https://github.com/git/git
Revision 0383bbb9015898cbc79abd7b64316484d7713b44 authored by Jeff King on 30 April 2018, 07:25:25 UTC, committed by Jeff King on 22 May 2018, 03:50:11 UTC
Submodule "names" come from the untrusted .gitmodules file,
but we blindly append them to $GIT_DIR/modules to create our
on-disk repo paths. This means you can do bad things by
putting "../" into the name (among other things).

Let's sanity-check these names to avoid building a path that
can be exploited. There are two main decisions:

  1. What should the allowed syntax be?

     It's tempting to reuse verify_path(), since submodule
     names typically come from in-repo paths. But there are
     two reasons not to:

       a. It's technically more strict than what we need, as
          we really care only about breaking out of the
          $GIT_DIR/modules/ hierarchy.  E.g., having a
          submodule named "foo/.git" isn't actually
          dangerous, and it's possible that somebody has
          manually given such a funny name.

       b. Since we'll eventually use this checking logic in
          fsck to prevent downstream repositories, it should
          be consistent across platforms. Because
          verify_path() relies on is_dir_sep(), it wouldn't
          block "foo\..\bar" on a non-Windows machine.

  2. Where should we enforce it? These days most of the
     .gitmodules reads go through submodule-config.c, so
     I've put it there in the reading step. That should
     cover all of the C code.

     We also construct the name for "git submodule add"
     inside the git-submodule.sh script. This is probably
     not a big deal for security since the name is coming
     from the user anyway, but it would be polite to remind
     them if the name they pick is invalid (and we need to
     expose the name-checker to the shell anyway for our
     test scripts).

     This patch issues a warning when reading .gitmodules
     and just ignores the related config entry completely.
     This will generally end up producing a sensible error,
     as it works the same as a .gitmodules file which is
     missing a submodule entry (so "submodule update" will
     barf, but "git clone --recurse-submodules" will print
     an error but not abort the clone.

     There is one minor oddity, which is that we print the
     warning once per malformed config key (since that's how
     the config subsystem gives us the entries). So in the
     new test, for example, the user would see three
     warnings. That's OK, since the intent is that this case
     should never come up outside of malicious repositories
     (and then it might even benefit the user to see the
     message multiple times).

Credit for finding this vulnerability and the proof of
concept from which the test script was adapted goes to
Etienne Stalmans.

Signed-off-by: Jeff King <peff@peff.net>
1 parent 42e6fde
Raw File
Tip revision: 0383bbb9015898cbc79abd7b64316484d7713b44 authored by Jeff King on 30 April 2018, 07:25:25 UTC
submodule-config: verify submodule names as paths
Tip revision: 0383bbb
lockfile.c
/*
 * Copyright (c) 2005, Junio C Hamano
 */

#include "cache.h"
#include "lockfile.h"

/*
 * path = absolute or relative path name
 *
 * Remove the last path name element from path (leaving the preceding
 * "/", if any).  If path is empty or the root directory ("/"), set
 * path to the empty string.
 */
static void trim_last_path_component(struct strbuf *path)
{
	int i = path->len;

	/* back up past trailing slashes, if any */
	while (i && path->buf[i - 1] == '/')
		i--;

	/*
	 * then go backwards until a slash, or the beginning of the
	 * string
	 */
	while (i && path->buf[i - 1] != '/')
		i--;

	strbuf_setlen(path, i);
}


/* We allow "recursive" symbolic links. Only within reason, though */
#define MAXDEPTH 5

/*
 * path contains a path that might be a symlink.
 *
 * If path is a symlink, attempt to overwrite it with a path to the
 * real file or directory (which may or may not exist), following a
 * chain of symlinks if necessary.  Otherwise, leave path unmodified.
 *
 * This is a best-effort routine.  If an error occurs, path will
 * either be left unmodified or will name a different symlink in a
 * symlink chain that started with the original path.
 */
static void resolve_symlink(struct strbuf *path)
{
	int depth = MAXDEPTH;
	static struct strbuf link = STRBUF_INIT;

	while (depth--) {
		if (strbuf_readlink(&link, path->buf, path->len) < 0)
			break;

		if (is_absolute_path(link.buf))
			/* absolute path simply replaces p */
			strbuf_reset(path);
		else
			/*
			 * link is a relative path, so replace the
			 * last element of p with it.
			 */
			trim_last_path_component(path);

		strbuf_addbuf(path, &link);
	}
	strbuf_reset(&link);
}

/* Make sure errno contains a meaningful value on error */
static int lock_file(struct lock_file *lk, const char *path, int flags)
{
	int fd;
	struct strbuf filename = STRBUF_INIT;

	strbuf_addstr(&filename, path);
	if (!(flags & LOCK_NO_DEREF))
		resolve_symlink(&filename);

	strbuf_addstr(&filename, LOCK_SUFFIX);
	fd = create_tempfile(&lk->tempfile, filename.buf);
	strbuf_release(&filename);
	return fd;
}

/*
 * Constants defining the gaps between attempts to lock a file. The
 * first backoff period is approximately INITIAL_BACKOFF_MS
 * milliseconds. The longest backoff period is approximately
 * (BACKOFF_MAX_MULTIPLIER * INITIAL_BACKOFF_MS) milliseconds.
 */
#define INITIAL_BACKOFF_MS 1L
#define BACKOFF_MAX_MULTIPLIER 1000

/*
 * Try locking path, retrying with quadratic backoff for at least
 * timeout_ms milliseconds. If timeout_ms is 0, try locking the file
 * exactly once. If timeout_ms is -1, try indefinitely.
 */
static int lock_file_timeout(struct lock_file *lk, const char *path,
			     int flags, long timeout_ms)
{
	int n = 1;
	int multiplier = 1;
	long remaining_ms = 0;
	static int random_initialized = 0;

	if (timeout_ms == 0)
		return lock_file(lk, path, flags);

	if (!random_initialized) {
		srand((unsigned int)getpid());
		random_initialized = 1;
	}

	if (timeout_ms > 0)
		remaining_ms = timeout_ms;

	while (1) {
		long backoff_ms, wait_ms;
		int fd;

		fd = lock_file(lk, path, flags);

		if (fd >= 0)
			return fd; /* success */
		else if (errno != EEXIST)
			return -1; /* failure other than lock held */
		else if (timeout_ms > 0 && remaining_ms <= 0)
			return -1; /* failure due to timeout */

		backoff_ms = multiplier * INITIAL_BACKOFF_MS;
		/* back off for between 0.75*backoff_ms and 1.25*backoff_ms */
		wait_ms = (750 + rand() % 500) * backoff_ms / 1000;
		sleep_millisec(wait_ms);
		remaining_ms -= wait_ms;

		/* Recursion: (n+1)^2 = n^2 + 2n + 1 */
		multiplier += 2*n + 1;
		if (multiplier > BACKOFF_MAX_MULTIPLIER)
			multiplier = BACKOFF_MAX_MULTIPLIER;
		else
			n++;
	}
}

void unable_to_lock_message(const char *path, int err, struct strbuf *buf)
{
	if (err == EEXIST) {
		strbuf_addf(buf, _("Unable to create '%s.lock': %s.\n\n"
		    "Another git process seems to be running in this repository, e.g.\n"
		    "an editor opened by 'git commit'. Please make sure all processes\n"
		    "are terminated then try again. If it still fails, a git process\n"
		    "may have crashed in this repository earlier:\n"
		    "remove the file manually to continue."),
			    absolute_path(path), strerror(err));
	} else
		strbuf_addf(buf, _("Unable to create '%s.lock': %s"),
			    absolute_path(path), strerror(err));
}

NORETURN void unable_to_lock_die(const char *path, int err)
{
	struct strbuf buf = STRBUF_INIT;

	unable_to_lock_message(path, err, &buf);
	die("%s", buf.buf);
}

/* This should return a meaningful errno on failure */
int hold_lock_file_for_update_timeout(struct lock_file *lk, const char *path,
				      int flags, long timeout_ms)
{
	int fd = lock_file_timeout(lk, path, flags, timeout_ms);
	if (fd < 0) {
		if (flags & LOCK_DIE_ON_ERROR)
			unable_to_lock_die(path, errno);
		if (flags & LOCK_REPORT_ON_ERROR) {
			struct strbuf buf = STRBUF_INIT;
			unable_to_lock_message(path, errno, &buf);
			error("%s", buf.buf);
			strbuf_release(&buf);
		}
	}
	return fd;
}

char *get_locked_file_path(struct lock_file *lk)
{
	struct strbuf ret = STRBUF_INIT;

	strbuf_addstr(&ret, get_tempfile_path(&lk->tempfile));
	if (ret.len <= LOCK_SUFFIX_LEN ||
	    strcmp(ret.buf + ret.len - LOCK_SUFFIX_LEN, LOCK_SUFFIX))
		die("BUG: get_locked_file_path() called for malformed lock object");
	/* remove ".lock": */
	strbuf_setlen(&ret, ret.len - LOCK_SUFFIX_LEN);
	return strbuf_detach(&ret, NULL);
}

int commit_lock_file(struct lock_file *lk)
{
	char *result_path = get_locked_file_path(lk);

	if (commit_lock_file_to(lk, result_path)) {
		int save_errno = errno;
		free(result_path);
		errno = save_errno;
		return -1;
	}
	free(result_path);
	return 0;
}
back to top