object-store.h

   1 #ifndef OBJECT_STORE_H
   2 #define OBJECT_STORE_H
   3
   4 #include "cache.h"
   5 #include "oidmap.h"
   6 #include "list.h"
   7 #include "oid-array.h"
   8 #include "strbuf.h"
   9 #include "thread-utils.h"
  10 #include "khash.h"
  11 #include "dir.h"
  12 #include "oidtree.h"
  13 #include "oidset.h"
  14
  15 struct object_directory {
  16         struct object_directory *next;
  17
  18         /*
  19          * Used to store the results of readdir(3) calls when we are OK
  20          * sacrificing accuracy due to races for speed. That includes
  21          * object existence with OBJECT_INFO_QUICK, as well as
  22          * our search for unique abbreviated hashes. Don't use it for tasks
  23          * requiring greater accuracy!
  24          *
  25          * Be sure to call odb_load_loose_cache() before using.
  26          */
  27         uint32_t loose_objects_subdir_seen[8]; /* 256 bits */
  28         struct oidtree *loose_objects_cache;
  29
  30         /*
  31          * This is a temporary object store created by the tmp_objdir
  32          * facility. Disable ref updates since the objects in the store
  33          * might be discarded on rollback.
  34          */
  35         int disable_ref_updates;
  36
  37         /*
  38          * This object store is ephemeral, so there is no need to fsync.
  39          */
  40         int will_destroy;
  41
  42         /*
  43          * Path to the alternative object store. If this is a relative path,
  44          * it is relative to the current working directory.
  45          */
  46         char *path;
  47 };
  48
  49 struct input_stream {
  50         const void *(*read)(struct input_stream *, unsigned long *len);
  51         void *data;
  52         int is_finished;
  53 };
  54
  55 KHASH_INIT(odb_path_map, const char * /* key: odb_path */,
  56         struct object_directory *, 1, fspathhash, fspatheq)
  57
  58 void prepare_alt_odb(struct repository *r);
  59 char *compute_alternate_path(const char *path, struct strbuf *err);
  60 struct object_directory *find_odb(struct repository *r, const char *obj_dir);
  61 typedef int alt_odb_fn(struct object_directory *, void *);
  62 int foreach_alt_odb(alt_odb_fn, void*);
  63 typedef void alternate_ref_fn(const struct object_id *oid, void *);
  64 void for_each_alternate_ref(alternate_ref_fn, void *);
  65
  66 /*
  67  * Add the directory to the on-disk alternates file; the new entry will also
  68  * take effect in the current process.
  69  */
  70 void add_to_alternates_file(const char *dir);
  71
  72 /*
  73  * Add the directory to the in-memory list of alternates (along with any
  74  * recursive alternates it points to), but do not modify the on-disk alternates
  75  * file.
  76  */
  77 void add_to_alternates_memory(const char *dir);
  78
  79 /*
  80  * Replace the current writable object directory with the specified temporary
  81  * object directory; returns the former primary object directory.
  82  */
  83 struct object_directory *set_temporary_primary_odb(const char *dir, int will_destroy);
  84
  85 /*
  86  * Restore a previous ODB replaced by set_temporary_main_odb.
  87  */
  88 void restore_primary_odb(struct object_directory *restore_odb, const char *old_path);
  89
  90 /*
  91  * Populate and return the loose object cache array corresponding to the
  92  * given object ID.
  93  */
  94 struct oidtree *odb_loose_cache(struct object_directory *odb,
  95                                   const struct object_id *oid);
  96
  97 /* Empty the loose object cache for the specified object directory. */
  98 void odb_clear_loose_cache(struct object_directory *odb);
  99
 100 /* Clear and free the specified object directory */
 101 void free_object_directory(struct object_directory *odb);
 102
 103 struct packed_git {
 104         struct hashmap_entry packmap_ent;
 105         struct packed_git *next;
 106         struct list_head mru;
 107         struct pack_window *windows;
 108         off_t pack_size;
 109         const void *index_data;
 110         size_t index_size;
 111         uint32_t num_objects;
 112         uint32_t crc_offset;
 113         struct oidset bad_objects;
 114         int index_version;
 115         time_t mtime;
 116         int pack_fd;
 117         int index;              /* for builtin/pack-objects.c */
 118         unsigned pack_local:1,
 119                  pack_keep:1,
 120                  pack_keep_in_core:1,
 121                  freshened:1,
 122                  do_not_close:1,
 123                  pack_promisor:1,
 124                  multi_pack_index:1,
 125                  is_cruft:1;
 126         unsigned char hash[GIT_MAX_RAWSZ];
 127         struct revindex_entry *revindex;
 128         const uint32_t *revindex_data;
 129         const uint32_t *revindex_map;
 130         size_t revindex_size;
 131         /*
 132          * mtimes_map points at the beginning of the memory mapped region of
 133          * this pack's corresponding .mtimes file, and mtimes_size is the size
 134          * of that .mtimes file
 135          */
 136         const uint32_t *mtimes_map;
 137         size_t mtimes_size;
 138         /* something like ".git/objects/pack/xxxxx.pack" */
 139         char pack_name[FLEX_ARRAY]; /* more */
 140 };
 141
 142 struct multi_pack_index;
 143
 144 static inline int pack_map_entry_cmp(const void *cmp_data UNUSED,
 145                                      const struct hashmap_entry *entry,
 146                                      const struct hashmap_entry *entry2,
 147                                      const void *keydata)
 148 {
 149         const char *key = keydata;
 150         const struct packed_git *pg1, *pg2;
 151
 152         pg1 = container_of(entry, const struct packed_git, packmap_ent);
 153         pg2 = container_of(entry2, const struct packed_git, packmap_ent);
 154
 155         return strcmp(pg1->pack_name, key ? key : pg2->pack_name);
 156 }
 157
 158 struct raw_object_store {
 159         /*
 160          * Set of all object directories; the main directory is first (and
 161          * cannot be NULL after initialization). Subsequent directories are
 162          * alternates.
 163          */
 164         struct object_directory *odb;
 165         struct object_directory **odb_tail;
 166         kh_odb_path_map_t *odb_by_path;
 167
 168         int loaded_alternates;
 169
 170         /*
 171          * A list of alternate object directories loaded from the environment;
 172          * this should not generally need to be accessed directly, but will
 173          * populate the "odb" list when prepare_alt_odb() is run.
 174          */
 175         char *alternate_db;
 176
 177         /*
 178          * Objects that should be substituted by other objects
 179          * (see git-replace(1)).
 180          */
 181         struct oidmap *replace_map;
 182         unsigned replace_map_initialized : 1;
 183         pthread_mutex_t replace_mutex; /* protect object replace functions */
 184
 185         struct commit_graph *commit_graph;
 186         unsigned commit_graph_attempted : 1; /* if loading has been attempted */
 187
 188         /*
 189          * private data
 190          *
 191          * should only be accessed directly by packfile.c and midx.c
 192          */
 193         struct multi_pack_index *multi_pack_index;
 194
 195         /*
 196          * private data
 197          *
 198          * should only be accessed directly by packfile.c
 199          */
 200
 201         struct packed_git *packed_git;
 202         /* A most-recently-used ordered version of the packed_git list. */
 203         struct list_head packed_git_mru;
 204
 205         struct {
 206                 struct packed_git **packs;
 207                 unsigned flags;
 208         } kept_pack_cache;
 209
 210         /*
 211          * A map of packfiles to packed_git structs for tracking which
 212          * packs have been loaded already.
 213          */
 214         struct hashmap pack_map;
 215
 216         /*
 217          * A fast, rough count of the number of objects in the repository.
 218          * These two fields are not meant for direct access. Use
 219          * approximate_object_count() instead.
 220          */
 221         unsigned long approximate_object_count;
 222         unsigned approximate_object_count_valid : 1;
 223
 224         /*
 225          * Whether packed_git has already been populated with this repository's
 226          * packs.
 227          */
 228         unsigned packed_git_initialized : 1;
 229 };
 230
 231 struct raw_object_store *raw_object_store_new(void);
 232 void raw_object_store_clear(struct raw_object_store *o);
 233
 234 /*
 235  * Put in `buf` the name of the file in the local object database that
 236  * would be used to store a loose object with the specified oid.
 237  */
 238 const char *loose_object_path(struct repository *r, struct strbuf *buf,
 239                               const struct object_id *oid);
 240
 241 void *map_loose_object(struct repository *r, const struct object_id *oid,
 242                        unsigned long *size);
 243
 244 void *repo_read_object_file(struct repository *r,
 245                             const struct object_id *oid,
 246                             enum object_type *type,
 247                             unsigned long *size);
 248 #ifndef NO_THE_REPOSITORY_COMPATIBILITY_MACROS
 249 #define read_object_file(oid, type, size) repo_read_object_file(the_repository, oid, type, size)
 250 #endif
 251
 252 /* Read and unpack an object file into memory, write memory to an object file */
 253 int oid_object_info(struct repository *r, const struct object_id *, unsigned long *);
 254
 255 void hash_object_file(const struct git_hash_algo *algo, const void *buf,
 256                       unsigned long len, enum object_type type,
 257                       struct object_id *oid);
 258
 259 int write_object_file_flags(const void *buf, unsigned long len,
 260                             enum object_type type, struct object_id *oid,
 261                             unsigned flags);
 262 static inline int write_object_file(const void *buf, unsigned long len,
 263                                     enum object_type type, struct object_id *oid)
 264 {
 265         return write_object_file_flags(buf, len, type, oid, 0);
 266 }
 267
 268 int write_object_file_literally(const void *buf, unsigned long len,
 269                                 const char *type, struct object_id *oid,
 270                                 unsigned flags);
 271 int stream_loose_object(struct input_stream *in_stream, size_t len,
 272                         struct object_id *oid);
 273
 274 /*
 275  * Add an object file to the in-memory object store, without writing it
 276  * to disk.
 277  *
 278  * Callers are responsible for calling write_object_file to record the
 279  * object in persistent storage before writing any other new objects
 280  * that reference it.
 281  */
 282 int pretend_object_file(void *, unsigned long, enum object_type,
 283                         struct object_id *oid);
 284
 285 int force_object_loose(const struct object_id *oid, time_t mtime);
 286
 287 /*
 288  * Open the loose object at path, check its hash, and return the contents,
 289  * use the "oi" argument to assert things about the object, or e.g. populate its
 290  * type, and size. If the object is a blob, then "contents" may return NULL,
 291  * to allow streaming of large blobs.
 292  *
 293  * Returns 0 on success, negative on error (details may be written to stderr).
 294  */
 295 int read_loose_object(const char *path,
 296                       const struct object_id *expected_oid,
 297                       struct object_id *real_oid,
 298                       void **contents,
 299                       struct object_info *oi);
 300
 301 /* Retry packed storage after checking packed and loose storage */
 302 #define HAS_OBJECT_RECHECK_PACKED 1
 303
 304 /*
 305  * Returns 1 if the object exists. This function will not lazily fetch objects
 306  * in a partial clone.
 307  */
 308 int has_object(struct repository *r, const struct object_id *oid,
 309                unsigned flags);
 310
 311 /*
 312  * These macros and functions are deprecated. If checking existence for an
 313  * object that is likely to be missing and/or whose absence is relatively
 314  * inconsequential (or is consequential but the caller is prepared to handle
 315  * it), use has_object(), which has better defaults (no lazy fetch in a partial
 316  * clone and no rechecking of packed storage). In the unlikely event that a
 317  * caller needs to assert existence of an object that it fully expects to
 318  * exist, and wants to trigger a lazy fetch in a partial clone, use
 319  * oid_object_info_extended() with a NULL struct object_info.
 320  *
 321  * These functions can be removed once all callers have migrated to
 322  * has_object() and/or oid_object_info_extended().
 323  */
 324 int repo_has_object_file(struct repository *r, const struct object_id *oid);
 325 int repo_has_object_file_with_flags(struct repository *r,
 326                                     const struct object_id *oid, int flags);
 327 #ifndef NO_THE_REPOSITORY_COMPATIBILITY_MACROS
 328 #define has_object_file(oid) repo_has_object_file(the_repository, oid)
 329 #define has_object_file_with_flags(oid, flags) repo_has_object_file_with_flags(the_repository, oid, flags)
 330 #endif
 331
 332 /*
 333  * Return true iff an alternate object database has a loose object
 334  * with the specified name.  This function does not respect replace
 335  * references.
 336  */
 337 int has_loose_object_nonlocal(const struct object_id *);
 338
 339 int has_loose_object(const struct object_id *);
 340
 341 /**
 342  * format_object_header() is a thin wrapper around s xsnprintf() that
 343  * writes the initial "<type> <obj-len>" part of the loose object
 344  * header. It returns the size that snprintf() returns + 1.
 345  */
 346 int format_object_header(char *str, size_t size, enum object_type type,
 347                          size_t objsize);
 348
 349 void assert_oid_type(const struct object_id *oid, enum object_type expect);
 350
 351 /*
 352  * Enabling the object read lock allows multiple threads to safely call the
 353  * following functions in parallel: repo_read_object_file(), read_object_file(),
 354  * read_object_with_reference(), oid_object_info() and oid_object_info_extended().
 355  *
 356  * obj_read_lock() and obj_read_unlock() may also be used to protect other
 357  * section which cannot execute in parallel with object reading. Since the used
 358  * lock is a recursive mutex, these sections can even contain calls to object
 359  * reading functions. However, beware that in these cases zlib inflation won't
 360  * be performed in parallel, losing performance.
 361  *
 362  * TODO: oid_object_info_extended()'s call stack has a recursive behavior. If
 363  * any of its callees end up calling it, this recursive call won't benefit from
 364  * parallel inflation.
 365  */
 366 void enable_obj_read_lock(void);
 367 void disable_obj_read_lock(void);
 368
 369 extern int obj_read_use_lock;
 370 extern pthread_mutex_t obj_read_mutex;
 371
 372 static inline void obj_read_lock(void)
 373 {
 374         if(obj_read_use_lock)
 375                 pthread_mutex_lock(&obj_read_mutex);
 376 }
 377
 378 static inline void obj_read_unlock(void)
 379 {
 380         if(obj_read_use_lock)
 381                 pthread_mutex_unlock(&obj_read_mutex);
 382 }
 383
 384 struct object_info {
 385         /* Request */
 386         enum object_type *typep;
 387         unsigned long *sizep;
 388         off_t *disk_sizep;
 389         struct object_id *delta_base_oid;
 390         struct strbuf *type_name;
 391         void **contentp;
 392
 393         /* Response */
 394         enum {
 395                 OI_CACHED,
 396                 OI_LOOSE,
 397                 OI_PACKED,
 398                 OI_DBCACHED
 399         } whence;
 400         union {
 401                 /*
 402                  * struct {
 403                  *      ... Nothing to expose in this case
 404                  * } cached;
 405                  * struct {
 406                  *      ... Nothing to expose in this case
 407                  * } loose;
 408                  */
 409                 struct {
 410                         struct packed_git *pack;
 411                         off_t offset;
 412                         unsigned int is_delta;
 413                 } packed;
 414         } u;
 415 };
 416
 417 /*
 418  * Initializer for a "struct object_info" that wants no items. You may
 419  * also memset() the memory to all-zeroes.
 420  */
 421 #define OBJECT_INFO_INIT { 0 }
 422
 423 /* Invoke lookup_replace_object() on the given hash */
 424 #define OBJECT_INFO_LOOKUP_REPLACE 1
 425 /* Allow reading from a loose object file of unknown/bogus type */
 426 #define OBJECT_INFO_ALLOW_UNKNOWN_TYPE 2
 427 /* Do not retry packed storage after checking packed and loose storage */
 428 #define OBJECT_INFO_QUICK 8
 429 /*
 430  * Do not attempt to fetch the object if missing (even if fetch_is_missing is
 431  * nonzero).
 432  */
 433 #define OBJECT_INFO_SKIP_FETCH_OBJECT 16
 434 /*
 435  * This is meant for bulk prefetching of missing blobs in a partial
 436  * clone. Implies OBJECT_INFO_SKIP_FETCH_OBJECT and OBJECT_INFO_QUICK
 437  */
 438 #define OBJECT_INFO_FOR_PREFETCH (OBJECT_INFO_SKIP_FETCH_OBJECT | OBJECT_INFO_QUICK)
 439
 440 /* Die if object corruption (not just an object being missing) was detected. */
 441 #define OBJECT_INFO_DIE_IF_CORRUPT 32
 442
 443 int oid_object_info_extended(struct repository *r,
 444                              const struct object_id *,
 445                              struct object_info *, unsigned flags);
 446
 447 /*
 448  * Iterate over the files in the loose-object parts of the object
 449  * directory "path", triggering the following callbacks:
 450  *
 451  *  - loose_object is called for each loose object we find.
 452  *
 453  *  - loose_cruft is called for any files that do not appear to be
 454  *    loose objects. Note that we only look in the loose object
 455  *    directories "objects/[0-9a-f]{2}/", so we will not report
 456  *    "objects/foobar" as cruft.
 457  *
 458  *  - loose_subdir is called for each top-level hashed subdirectory
 459  *    of the object directory (e.g., "$OBJDIR/f0"). It is called
 460  *    after the objects in the directory are processed.
 461  *
 462  * Any callback that is NULL will be ignored. Callbacks returning non-zero
 463  * will end the iteration.
 464  *
 465  * In the "buf" variant, "path" is a strbuf which will also be used as a
 466  * scratch buffer, but restored to its original contents before
 467  * the function returns.
 468  */
 469 typedef int each_loose_object_fn(const struct object_id *oid,
 470                                  const char *path,
 471                                  void *data);
 472 typedef int each_loose_cruft_fn(const char *basename,
 473                                 const char *path,
 474                                 void *data);
 475 typedef int each_loose_subdir_fn(unsigned int nr,
 476                                  const char *path,
 477                                  void *data);
 478 int for_each_file_in_obj_subdir(unsigned int subdir_nr,
 479                                 struct strbuf *path,
 480                                 each_loose_object_fn obj_cb,
 481                                 each_loose_cruft_fn cruft_cb,
 482                                 each_loose_subdir_fn subdir_cb,
 483                                 void *data);
 484 int for_each_loose_file_in_objdir(const char *path,
 485                                   each_loose_object_fn obj_cb,
 486                                   each_loose_cruft_fn cruft_cb,
 487                                   each_loose_subdir_fn subdir_cb,
 488                                   void *data);
 489 int for_each_loose_file_in_objdir_buf(struct strbuf *path,
 490                                       each_loose_object_fn obj_cb,
 491                                       each_loose_cruft_fn cruft_cb,
 492                                       each_loose_subdir_fn subdir_cb,
 493                                       void *data);
 494
 495 /* Flags for for_each_*_object() below. */
 496 enum for_each_object_flags {
 497         /* Iterate only over local objects, not alternates. */
 498         FOR_EACH_OBJECT_LOCAL_ONLY = (1<<0),
 499
 500         /* Only iterate over packs obtained from the promisor remote. */
 501         FOR_EACH_OBJECT_PROMISOR_ONLY = (1<<1),
 502
 503         /*
 504          * Visit objects within a pack in packfile order rather than .idx order
 505          */
 506         FOR_EACH_OBJECT_PACK_ORDER = (1<<2),
 507
 508         /* Only iterate over packs that are not marked as kept in-core. */
 509         FOR_EACH_OBJECT_SKIP_IN_CORE_KEPT_PACKS = (1<<3),
 510
 511         /* Only iterate over packs that do not have .keep files. */
 512         FOR_EACH_OBJECT_SKIP_ON_DISK_KEPT_PACKS = (1<<4),
 513 };
 514
 515 /*
 516  * Iterate over all accessible loose objects without respect to
 517  * reachability. By default, this includes both local and alternate objects.
 518  * The order in which objects are visited is unspecified.
 519  *
 520  * Any flags specific to packs are ignored.
 521  */
 522 int for_each_loose_object(each_loose_object_fn, void *,
 523                           enum for_each_object_flags flags);
 524
 525 /*
 526  * Iterate over all accessible packed objects without respect to reachability.
 527  * By default, this includes both local and alternate packs.
 528  *
 529  * Note that some objects may appear twice if they are found in multiple packs.
 530  * Each pack is visited in an unspecified order. By default, objects within a
 531  * pack are visited in pack-idx order (i.e., sorted by oid).
 532  */
 533 typedef int each_packed_object_fn(const struct object_id *oid,
 534                                   struct packed_git *pack,
 535                                   uint32_t pos,
 536                                   void *data);
 537 int for_each_object_in_pack(struct packed_git *p,
 538                             each_packed_object_fn, void *data,
 539                             enum for_each_object_flags flags);
 540 int for_each_packed_object(each_packed_object_fn, void *,
 541                            enum for_each_object_flags flags);
 542
 543 #endif /* OBJECT_STORE_H */