We had a thread-local varialbe `ModuleLoadFrame::sTopFrame` to track the topmost
stack frame of `LdrLoadDll`. However, our hook function `patched_LdrLoadDll` can
be called even before TLS is initialized. In such a case, accessing `sTopFrame`
causes AV.
This patch introduces `SafeThreadLocal` to safely access a thread-local varialbe.
If TLS is not initialized, it falls back to a global variable because in that
early stage there is only a single thread running.
Differential Revision: https://phabricator.services.mozilla.com/D55870
--HG--
extra : moz-landing-system : lando
2020-02-05 J.C. Jones <jjones@mozilla.com>
* lib/softoken/exports.gyp, lib/softoken/manifest.mn,
lib/softoken/pkcs11.c, lib/softoken/sftkdb.c,
lib/softoken/softoken.gyp:
Bug 1609673 - Conditionally compile out all libnssdbm glue if
NSS_DISABLE_DBM is set r=mt
Remove `lgglue` from compilation entirely if DBM is disabled
[b91bbf7a88c9] [NSS_3_50_BETA2] <NSS_3_50_BRANCH>
2020-02-04 Kevin Jacobs <kjacobs@mozilla.com>
* .hgtags:
Added tag NSS_3_50_BETA1 for changeset de6ba04bb1f4
[1201d0d89b72] <NSS_3_50_BRANCH>
Differential Revision: https://phabricator.services.mozilla.com/D61770
--HG--
extra : moz-landing-system : lando
We replace the a11y telemetry thread with a call to the new
`NS_DispatchBackgroundTask` infrastructure. We use `NS_DISPATCH_EVENT_MAY_BLOCK`
because the task performs some I/O to load the instantiator's version info.
I also took the opportunity to eliminate some redundant copying.
Differential Revision: https://phabricator.services.mozilla.com/D61769
--HG--
extra : moz-landing-system : lando
ScriptStencil centralizes the access from JSScript to BytecodeEmitter's fields.
Simple fileds are copied to it, and other fields are provided by methods.
Eventually, all those methods should go away, by having stencil structs for
the following, but that doesn't happen in this patch.
* GC things
* Atom
* the list of resume offsets
* the list of scope notes
* the list of try notes
* the list of inner functions
Those methods are virtual to utilize this also from rust frontend, by passing
another class derived from ScriptStencil to JSScript::fullyInitFromStencil.
Differential Revision: https://phabricator.services.mozilla.com/D61722
--HG--
extra : moz-landing-system : lando
Before now, we used XUL elements as the menu for device selection on RDM.
We introduced the icon on the XUL menuitem in bug 1306198, but the UI on Ubuntu
became weird looking. In order to fix this, we could think to modify Linux
gadget, but we replace to HTML menu components we are using in DevTools with
considering deXUL as well.
Differential Revision: https://phabricator.services.mozilla.com/D61149
--HG--
extra : moz-landing-system : lando
This patch uses the favicons as actually served by the sites. I tried to get 32x32 versions where possible, but a few are smaller and larger. There's also a mix of .ico and .png. I think the inconsistency is acceptable, and even desirable, seeing as these are the actual icons that would appear in the UrlbarView should they appear in the user's frecency results.
Differential Revision: https://phabricator.services.mozilla.com/D60421
--HG--
extra : moz-landing-system : lando
Otherwise it'd error, which is not great. We could add a custom attribute in
there if needed or something, or fix cbindgen to generate these for us as
described in the issue linked from the comments.
Do this in ServoStyleConstsInlines as that's where we have the guarantee that
the templates have been defined, and that the file is included every time
ServoStyleConsts is included.
Differential Revision: https://phabricator.services.mozilla.com/D61625
--HG--
extra : moz-landing-system : lando
Added long press as well as the old tooltip language back
and updated 2 tests.
Differential Revision: https://phabricator.services.mozilla.com/D61717
--HG--
extra : moz-landing-system : lando
Summary
Before the changes, if an mozilla addons theme was applied and an theme extension api has been called which call theme.reset() from background.js script then the browser theme was getting reset, which
should'nt happen. Code changes made in toolkit/components/extension/parent/ext-theme.js
Differential Revision: https://phabricator.services.mozilla.com/D58095
--HG--
extra : moz-landing-system : lando
We will no longer allow OMTP on 32-bit systems with less than 2GB of RAM
and 2 or fewer cores. These systems are unlikely to realize a benefit
from OMTP and see a higher than normal incidence of OOM crashes in the
content process due to OMTP having a higher memory footprint.
Differential Revision: https://phabricator.services.mozilla.com/D60058
--HG--
extra : moz-landing-system : lando
Toshihito Kikuchi
Lee Salzman
J.C. Jones
Csoregi Natalia
Aaron Klotz
Yura Zenevich
Robert Helmer
Cameron McCormack
Tooru Fujisawa
Emilio Cobos Álvarez
Daisuke Akatsuka
Harry Twyford
Boris Zbarsky
Matthew Gaudet
Jason Orendorff
Kirk Steuber
Sylvestre Ledru
Kendall Werts
ajitesh
MahakBansal
Andrew Osmond
Noemi Erli
Gabriel Luong
Shane Caraveo