jp9000
/
obs-studio
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
obs-studio/plugins/obs-ffmpeg/obs-ffmpeg.c

276 lines
6.3 KiB
C
Raw Normal View History

Revamp API and start using doxygen The API used to be designed in such a way to where it would expect exports for each individual source/output/encoder/etc. You would export functions for each and it would automatically load those functions based on a specific naming scheme from the module. The idea behind this was that I wanted to limit the usage of structures in the API so only functions could be used. It was an interesting idea in theory, but this idea turned out to be flawed in a number of ways: 1.) Requiring exports to create sources/outputs/encoders/etc meant that you could not create them by any other means, which meant that things like faruton's .net plugin would become difficult. 2.) Export function declarations could not be checked, therefore if you created a function with the wrong parameters and parameter types, the compiler wouldn't know how to check for that. 3.) Required overly complex load functions in libobs just to handle it. It makes much more sense to just have a load function that you call manually. Complexity is the bane of all good programs. 4.) It required that you have functions of specific names, which looked and felt somewhat unsightly. So, to fix these issues, I replaced it with a more commonly used API scheme, seen commonly in places like kernels and typical C libraries with abstraction. You simply create a structure that contains the callback definitions, and you pass it to a function to register that definition (such as obs_register_source), which you call in the obs_module_load of the module. It will also automatically check the structure size and ensure that it only loads the required values if the structure happened to add new values in an API change. The "main" source file for each module must include obs-module.h, and must use OBS_DECLARE_MODULE() within that source file. Also, started writing some doxygen documentation in to the main library headers. Will add more detailed documentation as I go.
2014-02-12 07:04:50 -08:00
#include <obs-module.h>
obs-ffmpeg: Add FFmpeg NVENC encoder
2016-04-18 10:19:49 -07:00
#include <util/platform.h>
obs-ffmpeg: Add linux VAAPI h.264 encoding support Closes obsproject/obs-studio#1482 Closes obsproject/obs-studio#1256
2018-04-08 12:05:22 -07:00
#include <libavutil/avutil.h>
obs-ffmpeg: Add FFmpeg NVENC encoder
2016-04-18 10:19:49 -07:00
#include <libavcodec/avcodec.h>
obs-ffmpeg: Call av_register_all before nvenc check Closes jp9000/obs-studio#1157
2018-01-18 01:34:15 -08:00
#include <libavformat/avformat.h>
obs-ffmpeg: Separate logging code Cleans up obs-ffmpeg.c and separates all the logging code to a different file, which can be enabled with ENABLE_FFMPEG_LOGGING.
2019-07-09 08:44:04 -07:00
#include "obs-ffmpeg-config.h"
obs-ffmpeg: Don't try to detect NVENC on mac
2018-01-17 09:22:13 -08:00
obs-ffmpeg: Implement NVENC video card blacklist This implements a blacklist of devices that ship with NVENC libraries, but do not support NVENC, which would cause the plugin to mistakenly think that NVENC was available when it was not. If these devices are the only NVIDIA devices on the system, consider NVENC unavailable. This is much more efficient than spawning an encoder on startup to see if NVENC is available, which would incur a ~500 millisecond hit on startup. Additionally, also much less crash-prone.
2019-02-05 00:33:48 -08:00
#ifdef _WIN32
#include <dxgi.h>
#include <util/dstr.h>
#include <util/windows/win-version.h>
#endif
Revamp API and start using doxygen The API used to be designed in such a way to where it would expect exports for each individual source/output/encoder/etc. You would export functions for each and it would automatically load those functions based on a specific naming scheme from the module. The idea behind this was that I wanted to limit the usage of structures in the API so only functions could be used. It was an interesting idea in theory, but this idea turned out to be flawed in a number of ways: 1.) Requiring exports to create sources/outputs/encoders/etc meant that you could not create them by any other means, which meant that things like faruton's .net plugin would become difficult. 2.) Export function declarations could not be checked, therefore if you created a function with the wrong parameters and parameter types, the compiler wouldn't know how to check for that. 3.) Required overly complex load functions in libobs just to handle it. It makes much more sense to just have a load function that you call manually. Complexity is the bane of all good programs. 4.) It required that you have functions of specific names, which looked and felt somewhat unsightly. So, to fix these issues, I replaced it with a more commonly used API scheme, seen commonly in places like kernels and typical C libraries with abstraction. You simply create a structure that contains the callback definitions, and you pass it to a function to register that definition (such as obs_register_source), which you call in the obs_module_load of the module. It will also automatically check the structure size and ensure that it only loads the required values if the structure happened to add new values in an API change. The "main" source file for each module must include obs-module.h, and must use OBS_DECLARE_MODULE() within that source file. Also, started writing some doxygen documentation in to the main library headers. Will add more detailed documentation as I go.
2014-02-12 07:04:50 -08:00
OBS_DECLARE_MODULE()
Add locale for modules
2014-07-09 22:12:57 -07:00
OBS_MODULE_USE_DEFAULT_LOCALE("obs-ffmpeg", "en-US")
Plugins: Add descriptions to modules
2018-09-11 01:51:38 -07:00
MODULE_EXPORT const char *obs_module_description(void)
{
return "FFmpeg based sources/outputs/encoders";
}
Add preliminary ffmpeg plugin (still testing) - Added some code for FFmpeg output that I'm still playing around with. Right now I'm just trying to get it to output to file and try to understand the FFmpeg/libav APIs. Hopefully in the future this plugin can be used for any sort of output to FFmpeg. - Fixed a cast warning in audio-io.c with size_t -> uint32_t - Renamed the 'video_info' and 'audio_info' structures to 'video_conver_info' and 'audio_convert_info' to better represent their actual purpose, and to avoid confusion with 'audio_output_info' and 'video_output_info' structures. - Removed a few macros from obs-def.h that were at one point going to be used but no longer going to be used (at least for now)
2014-01-19 02:16:41 -08:00
clang-format: Apply formatting Code submissions have continually suffered from formatting inconsistencies that constantly have to be addressed. Using clang-format simplifies this by making code formatting more consistent, and allows automation of the code formatting so that maintainers can focus more on the code itself instead of code formatting.
2019-06-22 22:13:45 -07:00
extern struct obs_source_info ffmpeg_source;
extern struct obs_output_info ffmpeg_output;
extern struct obs_output_info ffmpeg_muxer;
obs-ffmpeg: Use obs-ffmpeg-mux for mpegts network output This replaces the ffmpeg-encoded-output which had serious issues (missing headers, muxing issues with non compliant mpegts streams) with an output grafted from obs-ffmpeg-mux.
2020-03-24 19:27:07 -07:00
extern struct obs_output_info ffmpeg_mpegts_muxer;
clang-format: Apply formatting Code submissions have continually suffered from formatting inconsistencies that constantly have to be addressed. Using clang-format simplifies this by making code formatting more consistent, and allows automation of the code formatting so that maintainers can focus more on the code itself instead of code formatting.
2019-06-22 22:13:45 -07:00
extern struct obs_output_info replay_buffer;
Implement RTMP module (still needs drop code) - Implement the RTMP output module. This time around, we just use a simple FLV muxer, then just write to the stream with RTMP_Write. Easy and effective. - Fix the FLV muxer, the muxer now outputs proper FLV packets. - Output API: * When using encoders, automatically interleave encoded packets before sending it to the output. * Pair encoders and have them automatically wait for the other to start to ensure sync. * Change 'obs_output_signal_start_fail' to 'obs_output_signal_stop' because it was a bit confusing, and doing this makes a lot more sense for outputs that need to stop suddenly (disconnections/etc). - Encoder API: * Remove some unnecessary encoder functions from the actual API and make them internal. Most of the encoder functions are handled automatically by outputs anyway, so there's no real need to expose them and end up inadvertently confusing plugin writers. * Have audio encoders wait for the video encoder to get a frame, then start at the exact data point that the first video frame starts to ensure the most accrate sync of video/audio possible. * Add a required 'frame_size' callback for audio encoders that returns the expected number of frames desired to encode with. This way, the libobs encoder API can handle the circular buffering internally automatically for the encoder modules, so encoder writers don't have to do it themselves. - Fix a few bugs in the serializer interface. It was passing the wrong variable for the data in a few cases. - If a source has video, make obs_source_update defer the actual update callback until the tick function is called to prevent threading issues.
2014-04-07 22:00:10 -07:00
extern struct obs_encoder_info aac_encoder_info;
obs-ffmpeg: Add Opus audio encoder
2017-07-31 15:55:02 -07:00
extern struct obs_encoder_info opus_encoder_info;
obs-ffmpeg: Add FFmpeg NVENC encoder
2016-04-18 10:19:49 -07:00
extern struct obs_encoder_info nvenc_encoder_info;
Add preliminary ffmpeg plugin (still testing) - Added some code for FFmpeg output that I'm still playing around with. Right now I'm just trying to get it to output to file and try to understand the FFmpeg/libav APIs. Hopefully in the future this plugin can be used for any sort of output to FFmpeg. - Fixed a cast warning in audio-io.c with size_t -> uint32_t - Renamed the 'video_info' and 'audio_info' structures to 'video_conver_info' and 'audio_convert_info' to better represent their actual purpose, and to avoid confusion with 'audio_output_info' and 'video_output_info' structures. - Removed a few macros from obs-def.h that were at one point going to be used but no longer going to be used (at least for now)
2014-01-19 02:16:41 -08:00
obs-ffmpeg: Add linux VAAPI h.264 encoding support Closes obsproject/obs-studio#1482 Closes obsproject/obs-studio#1256
2018-04-08 12:05:22 -07:00
#if LIBAVUTIL_VERSION_INT >= AV_VERSION_INT(55, 27, 100)
#define LIBAVUTIL_VAAPI_AVAILABLE
#endif
#ifdef LIBAVUTIL_VAAPI_AVAILABLE
extern struct obs_encoder_info vaapi_encoder_info;
#endif
obs-ffmpeg: Don't try to detect NVENC on mac
2018-01-17 09:22:13 -08:00
#ifndef __APPLE__
obs-ffmpeg: Improve NVENC detection This improves detection by performing an actual check on encoder capability directly with the NVENC/CUDA libraries.
2018-01-17 04:20:34 -08:00
static const char *nvenc_check_name = "nvenc_check";
obs-ffmpeg: Implement NVENC video card blacklist This implements a blacklist of devices that ship with NVENC libraries, but do not support NVENC, which would cause the plugin to mistakenly think that NVENC was available when it was not. If these devices are the only NVIDIA devices on the system, consider NVENC unavailable. This is much more efficient than spawning an encoder on startup to see if NVENC is available, which would incur a ~500 millisecond hit on startup. Additionally, also much less crash-prone.
2019-02-05 00:33:48 -08:00
#ifdef _WIN32
obs-ffmpeg: Use vendor and device ID for NVENC blacklist
2019-10-16 03:19:09 -07:00
static const int blacklisted_adapters[] = {
0x1298, // GK208M [GeForce GT 720M]
0x1140, // GF117M [GeForce 610M/710M/810M/820M / GT 620M/625M/630M/720M]
0x1293, // GK208M [GeForce GT 730M]
0x1290, // GK208M [GeForce GT 730M]
0x0fe1, // GK107M [GeForce GT 730M]
0x0fdf, // GK107M [GeForce GT 740M]
0x1294, // GK208M [GeForce GT 740M]
0x1292, // GK208M [GeForce GT 740M]
0x0fe2, // GK107M [GeForce GT 745M]
0x0fe3, // GK107M [GeForce GT 745M]
0x1140, // GF117M [GeForce 610M/710M/810M/820M / GT 620M/625M/630M/720M]
0x0fed, // GK107M [GeForce 820M]
0x1340, // GM108M [GeForce 830M]
0x1393, // GM107M [GeForce 840M]
0x1341, // GM108M [GeForce 840M]
0x1398, // GM107M [GeForce 845M]
0x1390, // GM107M [GeForce 845M]
0x1344, // GM108M [GeForce 845M]
0x1299, // GK208BM [GeForce 920M]
0x134f, // GM108M [GeForce 920MX]
0x134e, // GM108M [GeForce 930MX]
0x1349, // GM108M [GeForce 930M]
0x1346, // GM108M [GeForce 930M]
0x179c, // GM107 [GeForce 940MX]
0x139c, // GM107M [GeForce 940M]
0x1347, // GM108M [GeForce 940M]
0x134d, // GM108M [GeForce 940MX]
0x134b, // GM108M [GeForce 940MX]
0x1399, // GM107M [GeForce 945M]
0x1348, // GM108M [GeForce 945M / 945A]
0x1d01, // GP108 [GeForce GT 1030]
0x0fc5, // GK107 [GeForce GT 1030]
0x174e, // GM108M [GeForce MX110]
0x174d, // GM108M [GeForce MX130]
0x1d10, // GP108M [GeForce MX150]
0x1d12, // GP108M [GeForce MX150]
0x1d11, // GP108M [GeForce MX230]
0x1d13, // GP108M [GeForce MX250]
0x1d52, // GP108BM [GeForce MX250]
0x137b, // GM108GLM [Quadro M520 Mobile]
0x1d33, // GP108GLM [Quadro P500 Mobile]
0x137a, // GM108GLM [Quadro K620M / Quadro M500M]
obs-ffmpeg: Implement NVENC video card blacklist This implements a blacklist of devices that ship with NVENC libraries, but do not support NVENC, which would cause the plugin to mistakenly think that NVENC was available when it was not. If these devices are the only NVIDIA devices on the system, consider NVENC unavailable. This is much more efficient than spawning an encoder on startup to see if NVENC is available, which would incur a ~500 millisecond hit on startup. Additionally, also much less crash-prone.
2019-02-05 00:33:48 -08:00
};
static const size_t num_blacklisted =
sizeof(blacklisted_adapters) / sizeof(blacklisted_adapters[0]);
obs-ffmpeg: Use vendor and device ID for NVENC blacklist
2019-10-16 03:19:09 -07:00
static bool is_blacklisted(const int device_id)
obs-ffmpeg: Implement NVENC video card blacklist This implements a blacklist of devices that ship with NVENC libraries, but do not support NVENC, which would cause the plugin to mistakenly think that NVENC was available when it was not. If these devices are the only NVIDIA devices on the system, consider NVENC unavailable. This is much more efficient than spawning an encoder on startup to see if NVENC is available, which would incur a ~500 millisecond hit on startup. Additionally, also much less crash-prone.
2019-02-05 00:33:48 -08:00
{
for (size_t i = 0; i < num_blacklisted; i++) {
obs-ffmpeg: Use vendor and device ID for NVENC blacklist
2019-10-16 03:19:09 -07:00
const int blacklisted_adapter = blacklisted_adapters[i];
if (device_id == blacklisted_adapter) {
obs-ffmpeg: Implement NVENC video card blacklist This implements a blacklist of devices that ship with NVENC libraries, but do not support NVENC, which would cause the plugin to mistakenly think that NVENC was available when it was not. If these devices are the only NVIDIA devices on the system, consider NVENC unavailable. This is much more efficient than spawning an encoder on startup to see if NVENC is available, which would incur a ~500 millisecond hit on startup. Additionally, also much less crash-prone.
2019-02-05 00:33:48 -08:00
return true;
}
}
return false;
}
clang-format: Apply formatting Code submissions have continually suffered from formatting inconsistencies that constantly have to be addressed. Using clang-format simplifies this by making code formatting more consistent, and allows automation of the code formatting so that maintainers can focus more on the code itself instead of code formatting.
2019-06-22 22:13:45 -07:00
typedef HRESULT(WINAPI *create_dxgi_proc)(const IID *, IDXGIFactory1 **);
obs-ffmpeg: Implement NVENC video card blacklist This implements a blacklist of devices that ship with NVENC libraries, but do not support NVENC, which would cause the plugin to mistakenly think that NVENC was available when it was not. If these devices are the only NVIDIA devices on the system, consider NVENC unavailable. This is much more efficient than spawning an encoder on startup to see if NVENC is available, which would incur a ~500 millisecond hit on startup. Additionally, also much less crash-prone.
2019-02-05 00:33:48 -08:00
static bool nvenc_device_available(void)
{
static HMODULE dxgi = NULL;
static create_dxgi_proc create = NULL;
IDXGIFactory1 *factory;
IDXGIAdapter1 *adapter;
bool available = false;
HRESULT hr;
UINT i = 0;
if (!dxgi) {
dxgi = GetModuleHandleW(L"dxgi");
if (!dxgi) {
dxgi = LoadLibraryW(L"dxgi");
if (!dxgi) {
return true;
}
}
}
if (!create) {
create = (create_dxgi_proc)GetProcAddress(dxgi,
clang-format: Apply formatting Code submissions have continually suffered from formatting inconsistencies that constantly have to be addressed. Using clang-format simplifies this by making code formatting more consistent, and allows automation of the code formatting so that maintainers can focus more on the code itself instead of code formatting.
2019-06-22 22:13:45 -07:00
"CreateDXGIFactory1");
obs-ffmpeg: Implement NVENC video card blacklist This implements a blacklist of devices that ship with NVENC libraries, but do not support NVENC, which would cause the plugin to mistakenly think that NVENC was available when it was not. If these devices are the only NVIDIA devices on the system, consider NVENC unavailable. This is much more efficient than spawning an encoder on startup to see if NVENC is available, which would incur a ~500 millisecond hit on startup. Additionally, also much less crash-prone.
2019-02-05 00:33:48 -08:00
if (!create) {
return true;
}
}
hr = create(&IID_IDXGIFactory1, &factory);
if (FAILED(hr)) {
return true;
}
while (factory->lpVtbl->EnumAdapters1(factory, i++, &adapter) == S_OK) {
DXGI_ADAPTER_DESC desc;
hr = adapter->lpVtbl->GetDesc(adapter, &desc);
adapter->lpVtbl->Release(adapter);
if (FAILED(hr)) {
continue;
}
obs-ffmpeg: Use vendor and device ID for NVENC blacklist
2019-10-16 03:19:09 -07:00
// 0x10de = NVIDIA Corporation
if (desc.VendorId == 0x10de && !is_blacklisted(desc.DeviceId)) {
obs-ffmpeg: Implement NVENC video card blacklist This implements a blacklist of devices that ship with NVENC libraries, but do not support NVENC, which would cause the plugin to mistakenly think that NVENC was available when it was not. If these devices are the only NVIDIA devices on the system, consider NVENC unavailable. This is much more efficient than spawning an encoder on startup to see if NVENC is available, which would incur a ~500 millisecond hit on startup. Additionally, also much less crash-prone.
2019-02-05 00:33:48 -08:00
available = true;
goto finish;
}
}
finish:
factory->lpVtbl->Release(factory);
return available;
}
#endif
obs-ffmpeg: Add texture-based NVENC encoder implementation Adds a texture-based NVENC implementation which passes OBS NV12 output textures directly to NVENC without downloading them off of the GPU, increasing NVENC performance by a significant margin. If NV12 textures are unavailable or the new encoder fails to initialize for whatever reason, it will fall back to the FFmpeg NVENC implementation safely.
2019-02-05 18:20:16 -08:00
#ifdef _WIN32
extern bool load_nvenc_lib(void);
#endif
obs-ffmpeg: Add FFmpeg NVENC encoder
2016-04-18 10:19:49 -07:00
static bool nvenc_supported(void)
{
libobs, obs-ffmpeg, win-dshow: Fix FFmpeg 4.0 deprecation Fixes FFmpeg 4.0 deprecation warnings.
2019-07-28 18:31:43 -07:00
#if LIBAVCODEC_VERSION_INT < AV_VERSION_INT(58, 9, 100)
obs-ffmpeg: Call av_register_all before nvenc check Closes jp9000/obs-studio#1157
2018-01-18 01:34:15 -08:00
av_register_all();
libobs, obs-ffmpeg, win-dshow: Fix FFmpeg 4.0 deprecation Fixes FFmpeg 4.0 deprecation warnings.
2019-07-28 18:31:43 -07:00
#endif
obs-ffmpeg: Call av_register_all before nvenc check Closes jp9000/obs-studio#1157
2018-01-18 01:34:15 -08:00
obs-ffmpeg: Improve NVENC detection This improves detection by performing an actual check on encoder capability directly with the NVENC/CUDA libraries.
2018-01-17 04:20:34 -08:00
profile_start(nvenc_check_name);
obs-ffmpeg: Call av_register_all before nvenc check Closes jp9000/obs-studio#1157
2018-01-18 01:34:15 -08:00
obs-ffmpeg: Add FFmpeg NVENC encoder
2016-04-18 10:19:49 -07:00
AVCodec *nvenc = avcodec_find_encoder_by_name("nvenc_h264");
void *lib = NULL;
obs-ffmpeg: Improve NVENC detection This improves detection by performing an actual check on encoder capability directly with the NVENC/CUDA libraries.
2018-01-17 04:20:34 -08:00
bool success = false;
obs-ffmpeg: Add FFmpeg NVENC encoder
2016-04-18 10:19:49 -07:00
obs-ffmpeg: Improve NVENC detection This improves detection by performing an actual check on encoder capability directly with the NVENC/CUDA libraries.
2018-01-17 04:20:34 -08:00
if (!nvenc) {
goto cleanup;
}
obs-ffmpeg: Add FFmpeg NVENC encoder
2016-04-18 10:19:49 -07:00
#if defined(_WIN32)
obs-ffmpeg: Implement NVENC video card blacklist This implements a blacklist of devices that ship with NVENC libraries, but do not support NVENC, which would cause the plugin to mistakenly think that NVENC was available when it was not. If these devices are the only NVIDIA devices on the system, consider NVENC unavailable. This is much more efficient than spawning an encoder on startup to see if NVENC is available, which would incur a ~500 millisecond hit on startup. Additionally, also much less crash-prone.
2019-02-05 00:33:48 -08:00
if (!nvenc_device_available()) {
goto cleanup;
}
obs-ffmpeg: Add texture-based NVENC encoder implementation Adds a texture-based NVENC implementation which passes OBS NV12 output textures directly to NVENC without downloading them off of the GPU, increasing NVENC performance by a significant margin. If NV12 textures are unavailable or the new encoder fails to initialize for whatever reason, it will fall back to the FFmpeg NVENC implementation safely.
2019-02-05 18:20:16 -08:00
if (load_nvenc_lib()) {
success = true;
goto finish;
obs-ffmpeg: Add FFmpeg NVENC encoder
2016-04-18 10:19:49 -07:00
}
#else
lib = os_dlopen("libnvidia-encode.so.1");
#endif
obs-ffmpeg: Improve NVENC detection This improves detection by performing an actual check on encoder capability directly with the NVENC/CUDA libraries.
2018-01-17 04:20:34 -08:00
/* ------------------------------------------- */
obs-ffmpeg: Remove NVENC detection code for now The reason this code is being reverted/removed is because this code is a risk to startup stability. This check will be added again in the future, however this code needs to be executed from a secondary piped process instead of directly in the process to reduce risk of driver/hardware issues impacting program startup. The same needs to be accomplished for the AMF plugin as well.
2018-01-20 05:31:58 -08:00
success = !!lib;
obs-ffmpeg: Improve NVENC detection This improves detection by performing an actual check on encoder capability directly with the NVENC/CUDA libraries.
2018-01-17 04:20:34 -08:00
cleanup:
obs-ffmpeg: Remove NVENC detection code for now The reason this code is being reverted/removed is because this code is a risk to startup stability. This check will be added again in the future, however this code needs to be executed from a secondary piped process instead of directly in the process to reduce risk of driver/hardware issues impacting program startup. The same needs to be accomplished for the AMF plugin as well.
2018-01-20 05:31:58 -08:00
if (lib)
os_dlclose(lib);
plugins: Clear all compiler warnings
2019-02-20 14:36:10 -08:00
#if defined(_WIN32)
obs-ffmpeg: Add texture-based NVENC encoder implementation Adds a texture-based NVENC implementation which passes OBS NV12 output textures directly to NVENC without downloading them off of the GPU, increasing NVENC performance by a significant margin. If NV12 textures are unavailable or the new encoder fails to initialize for whatever reason, it will fall back to the FFmpeg NVENC implementation safely.
2019-02-05 18:20:16 -08:00
finish:
plugins: Clear all compiler warnings
2019-02-20 14:36:10 -08:00
#endif
obs-ffmpeg: Improve NVENC detection This improves detection by performing an actual check on encoder capability directly with the NVENC/CUDA libraries.
2018-01-17 04:20:34 -08:00
profile_end(nvenc_check_name);
return success;
obs-ffmpeg: Add FFmpeg NVENC encoder
2016-04-18 10:19:49 -07:00
}
obs-ffmpeg: Don't try to detect NVENC on mac
2018-01-17 09:22:13 -08:00
#endif
obs-ffmpeg: Add linux VAAPI h.264 encoding support Closes obsproject/obs-studio#1482 Closes obsproject/obs-studio#1256
2018-04-08 12:05:22 -07:00
#ifdef LIBAVUTIL_VAAPI_AVAILABLE
static bool vaapi_supported(void)
{
AVCodec *vaenc = avcodec_find_encoder_by_name("h264_vaapi");
return !!vaenc;
}
#endif
obs-ffmpeg: Add texture-based NVENC encoder implementation Adds a texture-based NVENC implementation which passes OBS NV12 output textures directly to NVENC without downloading them off of the GPU, increasing NVENC performance by a significant margin. If NV12 textures are unavailable or the new encoder fails to initialize for whatever reason, it will fall back to the FFmpeg NVENC implementation safely.
2019-02-05 18:20:16 -08:00
#ifdef _WIN32
extern void jim_nvenc_load(void);
extern void jim_nvenc_unload(void);
#endif
obs-ffmpeg: Separate logging code Cleans up obs-ffmpeg.c and separates all the logging code to a different file, which can be enabled with ENABLE_FFMPEG_LOGGING.
2019-07-09 08:44:04 -07:00
#if ENABLE_FFMPEG_LOGGING
extern void obs_ffmpeg_load_logging(void);
extern void obs_ffmpeg_unload_logging(void);
#endif
Remove version parameter from obs_module_load Replaced by obs_get_version() API
2014-07-27 12:42:43 -07:00
bool obs_module_load(void)
Add planar audio support, improve test output - Add planar audio support. FFmpeg and libav use planar audio for many encoders, so it was somewhat necessary to add support in libobs itself. - Improve/adjust FFmpeg test output plugin. The exports were somewhat messed up (making me rethink how exports should be done). Not yet functional; it handles video properly, but it still does not handle audio properly. - Improve planar video code. The planar video code was not properly accounting for row sizes for each plane. Specifying row sizes for each plane has now been added. This will also make it more compatible with FFmpeg/libav. - Fixed a bug where callbacks wouldn't create properly in audio-io and video-io code. - Implement 'blogva' function to allow for va_list usage with libobs logging.
2014-02-07 02:03:54 -08:00
{
obs-ffmpeg: Add FFMpeg input source
2015-03-04 10:45:50 -08:00
obs_register_source(&ffmpeg_source);
Revamp API and start using doxygen The API used to be designed in such a way to where it would expect exports for each individual source/output/encoder/etc. You would export functions for each and it would automatically load those functions based on a specific naming scheme from the module. The idea behind this was that I wanted to limit the usage of structures in the API so only functions could be used. It was an interesting idea in theory, but this idea turned out to be flawed in a number of ways: 1.) Requiring exports to create sources/outputs/encoders/etc meant that you could not create them by any other means, which meant that things like faruton's .net plugin would become difficult. 2.) Export function declarations could not be checked, therefore if you created a function with the wrong parameters and parameter types, the compiler wouldn't know how to check for that. 3.) Required overly complex load functions in libobs just to handle it. It makes much more sense to just have a load function that you call manually. Complexity is the bane of all good programs. 4.) It required that you have functions of specific names, which looked and felt somewhat unsightly. So, to fix these issues, I replaced it with a more commonly used API scheme, seen commonly in places like kernels and typical C libraries with abstraction. You simply create a structure that contains the callback definitions, and you pass it to a function to register that definition (such as obs_register_source), which you call in the obs_module_load of the module. It will also automatically check the structure size and ensure that it only loads the required values if the structure happened to add new values in an API change. The "main" source file for each module must include obs-module.h, and must use OBS_DECLARE_MODULE() within that source file. Also, started writing some doxygen documentation in to the main library headers. Will add more detailed documentation as I go.
2014-02-12 07:04:50 -08:00
obs_register_output(&ffmpeg_output);
obs-ffmpeg: Add ffmpeg muxer In addition to the flv file format, this allows the ability to save to container formats such as mp4, ts, mkv, and any other containers that support the current codecs being used. It pipes the encoded data to the ffmpeg-mux process, which then safely muxes the file from the encoded data. If the main program unexpectedly terminates, the ffmpeg-mux piped program will safely close the file and write trailer data, preventing file corruption.
2015-05-28 23:11:37 -07:00
obs_register_output(&ffmpeg_muxer);
obs-ffmpeg: Use obs-ffmpeg-mux for mpegts network output This replaces the ffmpeg-encoded-output which had serious issues (missing headers, muxing issues with non compliant mpegts streams) with an output grafted from obs-ffmpeg-mux.
2020-03-24 19:27:07 -07:00
obs_register_output(&ffmpeg_mpegts_muxer);
obs-ffmpeg: Add replay buffer output This output allows buffering the encoded data, and then muxing the current buffer to a file on the spot via a hotkey the user sets.
2016-12-07 02:58:27 -08:00
obs_register_output(&replay_buffer);
Implement RTMP module (still needs drop code) - Implement the RTMP output module. This time around, we just use a simple FLV muxer, then just write to the stream with RTMP_Write. Easy and effective. - Fix the FLV muxer, the muxer now outputs proper FLV packets. - Output API: * When using encoders, automatically interleave encoded packets before sending it to the output. * Pair encoders and have them automatically wait for the other to start to ensure sync. * Change 'obs_output_signal_start_fail' to 'obs_output_signal_stop' because it was a bit confusing, and doing this makes a lot more sense for outputs that need to stop suddenly (disconnections/etc). - Encoder API: * Remove some unnecessary encoder functions from the actual API and make them internal. Most of the encoder functions are handled automatically by outputs anyway, so there's no real need to expose them and end up inadvertently confusing plugin writers. * Have audio encoders wait for the video encoder to get a frame, then start at the exact data point that the first video frame starts to ensure the most accrate sync of video/audio possible. * Add a required 'frame_size' callback for audio encoders that returns the expected number of frames desired to encode with. This way, the libobs encoder API can handle the circular buffering internally automatically for the encoder modules, so encoder writers don't have to do it themselves. - Fix a few bugs in the serializer interface. It was passing the wrong variable for the data in a few cases. - If a source has video, make obs_source_update defer the actual update callback until the tick function is called to prevent threading issues.
2014-04-07 22:00:10 -07:00
obs_register_encoder(&aac_encoder_info);
obs-ffmpeg: Add Opus audio encoder
2017-07-31 15:55:02 -07:00
obs_register_encoder(&opus_encoder_info);
obs-ffmpeg: Don't try to detect NVENC on mac
2018-01-17 09:22:13 -08:00
#ifndef __APPLE__
obs-ffmpeg: Add FFmpeg NVENC encoder
2016-04-18 10:19:49 -07:00
if (nvenc_supported()) {
blog(LOG_INFO, "NVENC supported");
obs-ffmpeg: Add texture-based NVENC encoder implementation Adds a texture-based NVENC implementation which passes OBS NV12 output textures directly to NVENC without downloading them off of the GPU, increasing NVENC performance by a significant margin. If NV12 textures are unavailable or the new encoder fails to initialize for whatever reason, it will fall back to the FFmpeg NVENC implementation safely.
2019-02-05 18:20:16 -08:00
#ifdef _WIN32
if (get_win_ver_int() > 0x0601) {
jim_nvenc_load();
obs-ffmpeg: Expose old NVENC on Windows 7 New NVENC isn't available on Windows 7 and the old encoder was marked as internal, giving users no option to use NVENC at all.
2020-03-26 16:16:59 -07:00
} else {
// if on Win 7, new nvenc isn't available so there's
// no nvenc encoder for the user to select, expose
// the old encoder directly
nvenc_encoder_info.caps &= ~OBS_ENCODER_CAP_INTERNAL;
obs-ffmpeg: Add texture-based NVENC encoder implementation Adds a texture-based NVENC implementation which passes OBS NV12 output textures directly to NVENC without downloading them off of the GPU, increasing NVENC performance by a significant margin. If NV12 textures are unavailable or the new encoder fails to initialize for whatever reason, it will fall back to the FFmpeg NVENC implementation safely.
2019-02-05 18:20:16 -08:00
}
#endif
obs-ffmpeg: Add FFmpeg NVENC encoder
2016-04-18 10:19:49 -07:00
obs_register_encoder(&nvenc_encoder_info);
}
obs-ffmpeg: Add linux VAAPI h.264 encoding support Closes obsproject/obs-studio#1482 Closes obsproject/obs-studio#1256
2018-04-08 12:05:22 -07:00
#if !defined(_WIN32) && defined(LIBAVUTIL_VAAPI_AVAILABLE)
if (vaapi_supported()) {
blog(LOG_INFO, "FFMPEG VAAPI supported");
obs_register_encoder(&vaapi_encoder_info);
}
#endif
obs-ffmpeg: Separate logging code Cleans up obs-ffmpeg.c and separates all the logging code to a different file, which can be enabled with ENABLE_FFMPEG_LOGGING.
2019-07-09 08:44:04 -07:00
#endif
#if ENABLE_FFMPEG_LOGGING
obs_ffmpeg_load_logging();
obs-ffmpeg: Don't try to detect NVENC on mac
2018-01-17 09:22:13 -08:00
#endif
Add planar audio support, improve test output - Add planar audio support. FFmpeg and libav use planar audio for many encoders, so it was somewhat necessary to add support in libobs itself. - Improve/adjust FFmpeg test output plugin. The exports were somewhat messed up (making me rethink how exports should be done). Not yet functional; it handles video properly, but it still does not handle audio properly. - Improve planar video code. The planar video code was not properly accounting for row sizes for each plane. Specifying row sizes for each plane has now been added. This will also make it more compatible with FFmpeg/libav. - Fixed a bug where callbacks wouldn't create properly in audio-io and video-io code. - Implement 'blogva' function to allow for va_list usage with libobs logging.
2014-02-07 02:03:54 -08:00
return true;
Add preliminary ffmpeg plugin (still testing) - Added some code for FFmpeg output that I'm still playing around with. Right now I'm just trying to get it to output to file and try to understand the FFmpeg/libav APIs. Hopefully in the future this plugin can be used for any sort of output to FFmpeg. - Fixed a cast warning in audio-io.c with size_t -> uint32_t - Renamed the 'video_info' and 'audio_info' structures to 'video_conver_info' and 'audio_convert_info' to better represent their actual purpose, and to avoid confusion with 'audio_output_info' and 'video_output_info' structures. - Removed a few macros from obs-def.h that were at one point going to be used but no longer going to be used (at least for now)
2014-01-19 02:16:41 -08:00
}
obs-ffmpeg: Sychronize logging on log context Different logging can occur from different contexts. This adds isolation for each context passed to the AV log.
2015-08-13 12:55:21 -07:00
void obs_module_unload(void)
{
obs-ffmpeg: Separate logging code Cleans up obs-ffmpeg.c and separates all the logging code to a different file, which can be enabled with ENABLE_FFMPEG_LOGGING.
2019-07-09 08:44:04 -07:00
#if ENABLE_FFMPEG_LOGGING
obs_ffmpeg_unload_logging();
obs-ffmpeg: Sychronize logging on log context Different logging can occur from different contexts. This adds isolation for each context passed to the AV log.
2015-08-13 12:55:21 -07:00
#endif
obs-ffmpeg: Add texture-based NVENC encoder implementation Adds a texture-based NVENC implementation which passes OBS NV12 output textures directly to NVENC without downloading them off of the GPU, increasing NVENC performance by a significant margin. If NV12 textures are unavailable or the new encoder fails to initialize for whatever reason, it will fall back to the FFmpeg NVENC implementation safely.
2019-02-05 18:20:16 -08:00
#ifdef _WIN32
jim_nvenc_unload();
#endif
obs-ffmpeg: Fix a race condition in create_or_fetch_log_context Another thread could be manipulating the active_log_contexts array while the current thread is trying to read it, resulting in an uninitialized memory crash as the da_push_back call was not protected by the mutex.
2015-10-12 13:41:18 -07:00
}