DM368開發 -- 編碼並實時播放
最近正好又用到 DM368 開發板，就將之前做的編解碼的專案總結一下。話說一年多沒碰，之前做的筆記全忘記是個什麼鬼了。還好整理了一下出影象了。不過再看看做的這個東西，真是夠渣的，只能作為參考了。
專案效果就是，編碼 encode 然後通過 rtsp 傳輸在 VLC 上實時播放。用的是sensor 是 MT9P031。
一、核心配置
讓其支援MT9P031
二、硬體設計錯誤排查
啟動資訊錯誤，無法檢測到MT9P031晶片
mt9p031 1-005d: No MT9P031 chip detected, register read ffffff87
vpfe-capture vpfe-capture: v4l2 sub device mt9p031 register fails通過示波器測試，i2c 的有時鐘訊號，通過資料線波形可以看到寫地址為 0xBA，但是沒有應答 ACK。等待一段時間後再發，還是沒有應答。 說明核心板有傳送裝置地址給 MT9P031 的，只不過沒有應答。經查 MT9P031 小板，第13 引腳應為高電平才會迴應裝置 ID 0xBA。但是我們的硬體設計為低電平了。問題找到。
正確的核心啟動資訊：
mt9p031 1-005d: Detected a MT9P031 chip ID 1801
mt9p031 1-005d: mt9p031 1-005d decoder driver registered !!
vpfe-capture vpfe-capture: v4l2 sub device mt9p031 registered
vpfe_register_ccdc_device: DM365 ISIF
DM365 ISIF is registered with vpfe.三、測試demo裡的encode
在 DVSDK 的資料夾 dvsdk-demos_4_02_00_01 下有編解碼的 demo 的。
其中 encode.txt 、decode.txt、encodedecode.txt 有介紹編解碼的使用。這裡只貼出 encode 的。
/*
* encode.txt
*
* This readme file explains the options and describes how to use 'encode demo'
* on DM365 platform.
*
* Copyright (C) 2011 Texas Instruments Incorporated - http://www.ti.com/
*
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
*
* Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
*
* Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the
* distribution.
*
* Neither the name of Texas Instruments Incorporated nor the names of
* its contributors may be used to endorse or promote products derived
* from this software without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
* "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
* LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
* A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
* OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
* SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
* LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
* DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
* THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
* OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*
*/
NAME
encode - encode video and/or audio or speech files
SYNOPSIS
encode [options...]
DESCRIPTION
This demo uses the Codec Engine to encode data from peripheral device
drivers to files. Video, audio and speech files are supported. The files
created will be encoded elementary streams of video/audio/speech.
You must supply at least one file for the demo to run.
The DM365MM and CMEM kernel modules need to be inserted for this demo
to run. Use the script 'loadmodule-rc' in the DVSDK to make sure both
kernel modules are loaded with adequate parameters.
OPTIONS
-v <video file>, --videofile <video file>
Encodes video data to the given file. The file will be created if
it doesn't exist, and truncated if it does exist. The demo
detects which type of video file is supplied using the file
extension. Supported video algorithm is MPEG4 SP, H.264 MP, MPEG2
(.mpeg4 or .m4v extension, .264, .m2v).
-s <speech file>, --speechfile <speech file>
Encodes speech data to the given file. The file will be created
if it doesn't exist, and truncated if it does exist. The demo
detects which type of speech file is supplied using the file
extension. The only supported speech algorithm as of now is
G.711 (.g711 extension).
-a <audio file>, --audiofile <audio file>
Encodes audio data to the given file. The file will be created
if it doesn't exist, and truncated if it does exist. The demo
detects which type of audio file is supplied using the file
extension. The only supported speech algorithm as of now is
AAC (.aac extension).
-y <1-5>, --display_standard <1-5>
Sets the resolution of the display. If the captured resolution
is larger than the display it will be center clamped, and if it
is smaller the image will be centered.
1 D1 @ 30 fps (NTSC)
2 D1 @ 25 fps (PAL)
3 720P @ 60 fps [Default]
5 1080I @ 30 fps - for DM368
-r <resolution>, --resolution <resolution>
The resolution of video to encode in the format 'width'x'height'.
Default is the size of the video standard (720x480 for NTSC,
720x576 for PAL, 1280x720 for 720P).
-b <bit rate>, --videobitrate <bit rate>
This option sets the bit rate with which the video will be
encoded. Use a negative value for variable bit rate. Default is
variable bit rate.
-p <bit rate>, --soundbitrate <bit rate>
This option sets the bit rate with which the audio will be
encoded. Use a negative value for variable bit rate. Default is
96000.
-u <sample rate>, --samplerate <sample rate>
This option sets the sample rate with which the video will be
encoded. Default is 44100 Hz.
-w, --preview_disable
Disable preview of captured video frames.
-f, --write_disable
Disable recording of encoded file. This helps to validate
performance without file I/O.
-I, --video_input
Video input source to use.
1 Composite
2 S-video
3 Component
4 Imager/Camera - for DM368
When not specified, the video input is chosen based on the display
video standard selected. NTSC/PAL use Composite, 720P uses
Component, and 1080I uses the Imager/Camera.
-l, --linein
This option makes the input device for sound recording be the
'line in' as opposed to the 'mic in' default.
-k, --keyboard
Enables the keyboard input mode which lets the user input
commands using the keyboard in addition to the QT-based OSD
interface. At the prompt type 'help' for a list of available
commands.
-t <seconds>, --time <seconds>
The number of seconds to run the demo. Defaults to infinite time.
-o, --osd
Enables the On Screen Display for configuration and data
visualization using a QT-based UI. If this option is not passed,
the data will be output to stdout instead.
-h, --help
This will print the usage of the demo.
EXAMPLE USAGE
First execute this script to load kernel modules required:
./loadmodules.sh
General usage:
./encode -h
H264 HP video encode only @720p resolution with OSD:
./encode -v test.264 -y 3 -o
H264 HP video encode from s-video and G.711 speech encode:
./encode -v test.264 -s test.g711 -I 2
MPEG4 SP video encode only in CIF NTSC resolution with OSD:
./encode -v test.mpeg4 -r 352x240 -o
MPEG4 SP video encode at 1Mbps with keyboard interface on D1 PAL display:
./encode -v test.mpeg4 -b 1000000 -k -y 2
COPYRIGHT
Copyright (c) Texas Instruments Inc 2011
Use of this software is controlled by the terms and conditions found in
the license agreement under which this software has been supplied or
provided.
KNOWN ISSUES
VERSION
4.02
CHANGELOG
from 4.01:
Modified data flow to eliminate frame copies to maximize
performance.
Modified thread priorities to enhance performance.
from 4.0:
Added support of 1080I display video standard for DM368
Added support of Camera input for DM368
Removed s-video (-x) option.
Added -I option to select video input.
Added options to disable preview and recording of encoded output.
from 3.10:
Replaced old remote-driven interface with QT-based interface. This
can be turned on using the '-o' flag.
Added audio thread to allow audio encode.
Added options for audio support ('-a', '-p', '-u').
Added MPEG2 support.
SEE ALSO
For documentation and release notes on the individual components see
the html files in the host installation directory.root@dm368-evm:/# ./encode_demo -v t.264 -I 4 -y 3 -b 1000 -w
Encode demo started.
###### output_store ######
###### davinci_get_cur_encoder ######
###### vpbe_encoder_initialize ######
###### output is COMPONENT,outindex is 1 ######
###### vpbe_encoder_setoutput ######
###### dm365 = 1 ######
###### mode_info->std is 1 ######
###### mode is 480P-60 ######
###### 22VPBE Encoder initialized ######
###### vpbe_encoder_setoutput ######
###### dm365 = 1 ######
###### mode_info->std is 1 ######
###### mode is 480P-60 ######
###### davinci_enc_set_output : next davinci_enc_set_mode_platform ######
###### davinci_enc_set_mode_platform : next davinci_enc_priv_setmode ######
###### output_show ######
###### davinci_enc_get_output ######
###### davinci_get_cur_encoder ######
###### mode_store ######
###### davinci_enc_get_mode ######
###### davinci_get_cur_encoder ######
###### davinci_enc_set_mode ######
###### davinci_get_cur_encoder ######
###### dm365 = 1 ######
###### mode_info->std is 1 ######
###### mode is 576P-50 ######
###### davinci_enc_set_mode : next davinci_enc_set_mode ######
###### davinci_enc_set_mode_platform : next davinci_enc_priv_setmode ######
###### mode_show ######
###### davinci_enc_get_mode ######
###### davinci_get_cur_encoder ######
###### davinci_enc_get_mode ######
###### davinci_get_cur_encoder ######
###### davinci_enc_get_mode ######
###### davinci_get_cur_encoder ######
davinci_resizer davinci_resizer.2: RSZ_G_CONFIG:0:1:124
davinci_previewer davinci_previewer.2: ipipe_set_preview_config
davinci_previewer davinci_previewer.2: ipipe_set_preview_config
vpfe-capture vpfe-capture: IPIPE Chained
vpfe-capture vpfe-capture: Resizer present
dm365evm_enable_pca9543a
dm365evm_enable_pca9543a, status = -121
EVM: switch to HD imager video input
######vpfe_dev->current_subdev->is_camera.
-----Exposure time = 2f2############
v4l2_device_call_until_err sdinfo->grp_id : 4.
-----Exposure time = 2f2
vpfe-capture vpfe-capture: width = 1280, height = 720, bpp = 1
vpfe-capture vpfe-capture: adjusted width = 1280, height = 720, bpp = 1, bytesperline = 1280, sizeimage = 1382400
vpfe-capture vpfe-capture: width = 1280, height = 720, bpp = 1
vpfe-capture vpfe-capture: adjusted width = 1280, height = 720, bpp = 1, bytesperline = 1280, sizeimage = 1382400
ARM Load: 28% Video fps: 31 fps Video bit rate: 42 kbps Sound bit rate: 0 kbps Time: 08:00:01 Demo: Encode Display: 720P 60Hz Video Codec: H.264 HP Resolution: 1280x720 Sound Codec: N/A Sampling Freq: N/A
ARM Load: 10% Video fps: 30 fps Video bit rate: 32 kbps Sound bit rate: 0 kbps Time: 08:00:02 Demo: Encode Display: 720P 60Hz Video Codec: H.264 HP Resolution: 1280x720 Sound Codec: N/A Sampling Freq: N/A
ARM Load: 61% Video fps: 31 fps Video bit rate: 28 kbps Sound bit rate: 0 kbps Time: 08:00:03 Demo: Encode Display: 720P 60Hz Video Codec: H.264 HP Resolution: 1280x720 Sound Codec: N/A Sampling Freq: N/A
ARM Load: 9% Video fps: 30 fps Video bit rate: 27 kbps Sound bit rate: 0 kbps Time: 08:00:04 Demo: Encode Display: 720P 60Hz Video Codec: H.264 HP Resolution: 1280x720 Sound Codec: N/A Sampling Freq: N/A
ARM Load: 62% Video fps: 30 fps Video bit rate: 35 kbps Sound bit rate: 0 kbps Time: 08:00:06 Demo: Encode Display: 720P 60Hz Video Codec: H.264 HP Resolution: 1280x720 Sound Codec: N/A Sampling Freq: N/A
ARM Load: 10% Video fps: 30 fps Video bit rate: 37 kbps Sound bit rate: 0 kbps Time: 08:00:07 Demo: Encode Display: 720P 60Hz Video Codec: H.264 HP Resolution: 1280x720 Sound Codec: N/A Sampling Freq: N/A
ARM Load: 45% Video fps: 30 fps Video bit rate: 28 kbps Sound bit rate: 0 kbps Time: 08:00:08 Demo: Encode Display: 720P 60Hz Video Codec: H.264 HP Resolution: 1280x720 Sound Codec: N/A Sampling Freq: N/A
ARM Load: 29% Video fps: 31 fps Video bit rate: 27 kbps Sound bit rate: 0 kbps Time: 08:00:09 Demo: Encode Display: 720P 60Hz Video Codec: H.264 HP Resolution: 1280x720 Sound Codec: N/A Sampling Freq: N/A
ARM Load: 23% Video fps: 29 fps Video bit rate: 29 kbps Sound bit rate: 0 kbps Time: 08:00:10 Demo: Encode Display: 720P 60Hz Video Codec: H.264 HP Resolution: 1280x720 Sound Codec: N/A Sampling Freq: N/A
ARM Load: 51% Video fps: 31 fps Video bit rate: 43 kbps Sound bit rate: 0 kbps Time: 08:00:12 Demo: Encode Display: 720P 60Hz Video Codec: H.264 HP Resolution: 1280x720 Sound Codec: N/A Sampling Freq: N/A 四、搭建 RTSP 伺服器
主要修改 writer.c 檔案，其中的原始碼部分的，這裡可以獲取一幀資料。
if (fwrite(Buffer_getUserPtr(hOutBuf),
Buffer_getNumBytesUsed(hOutBuf), 1, outFile) != 1) {
ERR("Error writing the encoded data to video file\n");
cleanup(THREAD_FAILURE);
}這協議過去這麼久我都忘了，直接上程式碼吧。
/*
* writer.c
*
* This source file has the implementations for the writer thread
* functions implemented for 'DVSDK encode demo' on DM365 platform
*
* Copyright (C) 2010 Texas Instruments Incorporated - http://www.ti.com/
*
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
*
* Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
*
* Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the
* distribution.
*
* Neither the name of Texas Instruments Incorporated nor the names of
* its contributors may be used to endorse or promote products derived
* from this software without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
* "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
* LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
* A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
* OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
* SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
* LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
* DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
* THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
* OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*/
//#include <stdio.h>
#include <xdc/std.h>
#include <ti/sdo/dmai/Fifo.h>
#include <ti/sdo/dmai/BufTab.h>
#include <ti/sdo/dmai/BufferGfx.h>
#include <ti/sdo/dmai/Rendezvous.h>
#include "writer.h"
#include "../demo.h"
/* Number of buffers in writer pipe */
#define NUM_WRITER_BUFS 10
#include <unistd.h>
#include <stdlib.h>
#include <stdio.h>
#include <string.h>
#ifdef WIN32
#include <conio.h>
#include <windows.h>
#else
#include <pthread.h>
#endif
#include "LiveServiceRTSPInvoke.h"
unsigned long dwDeliverHandle;
unsigned long dwDeliverOne;
int g_nVideoPayloadType = 96;
#define __TEST_SEED_H264_DATA__
typedef struct
{
unsigned long ulFifoMediaHandle;
int nDeliverIsOk;
int nGetIDR;
int nServiceRun;
char szVideoFmtp[256];
int nExpired;
#ifdef __TEST_SEED_H264_DATA__
/*************************************************/
/*testing*/
int nSeedTerminal;
#ifndef WIN32
pthread_t hThread;
#else
HANDLE hThread;
#endif
/************************************************/
#endif
}DATA_CONTEXT;
DATA_CONTEXT dataContext; // 資料上下文
int OnVideoExpiredCallback( void* lpData, unsigned long dwSize, int nExpired, unsigned long dwContext )
{
DATA_CONTEXT * lpContext = (DATA_CONTEXT*)dwContext;
if( !nExpired )
{
unsigned char * lpH264Slice = (unsigned char *)lpData;
if( *(lpH264Slice+0) == 0 &&
*(lpH264Slice+1) == 0 &&
*(lpH264Slice+2) == 0 &&
*(lpH264Slice+3) == 1 )
{
if( 0x07 == ((*(lpH264Slice+4))&0x1f) )
lpContext->nExpired = 0; /* 廢棄資料直到IDR*/
}
}
else
lpContext->nExpired = 1;
return lpContext->nExpired;
}
static const char cb64[]="ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/";
void live_base64encode(char *out,const unsigned char* in,int len)
{
for(;len >= 3; len -= 3)
{
*out ++ = cb64[ in[0]>>2];
*out ++ = cb64[ ((in[0]&0x03)<<4)|(in[1]>>4) ];
*out ++ = cb64[ ((in[1]&0x0f) <<2)|((in[2] & 0xc0)>>6) ];
*out ++ = cb64[ in[2]&0x3f];
in+=3;
}
if(len > 0)
{
unsigned char fragment;
*out ++ = cb64[ in[0]>>2];
fragment = (in[0] &0x03)<<4 ;
if(len > 1)
fragment |= in[1] >>4 ;
*out ++ = cb64[ fragment ];
*out ++ = (len <2) ? '=' : cb64[ ((in[1]&0x0f)<<2)];
*out ++ = '=';
}
*out = '\0';
}
/*---live_base64encode---*/
//視訊RTSP傳送
int ON_VIDEO_DELIVER_RTSP_SEND( void* lpStreamData, unsigned long * lpdwSize, unsigned long dwContext )
{
DATA_CONTEXT * lpDataContext = (DATA_CONTEXT*)dwContext;
unsigned char* lpDataBuf = NULL;
unsigned long dwSize = 0;
unsigned long dwTick = 0;
if( NULL == lpStreamData || NULL == lpdwSize )
{
/* THIS CHANNEL NOW TO DISABLE*/
lpDataContext->nDeliverIsOk = 0;
return 0;
}
*lpdwSize = 0;
lpDataContext->nDeliverIsOk = 1;
if( LiveDeliverRTSP_FifoMediaPopOne( lpDataContext->ulFifoMediaHandle, &lpDataBuf, &dwSize, &dwTick ) )
{
*lpdwSize = dwSize;
memcpy( lpStreamData, lpDataBuf, dwSize );
LiveDeliverRTSP_FifoMediaFreeOne( lpDataContext->ulFifoMediaHandle, lpDataBuf, dwSize );
}
#if 0 //cjk
else
{
*lpdwSize = 0;
}
#endif
return 0;
}
//檢測H264幀資料OK
void OnH264FrameDataOK( DATA_CONTEXT * lpDataContext, unsigned char * lpDataBuf, unsigned long ulSize )
{
if( lpDataContext && lpDataBuf && ulSize )
{
if( !lpDataContext->nGetIDR )
{
unsigned char * lpSPS = NULL;
unsigned char * lpSPS_E = NULL;
unsigned char * lpPPS = NULL;
unsigned char * lpPPS_E = NULL;
unsigned int fNalType = 0;
if( ( 0 == *(lpDataBuf+0) ) &&
( 0 == *(lpDataBuf+1) ) &&
( 0 == *(lpDataBuf+2) ) &&
( 1 == *(lpDataBuf+3) ) )
{
fNalType = (0x01f&(*(lpDataBuf+4)));
}
if( 7 == fNalType ) /*SPS*/
{
unsigned char * pNextNal = lpDataBuf+4;
unsigned char * pEnd = lpDataBuf+ulSize;
lpSPS = lpDataBuf;
while( pNextNal < pEnd )
{
if( ( 0 == *(pNextNal+0) ) &&
( 0 == *(pNextNal+1) ) &&
( 0 == *(pNextNal+2) ) &&
( 1 == *(pNextNal+3) ) )
{
if( !lpSPS_E )
lpSPS_E = pNextNal;
fNalType = (0x01f&(*(pNextNal+4)));
if( 8 == fNalType ) /*PPS*/
{
if( !lpPPS )
lpPPS = pNextNal;
}
else
{
if( lpPPS && !lpPPS_E )
lpPPS_E = pNextNal;
}
if( lpPPS_E )
break;
}
pNextNal++;
}
if( lpSPS && lpSPS_E && lpPPS && lpPPS_E )
{
char szSPS[128];
char szPPS[32];
unsigned int profile_level_id = 0x420029;
unsigned char spsWEB[3];
live_base64encode( szSPS, lpSPS+5, lpSPS_E-lpSPS-5 );
live_base64encode( szPPS, lpPPS+5, lpPPS_E-lpPPS-5 );
// sprintf( szSPS, "%s", "Z00AKZpmA8ARPyzUBAQFAAADA+gAAMNQBA==" );
// sprintf( szPPS, "%s", "aO48gA==" );
spsWEB[0] = *(lpSPS+5);
spsWEB[1] = *(lpSPS+6);
spsWEB[2] = *(lpSPS+7);
profile_level_id = (spsWEB[0]<<16) | (spsWEB[1]<<8) | spsWEB[2];
sprintf( lpDataContext->szVideoFmtp,
"a=fmtp:%d packetization-mode=1;profile-level-id=%06X;sprop-parameter-sets=%s,%s",
g_nVideoPayloadType,
profile_level_id,
szSPS,
szPPS );
lpDataContext->nGetIDR = 1;
}
}
}
//管道FIFO新的一個
LiveDeliverRTSP_FifoMediaNewOne( lpDataContext->ulFifoMediaHandle, lpDataBuf, ulSize );
}
}
void iRelease (int signo)
{
if( dwDeliverOne )
{
LiveDeliverRTSP_StopOne( dwDeliverHandle, dwDeliverOne );
LiveDeliverRTSP_ReleaseOne( dwDeliverHandle, &dwDeliverOne );
}
LiveDeliverRTSP_Exit( &dwDeliverHandle );
if( dataContext.ulFifoMediaHandle )
{
LiveDeliverRTSP_FifoMediaRelease( dataContext.ulFifoMediaHandle );
dataContext.ulFifoMediaHandle = 0;
}
printf ("iRelease ok!!!!!!!!!!!\n");
//exit (0);
}
/******************************************************************************
* writerThrFxn
******************************************************************************/
Void *writerThrFxn(Void *arg)
{
signal (SIGINT, iRelease); //cjk
sleep (2);
WriterEnv *envp = (WriterEnv *) arg;
Void *status = THREAD_SUCCESS;
FILE *outFile = NULL;
Buffer_Attrs bAttrs = Buffer_Attrs_DEFAULT;
BufTab_Handle hBufTab = NULL;
Buffer_Handle hOutBuf;
Int fifoRet;
Int bufIdx;
//建立FIFO 建立的函式不太一樣
dataContext.ulFifoMediaHandle = LiveDeliverRTSP_FifoMediaCreate( 1000, OnVideoExpiredCallback, (unsigned long)&dataContext );
dataContext.nDeliverIsOk = 0; //isok
dataContext.nGetIDR = 0; //IDR
dataContext.nExpired = 0;
memset( dataContext.szVideoFmtp, 0, sizeof(dataContext.szVideoFmtp) );
dwDeliverHandle = LiveDeliverRTSP_Init( 8765 );
dwDeliverOne = 0;
/* Open the output video file */
outFile = fopen(envp->videoFile, "w");
if (outFile == NULL) {
ERR("Failed to open %s for writing\n", envp->videoFile);
cleanup(THREAD_FAILURE);
}
/*
* Create a table of buffers for communicating buffers to
* and from the video thread.
*/
hBufTab = BufTab_create(NUM_WRITER_BUFS, envp->outBufSize, &bAttrs);
if (hBufTab == NULL) {
ERR("Failed to allocate contiguous buffers\n");
cleanup(THREAD_FAILURE);
}
/* Send all buffers to the video thread to be filled with encoded data */
for (bufIdx = 0; bufIdx < NUM_WRITER_BUFS; bufIdx++) {
if (Fifo_put(envp->hOutFifo, BufTab_getBuf(hBufTab, bufIdx)) < 0) {
ERR("Failed to send buffer to display thread\n");
cleanup(THREAD_FAILURE);
}
}
/* Signal that initialization is done and wait for other threads */
Rendezvous_meet(envp->hRendezvousInit);
int iRTSPCreateFlag = 0;
dataContext.nGetIDR = 0;
while (TRUE) {
if(0 == iRTSPCreateFlag)
{
if( dataContext.nGetIDR )
{
int nVideoBPS = 8*1024*1024;
dwDeliverOne = LiveDeliverRTSP_CreateOne( dwDeliverHandle, "pvm",
ON_VIDEO_DELIVER_RTSP_SEND, (unsigned long)&dataContext,
NULL, 0 );
LiveDeliverRTSP_SetupOne( dwDeliverHandle, dwDeliverOne, 1, g_nVideoPayloadType, "H264", 90000, dataContext.szVideoFmtp, nVideoBPS );
LiveDeliverRTSP_PlayOne( dwDeliverHandle, dwDeliverOne );
dataContext.nDeliverIsOk = 1;
dataContext.nServiceRun = 1;
iRTSPCreateFlag = 1;
//sleep (10);
//writerEnv.nDeliverIsOk = 0;
//writerEnv.nGetIDR = 0;
}
}
/* Get an encoded buffer from the video thread */
fifoRet = Fifo_get(envp->hInFifo, &hOutBuf);
if (fifoRet < 0) {
ERR("Failed to get buffer from video thread\n");
cleanup(THREAD_FAILURE);
}
/* Did the video thread flush the fifo? */
if (fifoRet == Dmai_EFLUSH) {
cleanup(THREAD_SUCCESS);
}
if (!envp->writeDisabled) {
/* Store the encoded frame to disk */
if (Buffer_getNumBytesUsed(hOutBuf)) {
#if 0
if (fwrite(Buffer_getUserPtr(hOutBuf),
Buffer_getNumBytesUsed(hOutBuf), 1, outFile) != 1) {
ERR("Error writing the encoded data to video file\n");
cleanup(THREAD_FAILURE);
#endif
unsigned char * lpReadBuf = (unsigned char *)Buffer_getUserPtr(hOutBuf);
unsigned long lpDateSize = (unsigned long)Buffer_getNumBytesUsed(hOutBuf);
unsigned char * lpFrameBuf = NULL;
unsigned char nFrameHeadFlag = 0;
while(!nFrameHeadFlag)
{
if((*lpReadBuf == 0 ) &&
(*(lpReadBuf+1) == 0 ) &&
(*(lpReadBuf+2) == 0 ) &&
(*(lpReadBuf+3) == 1 )
)
{
//int t = 0x1f &(*(lpReadBuf+4));
//if( t == 7 || t == 1 ){
nFrameHeadFlag = 1;
lpFrameBuf = lpReadBuf;
//}
}
else
{
lpDateSize -= 1;
lpReadBuf++;
}
OnH264FrameDataOK( &dataContext, lpReadBuf, lpDateSize );
usleep( 10000 );
}
}
else {
printf("Warning, writer received 0 byte encoded frame\n");
}
}
/* Return buffer to capture thread */
if (Fifo_put(envp->hOutFifo, hOutBuf) < 0) {
ERR("Failed to send buffer to display thread\n");
cleanup(THREAD_FAILURE);
}
}
cleanup:
/* Make sure the other threads aren't waiting for us */
Rendezvous_force(envp->hRendezvousInit);
Pause_off(envp->hPauseProcess);
Fifo_flush(envp->hOutFifo);
/* Meet up with other threads before cleaning up */
Rendezvous_meet(envp->hRendezvousCleanup);
/* Clean up the thread before exiting */
if (outFile) {
fclose(outFile);
}
if (hBufTab) {
BufTab_delete(hBufTab);
}
return status;
}五、專案下載
下載：編碼並實時播放
六、測試
開發板上執行：
./encode -v d.264 -I 4 -y 3 -w
在 VLC 輸入網路 ULR：
rtsp://192.168.2.xxx:8765/pvm
點選播放
七、問題
我說這個專案很渣吧，存在很多問題。
第一個問題：在 VLC 輸入網路 ULR，只能在執行encode的同時輸入才能有效。比如encode執行過程中都不行。
第二個問題：影象無顏色，因為sensor沒有加濾光片。
第三個問題：視訊有一點延時。
第四個問題：encode 多次執行會出現段錯誤。
第五個問題：硬體 主頻不是用的432Mhz
第六個問題：硬體 sensor對焦只能手動對焦
vpfe-capture vpfe-capture: width = 1280, height = 720, bpp = 1
vpfe-capture vpfe-capture: adjusted width = 1280, height = 720, bpp = 1, bytesperline = 1280, sizeimage = 1382400
vpfe-capture vpfe-capture: width = 1280, height = 720, bpp = 1
vpfe-capture vpfe-capture: adjusted width = 1280, height = 720, bpp = 1, bytesperline = 1280, sizeimage = 1382400
### EnableLiveOne - 1 ###
### EnableLiveOne - 2 ###
### EnableLiveOne - 3 ###
### EnableLiveOne - 5 ###
### EnableLiveOne - 6 ###
### EnableLiveOne - 7 ###
### EnableLiveOne - 8 ###
### EnableLiveOne - 9 ###
### m_lpLiveRTSPServerCore is null ###
Segmentation fault
相關文章
- Android音訊開發之AudioTrack實時播放Android音訊
- FFmpeg開發筆記（九）：ffmpeg解碼rtsp流並使用SDL同步播放筆記
- Android音訊實時傳輸與播放（三）：AMR硬編碼與硬解碼Android音訊
- FFmpeg開發筆記（八）：ffmpeg解碼音訊並使用SDL同步音訊播放筆記音訊
- Android MediaCodec硬解碼AAC音訊檔案（實時AAC音訊幀）並播放Android音訊
- iOS專案開發實戰——CoreLocation地理編碼和反地理編碼iOS
- 實時程式碼編輯器
- 用Gulp編譯Sass並實時重新整理編譯
- Android開發 海康威視 多路視訊播放（同時播放視訊）Android
- ffmpeg播放器開發 詳細記錄+程式碼實現3播放器
- DM6446的音訊編解碼及播放實現音訊
- 前端開發編碼規範前端
- pygame播放影片並實現音影片同步GAM
- 第二章 編譯FFmpeg並開啟H.264編碼編譯
- 您是否也有想在瀏覽器中實時的編輯程式碼並且渲染的想法？瀏覽器
- Django開發——003程式碼編寫Django
- iOS開發——OC編碼規範iOS
- C#開發編碼規範C#
- 前端開發中的字元編碼前端字元
- Gitee Webhook 實現自動拉取程式碼並編譯程式碼GiteeWebHook編譯
- Android 音樂播放器開發實錄（MediaSession)Android播放器Session
- iOS開發 整合ijkplayer實現直播#拉流#播放iOS
- 使用海康H5視訊播放器開發包實現監控播放H5播放器
- 實現零程式碼開發還需要多長時間
- 無程式碼開發如何實現？什麼時候考慮使用無程式碼開發平臺
- 實時流（直播流）播放、上牆（大屏播放）解決方案
- Qt開發Gif播放器QT播放器
- Objective-C開發編碼規範Object
- 用哪種編碼開發系統？
- 關於PHP開發編碼規範PHP
- 編碼風格和開發習慣
- CSS模組化編碼讓開發事半功倍CSS
- 【Android開發VR實戰】二.播放360°全景視訊AndroidVR
- Android音訊實時傳輸與播放（一）：寫在開頭Android音訊
- webrtc-streamer實時播放監控Web
- 實時顯示iOS編寫UI程式碼效果iOSUI
- 檢視並ORACLE的編碼方式Oracle
- Android 基於ffmpeg開發簡易播放器 - NDK交叉編譯ffmpegAndroid播放器編譯