Avoid using file names and file paths for compatibility with
out-of-tree builds.
Note make and ccache don't track blacklist dependencies,
add workarounds.
Document use of coverity, clang static analyzer, and clang dynamic
undefined behavior and address sanitizers in doc/HACKING.
Add clang dynamic sanitizer blacklist in
contrib/clang/sanitizer_blacklist.txt to exempt known undefined
behavior. Include detailed usage instructions in this blacklist file.
Patch by "teor".
There were following problems:
- configure.ac wrongly checked for defined HAVE_SYSTEMD; this
wasn't working, so the watchdog code was not compiled in.
Replace library search with explicit version check
- sd_notify() watchdog call was unsetting NOTIFY_SOCKET from env;
this means only first "watchdog ping" was delivered, each
subsequent one did not have socket to be sent to and systemd
was killing service
- after those fixes, enable Watchdog in systemd unit with one
minute intervals
If running under systemd, notify the supervisor about current PID
of Tor daemon. This makes systemd unit simpler and more robust:
it will do the right thing regardless of RunAsDaemon settings.
The rest of the filesystem is accessible for reading only. Still, quoting
systemd.exec(5):
Note that restricting access with these options does not extend to submounts
of a directory that are created later on.
Our current systemd unit uses "Type = simple", so systemd does not expect tor to
fork. If the user has "RunAsDaemon 1" in their torrc, then things won't work as
expected. This is e.g. the case on Debian (and derivatives), since there we pass
"--defaults-torrc /usr/share/tor/tor-service-defaults-torrc" (that contains
"RunAsDaemon 1") by default.
The only solution I could find is to explicitly pass "--RunAsDaemon 0" when
starting tor from the systemd unit file, which this commit does.
We've accumulated a lot of cruft in this directory over the years: so
much, that it passed the point of being so disorganized that we no
longer browsed through it to see how bad it had gotten.
This patch (based on changes by rl1987) tries to remove the most
useless items, and split the others into reasonable directories. It
creates a new scripts/ directory for maint and test scripts.
This patch was generated with the script below. No other changes are made in
this patch.
#############
# new directories
mkdir -p contrib/test-tools
mkdir -p contrib/or-tools
mkdir -p contrib/dirauth-tools
mkdir -p contrib/operator-tools
mkdir -p contrib/client-tools
mkdir -p contrib/test-tools
mkdir -p contrib/dist
mkdir -p contrib/dist/suse
mkdir -p contrib/win32build
mkdir -p scripts/maint
mkdir -p scripts/test
############
# Deleted -- nobody who wants this is going to be looking for it here any
# longer. Also, nobody wants it.
git rm contrib/auto-naming/README
# Deleted: We no longer do polipo.
git rm contrib/polipo/Makefile.mingw
git rm contrib/polipo/README
git rm contrib/polipo/polipo-mingw.nsi
# We haven't even tried to run this for ages. It is a relic of a bygone era
git rm contrib/mdd.py
# contrib/dir-tools/directory-archive/
# Tools for running a directory archive. No longer used - deleting them.
git rm contrib/directory-archive/crontab.sample
git rm contrib/directory-archive/fetch-all
git rm contrib/directory-archive/fetch-all-v3
git rm contrib/directory-archive/tar-them-up
git rm contrib/directory-archive/fetch-all-functions
git rm contrib/directory-archive/sort-into-month-folder
# This appears to be related to very old windows packaging stuff.
git rm contrib/bundle.nsi
git rm contrib/package_nsis-weasel.sh
git rm contrib/package_nsis.sh
git rm contrib/netinst.nsi
git rm contrib/torinst32.ico
git rm contrib/xenobite.ico
# This should not be needed for cross-compilation any more, should it?
git rm contrib/cross.sh
# I don't think anyone ever used this.
git rm contrib/make-signature.sh
# These are attempts to send tor controller commands from the command-line.
# They don't support modern authentication.
git rm contrib/tor-ctrl.sh
# this is for fetching about a tor server from a dirauth. But it
# doesn't authenticate the dirauth: yuck.
git rm contrib/sd
# wow, such unused, very perl4.
git rm contrib/tor-stress
####### contrib/dirauth-tools/
# Tools for running a directory authority
git mv contrib/add-tor contrib/dirauth-tools/
git mv contrib/nagios-check-tor-authority-cert contrib/dirauth-tools/
#######
# contrib/or-tools/
# Tools for examining relays
git mv contrib/check-tor contrib/or-tools/check-tor
git mv contrib/checksocks.pl contrib/or-tools/checksocks.pl
git mv contrib/exitlist contrib/or-tools/exitlist
#######
# contrib/operator-tools
# Tools for running a relay.
git mv contrib/linux-tor-prio.sh contrib/operator-tools/linux-tor-prio.sh
git mv contrib/tor-exit-notice.html contrib/operator-tools/tor-exit-notice.html
git mv contrib/tor.logrotate.in contrib/operator-tools/
######
# contrib/dist
git mv contrib/rc.subr contrib/dist/
git mv contrib/tor.sh.in contrib/dist/
git mv contrib/torctl.in contrib/dist/
git mv contrib/suse/* contrib/dist/suse/
######
# client-tools
git mv contrib/torify contrib/client-tools/torify
git mv contrib/tor-resolve.py contrib/client-tools/
######
# win32build
git mv contrib/package_nsis-mingw.sh contrib/win32build/
git mv contrib/tor.nsi.in contrib/win32build/
# Erinn didn't ask for this...
git mv contrib/tor-mingw.nsi.in contrib/win32build/
git mv contrib/tor.ico contrib/win32build/
######
# scripts/test
git mv contrib/cov-blame scripts/test/cov-blame
git mv contrib/cov-diff scripts/test/cov-diff
git mv contrib/coverage scripts/test/coverage
git mv contrib/scan-build.sh scripts/test/
######## scripts/maint
# Maintainance scripts
#
# These are scripts for developers to use when hacking on Tor. They mostly
# look at the Tor source in one way or another.
git mv contrib/findMergedChanges.pl scripts/maint/findMergedChanges.pl
git mv contrib/checkOptionDocs.pl scripts/maint/checkOptionDocs.pl
git mv contrib/checkSpace.pl scripts/maint/checkSpace.pl
git mv contrib/redox.py scripts/maint/redox.py
git mv contrib/updateVersions.pl scripts/maint/updateVersions.pl
git mv contrib/checkLogs.pl scripts/maint/checkLogs.pl
git mv contrib/format_changelog.py scripts/maint/
We can't just use "diff", since we don't care about changes in line
numbers, or changes in the exact number of times a line was called.
We just care about changes that make lines covered or non-coverd. So
pre-process the files before calling diff.
If you pass the --enable-coverage flag on the command line, we build
our testing binaries with appropriate options eo enable coverage
testing. We also build a "tor-cov" binary that has coverage enabled,
for integration tests.
On recent OSX versions, test coverage only works with clang, not gcc.
So we warn about that.
Also add a contrib/coverage script to actually run gcov with the
appropriate options to generate useful .gcov files. (Thanks to
automake, the .o files will not have the names that gcov expects to
find.)
Also, remove generated gcda and gcno files on clean.
Now the manpages no longer refer to tsocks or tsocks.conf, and we no
longer have or ship a tor-tsocks.conf. The only remaining instances
of "tsocks" in our repository are old ChangeLog and ReleaseNotes
entries, and the torify script saying that it doesn't support tsocks.
Fixes bug 8290.
This gives us a few benefits:
1) make -j clean all
this will start working, as it should. It currently doesn't.
2) increased parallel build
recursive make will max out at number of files in a directory,
non-recursive make doesn't have such a limitation
3) Removal of duplicate information in make files,
less error prone
I've also slightly updated how we call AM_INIT_AUTOMAKE, as the way
that was used was not only deprecated but will be *removed* in the next
major automake release (1.13).... so probably best that we can continue
to bulid tor without requiring old automake.
(see http://www.gnu.org/software/automake/manual/html_node/Public-Macros.html )
For more reasons why, see resources such as:
http://miller.emu.id.au/pmiller/books/rmch/
We now catch bare {s that should be on the previous line with a do,
while, if, or for, and elses that should share a line with their
preceding }.
That is,
if (foo)
{
and
if (foo) {
...
}
else
are now detected.
We should think about maybe making Tor uncrustify-clean some day,
but configuring uncrustify is an exercise in bizarreness, and
reformatting huge gobs of Tor is always painful.
Somewhere along the line, doxygen and tor changed their behavior a
little. The script is still a dreadful kludge, but now at least it
sorta works again.
The timercmp macro uses triggers a "space between function name and
opening parentheses" warning for the check spaces script. Work around
this by simply disabling the check for all "functions" named 'op()'.
Otherwise I need to figure out what it does and how to make it do it
every time I want to use it. It did that unixy thing where running
it with no arguments printed nothing and returned.
Having very long single lines with lots and lots of things in them
tends to make files hard to diff and hard to merge. Since our tools
are one-line-at-a-time, we should try to construct lists that way too,
within reason.
This incidentally turned up a few headers in configure.in that we were
for some reason searching for twice.
We decided to no longer ship expert packages for OS X because they're a
lot of trouble to keep maintained and confuse users. For those who want
a tor on OS X without Vidalia, macports is a fine option. Alternatively,
building from source is easy, too.
The polipo stuff that is still required for the Vidalia bundle build can
now be found in the torbrowser repository,
git://git.torproject.org/torbrowser.git.
