- Aug 21, 2023
-
-
-
The GeckoView AndroidManifest.xml is not preprocessed unlike Fennec's manifest, so we can't use the ifdef preprocessor guards around the permissions we do not want. Commenting the permissions is the next-best-thing.
-
-
Pier Angelo Vendrame authored
-
Bug 41117: using MOZ_PROXY_BYPASS_PROTECTION disable share menu on mac and windows, and restore email link on mac
-
Pier Angelo Vendrame authored
Bug 40857: Modified the fat .aar creation file This is a workaround to build fat .aars with the compiling enviornment disabled. Mozilla does not use a similar configuration, but either runs a Firefox build and discards its output, or uses artifacts build. We might switch to artifact builds too, and drop this patch, or write a better one to upstream. But until then we need this patch. See also https://bugzilla.mozilla.org/show_bug.cgi?id=1763770. Bug 41458: Prevent `mach package-multi-locale` from actually creating a package macOS builds need some files to be moved around with ./mach package-multi-locale to create multi-locale packages. The required command isn't exposed through any other mach command. So, we patch package-multi-locale both to prevent it from failing when doing official builds and to detect any future changes on it.
-
Pier Angelo Vendrame authored
Bug 17858: Cannot create incremental MARs for hardened builds. Define HOST_CFLAGS, etc. to avoid compiling programs such as mbsdiff (which is part of mar-tools and is not distributed to end-users) with ASan. Bug 21849: Don't allow SSL key logging. Bug 25741 - TBA: Disable features at compile-time Define MOZ_ANDROID_NETWORK_STATE and MOZ_ANDROID_LOCATION Bug 27623 - Export MOZILLA_OFFICIAL during desktop builds This fixes a problem where some preferences had the wrong default value. Also see bug 27472 where we made a similar fix for Android. Bug 29859: Disable HLS support for now Bug 30463: Explicitly disable MOZ_TELEMETRY_REPORTING Bug 32493: Disable MOZ_SERVICES_HEALTHREPORT Bug 33734: Set MOZ_NORMANDY to False Bug 33851: Omit Parental Controls. Bug 40252: Add --enable-rust-simd to our tor-browser mozconfig files Bug 41584: Move some configuration options to base-browser level
-
-
- fixes required to build dom/media/systemservices
-
- fixes required to build dom/media/webrtc
-
- fixes requried to build netwerk/sctp
-
- fixes required to build third_party/sipcc
-
- fixes required to build third_party/libwebrtc
-
- properly define NOMINMAX for just MSVC builds
-
Pier Angelo Vendrame authored
System fonts are an enormous fingerprinting vector. Even with font allow lists and with our custom configuration on Linux, which counter metrics measurements, getComputedStyle leaks several details. This patch counters both these kinds of attacks.
-
Remove this patch after upstream bugzilla bug 1790187
-
Temporary fix until mozilla bug 1799153 gets a patch upstream.
-
This reverts commit 1eb13643. This fixes tor-browser#40721, tor-browser#40698, and #40706. However, it is a temporary workaround, that we should revert once https://bugzilla.mozilla.org/show_bug.cgi?id=1744719 is fixed.
-
Depends on D177790 Differential Revision: https://phabricator.services.mozilla.com/D178339
-
Fullscreen/PointerLock warnings are initialized with hidden="true", but change to hidden="" after being shown and hidden again. I think this started happening when we began using HTML elements instead of XUL as they handle hidden attribute differently. Differential Revision: https://phabricator.services.mozilla.com/D177790
-
Bug 1781169: Ensure that the cached start/end containers contain characters before returning them in DocAccessibleParent::SelectionRanges. r=eeejay In bug 1772170, I added code to ensure that the cached offsets are valid. However, this didn't account for the possibility that one of the containers contained 0 characters. In that case, offset 0 is itself invalid. If there were also no children, this might in GetChildAtOffset being called and returning null, causing a crash when retrieving selection. Differential Revision: https://phabricator.services.mozilla.com/D154964
-
Bug 1772170: Ensure cached Accessibles and offsets are valid before returning them in DocAccessibleParent::SelectionRanges. r=eeejay Differential Revision: https://phabricator.services.mozilla.com/D148745
-
GCC 13s libstdc++ reduced its dependency on some headers like <cstdint>, so it's no longer transitively included through various headers. Include it explicitly. See also: https://gcc.gnu.org/gcc-13/porting_to.html#header-dep-changes 11:16.32 .../components/telemetry/pingsender/pingsender.cpp:30:7: error: unknown type name 'uint32_t' 11:16.33 const uint32_t kConnectionTimeoutMs = 30 * 1000; 11:16.33 ^ 11:16.33 ... /components/telemetry/pingsender/pingsender.cpp:76:9: error: unknown type name 'uint32_t' 11:16.33 const uint32_t kBufferSize = 4 * 1024 - 1; 11:16.34 ^ 11:16.44 2 errors generated. (et al) Differential Revision: https://phabricator.services.mozilla.com/D167491
-
Differential Revision: https://phabricator.services.mozilla.com/D178564
-
Pier Angelo Vendrame authored
This helps to prevent linkability of users in the same network. Differential Revision: https://phabricator.services.mozilla.com/D170373
-
Bug 1816158 - Part 2: Require no GC when giving out references to the realm's debugger vector r=sfink To prevent any other instances of this problme we can update the getDebuggers() methods on the global and the realm to require no GC. Depends on D169701 Differential Revision: https://phabricator.services.mozilla.com/D169702
-
GC can mutate this vector so don't allow that while we are iterating. I think it would be safe to use index-based iteration but it's safer to just ban it entirely. This fixes the crash produced by the testcase. Differential Revision: https://phabricator.services.mozilla.com/D169701
-
Pier Angelo Vendrame authored
Now it displays app display name, instead of having Firefox hardcoded. Differential Revision: https://phabricator.services.mozilla.com/D168562
-
Currently, `ReserveAndPutRaw` allocates a second span even if the data would be too big for the chunk. Here a second conditional is added to check if the block of data is too big in this scenario and silently discard the data if so. Differential Revision: https://phabricator.services.mozilla.com/D167167
-
Depends on D164656 Differential Revision: https://phabricator.services.mozilla.com/D166108
-
Bug 1792157 - Ensure correct font-visibility for bundled fonts on Linux when they 'shadow' system-installed families. r=gfx-reviewers,lsalzman Differential Revision: https://phabricator.services.mozilla.com/D175350
-
Bug 1790542 - [devtools] Convert browser_source_map-reload.js test from using .sjs files to createTestHTTPServer r=ochameau,devtools-reviewers .sjs files depend on redirects to load static files. In D168648 we block redirects on sourceMappingUrl, so this browser_source_map-reload.js fails because we use .sjs to serve the content. This patch refactors to no longer use .sjs files Differential Revision: https://phabricator.services.mozilla.com/D169739
-
Bug 1790542 - [devtools] Do not allow redirects on the 'sourceMappingUrl' r=jdescottes,devtools-reviewers - Block redirects on sourceMappingUrl - Add debugger test Differential Revision: https://phabricator.services.mozilla.com/D168648
-
On macOS versions prior to 10.15 os_unfair_locks cannot spin in kernel-space which degrades performance significantly. To obviate for this we spin in user-space like OSSpinLock does, for the same number of times and invoking x86-specific pause instructions in-between the locking attempts to avoid starving a thread that might be running on the same physical core. Differential Revision: https://phabricator.services.mozilla.com/D154205
-
Bug 1774458 - Use undocumented, non-public adaptive spinlocks on macOS 10.15+, revert to user-space spinlocks on older versions r=pbone Differential Revision: https://phabricator.services.mozilla.com/D149599
-
Bug 1769028: Disable DNS lookup in nsProfileLock when the proxy bypass protection is enabled r=mossop Instead of using the local computer's IP address within symlink-based profile lock signatures, always use 127.0.0.1 when the proxy bypass protection is enabled. Differential Revision: https://phabricator.services.mozilla.com/D161264
-
Differential Revision: https://phabricator.services.mozilla.com/D164656
-
Bug 1670885 - Replace deprecated NSSpinLocks with os_unfair_locks in the memory allocator r=glandium Depends on D148287 Differential Revision: https://phabricator.services.mozilla.com/D99280
-
Bug 1670885 - Fix post-fork() handlers for PHC/LogAlloc to work on macOS using unfair locks r=glandium macOS unfair locks enforce that a lock can only be released by the thread which locked it. On macOS 11+ this caused the fork()'d child process to raise a SIGILL signal. Confusingly enough this behavior seems to be different on macOS 10.15 and possibly interacted in odd ways with our exception handler if it was installed before fork()-ing. Differential Revision: https://phabricator.services.mozilla.com/D148287
-
Bug 1768907 - Part 2: nsTransferable disk-leaks when private browsing mode is enabled by default. r=handyman nsTransferable was modified to prevent disk leakings when copying data in private browsing mode with Bug 1123480. However, the context is nullptr when it is initialized, so it still leaks if PBM is enabled by default. Our solution is to check the browser.privatebrowsing.autostart in this condition. Differential Revision: https://phabricator.services.mozilla.com/D157800
-