Closed Bug 1189744 Opened 5 years ago Closed 5 years ago
_xul .dll!js::Call JSGetter Op
4.39 KB, text/plain
1.89 KB, patch
|Details | Diff | Splinter Review|
found via bughunter and reproduced with a windows 7 debug build based on m-c tip Steps to reproduce: -> Load http://shazam.pooltracktrends.com/?id=1525 -->> Crash signature CallGetter GetExistingProperty<int> NativeGetPropertyInline<int> js::NativeGetProperty js::GetProperty affects nightly and aurora !exploitable 126.96.36.199 Exploitability Classification: EXPLOITABLE Recommended Bug Title: Exploitable - Data Execution Prevention Violation starting at Unknown Symbol @ 0x0000000004497138 called from xul!js::CallJSGetterOp+0x0000000000000056 (Hash=0xff17ac06.0x050ceebd) User mode DEP access violations are exploitable.
[Tracking Requested - why for this release]: [Tracking Requested - why for this release]: Bisected it to: https://hg.mozilla.org/integration/mozilla-inbound/pushloghtml?fromchange=c0ab4a5c0acf&tochange=e6f863562d50 => bug 1125624 We're trying to invoke an old-style getter, but we crash because this pointer is a JSObject* getter. It reproduces with the JITs disabled FWIW.
Tracking for 41, 42 because regression.
Assignee: nobody → jorendorff
Status: NEW → ASSIGNED
Thanks for looking into this, Jan. Definitely needs to land on the branch.
Attachment #8643362 - Flags: review?(jdemooij) → review+
Comment on attachment 8643362 [details] [diff] [review] Fix crash after GetOwnPropertyDescriptor failed to populate all fields of desc [Security approval request comment] How easily could an exploit be constructed based on the patch? It'd be challenging, maybe impossible, but there's a very clear goal: try to make a JS object whose in-memory representation is also executable code. Do comments in the patch, the check-in comment, or tests included in the patch paint a bulls-eye on the security problem? To me, yeah. Which older supported branches are affected by this flaw? FF41 only If not all supported branches, which bug introduced the flaw? bug 1125624 Do you have backports for the affected branches? If not, how different, hard to create, and risky will they be? No, but easy and minimal risk. How likely is this patch to cause regressions; how much testing does it need? Very unlikely. Existing tests will suffice.
Attachment #8643362 - Flags: sec-approval?
Comment on attachment 8643362 [details] [diff] [review] Fix crash after GetOwnPropertyDescriptor failed to populate all fields of desc Sec-approval+. Please nominate for Aurora too so we can fix it there before it becomes Beta on Monday, if possible.
Attachment #8643362 - Flags: sec-approval? → sec-approval+
Comment on attachment 8643362 [details] [diff] [review] Fix crash after GetOwnPropertyDescriptor failed to populate all fields of desc Approval Request Comment [Feature/regressing bug #]: bug 1125624 [User impact if declined]: possibly exploitable bug [Describe test coverage new/current, TreeHerder]: unit tests are included [Risks and why]: minimal [String/UUID change made/needed]: none
Comment on attachment 8643362 [details] [diff] [review] Fix crash after GetOwnPropertyDescriptor failed to populate all fields of desc Approved for uplift to Aurora and Beta as it is a security fix. The patch is also simple with tests included.
url: https://hg.mozilla.org/integration/mozilla-inbound/rev/03b1eb0b1f9bcb470c1996dedc45992eb4acef59 changeset: 03b1eb0b1f9bcb470c1996dedc45992eb4acef59 user: Jason Orendorff <firstname.lastname@example.org> date: Tue Aug 04 18:16:08 2015 -0500 description: Bug 1189744 - Fix crash after GetOwnPropertyDescriptor failed to populate all fields of desc. r=jandem, a=abillings.
This was found by fuzzers in bug 1183448 >2 weeks before this one. If the 2 bugs are dupes, it would be nice to prevent this from happening again by working on such s-s fuzzer bugs asap.
5 years ago
Duplicate of this bug: 1183448
Reproduced the crash using the following builds: - http://inbound-archive.pub.build.mozilla.org/pub/mozilla.org/firefox/tinderbox-builds/mozilla-central-linux64-asan/1438369481/ - https://archive.mozilla.org/pub/firefox/nightly/2015/07/2015-07-31-mozilla-central-debug/ Went through verification using the following builds: - https://archive.mozilla.org/pub/firefox/nightly/2015-09-11-03-02-04-mozilla-central/ - http://inbound-archive.pub.build.mozilla.org/pub/mozilla.org/firefox/tinderbox-builds/mozilla-central-linux64-asan/1441965724/ - https://archive.mozilla.org/pub/firefox/nightly/2015-09-11-00-41-12-mozilla-aurora/ - https://archive.mozilla.org/pub/firefox/candidates/41.0b9-candidates/build1/ OS's Used: - OSX 10.10.5 x64 -> PASSED - Win 10 x64 (VM) -> PASSED - Ubuntu 14.04.3 (VM) x64 -> PASSED Test Cases: - Opened the link in several new tabs/windows using e10s - Opened the link in several new tabs/windows using non-e10s - Opened the link in several new private windows/tabs
You need to log in before you can comment on or make changes to this bug.