postgres 源码解析33 进程间通信--1

简介

  Postgres是基于进程实现的开源数据库，其进程间通信IPC主要通过共享内存和信号量实现，接下来会从源码角度学习postgres数据库中的IPC原理，借鉴其优秀的设计思路。内容很多，会分成多个小节进行讲解，同时若有理解不正确的地方，欢迎指正与探讨。

共享内存结构布局图

共享内存关键数据结构

1 全局变量

/* shared memory global variables */

static PGShmemHeader *ShmemSegHdr;	/* shared mem segment header */

static void *ShmemBase;			/* start address of shared memory */

static void *ShmemEnd;			/* end+1 address of shared memory */

slock_t    *ShmemLock;			/* spinlock for shared memory and LWLock
								 * allocation */

static HTAB *ShmemIndex = NULL; /* primary index hashtable for shmem */

2 PGShmemHeader
该结构体为共享内存的标准头信息，如校验魔数，总内存段大小、空闲空间的偏移量、动态共享控制段ID号以及ShememIndex表指针等

typedef struct PGShmemHeader	/* standard header for all Postgres shmem */
{
	int32		magic;			/* magic # to identify Postgres segments */
#define PGShmemMagic  679834894
	pid_t		creatorPID;		/* PID of creating process (set but unread) */
	Size		totalsize;		/* total size of segment */
	Size		freeoffset;		/* offset to first free space */
	dsm_handle	dsm_control;	/* ID of dynamic shared memory control seg */
	void	   *index;			/* pointer to ShmemIndex table */
#ifndef WIN32					/* Windows doesn't have useful inode#s */
	dev_t		device;			/* device data directory is on */
	ino_t		inode;			/* inode number of data directory */
#endif
} PGShmemHeader;

3 ShmemIndexEnt
哈希桶，该结构体包含索引建、位置、分配字节数等信息

/* size constants for the shmem index table */
 /* max size of data structure string name */
#define SHMEM_INDEX_KEYSIZE		 (48)
 /* estimated size of the shmem index table (not a hard limit) */
#define SHMEM_INDEX_SIZE		 (64)
// 队列
/* shmqueue.c */
typedef struct SHM_QUEUE
{
	struct SHM_QUEUE *prev;
	struct SHM_QUEUE *next;
} SHM_QUEUE;

/* this is a hash bucket in the shmem index table */
typedef struct
{
	char		key[SHMEM_INDEX_KEYSIZE];	/* string name */
	void	   *location;		/* location in shared mem */ // 此结构体在内存中的位置
	Size		size;			/* # bytes requested for the structure */
	Size		allocated_size; /* # bytes actually allocated */	// 分配字节数
} ShmemIndexEnt;

4 HTAB
该结构体为哈希索引表，加速各个模块共享内存位置的查找，包含控制信息、hash函数、内存上下文以及哈希键大小等信息

/*
 * Top control structure for a hashtable --- in a shared table, each backend
 * has its own copy (OK since no fields change at runtime)
 */
struct HTAB
{
	HASHHDR    *hctl;			/* => shared control information */	// 控制信息
	HASHSEGMENT *dir;			/* directory of segment starts */	
	HashValueFunc hash;			/* hash function */					
	HashCompareFunc match;		/* key comparison function */	
	HashCopyFunc keycopy;		/* key copying function */
	HashAllocFunc alloc;		/* memory allocator */
	MemoryContext hcxt;			/* memory context if default allocator used */
	char	   *tabname;		/* table name (for error messages) */
	bool		isshared;		/* true if table is in shared memory */
	bool		isfixed;		/* if true, don't enlarge */

	/* freezing a shared table isn't allowed, so we can keep state here */
	bool		frozen;			/* true = no more inserts allowed */

	/* We keep local copies of these fixed values to reduce contention */
	Size		keysize;		/* hash key length in bytes */
	long		ssize;			/* segment size --- must be power of 2 */
	int			sshift;			/* segment shift = log2(ssize) */
};

5 不同的操作系统其共享内存的分配方式也有所不同

/* Possible values for huge_pages */
typedef enum
{
	HUGE_PAGES_OFF,
	HUGE_PAGES_ON,
	HUGE_PAGES_TRY
}HugePagesType;

/* Possible values for shared_memory_type */
typedef enum
{
	SHMEM_TYPE_WINDOWS,
	SHMEM_TYPE_SYSV,
	SHMEM_TYPE_MMAP
}PGShmemType;

6 标识共享内存区块的相关字段

typedef key_t IpcMemoryKey;		/* shared memory key passed to shmget(2) */
typedef int IpcMemoryId;		/* shared memory ID returned by shmget(2) */

/*
 * How does a given IpcMemoryId relate to this PostgreSQL process?
 *
 * One could recycle unattached segments of different data directories if we
 * distinguished that case from other SHMSTATE_FOREIGN cases.  Doing so would
 * cause us to visit less of the key space, making us less likely to detect a
 * SHMSTATE_ATTACHED key.  It would also complicate the concurrency analysis,
 * in that postmasters of different data directories could simultaneously
 * attempt to recycle a given key.  We'll waste keys longer in some cases, but
 * avoiding the problems of the alternative justifies that loss.
 */
typedef enum
{
	SHMSTATE_ANALYSIS_FAILURE,	/* unexpected failure to analyze the ID */
	SHMSTATE_ATTACHED,			/* pertinent to DataDir, has attached PIDs */
	SHMSTATE_ENOENT,			/* no segment of that ID */
	SHMSTATE_FOREIGN,			/* exists, but not pertinent to DataDir */
	SHMSTATE_UNATTACHED			/* pertinent to DataDir, no attached PIDs */
} IpcMemoryState;

主流程

PostmasterMain
  |- - reset_shared    创建共享内存和信号量
    |- - CreateSharedMemoryAndSemaphores
        |- - PGSharedMemoryCreate   
        |- - PGReserveSemaphores   
        |- - SpinlockSemaInit      
        |- - CreateLWLocks       
        |- - InitShmemIndex       
        |- - dsm_shmem_init
        |- - XLOGShmemInit
        |- - CLOGShmemInit
        |- - CommitTsShmemInit
        |- - SUBTRANSShmemInit
        |- - MultiXactShmemInit
        |- - InitBufferPool
        |- - InitLocks
        |- - InitPredicateLocks
        |- - InitProcGlobal
        |- - CreateSharedProcArray
        |- - CreateSharedBackendStatus
        |- - TwoPhaseShmemInit
        |- - BackgroundWorkerShmemInit
        |- - CreateSharedInvalidationState
        |- - PMSignalShmemInit
        |- - ProcSignalShmemInit
        |- - CheckpointerShmemInit
        |- - AutoVacuumShmemInit
        |- - ReplicationSlotsShmemInit
        |- - ReplicationOriginShmemInit
        |- - WalSndShmemInit
        |- - WalRcvShmemInit
        |- - PgArchShmemInit
        |- - ApplyLauncherShmemInit
        |- - SnapMgrInit
        |- - BTreeShmemInit
        |- - SyncScanShmemInit
        |- - AsyncShmemInit