![[tokkee]](http://tokkee.org/images/avatar.png){kind=avatar}
diff --git a/sha1_file.c b/sha1_file.c
index c30452251943bbb058bc5064aaee4a0033ae22c3..2c870314d525ba0666470d53cf7901a2bac9e3c0 100644 (file)
--- a/sha1_file.c
+++ b/sha1_file.c
#endif
#endif
+#ifdef NO_C99_FORMAT
+#define SZ_FMT "lu"
+#else
+#define SZ_FMT "zu"
+#endif
+
const unsigned char null_sha1[20];
static unsigned int sha1_file_open_flag = O_NOATIME;
close(fd);
return;
}
- map = mmap(NULL, st.st_size, PROT_READ, MAP_PRIVATE, fd, 0);
+ map = xmmap(NULL, st.st_size, PROT_READ, MAP_PRIVATE, fd, 0);
close(fd);
- if (map == MAP_FAILED)
- return;
link_alt_odb_entries(map, map + st.st_size, '\n', relative_base, depth);
return NULL;
}
-static int pack_used_ctr;
-static unsigned long pack_mapped;
+static unsigned int pack_used_ctr;
+static unsigned int pack_mmap_calls;
+static unsigned int peak_pack_open_windows;
+static unsigned int pack_open_windows;
+static size_t peak_pack_mapped;
+static size_t pack_mapped;
struct packed_git *packed_git;
+void pack_report()
+{
+ fprintf(stderr,
+ "pack_report: getpagesize() = %10" SZ_FMT "\n"
+ "pack_report: core.packedGitWindowSize = %10" SZ_FMT "\n"
+ "pack_report: core.packedGitLimit = %10" SZ_FMT "\n",
+ (size_t) getpagesize(),
+ packed_git_window_size,
+ packed_git_limit);
+ fprintf(stderr,
+ "pack_report: pack_used_ctr = %10u\n"
+ "pack_report: pack_mmap_calls = %10u\n"
+ "pack_report: pack_open_windows = %10u / %10u\n"
+ "pack_report: pack_mapped = "
+ "%10" SZ_FMT " / %10" SZ_FMT "\n",
+ pack_used_ctr,
+ pack_mmap_calls,
+ pack_open_windows, peak_pack_open_windows,
+ pack_mapped, peak_pack_mapped);
+}
+
static int check_packed_git_idx(const char *path, unsigned long *idx_size_,
void **idx_map_)
{
void *idx_map;
- unsigned int *index;
+ uint32_t *index;
unsigned long idx_size;
int nr, i;
int fd = open(path, O_RDONLY);
return -1;
}
idx_size = st.st_size;
- idx_map = mmap(NULL, idx_size, PROT_READ, MAP_PRIVATE, fd, 0);
+ idx_map = xmmap(NULL, idx_size, PROT_READ, MAP_PRIVATE, fd, 0);
close(fd);
- if (idx_map == MAP_FAILED)
- return -1;
index = idx_map;
*idx_map_ = idx_map;
/* check index map */
if (idx_size < 4*256 + 20 + 20)
- return error("index file too small");
+ return error("index file %s is too small", path);
+
+ /* a future index format would start with this, as older git
+ * binaries would fail the non-monotonic index check below.
+ * give a nicer warning to the user if we can.
+ */
+ if (index[0] == htonl(PACK_IDX_SIGNATURE))
+ return error("index file %s is a newer version"
+ " and is not supported by this binary"
+ " (try upgrading GIT to a newer version)",
+ path);
+
nr = 0;
for (i = 0; i < 256; i++) {
unsigned int n = ntohl(index[i]);
if (n < nr)
- return error("non-monotonic index");
+ return error("non-monotonic index %s", path);
nr = n;
}
* - 20-byte SHA1 file checksum
*/
if (idx_size != 4*256 + nr * 24 + 20 + 20)
- return error("wrong index file size");
+ return error("wrong index file size in %s", path);
return 0;
}
-static int unuse_one_packed_git(void)
+static void scan_windows(struct packed_git *p,
+ struct packed_git **lru_p,
+ struct pack_window **lru_w,
+ struct pack_window **lru_l)
{
- struct packed_git *p, *lru = NULL;
+ struct pack_window *w, *w_l;
- for (p = packed_git; p; p = p->next) {
- if (!p->windows || p->windows->inuse_cnt)
- continue;
- if (!lru || p->windows->last_used < lru->windows->last_used)
- lru = p;
+ for (w_l = NULL, w = p->windows; w; w = w->next) {
+ if (!w->inuse_cnt) {
+ if (!*lru_w || w->last_used < (*lru_w)->last_used) {
+ *lru_p = p;
+ *lru_w = w;
+ *lru_l = w_l;
+ }
+ }
+ w_l = w;
}
- if (!lru)
- return 0;
- munmap(lru->windows->base, lru->windows->len);
- free(lru->windows);
- lru->windows = NULL;
- close(p->pack_fd);
- p->pack_fd = -1;
- return 1;
+}
+
+static int unuse_one_window(struct packed_git *current)
+{
+ struct packed_git *p, *lru_p = NULL;
+ struct pack_window *lru_w = NULL, *lru_l = NULL;
+
+ if (current)
+ scan_windows(current, &lru_p, &lru_w, &lru_l);
+ for (p = packed_git; p; p = p->next)
+ scan_windows(p, &lru_p, &lru_w, &lru_l);
+ if (lru_p) {
+ munmap(lru_w->base, lru_w->len);
+ pack_mapped -= lru_w->len;
+ if (lru_l)
+ lru_l->next = lru_w->next;
+ else {
+ lru_p->windows = lru_w->next;
+ if (!lru_p->windows && lru_p != current) {
+ close(lru_p->pack_fd);
+ lru_p->pack_fd = -1;
+ }
+ }
+ free(lru_w);
+ pack_open_windows--;
+ return 1;
+ }
+ return 0;
+}
+
+void release_pack_memory(size_t need)
+{
+ size_t cur = pack_mapped;
+ while (need >= (cur - pack_mapped) && unuse_one_window(NULL))
+ ; /* nothing */
}
void unuse_pack(struct pack_window **w_cursor)
}
}
-static void open_packed_git(struct packed_git *p)
+/*
+ * Do not call this directly as this leaks p->pack_fd on error return;
+ * call open_packed_git() instead.
+ */
+static int open_packed_git_1(struct packed_git *p)
{
struct stat st;
struct pack_header hdr;
unsigned char sha1[20];
unsigned char *idx_sha1;
+ long fd_flag;
p->pack_fd = open(p->pack_name, O_RDONLY);
if (p->pack_fd < 0 || fstat(p->pack_fd, &st))
- die("packfile %s cannot be opened", p->pack_name);
+ return -1;
/* If we created the struct before we had the pack we lack size. */
if (!p->pack_size) {
if (!S_ISREG(st.st_mode))
- die("packfile %s not a regular file", p->pack_name);
+ return error("packfile %s not a regular file", p->pack_name);
p->pack_size = st.st_size;
} else if (p->pack_size != st.st_size)
- die("packfile %s size changed", p->pack_name);
+ return error("packfile %s size changed", p->pack_name);
+
+ /* We leave these file descriptors open with sliding mmap;
+ * there is no point keeping them open across exec(), though.
+ */
+ fd_flag = fcntl(p->pack_fd, F_GETFD, 0);
+ if (fd_flag < 0)
+ return error("cannot determine file descriptor flags");
+ fd_flag |= FD_CLOEXEC;
+ if (fcntl(p->pack_fd, F_SETFD, fd_flag) == -1)
+ return error("cannot set FD_CLOEXEC");
/* Verify we recognize this pack file format. */
- read_or_die(p->pack_fd, &hdr, sizeof(hdr));
+ if (read_in_full(p->pack_fd, &hdr, sizeof(hdr)) != sizeof(hdr))
+ return error("file %s is far too short to be a packfile", p->pack_name);
if (hdr.hdr_signature != htonl(PACK_SIGNATURE))
- die("file %s is not a GIT packfile", p->pack_name);
+ return error("file %s is not a GIT packfile", p->pack_name);
if (!pack_version_ok(hdr.hdr_version))
- die("packfile %s is version %u and not supported"
+ return error("packfile %s is version %u and not supported"
" (try upgrading GIT to a newer version)",
p->pack_name, ntohl(hdr.hdr_version));
/* Verify the pack matches its index. */
if (num_packed_objects(p) != ntohl(hdr.hdr_entries))
- die("packfile %s claims to have %u objects"
+ return error("packfile %s claims to have %u objects"
" while index size indicates %u objects",
p->pack_name, ntohl(hdr.hdr_entries),
num_packed_objects(p));
if (lseek(p->pack_fd, p->pack_size - sizeof(sha1), SEEK_SET) == -1)
- die("end of packfile %s is unavailable", p->pack_name);
- read_or_die(p->pack_fd, sha1, sizeof(sha1));
+ return error("end of packfile %s is unavailable", p->pack_name);
+ if (read_in_full(p->pack_fd, sha1, sizeof(sha1)) != sizeof(sha1))
+ return error("packfile %s signature is unavailable", p->pack_name);
idx_sha1 = ((unsigned char *)p->index_base) + p->index_size - 40;
if (hashcmp(sha1, idx_sha1))
- die("packfile %s does not match index", p->pack_name);
+ return error("packfile %s does not match index", p->pack_name);
+ return 0;
+}
+
+static int open_packed_git(struct packed_git *p)
+{
+ if (!open_packed_git_1(p))
+ return 0;
+ if (p->pack_fd != -1) {
+ close(p->pack_fd);
+ p->pack_fd = -1;
+ }
+ return -1;
+}
+
+static int in_window(struct pack_window *win, unsigned long offset)
+{
+ /* We must promise at least 20 bytes (one hash) after the
+ * offset is available from this window, otherwise the offset
+ * is not actually in this window and a different window (which
+ * has that one hash excess) must be used. This is to support
+ * the object header and delta base parsing routines below.
+ */
+ off_t win_off = win->offset;
+ return win_off <= offset
+ && (offset + 20) <= (win_off + win->len);
}
unsigned char* use_pack(struct packed_git *p,
unsigned long offset,
unsigned int *left)
{
- struct pack_window *win = p->windows;
-
- if (p->pack_fd == -1)
- open_packed_git(p);
- if (!win) {
- pack_mapped += p->pack_size;
- while (packed_git_limit < pack_mapped && unuse_one_packed_git())
- ; /* nothing */
- win = xcalloc(1, sizeof(*win));
- win->len = p->pack_size;
- win->base = mmap(NULL, p->pack_size, PROT_READ, MAP_PRIVATE, p->pack_fd, 0);
- if (win->base == MAP_FAILED)
- die("packfile %s cannot be mapped.", p->pack_name);
- p->windows = win;
+ struct pack_window *win = *w_cursor;
+
+ if (p->pack_fd == -1 && open_packed_git(p))
+ die("packfile %s cannot be accessed", p->pack_name);
+
+ /* Since packfiles end in a hash of their content and its
+ * pointless to ask for an offset into the middle of that
+ * hash, and the in_window function above wouldn't match
+ * don't allow an offset too close to the end of the file.
+ */
+ if (offset > (p->pack_size - 20))
+ die("offset beyond end of packfile (truncated pack?)");
+
+ if (!win || !in_window(win, offset)) {
+ if (win)
+ win->inuse_cnt--;
+ for (win = p->windows; win; win = win->next) {
+ if (in_window(win, offset))
+ break;
+ }
+ if (!win) {
+ size_t window_align = packed_git_window_size / 2;
+ win = xcalloc(1, sizeof(*win));
+ win->offset = (offset / window_align) * window_align;
+ win->len = p->pack_size - win->offset;
+ if (win->len > packed_git_window_size)
+ win->len = packed_git_window_size;
+ pack_mapped += win->len;
+ while (packed_git_limit < pack_mapped
+ && unuse_one_window(p))
+ ; /* nothing */
+ win->base = xmmap(NULL, win->len,
+ PROT_READ, MAP_PRIVATE,
+ p->pack_fd, win->offset);
+ if (win->base == MAP_FAILED)
+ die("packfile %s cannot be mapped: %s",
+ p->pack_name,
+ strerror(errno));
+ pack_mmap_calls++;
+ pack_open_windows++;
+ if (pack_mapped > peak_pack_mapped)
+ peak_pack_mapped = pack_mapped;
+ if (pack_open_windows > peak_pack_open_windows)
+ peak_pack_open_windows = pack_open_windows;
+ win->next = p->windows;
+ p->windows = win;
+ }
}
if (win != *w_cursor) {
win->last_used = pack_used_ctr++;
win->inuse_cnt++;
*w_cursor = win;
}
+ offset -= win->offset;
if (left)
*left = win->len - offset;
return win->base + offset;
if (!has_extension(de->d_name, ".idx"))
continue;
- /* we have .idx. Is it a file we can map? */
+ /* Don't reopen a pack we already have. */
strcpy(path + len, de->d_name);
for (p = packed_git; p; p = p->next) {
if (!memcmp(path, p->pack_name, len + namelen - 4))
}
if (p)
continue;
+ /* See if it really is a valid .idx file with corresponding
+ * .pack file that we can map.
+ */
p = add_packed_git(path, len + namelen, local);
if (!p)
continue;
- p->next = packed_git;
- packed_git = p;
+ install_packed_git(p);
}
closedir(dir);
}
*/
sha1_file_open_flag = 0;
}
- map = mmap(NULL, st.st_size, PROT_READ, MAP_PRIVATE, fd, 0);
+ map = xmmap(NULL, st.st_size, PROT_READ, MAP_PRIVATE, fd, 0);
close(fd);
- if (map == MAP_FAILED)
- return NULL;
*size = st.st_size;
return map;
}
unsigned char *base_info = use_pack(p, w_curs, offset, NULL);
unsigned long base_offset;
- /* there must be at least 20 bytes left regardless of delta type */
- if (p->pack_size <= offset + 20)
- die("truncated pack file");
-
+ /* use_pack() assured us we have [base_info, base_info + 20)
+ * as a range that we can look at without walking off the
+ * end of the mapped window. Its actually the hash size
+ * that is assured. An OFS_DELTA longer than the hash size
+ * is stupid, as then a REF_DELTA would be smaller to store.
+ */
if (kind == OBJ_OFS_DELTA) {
unsigned used = 0;
unsigned char c = base_info[used++];
unsigned int left;
unsigned long used;
+ /* use_pack() assures us we have [base, base + 20) available
+ * as a range that we can look at at. (Its actually the hash
+ * size that is assured.) With our object header encoding
+ * the maximum deflated object size is 2^137, which is just
+ * insane, so we know won't exceed what we have been given.
+ */
base = use_pack(p, w_curs, offset, &left);
used = unpack_object_header_gently(base, left, type, sizep);
if (!used)
unsigned long find_pack_entry_one(const unsigned char *sha1,
struct packed_git *p)
{
- unsigned int *level1_ofs = p->index_base;
+ uint32_t *level1_ofs = p->index_base;
int hi = ntohl(level1_ofs[*sha1]);
int lo = ((*sha1 == 0x0) ? 0 : ntohl(level1_ofs[*sha1 - 1]));
void *index = p->index_base + 256;
int mi = (lo + hi) / 2;
int cmp = hashcmp((unsigned char *)index + (24 * mi) + 4, sha1);
if (!cmp)
- return ntohl(*((unsigned int *) ((char *) index + (24 * mi))));
+ return ntohl(*((uint32_t *)((char *)index + (24 * mi))));
if (cmp > 0)
hi = mi;
else
@@ -1265,6 +1421,18 @@ static int find_pack_entry(const unsigned char *sha1, struct pack_entry *e, cons
}
offset = find_pack_entry_one(sha1, p);
if (offset) {
+ /*
+ * We are about to tell the caller where they can
+ * locate the requested object. We better make
+ * sure the packfile is still here and can be
+ * accessed before supplying that answer, as
+ * it may have been deleted since the index
+ * was loaded!
+ */
+ if (p->pack_fd == -1 && open_packed_git(p)) {
+ error("packfile %s cannot be accessed", p->pack_name);
+ continue;
+ }
e->offset = offset;
e->p = p;
hashcpy(e->sha1, sha1);
@@ -1329,21 +1497,79 @@ static void *read_packed_sha1(const unsigned char *sha1, char *type, unsigned lo
{
struct pack_entry e;
- if (!find_pack_entry(sha1, &e, NULL)) {
- error("cannot read sha1_file for %s", sha1_to_hex(sha1));
+ if (!find_pack_entry(sha1, &e, NULL))
return NULL;
+ else
+ return unpack_entry(e.p, e.offset, type, size);
+}
+
+/*
+ * This is meant to hold a *small* number of objects that you would
+ * want read_sha1_file() to be able to return, but yet you do not want
+ * to write them into the object store (e.g. a browse-only
+ * application).
+ */
+static struct cached_object {
+ unsigned char sha1[20];
+ const char *type;
+ void *buf;
+ unsigned long size;
+} *cached_objects;
+static int cached_object_nr, cached_object_alloc;
+
+static struct cached_object *find_cached_object(const unsigned char *sha1)
+{
+ int i;
+ struct cached_object *co = cached_objects;
+
+ for (i = 0; i < cached_object_nr; i++, co++) {
+ if (!hashcmp(co->sha1, sha1))
+ return co;
+ }
+ return NULL;
+}
+
+int pretend_sha1_file(void *buf, unsigned long len, const char *type, unsigned char *sha1)
+{
+ struct cached_object *co;
+
+ hash_sha1_file(buf, len, type, sha1);
+ if (has_sha1_file(sha1) || find_cached_object(sha1))
+ return 0;
+ if (cached_object_alloc <= cached_object_nr) {
+ cached_object_alloc = alloc_nr(cached_object_alloc);
+ cached_objects = xrealloc(cached_objects,
+ sizeof(*cached_objects) *
+ cached_object_alloc);
}
- return unpack_entry(e.p, e.offset, type, size);
+ co = &cached_objects[cached_object_nr++];
+ co->size = len;
+ co->type = strdup(type);
+ co->buf = xmalloc(len);
+ memcpy(co->buf, buf, len);
+ hashcpy(co->sha1, sha1);
+ return 0;
}
-void * read_sha1_file(const unsigned char *sha1, char *type, unsigned long *size)
+void *read_sha1_file(const unsigned char *sha1, char *type, unsigned long *size)
{
unsigned long mapsize;
void *map, *buf;
- struct pack_entry e;
+ struct cached_object *co;
+
+ co = find_cached_object(sha1);
+ if (co) {
+ buf = xmalloc(co->size + 1);
+ memcpy(buf, co->buf, co->size);
+ ((char*)buf)[co->size] = 0;
+ strcpy(type, co->type);
+ *size = co->size;
+ return buf;
+ }
- if (find_pack_entry(sha1, &e, NULL))
- return read_packed_sha1(sha1, type, size);
+ buf = read_packed_sha1(sha1, type, size);
+ if (buf)
+ return buf;
map = map_sha1_file(sha1, &mapsize);
if (map) {
buf = unpack_sha1_file(map, mapsize, type, size);
@@ -1351,9 +1577,7 @@ void * read_sha1_file(const unsigned char *sha1, char *type, unsigned long *size
return buf;
}
reprepare_packed_git();
- if (find_pack_entry(sha1, &e, NULL))
- return read_packed_sha1(sha1, type, size);
- return NULL;
+ return read_packed_sha1(sha1, type, size);
}
void *read_object_with_reference(const unsigned char *sha1,
static int write_buffer(int fd, const void *buf, size_t len)
{
- while (len) {
- ssize_t size;
-
- size = write(fd, buf, len);
- if (!size)
- return error("file write: disk full");
- if (size < 0) {
- if (errno == EINTR || errno == EAGAIN)
- continue;
- return error("file write error (%s)", strerror(errno));
- }
- len -= size;
- buf = (char *) buf + size;
- }
+ if (write_in_full(fd, buf, len) < 0)
+ return error("file write error (%s)", strerror(errno));
return 0;
}
/* need to unpack and recompress it by itself */
unpacked = read_packed_sha1(sha1, type, &len);
+ if (!unpacked)
+ error("cannot read sha1_file for %s", sha1_to_hex(sha1));
hdrlen = sprintf(hdr, "%s %lu", type, len) + 1;
if (ret != Z_OK)
break;
}
- size = read(fd, buffer + *bufposn, bufsize - *bufposn);
+ size = xread(fd, buffer + *bufposn, bufsize - *bufposn);
if (size <= 0) {
close(local);
unlink(tmpfile);
@@ -1872,13 +2086,12 @@ int index_fd(unsigned char *sha1, int fd, struct stat *st, int write_object, con
buf = "";
if (size)
- buf = mmap(NULL, size, PROT_READ, MAP_PRIVATE, fd, 0);
+ buf = xmmap(NULL, size, PROT_READ, MAP_PRIVATE, fd, 0);
close(fd);
- if (buf == MAP_FAILED)
- return -1;
if (!type)
type = blob_type;
+ /* FIXME: CRLF -> LF conversion here for blobs! We'll need the path! */
if (write_object)
ret = write_sha1_file(buf, size, type, sha1);
else
@@ -1923,3 +2136,24 @@ int index_path(unsigned char *sha1, const char *path, struct stat *st, int write
}
return 0;
}
+
+int read_pack_header(int fd, struct pack_header *header)
+{
+ char *c = (char*)header;
+ ssize_t remaining = sizeof(struct pack_header);
+ do {
+ ssize_t r = xread(fd, c, remaining);
+ if (r <= 0)
+ /* "eof before pack header was fully read" */
+ return PH_ERROR_EOF;
+ remaining -= r;
+ c += r;
+ } while (remaining > 0);
+ if (header->hdr_signature != htonl(PACK_SIGNATURE))
+ /* "protocol error (pack signature mismatch detected)" */
+ return PH_ERROR_PACK_SIGNATURE;
+ if (!pack_version_ok(header->hdr_version))
+ /* "protocol error (pack version unsupported)" */
+ return PH_ERROR_PROTOCOL;
+ return 0;
+}