read/write operations will be moved to newly implemented wasapi2ringbuffer class and existing wasapi2client class will take care of device enumeration and activation parts only. Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2306>
2021-05-10 11:45:28 +00:00
HRESULT hr;
wasapi2: Rewrite plugin and implement audioringbuffer subclass ... based on MediaFoundation work queue API. By this commit, wasapi2 plugin will make use of pull mode scheduling with audioringbuffer subclass. There are several drawbacks of audiosrc/audiosink subclassing (not audiobasesrc/audiobasesink) for WASAPI API, which are: * audiosrc/audiosink classes try to set high priority to read/write thread via MMCSS (Multimedia Class Scheduler Service) but it's not allowed in case of UWP application. In order to use MMCSS in UWP, application should use MediaFoundation work queue indirectly. Since audiosrc/audiosink scheduling model is not compatible with MediaFoundation's work queue model, audioringbuffer subclassing is required. * WASAPI capture device might report larger packet size than expected (i.e., larger frames we can read than expected frame size per period). Meanwhile, in any case, application should drain all packets at that moment. In order to handle the case, wasapi/wasapi2 plugins were making use of GstAdapter which is obviously sub-optimal because it requires additional memory allocation and copy. By implementing audioringbuffer subclassing, we can avoid such inefficiency. In this commit, all the device read/write operations will be moved to newly implemented wasapi2ringbuffer class and existing wasapi2client class will take care of device enumeration and activation parts only. Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2306>
2021-05-10 11:45:28 +00:00
GST_OBJECT_LOCK (self);
mute = self->mute;
wasapi2: Rewrite plugin and implement audioringbuffer subclass ... based on MediaFoundation work queue API. By this commit, wasapi2 plugin will make use of pull mode scheduling with audioringbuffer subclass. There are several drawbacks of audiosrc/audiosink subclassing (not audiobasesrc/audiobasesink) for WASAPI API, which are: * audiosrc/audiosink classes try to set high priority to read/write thread via MMCSS (Multimedia Class Scheduler Service) but it's not allowed in case of UWP application. In order to use MMCSS in UWP, application should use MediaFoundation work queue indirectly. Since audiosrc/audiosink scheduling model is not compatible with MediaFoundation's work queue model, audioringbuffer subclassing is required. * WASAPI capture device might report larger packet size than expected (i.e., larger frames we can read than expected frame size per period). Meanwhile, in any case, application should drain all packets at that moment. In order to handle the case, wasapi/wasapi2 plugins were making use of GstAdapter which is obviously sub-optimal because it requires additional memory allocation and copy. By implementing audioringbuffer subclassing, we can avoid such inefficiency. In this commit, all the device read/write operations will be moved to newly implemented wasapi2ringbuffer class and existing wasapi2client class will take care of device enumeration and activation parts only. Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2306>
2021-05-10 11:45:28 +00:00
if (bsrc->ringbuffer) {
GstWasapi2RingBuffer *ringbuffer =
GST_WASAPI2_RING_BUFFER (bsrc->ringbuffer);
hr = gst_wasapi2_ring_buffer_get_mute (ringbuffer, &mute);
if (FAILED (hr)) {
wasapi2: Rewrite plugin and implement audioringbuffer subclass ... based on MediaFoundation work queue API. By this commit, wasapi2 plugin will make use of pull mode scheduling with audioringbuffer subclass. There are several drawbacks of audiosrc/audiosink subclassing (not audiobasesrc/audiobasesink) for WASAPI API, which are: * audiosrc/audiosink classes try to set high priority to read/write thread via MMCSS (Multimedia Class Scheduler Service) but it's not allowed in case of UWP application. In order to use MMCSS in UWP, application should use MediaFoundation work queue indirectly. Since audiosrc/audiosink scheduling model is not compatible with MediaFoundation's work queue model, audioringbuffer subclassing is required. * WASAPI capture device might report larger packet size than expected (i.e., larger frames we can read than expected frame size per period). Meanwhile, in any case, application should drain all packets at that moment. In order to handle the case, wasapi/wasapi2 plugins were making use of GstAdapter which is obviously sub-optimal because it requires additional memory allocation and copy. By implementing audioringbuffer subclassing, we can avoid such inefficiency. In this commit, all the device read/write operations will be moved to newly implemented wasapi2ringbuffer class and existing wasapi2client class will take care of device enumeration and activation parts only. Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2306>
2021-05-10 11:45:28 +00:00
GST_INFO_OBJECT (self, "Couldn't get mute");
} else {
self->mute = mute;
}
}
wasapi2: Rewrite plugin and implement audioringbuffer subclass ... based on MediaFoundation work queue API. By this commit, wasapi2 plugin will make use of pull mode scheduling with audioringbuffer subclass. There are several drawbacks of audiosrc/audiosink subclassing (not audiobasesrc/audiobasesink) for WASAPI API, which are: * audiosrc/audiosink classes try to set high priority to read/write thread via MMCSS (Multimedia Class Scheduler Service) but it's not allowed in case of UWP application. In order to use MMCSS in UWP, application should use MediaFoundation work queue indirectly. Since audiosrc/audiosink scheduling model is not compatible with MediaFoundation's work queue model, audioringbuffer subclassing is required. * WASAPI capture device might report larger packet size than expected (i.e., larger frames we can read than expected frame size per period). Meanwhile, in any case, application should drain all packets at that moment. In order to handle the case, wasapi/wasapi2 plugins were making use of GstAdapter which is obviously sub-optimal because it requires additional memory allocation and copy. By implementing audioringbuffer subclassing, we can avoid such inefficiency. In this commit, all the device read/write operations will be moved to newly implemented wasapi2ringbuffer class and existing wasapi2client class will take care of device enumeration and activation parts only. Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2306>
2021-05-10 11:45:28 +00:00
GST_OBJECT_UNLOCK (self);
return mute;
}
static void
gst_wasapi2_src_set_volume (GstWasapi2Src * self, gdouble volume)
{
wasapi2: Rewrite plugin and implement audioringbuffer subclass ... based on MediaFoundation work queue API. By this commit, wasapi2 plugin will make use of pull mode scheduling with audioringbuffer subclass. There are several drawbacks of audiosrc/audiosink subclassing (not audiobasesrc/audiobasesink) for WASAPI API, which are: * audiosrc/audiosink classes try to set high priority to read/write thread via MMCSS (Multimedia Class Scheduler Service) but it's not allowed in case of UWP application. In order to use MMCSS in UWP, application should use MediaFoundation work queue indirectly. Since audiosrc/audiosink scheduling model is not compatible with MediaFoundation's work queue model, audioringbuffer subclassing is required. * WASAPI capture device might report larger packet size than expected (i.e., larger frames we can read than expected frame size per period). Meanwhile, in any case, application should drain all packets at that moment. In order to handle the case, wasapi/wasapi2 plugins were making use of GstAdapter which is obviously sub-optimal because it requires additional memory allocation and copy. By implementing audioringbuffer subclassing, we can avoid such inefficiency. In this commit, all the device read/write operations will be moved to newly implemented wasapi2ringbuffer class and existing wasapi2client class will take care of device enumeration and activation parts only. Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2306>
2021-05-10 11:45:28 +00:00
GstAudioBaseSrc *bsrc = GST_AUDIO_BASE_SRC_CAST (self);
HRESULT hr;
GST_OBJECT_LOCK (self);
self->volume = volume;
/* clip volume value */
self->volume = MAX (0.0, self->volume);
self->volume = MIN (1.0, self->volume);
self->volume_changed = TRUE;
wasapi2: Rewrite plugin and implement audioringbuffer subclass ... based on MediaFoundation work queue API. By this commit, wasapi2 plugin will make use of pull mode scheduling with audioringbuffer subclass. There are several drawbacks of audiosrc/audiosink subclassing (not audiobasesrc/audiobasesink) for WASAPI API, which are: * audiosrc/audiosink classes try to set high priority to read/write thread via MMCSS (Multimedia Class Scheduler Service) but it's not allowed in case of UWP application. In order to use MMCSS in UWP, application should use MediaFoundation work queue indirectly. Since audiosrc/audiosink scheduling model is not compatible with MediaFoundation's work queue model, audioringbuffer subclassing is required. * WASAPI capture device might report larger packet size than expected (i.e., larger frames we can read than expected frame size per period). Meanwhile, in any case, application should drain all packets at that moment. In order to handle the case, wasapi/wasapi2 plugins were making use of GstAdapter which is obviously sub-optimal because it requires additional memory allocation and copy. By implementing audioringbuffer subclassing, we can avoid such inefficiency. In this commit, all the device read/write operations will be moved to newly implemented wasapi2ringbuffer class and existing wasapi2client class will take care of device enumeration and activation parts only. Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2306>
2021-05-10 11:45:28 +00:00
if (bsrc->ringbuffer) {
GstWasapi2RingBuffer *ringbuffer =
GST_WASAPI2_RING_BUFFER (bsrc->ringbuffer);
hr = gst_wasapi2_ring_buffer_set_volume (ringbuffer, (gfloat) self->volume);
if (FAILED (hr)) {
GST_INFO_OBJECT (self, "Couldn't set volume");
} else {
self->volume_changed = FALSE;
}
}
wasapi2: Rewrite plugin and implement audioringbuffer subclass ... based on MediaFoundation work queue API. By this commit, wasapi2 plugin will make use of pull mode scheduling with audioringbuffer subclass. There are several drawbacks of audiosrc/audiosink subclassing (not audiobasesrc/audiobasesink) for WASAPI API, which are: * audiosrc/audiosink classes try to set high priority to read/write thread via MMCSS (Multimedia Class Scheduler Service) but it's not allowed in case of UWP application. In order to use MMCSS in UWP, application should use MediaFoundation work queue indirectly. Since audiosrc/audiosink scheduling model is not compatible with MediaFoundation's work queue model, audioringbuffer subclassing is required. * WASAPI capture device might report larger packet size than expected (i.e., larger frames we can read than expected frame size per period). Meanwhile, in any case, application should drain all packets at that moment. In order to handle the case, wasapi/wasapi2 plugins were making use of GstAdapter which is obviously sub-optimal because it requires additional memory allocation and copy. By implementing audioringbuffer subclassing, we can avoid such inefficiency. In this commit, all the device read/write operations will be moved to newly implemented wasapi2ringbuffer class and existing wasapi2client class will take care of device enumeration and activation parts only. Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2306>
2021-05-10 11:45:28 +00:00
GST_OBJECT_UNLOCK (self);
}
static gdouble
gst_wasapi2_src_get_volume (GstWasapi2Src * self)
{
wasapi2: Rewrite plugin and implement audioringbuffer subclass ... based on MediaFoundation work queue API. By this commit, wasapi2 plugin will make use of pull mode scheduling with audioringbuffer subclass. There are several drawbacks of audiosrc/audiosink subclassing (not audiobasesrc/audiobasesink) for WASAPI API, which are: * audiosrc/audiosink classes try to set high priority to read/write thread via MMCSS (Multimedia Class Scheduler Service) but it's not allowed in case of UWP application. In order to use MMCSS in UWP, application should use MediaFoundation work queue indirectly. Since audiosrc/audiosink scheduling model is not compatible with MediaFoundation's work queue model, audioringbuffer subclassing is required. * WASAPI capture device might report larger packet size than expected (i.e., larger frames we can read than expected frame size per period). Meanwhile, in any case, application should drain all packets at that moment. In order to handle the case, wasapi/wasapi2 plugins were making use of GstAdapter which is obviously sub-optimal because it requires additional memory allocation and copy. By implementing audioringbuffer subclassing, we can avoid such inefficiency. In this commit, all the device read/write operations will be moved to newly implemented wasapi2ringbuffer class and existing wasapi2client class will take care of device enumeration and activation parts only. Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2306>
2021-05-10 11:45:28 +00:00
GstAudioBaseSrc *bsrc = GST_AUDIO_BASE_SRC_CAST (self);
gfloat volume;
wasapi2: Rewrite plugin and implement audioringbuffer subclass ... based on MediaFoundation work queue API. By this commit, wasapi2 plugin will make use of pull mode scheduling with audioringbuffer subclass. There are several drawbacks of audiosrc/audiosink subclassing (not audiobasesrc/audiobasesink) for WASAPI API, which are: * audiosrc/audiosink classes try to set high priority to read/write thread via MMCSS (Multimedia Class Scheduler Service) but it's not allowed in case of UWP application. In order to use MMCSS in UWP, application should use MediaFoundation work queue indirectly. Since audiosrc/audiosink scheduling model is not compatible with MediaFoundation's work queue model, audioringbuffer subclassing is required. * WASAPI capture device might report larger packet size than expected (i.e., larger frames we can read than expected frame size per period). Meanwhile, in any case, application should drain all packets at that moment. In order to handle the case, wasapi/wasapi2 plugins were making use of GstAdapter which is obviously sub-optimal because it requires additional memory allocation and copy. By implementing audioringbuffer subclassing, we can avoid such inefficiency. In this commit, all the device read/write operations will be moved to newly implemented wasapi2ringbuffer class and existing wasapi2client class will take care of device enumeration and activation parts only. Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2306>
2021-05-10 11:45:28 +00:00
HRESULT hr;
wasapi2: Rewrite plugin and implement audioringbuffer subclass ... based on MediaFoundation work queue API. By this commit, wasapi2 plugin will make use of pull mode scheduling with audioringbuffer subclass. There are several drawbacks of audiosrc/audiosink subclassing (not audiobasesrc/audiobasesink) for WASAPI API, which are: * audiosrc/audiosink classes try to set high priority to read/write thread via MMCSS (Multimedia Class Scheduler Service) but it's not allowed in case of UWP application. In order to use MMCSS in UWP, application should use MediaFoundation work queue indirectly. Since audiosrc/audiosink scheduling model is not compatible with MediaFoundation's work queue model, audioringbuffer subclassing is required. * WASAPI capture device might report larger packet size than expected (i.e., larger frames we can read than expected frame size per period). Meanwhile, in any case, application should drain all packets at that moment. In order to handle the case, wasapi/wasapi2 plugins were making use of GstAdapter which is obviously sub-optimal because it requires additional memory allocation and copy. By implementing audioringbuffer subclassing, we can avoid such inefficiency. In this commit, all the device read/write operations will be moved to newly implemented wasapi2ringbuffer class and existing wasapi2client class will take care of device enumeration and activation parts only. Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2306>
2021-05-10 11:45:28 +00:00
GST_OBJECT_LOCK (self);
volume = (gfloat) self->volume;
wasapi2: Rewrite plugin and implement audioringbuffer subclass ... based on MediaFoundation work queue API. By this commit, wasapi2 plugin will make use of pull mode scheduling with audioringbuffer subclass. There are several drawbacks of audiosrc/audiosink subclassing (not audiobasesrc/audiobasesink) for WASAPI API, which are: * audiosrc/audiosink classes try to set high priority to read/write thread via MMCSS (Multimedia Class Scheduler Service) but it's not allowed in case of UWP application. In order to use MMCSS in UWP, application should use MediaFoundation work queue indirectly. Since audiosrc/audiosink scheduling model is not compatible with MediaFoundation's work queue model, audioringbuffer subclassing is required. * WASAPI capture device might report larger packet size than expected (i.e., larger frames we can read than expected frame size per period). Meanwhile, in any case, application should drain all packets at that moment. In order to handle the case, wasapi/wasapi2 plugins were making use of GstAdapter which is obviously sub-optimal because it requires additional memory allocation and copy. By implementing audioringbuffer subclassing, we can avoid such inefficiency. In this commit, all the device read/write operations will be moved to newly implemented wasapi2ringbuffer class and existing wasapi2client class will take care of device enumeration and activation parts only. Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2306>
2021-05-10 11:45:28 +00:00
if (bsrc->ringbuffer) {
GstWasapi2RingBuffer *ringbuffer =
GST_WASAPI2_RING_BUFFER (bsrc->ringbuffer);
hr = gst_wasapi2_ring_buffer_get_volume (ringbuffer, &volume);
if (FAILED (hr)) {
wasapi2: Rewrite plugin and implement audioringbuffer subclass ... based on MediaFoundation work queue API. By this commit, wasapi2 plugin will make use of pull mode scheduling with audioringbuffer subclass. There are several drawbacks of audiosrc/audiosink subclassing (not audiobasesrc/audiobasesink) for WASAPI API, which are: * audiosrc/audiosink classes try to set high priority to read/write thread via MMCSS (Multimedia Class Scheduler Service) but it's not allowed in case of UWP application. In order to use MMCSS in UWP, application should use MediaFoundation work queue indirectly. Since audiosrc/audiosink scheduling model is not compatible with MediaFoundation's work queue model, audioringbuffer subclassing is required. * WASAPI capture device might report larger packet size than expected (i.e., larger frames we can read than expected frame size per period). Meanwhile, in any case, application should drain all packets at that moment. In order to handle the case, wasapi/wasapi2 plugins were making use of GstAdapter which is obviously sub-optimal because it requires additional memory allocation and copy. By implementing audioringbuffer subclassing, we can avoid such inefficiency. In this commit, all the device read/write operations will be moved to newly implemented wasapi2ringbuffer class and existing wasapi2client class will take care of device enumeration and activation parts only. Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2306>
2021-05-10 11:45:28 +00:00
GST_INFO_OBJECT (self, "Couldn't set volume");
} else {
self->volume = volume;
}
}
wasapi2: Rewrite plugin and implement audioringbuffer subclass ... based on MediaFoundation work queue API. By this commit, wasapi2 plugin will make use of pull mode scheduling with audioringbuffer subclass. There are several drawbacks of audiosrc/audiosink subclassing (not audiobasesrc/audiobasesink) for WASAPI API, which are: * audiosrc/audiosink classes try to set high priority to read/write thread via MMCSS (Multimedia Class Scheduler Service) but it's not allowed in case of UWP application. In order to use MMCSS in UWP, application should use MediaFoundation work queue indirectly. Since audiosrc/audiosink scheduling model is not compatible with MediaFoundation's work queue model, audioringbuffer subclassing is required. * WASAPI capture device might report larger packet size than expected (i.e., larger frames we can read than expected frame size per period). Meanwhile, in any case, application should drain all packets at that moment. In order to handle the case, wasapi/wasapi2 plugins were making use of GstAdapter which is obviously sub-optimal because it requires additional memory allocation and copy. By implementing audioringbuffer subclassing, we can avoid such inefficiency. In this commit, all the device read/write operations will be moved to newly implemented wasapi2ringbuffer class and existing wasapi2client class will take care of device enumeration and activation parts only. Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2306>
2021-05-10 11:45:28 +00:00
GST_OBJECT_UNLOCK (self);
volume = MAX (0.0, volume);
volume = MIN (1.0, volume);
return volume;
}