db/hash/hash.h

   1 /*-
   2  * Copyright (c) 1990, 1993, 1994
   3  *      The Regents of the University of California.  All rights reserved.
   4  *
   5  * This code is derived from software contributed to Berkeley by
   6  * Margo Seltzer.
   7  *
   8  * Redistribution and use in source and binary forms, with or without
   9  * modification, are permitted provided that the following conditions
  10  * are met:
  11  * 1. Redistributions of source code must retain the above copyright
  12  *    notice, this list of conditions and the following disclaimer.
  13  * 2. Redistributions in binary form must reproduce the above copyright
  14  *    notice, this list of conditions and the following disclaimer in the
  15  *    documentation and/or other materials provided with the distribution.
  16  * 3. All advertising materials mentioning features or use of this software
  17  *    must display the following acknowledgement:
  18  *      This product includes software developed by the University of
  19  *      California, Berkeley and its contributors.
  20  * 4. Neither the name of the University nor the names of its contributors
  21  *    may be used to endorse or promote products derived from this software
  22  *    without specific prior written permission.
  23  *
  24  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
  25  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
  26  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
  27  * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
  28  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
  29  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
  30  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
  31  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
  32  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
  33  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
  34  * SUCH DAMAGE.
  35  *
  36  *      @(#)hash.h      8.3 (Berkeley) 5/31/94
  37  * $FreeBSD: src/lib/libc/db/hash/hash.h,v 1.6 2002/03/21 22:46:26 obrien Exp $
  38  */
  39
  40 /* Operations */
  41 typedef enum {
  42         HASH_GET, HASH_PUT, HASH_PUTNEW, HASH_DELETE, HASH_FIRST, HASH_NEXT
  43 } ACTION;
  44
  45 /* Buffer Management structures */
  46 typedef struct _bufhead BUFHEAD;
  47
  48 struct _bufhead {
  49         BUFHEAD         *prev;          /* LRU links */
  50         BUFHEAD         *next;          /* LRU links */
  51         BUFHEAD         *ovfl;          /* Overflow page buffer header */
  52         u_int32_t        addr;          /* Address of this page */
  53         char            *page;          /* Actual page data */
  54         char            flags;
  55 #define BUF_MOD         0x0001
  56 #define BUF_DISK        0x0002
  57 #define BUF_BUCKET      0x0004
  58 #define BUF_PIN         0x0008
  59 };
  60
  61 #define IS_BUCKET(X)    ((X) & BUF_BUCKET)
  62
  63 typedef BUFHEAD **SEGMENT;
  64
  65 /* Hash Table Information */
  66 typedef struct hashhdr {                /* Disk resident portion */
  67         int             magic;          /* Magic NO for hash tables */
  68         int             version;        /* Version ID */
  69         u_int32_t       lorder;         /* Byte Order */
  70         int             bsize;          /* Bucket/Page Size */
  71         int             bshift;         /* Bucket shift */
  72         int             dsize;          /* Directory Size */
  73         int             ssize;          /* Segment Size */
  74         int             sshift;         /* Segment shift */
  75         int             ovfl_point;     /* Where overflow pages are being
  76                                          * allocated */
  77         int             last_freed;     /* Last overflow page freed */
  78         int             max_bucket;     /* ID of Maximum bucket in use */
  79         int             high_mask;      /* Mask to modulo into entire table */
  80         int             low_mask;       /* Mask to modulo into lower half of
  81                                          * table */
  82         int             ffactor;        /* Fill factor */
  83         int             nkeys;          /* Number of keys in hash table */
  84         int             hdrpages;       /* Size of table header */
  85         int             h_charkey;      /* value of hash(CHARKEY) */
  86 #define NCACHED 32                      /* number of bit maps and spare
  87                                          * points */
  88         int             spares[NCACHED];/* spare pages for overflow */
  89         u_int16_t       bitmaps[NCACHED];       /* address of overflow page
  90                                                  * bitmaps */
  91 } HASHHDR;
  92
  93 typedef struct htab      {              /* Memory resident data structure */
  94         HASHHDR         hdr;            /* Header */
  95         int             nsegs;          /* Number of allocated segments */
  96         int             exsegs;         /* Number of extra allocated
  97                                          * segments */
  98         u_int32_t                       /* Hash function */
  99             (*hash)(const void *, size_t);
 100         int             flags;          /* Flag values */
 101         int             fp;             /* File pointer */
 102         char            *tmp_buf;       /* Temporary Buffer for BIG data */
 103         char            *tmp_key;       /* Temporary Buffer for BIG keys */
 104         BUFHEAD         *cpage;         /* Current page */
 105         int             cbucket;        /* Current bucket */
 106         int             cndx;           /* Index of next item on cpage */
 107         int             error;          /* Error Number -- for DBM
 108                                          * compatibility */
 109         int             new_file;       /* Indicates if fd is backing store
 110                                          * or no */
 111         int             save_file;      /* Indicates whether we need to flush
 112                                          * file at
 113                                          * exit */
 114         u_int32_t       *mapp[NCACHED]; /* Pointers to page maps */
 115         int             nmaps;          /* Initial number of bitmaps */
 116         int             nbufs;          /* Number of buffers left to
 117                                          * allocate */
 118         BUFHEAD         bufhead;        /* Header of buffer lru list */
 119         SEGMENT         *dir;           /* Hash Bucket directory */
 120                                         /* other flags */
 121         int             nextkey_eof :1; /* dbm_nextkey() reached EOF */
 122 } HTAB;
 123
 124 /*
 125  * Constants
 126  */
 127 #define MAX_BSIZE               65536           /* 2^16 */
 128 #define MIN_BUFFERS             6
 129 #define MINHDRSIZE              512
 130 #define DEF_BUFSIZE             65536           /* 64 K */
 131 #define DEF_BUCKET_SIZE         4096
 132 #define DEF_BUCKET_SHIFT        12              /* log2(BUCKET) */
 133 #define DEF_SEGSIZE             256
 134 #define DEF_SEGSIZE_SHIFT       8               /* log2(SEGSIZE)         */
 135 #define DEF_DIRSIZE             256
 136 #define DEF_FFACTOR             65536
 137 #define MIN_FFACTOR             4
 138 #define SPLTMAX                 8
 139 #define CHARKEY                 "%$sniglet^&"
 140 #define NUMKEY                  1038583
 141 #define BYTE_SHIFT              3
 142 #define INT_TO_BYTE             2
 143 #define INT_BYTE_SHIFT          5
 144 #define ALL_SET                 ((u_int32_t)0xFFFFFFFF)
 145 #define ALL_CLEAR               0
 146
 147 #define PTROF(X)        ((BUFHEAD *)((ptrdiff_t)(X)&~0x3))
 148 #define ISMOD(X)        ((u_int32_t)(ptrdiff_t)(X)&0x1)
 149 #define DOMOD(X)        ((X) = (char *)((ptrdiff_t)(X)|0x1))
 150 #define ISDISK(X)       ((u_int32_t)(ptrdiff_t)(X)&0x2)
 151 #define DODISK(X)       ((X) = (char *)((ptrdiff_t)(X)|0x2))
 152
 153 #define BITS_PER_MAP    32
 154
 155 /* Given the address of the beginning of a big map, clear/set the nth bit */
 156 #define CLRBIT(A, N)    ((A)[(N)/BITS_PER_MAP] &= ~(1<<((N)%BITS_PER_MAP)))
 157 #define SETBIT(A, N)    ((A)[(N)/BITS_PER_MAP] |= (1<<((N)%BITS_PER_MAP)))
 158 #define ISSET(A, N)     ((A)[(N)/BITS_PER_MAP] & (1<<((N)%BITS_PER_MAP)))
 159
 160 /* Overflow management */
 161 /*
 162  * Overflow page numbers are allocated per split point.  At each doubling of
 163  * the table, we can allocate extra pages.  So, an overflow page number has
 164  * the top 5 bits indicate which split point and the lower 11 bits indicate
 165  * which page at that split point is indicated (pages within split points are
 166  * numberered starting with 1).
 167  */
 168
 169 #define SPLITSHIFT      11
 170 #define SPLITMASK       0x7FF
 171 #define SPLITNUM(N)     (((u_int32_t)(N)) >> SPLITSHIFT)
 172 #define OPAGENUM(N)     ((N) & SPLITMASK)
 173 #define OADDR_OF(S,O)   ((u_int32_t)((u_int32_t)(S) << SPLITSHIFT) + (O))
 174
 175 #define BUCKET_TO_PAGE(B) \
 176         (B) + hashp->HDRPAGES + ((B) ? hashp->SPARES[__log2((B)+1)-1] : 0)
 177 #define OADDR_TO_PAGE(B)        \
 178         BUCKET_TO_PAGE ( (1 << SPLITNUM((B))) -1 ) + OPAGENUM((B));
 179
 180 /*
 181  * page.h contains a detailed description of the page format.
 182  *
 183  * Normally, keys and data are accessed from offset tables in the top of
 184  * each page which point to the beginning of the key and data.  There are
 185  * four flag values which may be stored in these offset tables which indicate
 186  * the following:
 187  *
 188  *
 189  * OVFLPAGE     Rather than a key data pair, this pair contains
 190  *              the address of an overflow page.  The format of
 191  *              the pair is:
 192  *                  OVERFLOW_PAGE_NUMBER OVFLPAGE
 193  *
 194  * PARTIAL_KEY  This must be the first key/data pair on a page
 195  *              and implies that page contains only a partial key.
 196  *              That is, the key is too big to fit on a single page
 197  *              so it starts on this page and continues on the next.
 198  *              The format of the page is:
 199  *                  KEY_OFF PARTIAL_KEY OVFL_PAGENO OVFLPAGE
 200  *
 201  *                  KEY_OFF -- offset of the beginning of the key
 202  *                  PARTIAL_KEY -- 1
 203  *                  OVFL_PAGENO - page number of the next overflow page
 204  *                  OVFLPAGE -- 0
 205  *
 206  * FULL_KEY     This must be the first key/data pair on the page.  It
 207  *              is used in two cases.
 208  *
 209  *              Case 1:
 210  *                  There is a complete key on the page but no data
 211  *                  (because it wouldn't fit).  The next page contains
 212  *                  the data.
 213  *
 214  *                  Page format it:
 215  *                  KEY_OFF FULL_KEY OVFL_PAGENO OVFL_PAGE
 216  *
 217  *                  KEY_OFF -- offset of the beginning of the key
 218  *                  FULL_KEY -- 2
 219  *                  OVFL_PAGENO - page number of the next overflow page
 220  *                  OVFLPAGE -- 0
 221  *
 222  *              Case 2:
 223  *                  This page contains no key, but part of a large
 224  *                  data field, which is continued on the next page.
 225  *
 226  *                  Page format it:
 227  *                  DATA_OFF FULL_KEY OVFL_PAGENO OVFL_PAGE
 228  *
 229  *                  KEY_OFF -- offset of the beginning of the data on
 230  *                              this page
 231  *                  FULL_KEY -- 2
 232  *                  OVFL_PAGENO - page number of the next overflow page
 233  *                  OVFLPAGE -- 0
 234  *
 235  * FULL_KEY_DATA
 236  *              This must be the first key/data pair on the page.
 237  *              There are two cases:
 238  *
 239  *              Case 1:
 240  *                  This page contains a key and the beginning of the
 241  *                  data field, but the data field is continued on the
 242  *                  next page.
 243  *
 244  *                  Page format is:
 245  *                  KEY_OFF FULL_KEY_DATA OVFL_PAGENO DATA_OFF
 246  *
 247  *                  KEY_OFF -- offset of the beginning of the key
 248  *                  FULL_KEY_DATA -- 3
 249  *                  OVFL_PAGENO - page number of the next overflow page
 250  *                  DATA_OFF -- offset of the beginning of the data
 251  *
 252  *              Case 2:
 253  *                  This page contains the last page of a big data pair.
 254  *                  There is no key, only the  tail end of the data
 255  *                  on this page.
 256  *
 257  *                  Page format is:
 258  *                  DATA_OFF FULL_KEY_DATA <OVFL_PAGENO> <OVFLPAGE>
 259  *
 260  *                  DATA_OFF -- offset of the beginning of the data on
 261  *                              this page
 262  *                  FULL_KEY_DATA -- 3
 263  *                  OVFL_PAGENO - page number of the next overflow page
 264  *                  OVFLPAGE -- 0
 265  *
 266  *                  OVFL_PAGENO and OVFLPAGE are optional (they are
 267  *                  not present if there is no next page).
 268  */
 269
 270 #define OVFLPAGE        0
 271 #define PARTIAL_KEY     1
 272 #define FULL_KEY        2
 273 #define FULL_KEY_DATA   3
 274 #define REAL_KEY        4
 275
 276 /* Short hands for accessing structure */
 277 #define BSIZE           hdr.bsize
 278 #define BSHIFT          hdr.bshift
 279 #define DSIZE           hdr.dsize
 280 #define SGSIZE          hdr.ssize
 281 #define SSHIFT          hdr.sshift
 282 #define LORDER          hdr.lorder
 283 #define OVFL_POINT      hdr.ovfl_point
 284 #define LAST_FREED      hdr.last_freed
 285 #define MAX_BUCKET      hdr.max_bucket
 286 #define FFACTOR         hdr.ffactor
 287 #define HIGH_MASK       hdr.high_mask
 288 #define LOW_MASK        hdr.low_mask
 289 #define NKEYS           hdr.nkeys
 290 #define HDRPAGES        hdr.hdrpages
 291 #define SPARES          hdr.spares
 292 #define BITMAPS         hdr.bitmaps
 293 #define VERSION         hdr.version
 294 #define MAGIC           hdr.magic
 295 #define NEXT_FREE       hdr.next_free
 296 #define H_CHARKEY       hdr.h_charkey