Skip to content
GitLab
Commit 76406e9f authored by Seungha Yang's avatar Seungha Yang 🐑
Browse files

nvdec: Port to GstCudaContext 

... and add support CUDA context sharing similar to glcontext sharing.
parent b83cbd9f
Hide whitespace changes
Inline Side-by-side
sys/nvcodec/gstnvdec.c
View file @ 76406e9f
......@@ -30,6 +30,7 @@
#endif
#include "gstnvdec.h"
#include "gstcudautils.h"
typedef enum
{
......@@ -47,73 +48,10 @@ typedef struct _GstNvDecQueueItem
GST_DEBUG_CATEGORY_STATIC (gst_nvdec_debug_category);
#define GST_CAT_DEFAULT gst_nvdec_debug_category
static inline gboolean
cuda_OK (CUresult result)
{
const gchar *error_name, *error_text;
if (result != CUDA_SUCCESS) {
CuGetErrorName (result, &error_name);
CuGetErrorString (result, &error_text);
GST_WARNING ("CUDA call failed: %s, %s", error_name, error_text);
return FALSE;
}
return TRUE;
}
G_DEFINE_TYPE (GstNvDecCudaContext, gst_nvdec_cuda_context, G_TYPE_OBJECT);
static void
gst_nvdec_cuda_context_finalize (GObject * object)
{
GstNvDecCudaContext *self = (GstNvDecCudaContext *) object;
if (self->lock) {
GST_DEBUG ("destroying CUDA context lock");
if (cuda_OK (CuvidCtxLockDestroy (self->lock)))
self->lock = NULL;
else
GST_ERROR ("failed to destroy CUDA context lock");
}
if (self->context) {
GST_DEBUG ("destroying CUDA context");
if (cuda_OK (CuCtxDestroy (self->context)))
self->context = NULL;
else
GST_ERROR ("failed to destroy CUDA context");
}
G_OBJECT_CLASS (gst_nvdec_cuda_context_parent_class)->finalize (object);
}
static void
gst_nvdec_cuda_context_class_init (GstNvDecCudaContextClass * klass)
{
G_OBJECT_CLASS (klass)->finalize = gst_nvdec_cuda_context_finalize;
}
static void
gst_nvdec_cuda_context_init (GstNvDecCudaContext * self)
{
if (!cuda_OK (CuInit (0)))
GST_ERROR ("failed to init CUDA");
if (!cuda_OK (CuCtxCreate (&self->context, CU_CTX_SCHED_AUTO, 0)))
GST_ERROR ("failed to create CUDA context");
if (!cuda_OK (CuCtxPopCurrent (NULL)))
GST_ERROR ("failed to pop current CUDA context");
if (!cuda_OK (CuvidCtxLockCreate (&self->lock, self->context)))
GST_ERROR ("failed to create CUDA context lock");
}
typedef struct _GstNvDecCudaGraphicsResourceInfo
{
GstGLContext *gl_context;
GstNvDecCudaContext *cuda_context;
GstCudaContext *cuda_context;
CUgraphicsResource resource;
} GstNvDecCudaGraphicsResourceInfo;
......@@ -125,22 +63,24 @@ register_cuda_resource (GstGLContext * context, gpointer * args)
(GstNvDecCudaGraphicsResourceInfo *) args[1];
GstMapInfo map_info = GST_MAP_INFO_INIT;
guint texture_id;
GstCudaContext *ctx = cgr_info->cuda_context;
if (!cuda_OK (CuvidCtxLock (cgr_info->cuda_context->lock, 0)))
if (!gst_cuda_context_push (ctx))
GST_WARNING ("failed to lock CUDA context");
if (gst_memory_map (mem, &map_info, GST_MAP_READ | GST_MAP_GL)) {
texture_id = *(guint *) map_info.data;
if (!cuda_OK (CuGraphicsGLRegisterImage (&cgr_info->resource, texture_id,
GL_TEXTURE_2D, CU_GRAPHICS_REGISTER_FLAGS_WRITE_DISCARD)))
if (!gst_cuda_result (CuGraphicsGLRegisterImage (&cgr_info->resource,
texture_id, GL_TEXTURE_2D,
CU_GRAPHICS_REGISTER_FLAGS_WRITE_DISCARD)))
GST_WARNING ("failed to register texture with CUDA");
gst_memory_unmap (mem, &map_info);
} else
GST_WARNING ("failed to map memory");
if (!cuda_OK (CuvidCtxUnlock (cgr_info->cuda_context->lock, 0)))
if (!gst_cuda_context_pop ())
GST_WARNING ("failed to unlock CUDA context");
}
......@@ -148,14 +88,16 @@ static void
unregister_cuda_resource (GstGLContext * context,
GstNvDecCudaGraphicsResourceInfo * cgr_info)
{
if (!cuda_OK (CuvidCtxLock (cgr_info->cuda_context->lock, 0)))
GstCudaContext *ctx = cgr_info->cuda_context;
if (!gst_cuda_context_push (ctx))
GST_WARNING ("failed to lock CUDA context");
if (!cuda_OK (CuGraphicsUnregisterResource ((const CUgraphicsResource)
if (!gst_cuda_result (CuGraphicsUnregisterResource ((const CUgraphicsResource)
cgr_info->resource)))
GST_WARNING ("failed to unregister resource");
if (!cuda_OK (CuvidCtxUnlock (cgr_info->cuda_context->lock, 0)))
if (!gst_cuda_context_pop ())
GST_WARNING ("failed to unlock CUDA context");
}
......@@ -170,8 +112,7 @@ free_cgr_info (GstNvDecCudaGraphicsResourceInfo * cgr_info)
}
static CUgraphicsResource
ensure_cuda_graphics_resource (GstMemory * mem,
GstNvDecCudaContext * cuda_context)
ensure_cuda_graphics_resource (GstMemory * mem, GstCudaContext * cuda_context)
{
static GQuark quark = 0;
GstNvDecCudaGraphicsResourceInfo *cgr_info;
......@@ -202,8 +143,10 @@ ensure_cuda_graphics_resource (GstMemory * mem,
return cgr_info->resource;
}
static gboolean gst_nvdec_open (GstVideoDecoder * decoder);
static gboolean gst_nvdec_start (GstVideoDecoder * decoder);
static gboolean gst_nvdec_stop (GstVideoDecoder * decoder);
static gboolean gst_nvdec_close (GstVideoDecoder * decoder);
static gboolean gst_nvdec_set_format (GstVideoDecoder * decoder,
GstVideoCodecState * state);
static GstFlowReturn gst_nvdec_handle_frame (GstVideoDecoder * decoder,
......@@ -251,8 +194,10 @@ gst_nvdec_class_init (GstNvDecClass * klass)
"Codec/Decoder/Video/Hardware", "NVDEC video decoder",
"Ericsson AB, http://www.ericsson.com");
video_decoder_class->open = GST_DEBUG_FUNCPTR (gst_nvdec_open);
video_decoder_class->start = GST_DEBUG_FUNCPTR (gst_nvdec_start);
video_decoder_class->stop = GST_DEBUG_FUNCPTR (gst_nvdec_stop);
video_decoder_class->close = GST_DEBUG_FUNCPTR (gst_nvdec_close);
video_decoder_class->set_format = GST_DEBUG_FUNCPTR (gst_nvdec_set_format);
video_decoder_class->handle_frame =
GST_DEBUG_FUNCPTR (gst_nvdec_handle_frame);
......@@ -279,20 +224,21 @@ parser_sequence_callback (GstNvDec * nvdec, CUVIDEOFORMAT * format)
guint width, height;
CUVIDDECODECREATEINFO create_info = { 0, };
gboolean ret = TRUE;
GstCudaContext *ctx = nvdec->cuda_context;
width = format->display_area.right - format->display_area.left;
height = format->display_area.bottom - format->display_area.top;
GST_DEBUG_OBJECT (nvdec, "width: %u, height: %u", width, height);
if (!nvdec->decoder || (nvdec->width != width || nvdec->height != height)) {
if (!cuda_OK (CuvidCtxLock (nvdec->cuda_context->lock, 0))) {
if (!gst_cuda_context_push (ctx)) {
GST_ERROR_OBJECT (nvdec, "failed to lock CUDA context");
return FALSE;
}
if (nvdec->decoder) {
GST_DEBUG_OBJECT (nvdec, "destroying decoder");
if (!cuda_OK (CuvidDestroyDecoder (nvdec->decoder))) {
if (!gst_cuda_result (CuvidDestroyDecoder (nvdec->decoder))) {
GST_ERROR_OBJECT (nvdec, "failed to destroy decoder");
ret = FALSE;
} else
......@@ -315,19 +261,19 @@ parser_sequence_callback (GstNvDec * nvdec, CUVIDEOFORMAT * format)
create_info.ulTargetWidth = width;
create_info.ulTargetHeight = height;
create_info.ulNumOutputSurfaces = 1;
create_info.vidLock = nvdec->cuda_context->lock;
create_info.target_rect.left = 0;
create_info.target_rect.top = 0;
create_info.target_rect.right = width;
create_info.target_rect.bottom = height;
if (nvdec->decoder
|| !cuda_OK (CuvidCreateDecoder (&nvdec->decoder, &create_info))) {
|| !gst_cuda_result (CuvidCreateDecoder (&nvdec->decoder,
&create_info))) {
GST_ERROR_OBJECT (nvdec, "failed to create decoder");
ret = FALSE;
}
if (!cuda_OK (CuvidCtxUnlock (nvdec->cuda_context->lock, 0))) {
if (!gst_cuda_context_pop ()) {
GST_ERROR_OBJECT (nvdec, "failed to unlock CUDA context");
ret = FALSE;
}
......@@ -345,16 +291,17 @@ static gboolean
parser_decode_callback (GstNvDec * nvdec, CUVIDPICPARAMS * params)
{
GstNvDecQueueItem *item;
GstCudaContext *ctx = nvdec->cuda_context;
GST_LOG_OBJECT (nvdec, "picture index: %u", params->CurrPicIdx);
if (!cuda_OK (CuvidCtxLock (nvdec->cuda_context->lock, 0)))
if (!gst_cuda_context_push (ctx))
GST_WARNING_OBJECT (nvdec, "failed to lock CUDA context");
if (!cuda_OK (CuvidDecodePicture (nvdec->decoder, params)))
if (!gst_cuda_result (CuvidDecodePicture (nvdec->decoder, params)))
GST_WARNING_OBJECT (nvdec, "failed to decode picture");
if (!cuda_OK (CuvidCtxUnlock (nvdec->cuda_context->lock, 0)))
if (!gst_cuda_context_pop ())
GST_WARNING_OBJECT (nvdec, "failed to unlock CUDA context");
item = g_slice_new (GstNvDecQueueItem);
......@@ -383,49 +330,64 @@ parser_display_callback (GstNvDec * nvdec, CUVIDPARSERDISPINFO * dispinfo)
}
static gboolean
gst_nvdec_start (GstVideoDecoder * decoder)
gst_nvdec_open (GstVideoDecoder * decoder)
{
GstNvDec *nvdec = GST_NVDEC (decoder);
GST_DEBUG_OBJECT (nvdec, "creating CUDA context");
nvdec->cuda_context = g_object_new (gst_nvdec_cuda_context_get_type (), NULL);
nvdec->decode_queue = g_async_queue_new ();
GST_DEBUG_OBJECT (nvdec, "open");
if (!nvdec->cuda_context->context || !nvdec->cuda_context->lock) {
GST_ERROR_OBJECT (nvdec, "failed to create CUDA context or lock");
/* FIXME: set device id */
if (!gst_cuda_ensure_element_context (GST_ELEMENT_CAST (decoder),
&nvdec->cuda_context, 0)) {
GST_ERROR_OBJECT (nvdec, "failed to create CUDA context");
return FALSE;
}
return TRUE;
}
static gboolean
gst_nvdec_start (GstVideoDecoder * decoder)
{
GstNvDec *nvdec = GST_NVDEC (decoder);
if (!nvdec->cuda_context) {
GST_ERROR_OBJECT (nvdec, "No available CUDA context");
return FALSE;
}
nvdec->decode_queue = g_async_queue_new ();
return TRUE;
}
static gboolean
maybe_destroy_decoder_and_parser (GstNvDec * nvdec)
{
gboolean ret = TRUE;
if (!cuda_OK (CuvidCtxLock (nvdec->cuda_context->lock, 0))) {
if (!gst_cuda_context_push (nvdec->cuda_context)) {
GST_ERROR_OBJECT (nvdec, "failed to lock CUDA context");
return FALSE;
}
if (nvdec->decoder) {
GST_DEBUG_OBJECT (nvdec, "destroying decoder");
ret = cuda_OK (CuvidDestroyDecoder (nvdec->decoder));
ret = gst_cuda_result (CuvidDestroyDecoder (nvdec->decoder));
if (ret)
nvdec->decoder = NULL;
else
GST_ERROR_OBJECT (nvdec, "failed to destroy decoder");
}
if (!cuda_OK (CuvidCtxUnlock (nvdec->cuda_context->lock, 0))) {
if (!gst_cuda_context_pop ()) {
GST_ERROR_OBJECT (nvdec, "failed to unlock CUDA context");
return FALSE;
}
if (nvdec->parser) {
GST_DEBUG_OBJECT (nvdec, "destroying parser");
if (!cuda_OK (CuvidDestroyVideoParser (nvdec->parser))) {
if (!gst_cuda_result (CuvidDestroyVideoParser (nvdec->parser))) {
GST_ERROR_OBJECT (nvdec, "failed to destroy parser");
return FALSE;
}
......@@ -487,6 +449,18 @@ gst_nvdec_stop (GstVideoDecoder * decoder)
return TRUE;
}
static gboolean
gst_nvdec_close (GstVideoDecoder * decoder)
{
GstNvDec *nvdec = GST_NVDEC (decoder);
GST_DEBUG_OBJECT (nvdec, "close");
gst_clear_object (&nvdec->cuda_context);
return TRUE;
}
static gboolean
gst_nvdec_set_format (GstVideoDecoder * decoder, GstVideoCodecState * state)
{
......@@ -556,7 +530,8 @@ gst_nvdec_set_format (GstVideoDecoder * decoder, GstVideoCodecState * state)
(PFNVIDDISPLAYCALLBACK) parser_display_callback;
GST_DEBUG_OBJECT (nvdec, "creating parser");
if (!cuda_OK (CuvidCreateVideoParser (&nvdec->parser, &parser_params))) {
if (!gst_cuda_result (CuvidCreateVideoParser (&nvdec->parser,
&parser_params))) {
GST_ERROR_OBJECT (nvdec, "failed to create parser");
return FALSE;
}
......@@ -583,18 +558,19 @@ copy_video_frame_to_gl_textures (GstGLContext * context, gpointer * args)
proc_params.top_field_first = dispinfo->top_field_first;
proc_params.unpaired_field = dispinfo->repeat_first_field == -1;
if (!cuda_OK (CuvidCtxLock (nvdec->cuda_context->lock, 0))) {
if (!gst_cuda_context_push (nvdec->cuda_context)) {
GST_WARNING_OBJECT (nvdec, "failed to lock CUDA context");
return;
}
if (!cuda_OK (CuvidMapVideoFrame (nvdec->decoder, dispinfo->picture_index,
&dptr, &pitch, &proc_params))) {
if (!gst_cuda_result (CuvidMapVideoFrame (nvdec->decoder,
dispinfo->picture_index, &dptr, &pitch, &proc_params))) {
GST_WARNING_OBJECT (nvdec, "failed to map CUDA video frame");
goto unlock_cuda_context;
}
if (!cuda_OK (CuGraphicsMapResources (num_resources, resources, NULL))) {
if (!gst_cuda_result (CuGraphicsMapResources (num_resources, resources,
NULL))) {
GST_WARNING_OBJECT (nvdec, "failed to map CUDA resources");
goto unmap_video_frame;
}
......@@ -606,8 +582,8 @@ copy_video_frame_to_gl_textures (GstGLContext * context, gpointer * args)
mcpy2d.WidthInBytes = nvdec->width;
for (i = 0; i < num_resources; i++) {
if (!cuda_OK (CuGraphicsSubResourceGetMappedArray (&array, resources[i], 0,
0))) {
if (!gst_cuda_result (CuGraphicsSubResourceGetMappedArray (&array,
resources[i], 0, 0))) {
GST_WARNING_OBJECT (nvdec, "failed to map CUDA array");
break;
}
......@@ -616,19 +592,20 @@ copy_video_frame_to_gl_textures (GstGLContext * context, gpointer * args)
mcpy2d.dstArray = array;
mcpy2d.Height = nvdec->height / (i + 1);
if (!cuda_OK (CuMemcpy2D (&mcpy2d)))
if (!gst_cuda_result (CuMemcpy2D (&mcpy2d)))
GST_WARNING_OBJECT (nvdec, "memcpy to mapped array failed");
}
if (!cuda_OK (CuGraphicsUnmapResources (num_resources, resources, NULL)))
if (!gst_cuda_result (CuGraphicsUnmapResources (num_resources, resources,
NULL)))
GST_WARNING_OBJECT (nvdec, "failed to unmap CUDA resources");
unmap_video_frame:
if (!cuda_OK (CuvidUnmapVideoFrame (nvdec->decoder, dptr)))
if (!gst_cuda_result (CuvidUnmapVideoFrame (nvdec->decoder, dptr)))
GST_WARNING_OBJECT (nvdec, "failed to unmap CUDA video frame");
unlock_cuda_context:
if (!cuda_OK (CuvidCtxUnlock (nvdec->cuda_context->lock, 0)))
if (!gst_cuda_context_pop ())
GST_WARNING_OBJECT (nvdec, "failed to unlock CUDA context");
}
......@@ -963,7 +940,7 @@ gst_nvdec_handle_frame (GstVideoDecoder * decoder, GstVideoCodecFrame * frame)
if (GST_BUFFER_IS_DISCONT (frame->input_buffer))
packet.flags |= CUVID_PKT_DISCONTINUITY;
if (!cuda_OK (CuvidParseVideoData (nvdec->parser, &packet)))
if (!gst_cuda_result (CuvidParseVideoData (nvdec->parser, &packet)))
GST_WARNING_OBJECT (nvdec, "parser failed");
gst_buffer_unmap (frame->input_buffer, &map_info);
......@@ -984,7 +961,7 @@ gst_nvdec_flush (GstVideoDecoder * decoder)
packet.payload = NULL;
packet.flags = CUVID_PKT_ENDOFSTREAM;
if (!cuda_OK (CuvidParseVideoData (nvdec->parser, &packet)))
if (!gst_cuda_result (CuvidParseVideoData (nvdec->parser, &packet)))
GST_WARNING_OBJECT (nvdec, "parser failed");
handle_pending_frames (nvdec);
......@@ -1004,7 +981,7 @@ gst_nvdec_drain (GstVideoDecoder * decoder)
packet.payload = NULL;
packet.flags = CUVID_PKT_ENDOFSTREAM;
if (!cuda_OK (CuvidParseVideoData (nvdec->parser, &packet)))
if (!gst_cuda_result (CuvidParseVideoData (nvdec->parser, &packet)))
GST_WARNING_OBJECT (nvdec, "parser failed");
return handle_pending_frames (nvdec);
......@@ -1095,6 +1072,9 @@ gst_nvdec_src_query (GstVideoDecoder * decoder, GstQuery * query)
if (gst_gl_handle_context_query (GST_ELEMENT (decoder), query,
nvdec->gl_display, nvdec->gl_context, nvdec->other_gl_context))
return TRUE;
else if (gst_cuda_handle_context_query (GST_ELEMENT (decoder),
query, nvdec->cuda_context))
return TRUE;
break;
default:
break;
......@@ -1108,10 +1088,17 @@ static void
gst_nvdec_set_context (GstElement * element, GstContext * context)
{
GstNvDec *nvdec = GST_NVDEC (element);
GST_DEBUG_OBJECT (nvdec, "set context");
GST_DEBUG_OBJECT (nvdec, "set context %s",
gst_context_get_context_type (context));
/* FIXME: Add device-id */
if (gst_cuda_handle_set_context (element, context, &nvdec->cuda_context, 0)) {
goto done;
}
gst_gl_handle_set_context (element, context, &nvdec->gl_display,
&nvdec->other_gl_context);
done:
GST_ELEMENT_CLASS (gst_nvdec_parent_class)->set_context (element, context);
}
sys/nvcodec/gstnvdec.h
View file @ 76406e9f
......@@ -32,29 +32,11 @@
#include <gst/gl/gstglfuncs.h>
#include "gstcudaloader.h"
#include "gstcuvidloader.h"
#include "gstcudacontext.h"
#include <cudaGL.h>
G_BEGIN_DECLS
typedef struct _GstNvDecCudaContext GstNvDecCudaContext;
typedef struct _GstNvDecCudaContextClass GstNvDecCudaContextClass;
struct _GstNvDecCudaContext
{
GObject parent;
CUcontext context;
CUvideoctxlock lock;
};
struct _GstNvDecCudaContextClass
{
GObjectClass parent_class;
};
GType gst_nvdec_cuda_context_get_type (void);
#define GST_TYPE_NVDEC (gst_nvdec_get_type())
#define GST_NVDEC(obj) (G_TYPE_CHECK_INSTANCE_CAST((obj), GST_TYPE_NVDEC, GstNvDec))
#define GST_NVDEC_CLASS(klass) (G_TYPE_CHECK_CLASS_CAST((klass), GST_TYPE_NVDEC, GstNvDecClass))
......@@ -72,7 +54,7 @@ struct _GstNvDec
GstGLContext *gl_context;
GstGLContext *other_gl_context;
GstNvDecCudaContext *cuda_context;
GstCudaContext *cuda_context;
CUvideoparser parser;
CUvideodecoder decoder;
GAsyncQueue *decode_queue;
......
Supports Markdown
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment