| Commit message (Collapse) | Author | Age | Lines |
|
|
|
|
|
| |
HKDF-Expand enforces a maximum output length much shorter than stated in
the RFC. This patch aligns the implementation with the RFC by allocating
more output space when necessary.
|
|
|
|
|
|
|
|
| |
On MacOS, particularly large allocations within the platform limits
(1.2G+) will fail and crash. This adds a specific size check for that
when working around driver bugs (default).
While there, added a generic size_t limited size check for the
platform, and reporting OOM if too large.
|
| |
|
| |
|
| |
|
| |
|
| |
|
|
|
|
|
| |
No longer follow the value's prototype chain when creating index updates
in IndexedDB.
|
| |
|
|
|
|
|
|
|
| |
This also addresses clearing of document dependent JS slots which might
get out of sync with innerWindow navigation; relevant comments added.
This resolves #1253
|
|
|
|
| |
GCC 9 compiler does not like the way we have it in XPCWrappedNative.cpp
|
| |
|
|
|
|
|
|
|
| |
When Mozilla implemented this initially, only Windows 8 existed.
Because of the strict equal check, 8.1 and 10 didn't get HWA over RDP
while they are perfectly capable of doing so with RemoteFX. This change
allows any version of Windows from 8.0 onwards to use HWA over RDP.
|
| |
|
|
|
|
|
| |
360 Safeguard/360 Total Security (Qihoo) causes crashes in a11y components.
This adds the offending dll to the injection blocklist.
See BZ bug 1536227 for details.
|
|
|
|
| |
Add NSS_CMSSignedData_GetDigestAlgs and NSS_CMSSignedData_hasDigests which are required for security patches in mailnews applications. Ref: m-c bug 1526473
|
| |
|
|
|
|
|
| |
General consensus seems to be that color management on Linux desktops
is not mature enough to enable by default.
|
|\ |
|
| |
| |
| |
| |
| |
| | |
Although it's enabled by default in gcc (should be, anyway!), we're
being explicit here for 64-bit x86 platforms here also. This matches the
old behavior.
|
| |
| |
| |
| |
| | |
This only adds SSE2 flags when the CPU architecture is correct for it.
Resolves #1226
|
|\| |
|
| | |
|
| |
| |
| |
| |
| |
| |
| |
| | |
Previously, HSTS preload list values could be overridden temporarily due
to counter-intuitive behavior of the API's removeState function.
This adds an explicit flag to the API for writing knockout values to
the Site Security Service, with the default resetting to whatever the
preload list state is.
|
| |
| |
| |
| |
| |
| |
| | |
This fixes a rare crash/CTD in JS.
This adds information about the constraints to a new RAII class so we
can finish all constraints at the end.
Based on changes in BZ 1568397
|
| | |
|
| | |
|
| | |
|
| | |
|
| |
| |
| |
| |
| | |
java->C++ automated translation is not something we want to deal with
now or in the future.
|
| | |
|
| | |
|
| |\ |
|
| | | |
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
entire polygon.
This fixes a bug that was introduced three years ago in BZ bug 1268854.
What happened was that the final pass over the polygon assumed that the
current polygon was living in plane[0]. But due to the double buffering,
the "current" polygon alternates between plane[0] and plane[1].
The bug had also introduced an early exit so that we could hit the final
pass at a time where the current, now empty, polygon was in plane[1]. So
we would incorrectly treat all 32 points in plane[0] as part of the
final polygon.
This bug was responsible for intermittently unreasonable numbers in
CompositorOGL's fill rate / overdraw overlay.
This fixes a regression caused by the fix for CVE-2016-5252.
|
| |/
| |
| |
| | |
This reverts commit 09a8b2f19689b679b1268a3004ec5e3f37b9732a.
|
| |\ |
|
| | |
| | |
| | |
| | |
| | |
| | | |
Favor fallback content in that case (if present). Fallback is always
considered "good" in this case so may end up doing nothing which is what
we'd want for corner cases that hammer this routine with no content.
|
| | |
| | |
| | |
| | |
| | |
| | | |
entire polygon.
This fixes a regression caused by the fix for CVE-2016-5252
|
| |/ |
|
| | |
|
| |
| |
| |
| |
| |
| |
| | |
nsXMLContentSerializer::CheckElementEnd so that we can properly
determine whether it has children.
This resolves #1221
|
|/
|
|
|
|
| |
and Basilisk
Also removes the redundant branding version of app.update.url in Pale Moon that was missed when many prefs were merged back into application preferences
|
|\
| |
| | |
Allow the use of Xdg desktop portal dialogs
|
| | |
|
| | |
|
| | |
|
| |
| |
| |
| |
| |
| |
| | |
Sync will not do anything unless specifically set up to do so and at
least one engine is enabled, so there's no need for this "master switch"
to force it disabled based on engines being disabled (which was its
previous function to shortcut syncing in that situation).
|
| |
| |
| |
| |
| | |
- `OS.Path.join` can throw, so we always need to try/catch it.
- Also do a sanity check to make sure `callback` is defined before use
|
| | |
|