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  * 4. Neither the name of the University nor the names of its contributors
  17  *    may be used to endorse or promote products derived from this software
  18  *    without specific prior written permission.
  19  *
  20  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
  21  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
  22  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
  23  * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
  24  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
  25  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
  26  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
  27  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
  28  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
  29  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
  30  * SUCH DAMAGE.
  31  *
  32  *      @(#)hash.h      8.3 (Berkeley) 5/31/94
  33  * $FreeBSD: src/lib/libc/db/hash/hash.h,v 1.9 2009/03/28 05:45:29 delphij Exp $
  34  */
  35
  36 /* Operations */
  37 typedef enum {
  38         HASH_GET, HASH_PUT, HASH_PUTNEW, HASH_DELETE, HASH_FIRST, HASH_NEXT
  39 } ACTION;
  40
  41 /* Buffer Management structures */
  42 typedef struct _bufhead BUFHEAD;
  43
  44 struct _bufhead {
  45         BUFHEAD         *prev;          /* LRU links */
  46         BUFHEAD         *next;          /* LRU links */
  47         BUFHEAD         *ovfl;          /* Overflow page buffer header */
  48         u_int32_t        addr;          /* Address of this page */
  49         char            *page;          /* Actual page data */
  50         char            flags;
  51 #define BUF_MOD         0x0001
  52 #define BUF_DISK        0x0002
  53 #define BUF_BUCKET      0x0004
  54 #define BUF_PIN         0x0008
  55 };
  56
  57 #define IS_BUCKET(X)    ((X) & BUF_BUCKET)
  58
  59 typedef BUFHEAD **SEGMENT;
  60
  61 /* Hash Table Information */
  62 typedef struct hashhdr {                /* Disk resident portion */
  63         int32_t         magic;          /* Magic NO for hash tables */
  64         int32_t         version;        /* Version ID */
  65         u_int32_t       lorder;         /* Byte Order */
  66         int32_t         bsize;          /* Bucket/Page Size */
  67         int32_t         bshift;         /* Bucket shift */
  68         int32_t         dsize;          /* Directory Size */
  69         int32_t         ssize;          /* Segment Size */
  70         int32_t         sshift;         /* Segment shift */
  71         int32_t         ovfl_point;     /* Where overflow pages are being
  72                                          * allocated */
  73         int32_t         last_freed;     /* Last overflow page freed */
  74         u_int32_t       max_bucket;     /* ID of Maximum bucket in use */
  75         u_int32_t       high_mask;      /* Mask to modulo into entire table */
  76         u_int32_t       low_mask;       /* Mask to modulo into lower half of
  77                                          * table */
  78         u_int32_t       ffactor;        /* Fill factor */
  79         int32_t         nkeys;          /* Number of keys in hash table */
  80         int32_t         hdrpages;       /* Size of table header */
  81         int32_t         h_charkey;      /* value of hash(CHARKEY) */
  82 #define NCACHED 32                      /* number of bit maps and spare
  83                                          * points */
  84         int32_t         spares[NCACHED];/* spare pages for overflow */
  85         u_int16_t       bitmaps[NCACHED];       /* address of overflow page
  86                                                  * bitmaps */
  87 } HASHHDR;
  88
  89 typedef struct htab      {              /* Memory resident data structure */
  90         HASHHDR         hdr;            /* Header */
  91         int             nsegs;          /* Number of allocated segments */
  92         int             exsegs;         /* Number of extra allocated
  93                                          * segments */
  94         u_int32_t                       /* Hash function */
  95             (*hash)(const void *, size_t);
  96         int             flags;          /* Flag values */
  97         int             fp;             /* File pointer */
  98         char            *tmp_buf;       /* Temporary Buffer for BIG data */
  99         char            *tmp_key;       /* Temporary Buffer for BIG keys */
 100         BUFHEAD         *cpage;         /* Current page */
 101         int             cbucket;        /* Current bucket */
 102         int             cndx;           /* Index of next item on cpage */
 103         int             error;          /* Error Number -- for DBM
 104                                          * compatibility */
 105         int             new_file;       /* Indicates if fd is backing store
 106                                          * or no */
 107         int             save_file;      /* Indicates whether we need to flush
 108                                          * file at
 109                                          * exit */
 110         u_int32_t       *mapp[NCACHED]; /* Pointers to page maps */
 111         int             nmaps;          /* Initial number of bitmaps */
 112         int             nbufs;          /* Number of buffers left to
 113                                          * allocate */
 114         BUFHEAD         bufhead;        /* Header of buffer lru list */
 115         SEGMENT         *dir;           /* Hash Bucket directory */
 116                                         /* other flags */
 117         int             nextkey_eof :1; /* dbm_nextkey() reached EOF */
 118 } HTAB;
 119
 120 /*
 121  * Constants
 122  */
 123 #define MAX_BSIZE               65536           /* 2^16 */
 124 #define MIN_BUFFERS             6
 125 #define MINHDRSIZE              512
 126 #define DEF_BUFSIZE             65536           /* 64 K */
 127 #define DEF_BUCKET_SIZE         4096
 128 #define DEF_BUCKET_SHIFT        12              /* log2(BUCKET) */
 129 #define DEF_SEGSIZE             256
 130 #define DEF_SEGSIZE_SHIFT       8               /* log2(SEGSIZE)         */
 131 #define DEF_DIRSIZE             256
 132 #define DEF_FFACTOR             65536
 133 #define MIN_FFACTOR             4
 134 #define SPLTMAX                 8
 135 #define CHARKEY                 "%$sniglet^&"
 136 #define NUMKEY                  1038583
 137 #define BYTE_SHIFT              3
 138 #define INT_TO_BYTE             2
 139 #define INT_BYTE_SHIFT          5
 140 #define ALL_SET                 ((u_int32_t)0xFFFFFFFF)
 141 #define ALL_CLEAR               0
 142
 143 #define PTROF(X)        ((BUFHEAD *)((ptrdiff_t)(X)&~0x3))
 144 #define ISMOD(X)        ((u_int32_t)(ptrdiff_t)(X)&0x1)
 145 #define DOMOD(X)        ((X) = (char *)((ptrdiff_t)(X)|0x1))
 146 #define ISDISK(X)       ((u_int32_t)(ptrdiff_t)(X)&0x2)
 147 #define DODISK(X)       ((X) = (char *)((ptrdiff_t)(X)|0x2))
 148
 149 #define BITS_PER_MAP    32
 150
 151 /* Given the address of the beginning of a big map, clear/set the nth bit */
 152 #define CLRBIT(A, N)    ((A)[(N)/BITS_PER_MAP] &= ~(1<<((N)%BITS_PER_MAP)))
 153 #define SETBIT(A, N)    ((A)[(N)/BITS_PER_MAP] |= (1<<((N)%BITS_PER_MAP)))
 154 #define ISSET(A, N)     ((A)[(N)/BITS_PER_MAP] & (1<<((N)%BITS_PER_MAP)))
 155
 156 /* Overflow management */
 157 /*
 158  * Overflow page numbers are allocated per split point.  At each doubling of
 159  * the table, we can allocate extra pages.  So, an overflow page number has
 160  * the top 5 bits indicate which split point and the lower 11 bits indicate
 161  * which page at that split point is indicated (pages within split points are
 162  * numberered starting with 1).
 163  */
 164
 165 #define SPLITSHIFT      11
 166 #define SPLITMASK       0x7FF
 167 #define SPLITNUM(N)     (((u_int32_t)(N)) >> SPLITSHIFT)
 168 #define OPAGENUM(N)     ((N) & SPLITMASK)
 169 #define OADDR_OF(S,O)   ((u_int32_t)((u_int32_t)(S) << SPLITSHIFT) + (O))
 170
 171 #define BUCKET_TO_PAGE(B) \
 172         (B) + hashp->HDRPAGES + ((B) ? hashp->SPARES[__log2((B)+1)-1] : 0)
 173 #define OADDR_TO_PAGE(B)        \
 174         BUCKET_TO_PAGE ( (1 << SPLITNUM((B))) -1 ) + OPAGENUM((B));
 175
 176 /*
 177  * page.h contains a detailed description of the page format.
 178  *
 179  * Normally, keys and data are accessed from offset tables in the top of
 180  * each page which point to the beginning of the key and data.  There are
 181  * four flag values which may be stored in these offset tables which indicate
 182  * the following:
 183  *
 184  *
 185  * OVFLPAGE     Rather than a key data pair, this pair contains
 186  *              the address of an overflow page.  The format of
 187  *              the pair is:
 188  *                  OVERFLOW_PAGE_NUMBER OVFLPAGE
 189  *
 190  * PARTIAL_KEY  This must be the first key/data pair on a page
 191  *              and implies that page contains only a partial key.
 192  *              That is, the key is too big to fit on a single page
 193  *              so it starts on this page and continues on the next.
 194  *              The format of the page is:
 195  *                  KEY_OFF PARTIAL_KEY OVFL_PAGENO OVFLPAGE
 196  *
 197  *                  KEY_OFF -- offset of the beginning of the key
 198  *                  PARTIAL_KEY -- 1
 199  *                  OVFL_PAGENO - page number of the next overflow page
 200  *                  OVFLPAGE -- 0
 201  *
 202  * FULL_KEY     This must be the first key/data pair on the page.  It
 203  *              is used in two cases.
 204  *
 205  *              Case 1:
 206  *                  There is a complete key on the page but no data
 207  *                  (because it wouldn't fit).  The next page contains
 208  *                  the data.
 209  *
 210  *                  Page format it:
 211  *                  KEY_OFF FULL_KEY OVFL_PAGENO OVFL_PAGE
 212  *
 213  *                  KEY_OFF -- offset of the beginning of the key
 214  *                  FULL_KEY -- 2
 215  *                  OVFL_PAGENO - page number of the next overflow page
 216  *                  OVFLPAGE -- 0
 217  *
 218  *              Case 2:
 219  *                  This page contains no key, but part of a large
 220  *                  data field, which is continued on the next page.
 221  *
 222  *                  Page format it:
 223  *                  DATA_OFF FULL_KEY OVFL_PAGENO OVFL_PAGE
 224  *
 225  *                  KEY_OFF -- offset of the beginning of the data on
 226  *                              this page
 227  *                  FULL_KEY -- 2
 228  *                  OVFL_PAGENO - page number of the next overflow page
 229  *                  OVFLPAGE -- 0
 230  *
 231  * FULL_KEY_DATA
 232  *              This must be the first key/data pair on the page.
 233  *              There are two cases:
 234  *
 235  *              Case 1:
 236  *                  This page contains a key and the beginning of the
 237  *                  data field, but the data field is continued on the
 238  *                  next page.
 239  *
 240  *                  Page format is:
 241  *                  KEY_OFF FULL_KEY_DATA OVFL_PAGENO DATA_OFF
 242  *
 243  *                  KEY_OFF -- offset of the beginning of the key
 244  *                  FULL_KEY_DATA -- 3
 245  *                  OVFL_PAGENO - page number of the next overflow page
 246  *                  DATA_OFF -- offset of the beginning of the data
 247  *
 248  *              Case 2:
 249  *                  This page contains the last page of a big data pair.
 250  *                  There is no key, only the  tail end of the data
 251  *                  on this page.
 252  *
 253  *                  Page format is:
 254  *                  DATA_OFF FULL_KEY_DATA <OVFL_PAGENO> <OVFLPAGE>
 255  *
 256  *                  DATA_OFF -- offset of the beginning of the data on
 257  *                              this page
 258  *                  FULL_KEY_DATA -- 3
 259  *                  OVFL_PAGENO - page number of the next overflow page
 260  *                  OVFLPAGE -- 0
 261  *
 262  *                  OVFL_PAGENO and OVFLPAGE are optional (they are
 263  *                  not present if there is no next page).
 264  */
 265
 266 #define OVFLPAGE        0
 267 #define PARTIAL_KEY     1
 268 #define FULL_KEY        2
 269 #define FULL_KEY_DATA   3
 270 #define REAL_KEY        4
 271
 272 /* Short hands for accessing structure */
 273 #define BSIZE           hdr.bsize
 274 #define BSHIFT          hdr.bshift
 275 #define DSIZE           hdr.dsize
 276 #define SGSIZE          hdr.ssize
 277 #define SSHIFT          hdr.sshift
 278 #define LORDER          hdr.lorder
 279 #define OVFL_POINT      hdr.ovfl_point
 280 #define LAST_FREED      hdr.last_freed
 281 #define MAX_BUCKET      hdr.max_bucket
 282 #define FFACTOR         hdr.ffactor
 283 #define HIGH_MASK       hdr.high_mask
 284 #define LOW_MASK        hdr.low_mask
 285 #define NKEYS           hdr.nkeys
 286 #define HDRPAGES        hdr.hdrpages
 287 #define SPARES          hdr.spares
 288 #define BITMAPS         hdr.bitmaps
 289 #define VERSION         hdr.version
 290 #define MAGIC           hdr.magic
 291 #define NEXT_FREE       hdr.next_free
 292 #define H_CHARKEY       hdr.h_charkey