1 /*
2  * Copyright (c) 2018 Stefan Sperling <stsp@openbsd.org>
3  *
4  * Permission to use, copy, modify, and distribute this software for any
5  * purpose with or without fee is hereby granted, provided that the above
6  * copyright notice and this permission notice appear in all copies.
7  *
8  * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
9  * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
10  * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
11  * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
12  * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
13  * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
14  * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
15  */
16 
17 #define GOT_GIT_DIR	".git"
18 
19 /* Mandatory files and directories inside the git directory. */
20 #define GOT_OBJECTS_DIR		"objects"
21 #define GOT_REFS_DIR		"refs"
22 #define GOT_HEAD_FILE		"HEAD"
23 #define GOT_GITCONFIG		"config"
24 
25 /* Other files and directories inside the git directory. */
26 #define GOT_FETCH_HEAD_FILE	"FETCH_HEAD"
27 #define GOT_ORIG_HEAD_FILE	"ORIG_HEAD"
28 #define GOT_OBJECTS_PACK_DIR	"objects/pack"
29 #define GOT_PACKED_REFS_FILE	"packed-refs"
30 
31 #define GOT_PACK_CACHE_SIZE	32
32 
33 /*
34  * An open repository needs this many temporary files.
35  * This limit sets an upper bound on how many raw objects or blobs can
36  * be kept open in parallel.
37  */
38 #define GOT_REPO_NUM_TEMPFILES 32
39 
40 struct got_packidx_bloom_filter {
41 	RB_ENTRY(got_packidx_bloom_filter) entry;
42 	char path[PATH_MAX]; /* on-disk path */
43 	size_t path_len;
44 	struct bloom *bloom;
45 };
46 
47 RB_HEAD(got_packidx_bloom_filter_tree, got_packidx_bloom_filter);
48 
49 static inline int
50 got_packidx_bloom_filter_cmp(const struct got_packidx_bloom_filter *f1,
51     const struct got_packidx_bloom_filter *f2)
52 {
53 	return got_path_cmp(f1->path, f2->path, f1->path_len, f2->path_len);
54 }
55 
56 struct got_repo_privsep_child {
57 	int imsg_fd;
58 	pid_t pid;
59 	struct imsgbuf *ibuf;
60 };
61 
62 struct got_repository {
63 	char *path;
64 	char *path_git_dir;
65 	int gitdir_fd;
66 	enum got_hash_algorithm algo;
67 
68 	struct got_pathlist_head packidx_paths;
69 	struct timespec pack_path_mtime;
70 
71 	/* The pack index cache speeds up search for packed objects. */
72 	struct got_packidx *packidx_cache[GOT_PACK_CACHE_SIZE];
73 
74 	/*
75 	 * List of bloom filters for pack index files.
76 	 * Used to avoid opening a pack index in search of an
77 	 * object ID which is not contained in this pack index.
78 	 */
79 	struct got_packidx_bloom_filter_tree packidx_bloom_filters;
80 
81 	/*
82 	 * Open file handles for pack files. Each struct got_pack uses
83 	 * a number of file descriptors. See struct got_pack for details.
84 	 */
85 	struct got_pack packs[GOT_PACK_CACHE_SIZE];
86 
87 	/* Open file handles for temporary files. */
88 	int tempfiles[GOT_REPO_NUM_TEMPFILES];
89 	uint32_t tempfile_use_mask;
90 
91 	/*
92 	 * The cache size limit may be lower than GOT_PACK_CACHE_SIZE,
93 	 * depending on resource limits.
94 	 */
95 	int pack_cache_size;
96 
97 	/*
98 	 * Index to cache entries which are pinned to avoid eviction.
99 	 * This may be used to keep one got-index-pack process alive
100 	 * across searches for arbitrary objects which may be stored
101 	 * in other pack files.
102 	 */
103 	int pinned_pack;
104 	pid_t pinned_pid;
105 	int pinned_packidx;
106 
107 	/* Handles to child processes for reading loose objects. */
108 	struct got_repo_privsep_child privsep_children[5];
109 #define GOT_REPO_PRIVSEP_CHILD_OBJECT	0
110 #define GOT_REPO_PRIVSEP_CHILD_COMMIT	1
111 #define GOT_REPO_PRIVSEP_CHILD_TREE	2
112 #define GOT_REPO_PRIVSEP_CHILD_BLOB	3
113 #define GOT_REPO_PRIVSEP_CHILD_TAG	4
114 
115 	/* Caches for open objects. */
116 	struct got_object_cache objcache;
117 	struct got_object_cache treecache;
118 	struct got_object_cache commitcache;
119 	struct got_object_cache tagcache;
120 	struct got_object_cache rawcache;
121 
122 	/* Settings read from Git configuration files. */
123 	int gitconfig_repository_format_version;
124 	char *gitconfig_author_name;
125 	char *gitconfig_author_email;
126 	char *global_gitconfig_author_name;
127 	char *global_gitconfig_author_email;
128 	int ngitconfig_remotes;
129 	struct got_remote_repo *gitconfig_remotes;
130 	char *gitconfig_owner;
131 	char **extnames;
132 	char **extvals;
133 	int nextensions;
134 
135 	/* Settings read from got.conf. */
136 	struct got_gotconfig *gotconfig;
137 };
138 
139 const struct got_error*got_repo_cache_object(struct got_repository *,
140     struct got_object_id *, struct got_object *);
141 struct got_object *got_repo_get_cached_object(struct got_repository *,
142     struct got_object_id *);
143 const struct got_error*got_repo_cache_tree(struct got_repository *,
144     struct got_object_id *, struct got_tree_object *);
145 struct got_tree_object *got_repo_get_cached_tree(struct got_repository *,
146     struct got_object_id *);
147 const struct got_error*got_repo_cache_commit(struct got_repository *,
148     struct got_object_id *, struct got_commit_object *);
149 struct got_commit_object *got_repo_get_cached_commit(struct got_repository *,
150     struct got_object_id *);
151 const struct got_error*got_repo_cache_tag(struct got_repository *,
152     struct got_object_id *, struct got_tag_object *);
153 struct got_tag_object *got_repo_get_cached_tag(struct got_repository *,
154     struct got_object_id *);
155 const struct got_error*got_repo_cache_raw_object(struct got_repository *,
156     struct got_object_id *, struct got_raw_object *);
157 struct got_raw_object *got_repo_get_cached_raw_object(struct got_repository *,
158     struct got_object_id *);
159 int got_repo_is_packidx_filename(const char *, size_t);
160 int got_repo_check_packidx_bloom_filter(struct got_repository *,
161     const char *, struct got_object_id *);
162 const struct got_error *got_repo_search_packidx(struct got_packidx **, int *,
163     struct got_repository *, struct got_object_id *);
164 const struct got_error *got_repo_list_packidx(struct got_pathlist_head *,
165     struct got_repository *);
166 const struct got_error *got_repo_get_packidx(struct got_packidx **, const char *,
167     struct got_repository *);
168 const struct got_error *got_repo_cache_pack(struct got_pack **,
169     struct got_repository *, const char *, struct got_packidx *);
170 struct got_pack *got_repo_get_cached_pack(struct got_repository *,
171     const char *);
172 const struct got_error *got_repo_pin_pack(struct got_repository *,
173     struct got_packidx *, struct got_pack *);
174 struct got_pack *got_repo_get_pinned_pack(struct got_repository *);
175 void got_repo_unpin_pack(struct got_repository *);
176 
177 const struct got_error *got_repo_read_gitconfig(int *, char **, char **,
178     struct got_remote_repo **, int *, char **, char ***, char ***, int *,
179     const char *);
180 
181 const struct got_error *got_repo_temp_fds_get(int *, int *,
182     struct got_repository *);
183 void got_repo_temp_fds_put(int, struct got_repository *);