June 2015 - tbb-dev - lists.torproject.org

Using topic branches for Tor Browser patches
by Nicolas Vigier 14 Dec '15

14 Dec '15

Hello, Currently on tor-browser.git, each Tor Browser release has a single branch that includes all our patches. This is simple, however it makes it difficult to test each patch individually, and see which patch is responsible for which test failure. For instance, the patch to add preference overrides causes many tests to fail, probably because many tests don't expect the default preferences to change: 31.7.0esr pass all tests: https://treeherder.mozilla.org/#/jobs?repo=try&revision=5bbe47119ac0 31.7.0esr + preference overrides patch fails many tests: https://treeherder.mozilla.org/#/jobs?repo=try&revision=fdc349fbbf5f After discussing about this quickly with intrigeri last week, I have been thinking about a different way to organize the patches: maybe we could use one branch for each topic, and merge all those branches in a single branch when we prepare the release. The naming of the branches could be: - ${firefox_version}/base: the commit for the firefox version we are using. - ${firefox_version}/${topic_name}: A branch containing patch(es) on a specific topic. $topic_name can be a bug number and/or a short description. We can submit those branches to Mozilla Try to see if they break some unit tests. - ${firefox_version}/${tbb_version}/${topic_name}: The same as the previous one, but for a topic that we only want to include in a specific Tor Browser version. - tor-browser-${firefox_version}-${tbb_version}-1: The branch that we use to build Tor Browser. This branch is based on ${firefox_version}/base, and includes merges of all branches matching ${firefox_version}/[^/]* or ${firefox_version}/${tbb_version}/[^/]*. As an exemple, I have converted a few patches from tor-browser-31.7.0esr-4.5-1 to this new naming in this git repository: https://github.com/boklm/gecko-dev/branches/all With all topic branches merged to this tor-browser-31.7.0esr-4.5-1 branch: https://github.com/boklm/gecko-dev/commits/tor-browser-31.7.0esr-4.5-1 I also started submitting some of them to Mozilla Try: 31.7.0esr/2874-Block-Components.interfaces-from-content: https://treeherder.mozilla.org/#/jobs?repo=try&revision=91cd13cf0ac7 (some mochitest tests failing) 31.7.0esr/2950-Make-Permissions-Manager-memory-only: https://treeherder.mozilla.org/#/jobs?repo=try&revision=3f8cb0c1984e (tbb-tests/browser_tor_bug2950.js failing) 31.7.0esr/2949-Make-Intermediate-Cert-Store-memory-only: https://treeherder.mozilla.org/#/jobs?repo=try&revision=139538724d04 (no test failing) 31.7.0esr/3547-Block-all-plugins-except-flash: https://treeherder.mozilla.org/#/jobs?repo=try&revision=c75fb4685994 (some xpcshell, mochitest, mochitest-chrome, reftest failing) 31.7.0esr/3229-Make-content-pref-service-memory-only-clearable: https://treeherder.mozilla.org/#/jobs?repo=try&revision=344873ae1513 (1 xpcshell test failing) 31.7.0esr/TB2-Provide-an-observer-event-to-close-persistent-connections: https://treeherder.mozilla.org/#/jobs?repo=try&revision=93af45d5111a (no test failing) 31.7.0esr/2875-Limit-device-and-system-specific-CSS-Media-Queries: https://treeherder.mozilla.org/#/jobs?repo=try&revision=32244f089736 (still running at the moment) 31.7.0esr/2872-Limit-the-number-of-fonts-per-document: https://treeherder.mozilla.org/#/jobs?repo=try&revision=4abfb45bba86 (still running at the moment) We could also have some helpers scripts to make management of all those branches easier: - a script to merge / list all topic branches that have not yet been merged into their tor-browser-${firefox_version}-${tbb_version}-1 branch. - a script to push / list all branches that need to be pushed for a firefox or tbb version to a remote git repository. - a script to rebase all patches from a previous firefox version to a new firefox version. What do you think about this idea ? Nicolas

1 7

Remaining loose ends for 5.0a3
by Mike Perry 25 Jun '15

25 Jun '15

Hello all, I went on a merge rampage earlier today merging everything that seemed ready for 5.0a3. A few loose ends remain, though. Here's a quick list of the remaining things we need to wrap up somehow, broken down by person. Arthur: * https://trac.torproject.org/projects/tor/ticket/13670 - Probably shouldn't be merged until mcs's concerned are fixed (ideally with the proper URI-like cache key). Either that, or we just let it sit out. * https://trac.torproject.org/projects/tor/ticket/15646 - We should probably go with this patch as-is for now and fix it later, unless Arthur has time to update it tomorrow. My preference is for fixing up #13670 for it to be good enough to go in 5.0a3 though. Pearl Crescent: * https://trac.torproject.org/projects/tor/ticket/16397 - This patch did not apply against FF38. It seemed to only be against the stable? * https://trac.torproject.org/projects/tor/ticket/16300 - Either needs revision or we need to disable the BroadcastChannel API via pref. Either one is fine with me. My preference is on #16397 getting done first. Phoul: * https://trac.torproject.org/projects/tor/ticket/16268 - Blocking on transifex Georg: * https://trac.torproject.org/projects/tor/ticket/16351 - Needs binutils patch? Unclear on what we need here. Georg/Myself: * https://trac.torproject.org/projects/tor/ticket/16005 - Is the patch breaking the game, minimal mode itself, or something else? If disabling webgl.min_capability_mode doesn't make the game work, then either something about our build process or your system is at fault, and not this patch. I say merge it in that case (or ask for secondary review). * Still need to fix up Torbutton somehow.. Should we create a branch from some commit in the history, or are all the merged changes for FF38 OK for FF31 and 4.5.x? - Either way, we'll need to handle the window resizing pref for 4.5.3. Myself: * https://trac.torproject.org/projects/tor/ticket/16403 - Need localization to propagate the new params - How do we handle localization of the stable torbutton? New branch? * Still need to rebase onto the June 30th Firefox release tags. Anything I missed? -- Mike Perry

2 1

Hardening checks exceptions
by Nicolas Vigier 22 Jun '15

22 Jun '15

Hello, In the tests that we are doing on Tor Browser releases to check the hardenning of binary files, I am planning to add exceptions for the following files for the following tests (which are currently failing): RELRO: TorBrowser/Tor/PluggableTransports/meek-client TorBrowser/Tor/PluggableTransports/meek-client-torbrowser TorBrowser/Tor/PluggableTransports/meek-client-torbrowser TorBrowser/Tor/PluggableTransports/obfs4proxy stack_canary: libmozalloc.so libnssckbi.so libplc4.so libplds4.so TorBrowser/Tor/libstdc++.so.6 TorBrowser/Tor/PluggableTransports/Crypto/Cipher/_ARC4.so TorBrowser/Tor/PluggableTransports/Crypto/Cipher/_XOR.so TorBrowser/Tor/PluggableTransports/Crypto/Util/_counter.so TorBrowser/Tor/PluggableTransports/fte/cDFA.so TorBrowser/Tor/PluggableTransports/meek-client-torbrowser TorBrowser/Tor/PluggableTransports/twisted/python/_initgroups.so TorBrowser/Tor/PluggableTransports/twisted/runner/portmap.so TorBrowser/Tor/PluggableTransports/twisted/test/raiser.so TorBrowser/Tor/PluggableTransports/zope/interface/_zope_interface_coptimizations.so PIE: TorBrowser/Tor/PluggableTransports/meek-client TorBrowser/Tor/PluggableTransports/meek-client-torbrowser TorBrowser/Tor/PluggableTransports/obfs4proxy DEP/ASLR (Windows): TorBrowser/Tor/PluggableTransports/_ctypes.pyd TorBrowser/Tor/PluggableTransports/_hashlib.pyd TorBrowser/Tor/PluggableTransports/_socket.pyd TorBrowser/Tor/PluggableTransports/_ssl.pyd TorBrowser/Tor/PluggableTransports/bz2.pyd TorBrowser/Tor/PluggableTransports/Crypto.Cipher._AES.pyd TorBrowser/Tor/PluggableTransports/Crypto.Hash._SHA256.pyd TorBrowser/Tor/PluggableTransports/Crypto.Hash._SHA512.pyd TorBrowser/Tor/PluggableTransports/Crypto.Random.OSRNG.winrandom.pyd TorBrowser/Tor/PluggableTransports/Crypto.Util._counter.pyd TorBrowser/Tor/PluggableTransports/Crypto.Util.strxor.pyd TorBrowser/Tor/PluggableTransports/flashproxy-client.exe TorBrowser/Tor/PluggableTransports/flashproxy-reg-appspot.exe TorBrowser/Tor/PluggableTransports/flashproxy-reg-email.exe TorBrowser/Tor/PluggableTransports/flashproxy-reg-http.exe TorBrowser/Tor/PluggableTransports/flashproxy-reg-url.exe TorBrowser/Tor/PluggableTransports/fte.cDFA.pyd TorBrowser/Tor/PluggableTransports/fteproxy.exe TorBrowser/Tor/PluggableTransports/M2Crypto.__m2crypto.pyd TorBrowser/Tor/PluggableTransports/meek-client-torbrowser.exe TorBrowser/Tor/PluggableTransports/meek-client.exe TorBrowser/Tor/PluggableTransports/obfs4proxy.exe TorBrowser/Tor/PluggableTransports/obfsproxy.exe TorBrowser/Tor/PluggableTransports/pyexpat.pyd TorBrowser/Tor/PluggableTransports/python27.dll TorBrowser/Tor/PluggableTransports/select.pyd TorBrowser/Tor/PluggableTransports/terminateprocess-buffer.exe TorBrowser/Tor/PluggableTransports/unicodedata.pyd TorBrowser/Tor/PluggableTransports/w9xpopen.exe TorBrowser/Tor/PluggableTransports/zope.interface._zope_interface_coptimizations.pyd

2 2

Prioritization/timeline tags for ff38-esr work
by Mike Perry 12 Jun '15

12 Jun '15

Georg and I have finally filed the full set of tags for the ff38-esr work, based on the developer documentation review and the Mozilla bugtracker review. Everything should be in this tag now: https://trac.torproject.org/projects/tor/query?status=!closed&keywords=~ff3… I also broke out the set of things we really need to have ready for 5.0a3 on June 30th: https://trac.torproject.org/projects/tor/query?status=!closed&keywords=~tbb… Those tickets represent major proxy safety audit work, patch rebase work, build system bugs, and major functionality bugs. Basically those bugs are blockers for us deploying a safe, usable, functioning TBB 5.0a3 based on Firefox 38. The set of bugs that really want to get fixed in some kind of alpha release before 5.0-stable are: https://trac.torproject.org/projects/tor/query?status=!closed&keywords=~tbb… We should aim to get those bugs fixed for 5.0a3, but if a lot of them still end up remaining after 5.0a3, we'll want to consider either another alpha sometime in July, or do a "soft launch" release of TBB 5.0 before August 11th, so we don't have to autoupgrade 4.5.x users to it right away. The code freeze for 5.0a3 will be June 23, though we can probably slip a couple things to as late as June 25. In other words, we've got about 2 weeks left to get through that essential list and make a dent in tbb-5.0a-highrisk. -- Mike Perry

1 0

MathML disabling in Tor Browser
by Arthur D. Edelstein 10 Jun '15

10 Jun '15

Frédéric Wang posted the following comment on bugzilla.mozilla.org about Tor Browser's disabling of MathML. I'm posting it here in case it's useful: > I'm curious to know what was the reasoning to put MathML in that category [Medium-Low] and whether the Mozilla MathML/Security teams should do something to provide more security guarantee on MathML to Tor people. > > The "Medium-Low" level seems to only disable features related to executable code (javascript & java) while MathML is essentially a complex rendering of text so it should be treated at the same level as layout, graphics & fonts (e.g. the latest public critical issue I'm aware of is https://www.mozilla.org/en-US/security/advisories/mfsa2014-59/, which is actually really from the "DirectWrite font handling"). > > The iSEC study does not even mention vulnerability of MathML while it says that "the SVG components have been the host of several exploitable bugs in the past several years" and recommends to "disable at the Low or Medium security level"... but your link says it is only disabled in High mode. > > Also, the iSEC study says it rely on the exploit analysis, but a quick search on https://www.mozilla.org/en-US/security returns far less results (two) for MathML than for SVG. And actually a search for "graphite" also returns two crashes too: https://www.mozilla.org/en-US/security/advisories/mfsa2012-64/ > > Finally, the iSEC study seems to take into account the number of websites using a given feature, but MathML does not seem less popular than graphite or svg opentype fonts.

1 1

Orfox Updates
by Amogh Pradeep 04 Jun '15

04 Jun '15

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256 Build updates with Orfox. Building Firefox for Android on the current tor browser dev branch[1] caused some problems with the app that it produced. Finding a fix to this seemed almost impossible with the limited log that the app produced and so I decided to apply patches manually from the GECKO380esr_2015050513_RELBRANCH to get the tb branch. A set of news updates related to this process can be found at the Orfox news [2] page. Eventually, after going through all the patches and testing out the applications it was producing, I managed to boil it down to one file that was causing the problems, toolkit/xre/nsXREDirProvider.cpp . This file seems to be modified as a patch to bug 9173 [3] . Currently I've replaced this with the esr38 release branch file. I'd like to know whether this is fine, if not, what other fix can be found for this problem. My current working branch can be found on my github [4] and the branch for Orfox [5] is there as well. Firefox for Android has multiple network implementations that work at the Java/Android layer, a report of that can be found on the Orfox project page [6], [7]. I am working alongside mozilla to speed up fixing these issues as well [8]. [1] https://github.com/arthuredelstein/tor-browser/tree/tb_GECKO380esr_20150505… [2] https://dev.guardianproject.info/projects/orfox-private-browser/news [3] https://trac.torproject.org/projects/tor/ticket/9173 [4] https://github.com/amoghbl1/tor-browser/tree/orfox-tb_GECKO380esr_201505051… [5] https://github.com/amoghbl1/Orfox [6] https://dev.guardianproject.info/attachments/download/1580/firefox-for-andr… [7] https://dev.guardianproject.info/news/219 [8] https://bugzilla.mozilla.org/show_bug.cgi?id=1169421 -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iQEcBAEBCAAGBQJVb9nxAAoJELm/SSKdtjUkSe8H/3fpinRtQBYVBKDbfT2Fw9hF Hp0GT82qdg6rqsouDeL8AVrjfHkf5GFr25i0RuFKxF14O3L7qhC4U2NbKbq5oEPH GfJtNs4EGJbk0x1/Dn5aOZMAU/fmgnH6wf/vrPDl/kTBzp/U4mRsPTSOcD0lUEZP kp7Dsu5YYBGcbmb7oB32fUCim4bWYPi8ScYcBEsT7ENFyOt1msGcNOj/bAKx9+6I lyhrtbuiAkf6jPVjC4a46XB5WWtC5S01Ht05Y+AHlL/EsmwKM9Tvxpas7HQIMW2F EQdKjVKwO20zUn185JsiKKXiEW6kUlqSqnpSQhIGOBZRt4oivAOZLZPNaFsqzkI= =JRkf -----END PGP SIGNATURE-----

2 1

Build Failures on ESR38
by Tom Ritter 03 Jun '15

03 Jun '15

Per irc today. I confirmed I did not run out of disk space on the vm. I have 11G free of 15. I have 2 GB of RAM, I'm not sure if that is the problem or not. Below are five last-many-lines of logs of failures of tor-browser: http://pastebin.com/LyMRhKdV http://pastebin.com/mRMP2FVH http://pastebin.com/8L3cFMU5 http://pastebin.com/D7eLB65y http://filebin.ca/23mLDScVCsKJ/build.log (most recent, full log) -tom

2 2