Postgres 源码学习 3—Postgres 存储管理器

/*
 * Stuff for fork names.
 *
 * The physical storage of a relation consists of one or more forks.
 * The main fork is always created, but in addition to that there can be
 * additional forks for storing various metadata. ForkNumber is used when
 * we need to refer to a specific fork in a relation.
 */
typedef enum ForkNumber
{
    InvalidForkNumber = -1,
    MAIN_FORKNUM = 0,
    FSM_FORKNUM,
    VISIBILITYMAP_FORKNUM,
    INIT_FORKNUM,

    /*
     * NOTE: if you add a new fork, change MAX_FORKNUM and possibly
     * FORKNAMECHARS below, and update the forkNames array in
     * src/common/relpath.c
     */
} ForkNumber;

/* Size of a disk block --- this also limits the size of a tuple. You can set
   it bigger if you need bigger tuples (although TOAST should reduce the need
   to have large tuples, since fields can be spread across multiple tuples).
   BLCKSZ must be a power of 2. The maximum possible value of BLCKSZ is
   currently 2^15 (32768). This is determined by the 15-bit widths of the
   lp_off and lp_len fields in ItemIdData (see include/storage/itemid.h).
   Changing BLCKSZ requires an initdb. */
#define BLCKSZ 8192

/* RELSEG_SIZE is the maximum number of blocks allowed in one disk file. Thus,
   the maximum size of a single file is RELSEG_SIZE * BLCKSZ; relations bigger
   than that are divided into multiple files. RELSEG_SIZE * BLCKSZ must be
   less than your OS' limit on file size. This is often 2 GB or 4GB in a
   32-bit operating system, unless you have large file support enabled. By
   default, we make the limit 1 GB to avoid any possible integer-overflow
   problems within the OS. A limit smaller than necessary only means we divide
   a large relation into more chunks than necessary, so it seems best to err
   in the direction of a small limit. A power-of-2 value is recommended to
   save a few cycles in md.c, but is not absolutely required. Changing
   RELSEG_SIZE requires an initdb. */
#define RELSEG_SIZE 131072

/*
 * smgr.c maintains a table of SMgrRelation objects, which are essentially
 * cached file handles.  An SMgrRelation is created (if not already present)
 * by smgropen(), and destroyed by smgrclose().  Note that neither of these
 * operations imply I/O, they just create or destroy a hashtable entry.
 * (But smgrclose() may release associated resources, such as OS-level file
 * descriptors.)
 *
 * An SMgrRelation may have an "owner", which is just a pointer to it from
 * somewhere else; smgr.c will clear this pointer if the SMgrRelation is
 * closed.  We use this to avoid dangling pointers from relcache to smgr
 * without having to make the smgr explicitly aware of relcache.  There
 * can't be more than one "owner" pointer per SMgrRelation, but that's
 * all we need.
 *
 * SMgrRelations that do not have an "owner" are considered to be transient,
 * and are deleted at end of transaction.
 */
typedef struct SMgrRelationData
{
    /* rlocator is the hashtable lookup key, so it must be first! */
    RelFileLocatorBackend smgr_rlocator;    /* relation physical identifier */

    /* pointer to owning pointer, or NULL if none */
    struct SMgrRelationData **smgr_owner;

    /*
     * The following fields are reset to InvalidBlockNumber upon a cache flush
     * event, and hold the last known size for each fork.  This information is
     * currently only reliable during recovery, since there is no cache
     * invalidation for fork extension.
     */
    BlockNumber smgr_targblock; /* current insertion target block */
    BlockNumber smgr_cached_nblocks[MAX_FORKNUM + 1];   /* last known size */

    /* additional public fields may someday exist here */

    /*
     * Fields below here are intended to be private to smgr.c and its
     * submodules.  Do not touch them from elsewhere.
     */
    int         smgr_which;     /* storage manager selector */

    /*
     * for md.c; per-fork arrays of the number of open segments
     * (md_num_open_segs) and the segments themselves (md_seg_fds).
     */
    int         md_num_open_segs[MAX_FORKNUM + 1];
    struct _MdfdVec *md_seg_fds[MAX_FORKNUM + 1];

    /* if unowned, list link in list of all unowned SMgrRelations */
    dlist_node  node;
} SMgrRelationData;

typedef struct RelFileLocator
{
    Oid         spcOid;         /* tablespace */
    Oid         dbOid;          /* database */
    RelFileNumber relNumber;    /* relation */
} RelFileLocator;

/*
 * Augmenting a relfilelocator with the backend ID provides all the information
 * we need to locate the physical storage.  The backend ID is InvalidBackendId
 * for regular relations (those accessible to more than one backend), or the
 * owning backend's ID for backend-local relations.  Backend-local relations
 * are always transient and removed in case of a database crash; they are
 * never WAL-logged or fsync'd.
 */
typedef struct RelFileLocatorBackend
{
    RelFileLocator locator;
    BackendId   backend;
} RelFileLocatorBackend;

    TablespaceCreateDbspace(reln->smgr_rlocator.locator.spcOid,
                            reln->smgr_rlocator.locator.dbOid,
                            isRedo);

    path = relpath(reln->smgr_rlocator, forknum);

    fd = PathNameOpenFile(path, _mdfd_open_flags() | O_CREAT | O_EXCL);

    if (fd < 0)
    {
        int         save_errno = errno;

        if (isRedo)
            fd = PathNameOpenFile(path, _mdfd_open_flags());
        if (fd < 0)
        {
            /* be sure to report the error reported by create, not open */
            errno = save_errno;
            ereport(ERROR,
                    (errcode_for_file_access(),
                     errmsg("could not create file \"%s\": %m", path)));
        }
    }

    pfree(path);

    _fdvec_resize(reln, forknum, 1);
    mdfd = &reln->md_seg_fds[forknum][0];
    mdfd->mdfd_vfd = fd;
    mdfd->mdfd_segno = 0;

    if (!SmgrIsTemp(reln))
        register_dirty_segment(reln, forknum, mdfd);

/*
 * smgropen() -- Return an SMgrRelation object, creating it if need be.
 *
 * This does not attempt to actually open the underlying file.
 */
SMgrRelation
smgropen(RelFileLocator rlocator, BackendId backend)
{
    RelFileLocatorBackend brlocator;
    SMgrRelation reln;
    bool        found;

    if (SMgrRelationHash == NULL)
    {
        /* First time through: initialize the hash table */
        HASHCTL     ctl;

        ctl.keysize = sizeof(RelFileLocatorBackend);
        ctl.entrysize = sizeof(SMgrRelationData);
        SMgrRelationHash = hash_create("smgr relation table", 400,
                                       &ctl, HASH_ELEM | HASH_BLOBS);
        dlist_init(&unowned_relns);
    }

    /* Look up or create an entry */
    brlocator.locator = rlocator;
    brlocator.backend = backend;
    reln = (SMgrRelation) hash_search(SMgrRelationHash,
                                      &brlocator,
                                      HASH_ENTER, &found);

    /* Initialize it if not present before */
    if (!found)
    {
        /* hash_search already filled in the lookup key */
        reln->smgr_owner = NULL;
        reln->smgr_targblock = InvalidBlockNumber;
        for (int i = 0; i <= MAX_FORKNUM; ++i)
            reln->smgr_cached_nblocks[i] = InvalidBlockNumber;
        reln->smgr_which = 0;   /* we only have md.c at present */

        /* implementation-specific initialization */
        smgrsw[reln->smgr_which].smgr_open(reln);

        /* it has no owner yet */
        dlist_push_tail(&unowned_relns, &reln->node);
    }

    return reln;
}

/*
 * mdopen() -- Initialize newly-opened relation.
 */
void
mdopen(SMgrRelation reln)
{
    /* mark it not open */
    for (int forknum = 0; forknum <= MAX_FORKNUM; forknum++)
        reln->md_num_open_segs[forknum] = 0;
}

/*
 * mdclose() -- Close the specified relation, if it isn't closed already.
 */
void
mdclose(SMgrRelation reln, ForkNumber forknum)
{
    int         nopensegs = reln->md_num_open_segs[forknum];

    /* No work if already closed */
    if (nopensegs == 0)
        return;

    /* close segments starting from the end */
    while (nopensegs > 0)
    {
        MdfdVec    *v = &reln->md_seg_fds[forknum][nopensegs - 1];

        FileClose(v->mdfd_vfd);
        _fdvec_resize(reln, forknum, nopensegs - 1);
        nopensegs--;
    }
}

    /*
     * create an array which contains all relations to be dropped, and close
     * each relation's forks at the smgr level while at it
     */
    rlocators = palloc(sizeof(RelFileLocatorBackend) * nrels);
    for (i = 0; i < nrels; i++)
    {
        RelFileLocatorBackend rlocator = rels[i]->smgr_rlocator;
        int         which = rels[i]->smgr_which;

        rlocators[i] = rlocator;

        /* Close the forks at smgr level */
        for (forknum = 0; forknum <= MAX_FORKNUM; forknum++)
            smgrsw[which].smgr_close(rels[i], forknum);
    }

    for (i = 0; i < nrels; i++)
    {
        int         which = rels[i]->smgr_which;

        for (forknum = 0; forknum <= MAX_FORKNUM; forknum++)
            smgrsw[which].smgr_unlink(rlocators[i], forknum, isRedo);
    }

/* Prevent other backends' fds from holding on to the disk space */
ret = do_truncate(path);

/* Register request to unlink first segment later */
save_errno = errno;
register_unlink_segment(rlocator, forknum, 0 /* first seg */ );
errno = save_errno;

    v = _mdfd_getseg(reln, forknum, blocknum, skipFsync, EXTENSION_CREATE);

    seekpos = (off_t) BLCKSZ * (blocknum % ((BlockNumber) RELSEG_SIZE));

    Assert(seekpos < (off_t) BLCKSZ * RELSEG_SIZE);

    if ((nbytes = FileWrite(v->mdfd_vfd, buffer, BLCKSZ, seekpos, WAIT_EVENT_DATA_FILE_EXTEND)) != BLCKSZ)
    {
        if (nbytes < 0)
            ereport(ERROR,
                    (errcode_for_file_access(),
                     errmsg("could not extend file \"%s\": %m",
                            FilePathName(v->mdfd_vfd)),
                     errhint("Check free disk space.")));
        /* short write: complain appropriately */
        ereport(ERROR,
                (errcode(ERRCODE_DISK_FULL),
                 errmsg("could not extend file \"%s\": wrote only %d of %d bytes at block %u",
                        FilePathName(v->mdfd_vfd),
                        nbytes, BLCKSZ, blocknum),
                 errhint("Check free disk space.")));
    }

    if (!skipFsync && !SmgrIsTemp(reln))
        register_dirty_segment(reln, forknum, v);

/*
 * mdread() -- Read the specified block from a relation.
 */
void
mdread(SMgrRelation reln, ForkNumber forknum, BlockNumber blocknum,
       void *buffer)
{
    off_t       seekpos;
    int         nbytes;
    MdfdVec    *v;

    /* If this build supports direct I/O, the buffer must be I/O aligned. */
    if (PG_O_DIRECT != 0 && PG_IO_ALIGN_SIZE <= BLCKSZ)
        Assert((uintptr_t) buffer == TYPEALIGN(PG_IO_ALIGN_SIZE, buffer));

    TRACE_POSTGRESQL_SMGR_MD_READ_START(forknum, blocknum,
                                        reln->smgr_rlocator.locator.spcOid,
                                        reln->smgr_rlocator.locator.dbOid,
                                        reln->smgr_rlocator.locator.relNumber,
                                        reln->smgr_rlocator.backend);

    v = _mdfd_getseg(reln, forknum, blocknum, false,
                     EXTENSION_FAIL | EXTENSION_CREATE_RECOVERY);

    seekpos = (off_t) BLCKSZ * (blocknum % ((BlockNumber) RELSEG_SIZE));

    Assert(seekpos < (off_t) BLCKSZ * RELSEG_SIZE);

    nbytes = FileRead(v->mdfd_vfd, buffer, BLCKSZ, seekpos, WAIT_EVENT_DATA_FILE_READ);
}

    v = _mdfd_getseg(reln, forknum, blocknum, skipFsync,
                     EXTENSION_FAIL | EXTENSION_CREATE_RECOVERY);

    seekpos = (off_t) BLCKSZ * (blocknum % ((BlockNumber) RELSEG_SIZE));

    Assert(seekpos < (off_t) BLCKSZ * RELSEG_SIZE);

    nbytes = FileWrite(v->mdfd_vfd, buffer, BLCKSZ, seekpos, WAIT_EVENT_DATA_FILE_WRITE);