client的TCP协议栈发送 ZeroWindow 标志给服务端。也就是接收端的 buffer 堆满了（但是服务端这个时候看到的bytes in fly是0，因为都ack了）

2. The Framing Method

一些显而易见的访问无效内存操作可能会被编译器优化而会漏报。

然后使用我发布的 public key 对 (1) 拿到的秘文进行解密，对比 hash 值，如果一样，说明确实 private key 的持有者加密的；

使用 private key 加密

an integral (积分) hash is computed for each item, then this hash is used as an index in a table where the items are stored. If many items end up in the same slot (槽) , they are chained, as in a linked list. Conceptually, it looks like this:

Just like in the initial code, we search the map using std::map::find, but now we store the result in a local variable. The result is an iterator that we can compare with entries.end(), as before. Now, if the iterator is not the end iterator, then we can get the value from the map by dereferencing the iterator and accessing its second member.

Sometimes, a very small Time-To-Live (TTL) can result in important data being removed from the CDN, resulting in longer round-trips back to your video origin server.

A good rule of thumb (拇指) is to start with the lowest bitrate rendition and slowly increase the quality based on the network condition, and then, you can expect low startup delays.

If you are using ABR video streaming, the video probably has four or five renditions and the player will pick one of the renditions after inspecting (检查) the network conditions. You can expect a long startup delay if this rendition is wrong (w.r.t the network condition).

如果你查看了我们之前输出的帧相关信息，你会发现图像类型信息包含在pict_type 中。

The new standard will be called MPEG-5 Essential (必不可少) Video Coding (EVC) and is expected to reach FDIS in January 2020

Thirteen is a large number for video coding standards but this number should be measured (措施) against the number of years covered – close to 40. In this long period of time we have gone from 3 initial (初始) standards that were mostly application/industry-specific (H.120, MPEG-1 and H.261) to a series of generic (i.e. industry-neutral) standards (MPEG-2, MPEG-4 Visual, MPEG-4 AVC and HEVC) and then to a group of standards that sought (寻求) to achieve Option 1 status (WebVC, IVC and VCB). Other proprietary video coding formats that have found significant (显著) use in the market point to the fact that MPEG cannot stay forever (永远) in its ivory tower of “best video coding standards no matter what”. MPEG has to face the reality of a market that becomes more and more diversified and where – unlike the golden age of a single coding standard – there is no longer one size that fits all

For the third time MPEG and ITU are collaborating in the development of a new video coding standard with the target of a 50% reduction (还原) of bitrate compared to HEVC. The development of Versatile (多才多艺) Video Coding (VVC),

After that HEVC has been enriched with a number of features that at the time of development were not supported by previous standards such as High Dynamic Range (HDR) and Wide Colour Gamut (WCG)

the HEVC Verification Tests showed that the standard had achieved 60% improvement on AVC, 10% more than originally planned

In 2010 MPEG and VCEG extended their collaboration (协作) to a new project: High Efficiency (效率) Video Coding (HEVC)

The standard is called Advanced Video Coding by both MPEG and ITU, but is labelled as AVC by MPEG and as H.264 by ITU-T.

MPEG-4 Visual suffered (受苦) a devastating (摧毁) blow (吹) by a patent pool that decided to impose (施加) “content fees (费) ” on their licensing term

H.263 focused on low bitrate video communication, while MPEG-4 Visual kept on making real the vision (视觉) of extending video coding to more industries

The conjunction (连词) of the two standards is a very tiny code that simply tells the decoder that a bitstream is H.263 or MPEG-4 Visual

The first international standard that used video coding techniques (技术) – ITU-T Recommendation (推荐) H.120

The memfd_create function returns a file descriptor which can be used to create memory mappings using the mmap function. It is similar to the shm_open function in the sense that these mappings are not backed by actual files. However, the descriptor returned by memfd_create does not correspond to a named object; the name argument is used for debugging purposes only (e.g., will appear in /proc), and separate invocations of memfd_create with the same name will not return descriptors for the same region of memory. The descriptor can also be used to create alias mappings within the same process

This function returns a file descriptor that can be used to allocate shared memory via mmap. Unrelated processes can use same name to create or open existing shared memory objects.

This function can be used to provide the system with advice about the intended usage patterns of the memory region starting at addr and extending length bytes.

Not all file descriptors may be mapped. Sockets, pipes, and most devices only allow sequential access and do not fit into the mapping abstraction. In addition, some regular files may not be mmapable, and older kernels may not support mapping at all. Thus, programs using mmap should have a fallback method to use should it fail.

This function is only available on a few systems. Except for performing optional optimizations one should not rely on this function.

This function can be used to change the size of an existing memory area. address and length must cover a region entirely mapped in the same mmap statement. A new mapping with the same characteristics will be returned with the length new_length.

only entire pages can be removed. If length is not an even number of pages, it will be rounded up

To determine the default size of a page the machine uses one should use

Processes can insert control over this issue by using the msync() function. This function takes a flags parameter that can initiate a synchronous, blocking write (MS_SYNC) or an asynchronous, non-blocking one (MS_ASYNC). In the case of the asynchronous write, the data will get copied to disk at a later point; however, the updated data would be immediately available to any process that reads from the file with read().

When a file is mapped into memory with mmap(), the kernel will occasionally (偶尔) trigger (触发) a write to copy updated portions (部分) back to disk

Note that setting up the regions in both processes is an expensive operation in terms of execution time; however, once the region is set up, data is exchanged immediately

That is, when one process writes to the region, that data is immediately accessible by the other process without having to invoke a system call.

Memory-mapped files allow for multiple processes to share read-only access to a common file

This is offset from where the file mapping started. In simple terms, the mapping connects to (offset) to (offset+length-1) bytes for the file open on filedes descriptor.

When this flag is used, the system has to be forced to use the exact (确切) mapping address specified in the address If this is not possible, then the mapping will be failed.

Anonymous mapping means the mapping is not connected to any files. This mapping is used as the basic primitive (原始) to extend the heap .

MAP_PRIVATE: When this flag is used, the mapping will not be seen by any other processes, and the changes made will not be written to the file.

The standard also describes the process of registering new payload types with IANA; additional payload formats and payload types are defined in the following specifications:

WebRTC, on the other hand, is becoming more and more popular. However, it still faces one major limitation : it is not widely supported by most encoders. The same goes for other streaming software, like production (生产) and mixing tools.

Different circumstances (情况) call for unique streaming setups and protocols. Currently, HLS delivery with RTMP ingest is the chosen combination (组合) in many streaming setups. That is because it ticks (蜱) the boxes of low latency, ultra-compatibility, and affordability.

Aside (一旁) from streaming with real-time latency, WebRTC is very secure . It is encrypted with SRTP and other gold-standard security measures (措施) . Like HLS, WebRTC is capable of adaptive bitrate streaming, so you can deliver multiple (多个) renditions of your stream with the optimal quality (质量) for each viewer .

Although WebRTC is technically a project, it is typically lumped (肿块) in with protocols since their functionality is very similar .

It is also important to point out one major downfall of HLS: when used alone, it causes a latency of 15-30 seconds, which means that the HLS delivery/HLS ingest combination (组合) is not capable of streaming with latency as low as some as other setups.

HLS can be used for both delivery and ingest, but it is currently more prominent (突出) for delivery since it is not widely compatible with most encoders.

This protocol also supports adaptive bitrate streaming, which is very important at the professional broadcasting level. In addition to being adaptive, HLS is also dynamic . That means that at any given moment, the bitrate of each individual viewer’s stream will adjust to best suit the connection conditions .

HLS is absolutely essential in most modern streaming setups since the HTML5 video player is the only type of universally compatible video player. Since HTML5 video players come with so many benefits , most broadcasters consider it the only viable (可行) option .

Unfortunately, RTMP is not compatible with more modern video players, including HTML5, so it is no longer used for delivery in any streaming setups. However, it is very effective in its ingestion role , so it is quite valuable in streaming. As with many types of technology, the role of RTMP continues to change.

There are several variations of RTMP that serve slightly (稍) different purposes. (change formatting) 

It is now used for ingesting media from the encoder or other source to a video streaming platform.

Real-Time Messaging Protocol (RTMP) is a streaming protocol that was originally used to deliver video content to Adobe’s Flash player. This protocol is known for its capability of low latency streaming.

Prior (事先) to making any SNMP request from an authoritative SNMP engine, you will need to send it a discovery packet, basically an empty SNMP version 3 packet and wait for the REPORT message agent will send you. REPORT that you will receive includes authoritative SNMP engine ID, SNMP engine boots and SNMP engine time values that you need to use in subsequent requests.

分块例子

The FFmpeg's "-re" flag means to "Read input at native frame rate.

PTHREAD_CANCEL_DEFERRED

PTHREAD_CANCEL_ASYNCHRONOUS

ARM、PowerPC (除PowerPC 970外)、DEC Alpha，SPARC V9，MIPS，PA-RISC and IA64等的字节序是可配置的

asctime()将参数timeptr所指的tm结构中的信息转换成真实世界所使用的时间日期表示方法，然后将结果以字符串形态返回

ctime()将参数timep所指的time_t结构中的信息转换成真实世界所使用的时间日期表示方法，然后将结果以字符串形态返回

gettimeofday（取得目前的时间）

因程序逻辑需要而产生的信号(如SIGUSR1/ SIGUSR2和实时信号)，被处理后程序继续正常运行，可考虑使用sigwait同步模型规避信号处理函数执行上下文不确定性带来的潜在风险。而对于硬件致命错误等导致程序运行终止的信号(如SIGSEGV)，必须按照传统的异步方式使用 signal()或sigaction()注册信号处理函数进行非阻塞处理，以提高响应的实时性。在应用程序中，可根据所处理信号的不同而同时使用这两种信号处理模型。

Signal 6 (SIGABRT) = SIGABRT is commonly used by libc and other libraries to abort the program in case of critical errors. For example, glibc sends an SIGABRT in case of a detected double-free or other heap corruptions

The LD_LIBRARY_PATH environment variable defines the path to libraries

The ldd utility prints shared object dependencies

-l (lowercase L) sets the name of the library you want to link against

-L (capital L) adds an additional library path to GCC's search locations

The -c option prevents GCC from running a linker

By default, GCC assumes that libraries are in a system library path, such as /lib64 and /usr/lib64

GB28181中对PT的定义

This model describes a mechanism by which two RTP participants agree to a common subset of SDP codecs and settings

Multiple media & Payload Types

It describes things such as the peer's host IP address, time bases, and summary description. Most of these values are optional, so they can be set to zero (0) or empty strings with a dash (-)

These features might or might not be supported by both peers in an RTP session, and must be explicitly negotiated and enabled. This is typically done with the SDP negotiation, that we'll cover next.

Again, this starts being an arbitrary random number, and then grows monotonically at the speed given by the media clock rate (defined by the Payload Type). Represents the instant of time when the media source was packetized into the RTP packet; the protocol doesn't use absolute timestamp values, but it uses differences between timestamps to calculate elapsed time between packets, which allows synchronization of multiple media streams (think lip sync between video and audio tracks), and also to calculate network latency and jitter

This starts as an arbitrary random number, which then increments by one for each RTP data packet sent. Receivers can use these numbers to detect packet loss and to sort packets in case they are received out of order.

Payload Type Definitions

Bash Conditional Expressions

AES uses a fixed block size of 16-bytes. If a file is not a multiple of a block size, then AES uses padding to complete the block

Anecdotal evidence suggests that files larger than 1 MB encrypted with AES tend to be around 35% larger than before encryption