Diffstat (limited to 'cache.c')
| -rw-r--r-- | cache.c | 497 |
1 files changed, 58 insertions, 439 deletions
@@ -1,475 +1,94 @@ /* cache.c: cache management * - * Copyright (C) 2006-2014 cgit Development Team <cgit@lists.zx2c4.com> + * Copyright (C) 2006 Lars Hjemli * * Licensed under GNU General Public License v2 * (see COPYING for full license text) - * - * - * The cache is just a directory structure where each file is a cache slot, - * and each filename is based on the hash of some key (e.g. the cgit url). - * Each file contains the full key followed by the cached content for that - * key. - * */ #include "cgit.h" -#include "cache.h" -#include "html.h" -#ifdef HAVE_LINUX_SENDFILE -#include <sys/sendfile.h> -#endif -#define CACHE_BUFSIZE (1024 * 4) +const int NOLOCK = -1; -struct cache_slot { - const char *key; - size_t keylen; - int ttl; - cache_fill_fn fn; - int cache_fd; - int lock_fd; - int stdout_fd; - const char *cache_name; - const char *lock_name; - int match; - struct stat cache_st; - int bufsize; - char buf[CACHE_BUFSIZE]; -}; - -/* Open an existing cache slot and fill the cache buffer with - * (part of) the content of the cache file. Return 0 on success - * and errno otherwise. - */ -static int open_slot(struct cache_slot *slot) +int cache_lookup(struct cacheitem *item) { - char *bufz; - ssize_t bufkeylen = -1; - - slot->cache_fd = open(slot->cache_name, O_RDONLY); - if (slot->cache_fd == -1) - return errno; - - if (fstat(slot->cache_fd, &slot->cache_st)) - return errno; - - slot->bufsize = xread(slot->cache_fd, slot->buf, sizeof(slot->buf)); - if (slot->bufsize < 0) - return errno; - - bufz = memchr(slot->buf, 0, slot->bufsize); - if (bufz) - bufkeylen = bufz - slot->buf; - - if (slot->key) - slot->match = bufkeylen == slot->keylen && - !memcmp(slot->key, slot->buf, bufkeylen + 1); - - return 0; -} - -/* Close the active cache slot */ -static int close_slot(struct cache_slot *slot) -{ - int err = 0; - if (slot->cache_fd > 0) { - if (close(slot->cache_fd)) - err = errno; + if (!cgit_query_repo) { + item->name = xstrdup(fmt("%s/index.html", cgit_cache_root)); + item->ttl = cgit_cache_root_ttl; + } else if (!cgit_query_page) { + item->name = xstrdup(fmt("%s/%s/index.html", cgit_cache_root, + cgit_query_repo)); + item->ttl = cgit_cache_repo_ttl; + } else { + item->name = xstrdup(fmt("%s/%s/%s/%s.html", cgit_cache_root, + cgit_query_repo, cgit_query_page, + cgit_querystring)); + if (cgit_query_has_symref) + item->ttl = cgit_cache_dynamic_ttl; + else if (cgit_query_has_sha1) + item->ttl = cgit_cache_static_ttl; else - slot->cache_fd = -1; + item->ttl = cgit_cache_repo_ttl; } - return err; -} - -/* Print the content of the active cache slot (but skip the key). */ -static int print_slot(struct cache_slot *slot) -{ - off_t off; -#ifdef HAVE_LINUX_SENDFILE - off_t size; -#endif - - off = slot->keylen + 1; - -#ifdef HAVE_LINUX_SENDFILE - size = slot->cache_st.st_size; - - do { - ssize_t ret; - ret = sendfile(STDOUT_FILENO, slot->cache_fd, &off, size - off); - if (ret < 0) { - if (errno == EAGAIN || errno == EINTR) - continue; - /* Fall back to read/write on EINVAL or ENOSYS */ - if (errno == EINVAL || errno == ENOSYS) - break; - return errno; - } - if (off == size) - return 0; - } while (1); -#endif - - if (lseek(slot->cache_fd, off, SEEK_SET) != off) - return errno; - - do { - ssize_t ret; - ret = xread(slot->cache_fd, slot->buf, sizeof(slot->buf)); - if (ret < 0) - return errno; - if (ret == 0) - return 0; - if (write_in_full(STDOUT_FILENO, slot->buf, ret) < 0) - return errno; - } while (1); -} - -/* Check if the slot has expired */ -static int is_expired(struct cache_slot *slot) -{ - if (slot->ttl < 0) + if (stat(item->name, &item->st)) { + item->st.st_mtime = 0; return 0; - else - return slot->cache_st.st_mtime + slot->ttl * 60 < time(NULL); -} - -/* Check if the slot has been modified since we opened it. - * NB: If stat() fails, we pretend the file is modified. - */ -static int is_modified(struct cache_slot *slot) -{ - struct stat st; - - if (stat(slot->cache_name, &st)) - return 1; - return (st.st_ino != slot->cache_st.st_ino || - st.st_mtime != slot->cache_st.st_mtime || - st.st_size != slot->cache_st.st_size); -} - -/* Close an open lockfile */ -static int close_lock(struct cache_slot *slot) -{ - int err = 0; - if (slot->lock_fd > 0) { - if (close(slot->lock_fd)) - err = errno; - else - slot->lock_fd = -1; } - return err; + return 1; } -/* Create a lockfile used to store the generated content for a cache - * slot, and write the slot key + \0 into it. - * Returns 0 on success and errno otherwise. - */ -static int lock_slot(struct cache_slot *slot) +int cache_create_dirs() { - struct flock lock = { - .l_type = F_WRLCK, - .l_whence = SEEK_SET, - .l_start = 0, - .l_len = 0, - }; - - slot->lock_fd = open(slot->lock_name, O_RDWR | O_CREAT, - S_IRUSR | S_IWUSR); - if (slot->lock_fd == -1) - return errno; - if (fcntl(slot->lock_fd, F_SETLK, &lock) < 0) { - int saved_errno = errno; - close(slot->lock_fd); - slot->lock_fd = -1; - return saved_errno; - } - if (ftruncate(slot->lock_fd, 0) < 0) - return errno; - if (xwrite(slot->lock_fd, slot->key, slot->keylen + 1) < 0) - return errno; - return 0; -} - -/* Release the current lockfile. If `replace_old_slot` is set the - * lockfile replaces the old cache slot, otherwise the lockfile is - * just deleted. - */ -static int unlock_slot(struct cache_slot *slot, int replace_old_slot) -{ - int err; - - if (replace_old_slot) - err = rename(slot->lock_name, slot->cache_name); - else - err = unlink(slot->lock_name); - - /* Restore stdout and close the temporary FD. */ - if (slot->stdout_fd >= 0) { - dup2(slot->stdout_fd, STDOUT_FILENO); - close(slot->stdout_fd); - slot->stdout_fd = -1; - } + char *path; - if (err) - return errno; - - return 0; -} - -/* Generate the content for the current cache slot by redirecting - * stdout to the lock-fd and invoking the callback function - */ -static int fill_slot(struct cache_slot *slot) -{ - /* Preserve stdout */ - slot->stdout_fd = dup(STDOUT_FILENO); - if (slot->stdout_fd == -1) - return errno; - - /* Redirect stdout to lockfile */ - if (dup2(slot->lock_fd, STDOUT_FILENO) == -1) - return errno; - - /* Generate cache content */ - slot->fn(); - - /* Make sure any buffered data is flushed to the file */ - if (fflush(stdout)) - return errno; - - /* update stat info */ - if (fstat(slot->lock_fd, &slot->cache_st)) - return errno; - - return 0; -} - -/* Crude implementation of 32-bit FNV-1 hash algorithm, - * see http://www.isthe.com/chongo/tech/comp/fnv/ for details - * about the magic numbers. - */ -#define FNV_OFFSET 0x811c9dc5 -#define FNV_PRIME 0x01000193 - -unsigned long hash_str(const char *str) -{ - unsigned long h = FNV_OFFSET; - unsigned char *s = (unsigned char *)str; - - if (!s) - return h; - - while (*s) { - h *= FNV_PRIME; - h ^= *s++; - } - return h; -} - -static int process_slot(struct cache_slot *slot) -{ - int err; - - err = open_slot(slot); - if (!err && slot->match) { - if (is_expired(slot)) { - if (!lock_slot(slot)) { - /* If the cachefile has been replaced between - * `open_slot` and `lock_slot`, we'll just - * serve the stale content from the original - * cachefile. This way we avoid pruning the - * newly generated slot. The same code-path - * is chosen if fill_slot() fails for some - * reason. - * - * TODO? check if the new slot contains the - * same key as the old one, since we would - * prefer to serve the newest content. - * This will require us to open yet another - * file-descriptor and read and compare the - * key from the new file, so for now we're - * lazy and just ignore the new file. - */ - if (is_modified(slot) || fill_slot(slot)) { - unlock_slot(slot, 0); - close_lock(slot); - } else { - close_slot(slot); - unlock_slot(slot, 1); - slot->cache_fd = slot->lock_fd; - } - } - } - if ((err = print_slot(slot)) != 0) { - cache_log("[cgit] error printing cache %s: %s (%d)\n", - slot->cache_name, - strerror(err), - err); - } - close_slot(slot); - return err; - } - - /* If the cache slot does not exist (or its key doesn't match the - * current key), lets try to create a new cache slot for this - * request. If this fails (for whatever reason), lets just generate - * the content without caching it and fool the caller to believe - * everything worked out (but print a warning on stdout). - */ - - close_slot(slot); - if ((err = lock_slot(slot)) != 0) { - cache_log("[cgit] Unable to lock slot %s: %s (%d)\n", - slot->lock_name, strerror(err), err); - slot->fn(); + if (!cgit_query_repo) return 0; - } - if ((err = fill_slot(slot)) != 0) { - cache_log("[cgit] Unable to fill slot %s: %s (%d)\n", - slot->lock_name, strerror(err), err); - unlock_slot(slot, 0); - close_lock(slot); - slot->fn(); + path = fmt("%s/%s", cgit_cache_root, cgit_query_repo); + if (mkdir(path, S_IRWXU) && errno!=EEXIST) return 0; + + if (cgit_query_page) { + path = fmt("%s/%s/%s", cgit_cache_root, cgit_query_repo, + cgit_query_page); + if (mkdir(path, S_IRWXU) && errno!=EEXIST) + return 0; } - // We've got a valid cache slot in the lock file, which - // is about to replace the old cache slot. But if we - // release the lockfile and then try to open the new cache - // slot, we might get a race condition with a concurrent - // writer for the same cache slot (with a different key). - // Lets avoid such a race by just printing the content of - // the lock file. - slot->cache_fd = slot->lock_fd; - unlock_slot(slot, 1); - if ((err = print_slot(slot)) != 0) { - cache_log("[cgit] error printing cache %s: %s (%d)\n", - slot->cache_name, - strerror(err), - err); - } - close_slot(slot); - return err; + return 1; } -/* Print cached content to stdout, generate the content if necessary. */ -int cache_process(int size, const char *path, const char *key, int ttl, - cache_fill_fn fn) +int cache_lock(struct cacheitem *item) { - unsigned long hash; - int i; - struct strbuf filename = STRBUF_INIT; - struct strbuf lockname = STRBUF_INIT; - struct cache_slot slot; - int result; + int ret; + char *lockfile = fmt("%s.lock", item->name); - /* If the cache is disabled, just generate the content */ - if (size <= 0 || ttl == 0) { - fn(); + top: + item->fd = open(lockfile, O_WRONLY | O_CREAT | O_EXCL, S_IRUSR|S_IWUSR); + if (item->fd == NOLOCK && errno == ENOENT && cache_create_dirs()) + goto top; + if (item->fd == NOLOCK && errno == EEXIST) { + struct stat st; + time_t t; + if (stat(lockfile, &st)) + return ret; + t = time(NULL); + if (t-st.st_mtime > cgit_cache_max_create_time && + !unlink(lockfile)) + goto top; return 0; } - - /* Verify input, calculate filenames */ - if (!path) { - cache_log("[cgit] Cache path not specified, caching is disabled\n"); - fn(); - return 0; - } - if (!key) - key = ""; - hash = hash_str(key) % size; - strbuf_addstr(&filename, path); - strbuf_ensure_end(&filename, '/'); - for (i = 0; i < 8; i++) { - strbuf_addf(&filename, "%x", (unsigned char)(hash & 0xf)); - hash >>= 4; - } - strbuf_addbuf(&lockname, &filename); - strbuf_addstr(&lockname, ".lock"); - slot.fn = fn; - slot.ttl = ttl; - slot.stdout_fd = -1; - slot.cache_name = filename.buf; - slot.lock_name = lockname.buf; - slot.key = key; - slot.keylen = strlen(key); - result = process_slot(&slot); - - strbuf_release(&filename); - strbuf_release(&lockname); - return result; + return (item->fd > 0); } -/* Return a strftime formatted date/time - * NB: the result from this function is to shared memory - */ -static char *sprintftime(const char *format, time_t time) +int cache_unlock(struct cacheitem *item) { - static char buf[64]; - struct tm tm; - - if (!time) - return NULL; - gmtime_r(&time, &tm); - strftime(buf, sizeof(buf)-1, format, &tm); - return buf; + close(item->fd); + return (rename(fmt("%s.lock", item->name), item->name) == 0); } -int cache_ls(const char *path) +int cache_expired(struct cacheitem *item) { - DIR *dir; - struct dirent *ent; - int err = 0; - struct cache_slot slot = { NULL }; - struct strbuf fullname = STRBUF_INIT; - size_t prefixlen; - - if (!path) { - cache_log("[cgit] cache path not specified\n"); - return -1; - } - dir = opendir(path); - if (!dir) { - err = errno; - cache_log("[cgit] unable to open path %s: %s (%d)\n", - path, strerror(err), err); - return err; - } - strbuf_addstr(&fullname, path); - strbuf_ensure_end(&fullname, '/'); - prefixlen = fullname.len; - while ((ent = readdir(dir)) != NULL) { - if (strlen(ent->d_name) != 8) - continue; - strbuf_setlen(&fullname, prefixlen); - strbuf_addstr(&fullname, ent->d_name); - slot.cache_name = fullname.buf; - if ((err = open_slot(&slot)) != 0) { - cache_log("[cgit] unable to open path %s: %s (%d)\n", - fullname.buf, strerror(err), err); - continue; - } - htmlf("%s %s %10"PRIuMAX" %s\n", - fullname.buf, - sprintftime("%Y-%m-%d %H:%M:%S", - slot.cache_st.st_mtime), - (uintmax_t)slot.cache_st.st_size, - slot.buf); - close_slot(&slot); - } - closedir(dir); - strbuf_release(&fullname); - return 0; -} - -/* Print a message to stdout */ -void cache_log(const char *format, ...) -{ - va_list args; - va_start(args, format); - vfprintf(stderr, format, args); - va_end(args); + if (item->ttl < 0) + return 0; + return item->st.st_mtime + item->ttl * 60 < time(NULL); } - |