We're just shipping this in 72 on macOS/Linux, and I think are still tweaking some things, so it seems premature for the telemetry to expire.
Differential Revision: https://phabricator.services.mozilla.com/D58545
--HG--
extra : moz-landing-system : lando
nsIMutable is used only by DOM Blob/File. But Blobs are immutable by spec.
FileBlobImpl has a couple of lazy member values, but those are called when
the object is cloned in order to be sent to a different thread/process.
Differential Revision: https://phabricator.services.mozilla.com/D58946
--HG--
extra : moz-landing-system : lando
nsIProcess is the tried-and-true method for launching utility subprocesses on
Firefox Desktop's supported platforms. Use that.
Depends on D57977
Differential Revision: https://phabricator.services.mozilla.com/D58809
--HG--
extra : moz-landing-system : lando
Since we're the only one sending data, and we're doing so infrequently, let's
get the pref value before each ping send instead of building a pref observer
right this second.
Differential Revision: https://phabricator.services.mozilla.com/D57107
--HG--
extra : moz-landing-system : lando
Revert to using CSS pseudo-elements for the +/- diff markers in the Changes panel so they don't get copied over as plain text to the clipboard.
Differential Revision: https://phabricator.services.mozilla.com/D58768
--HG--
extra : moz-landing-system : lando
The remote agent used to not compile on Windows AArch64 due to iovec's
dependency on a version of winapi without support for this architecture.
Now that the remote agent has upgraded to http 0.2, which depends on
a version of the bytes crate that has moved away from iovec in favour
of std::io::IoSlice, we are able to turn on support for Windows AArch64.
This in turn will also fix bug 1606935 because the browser-chrome
test manifest for M(remote) will no longer be empty. It was a
regression caused by 1603930 where we fixed a logic error causnig
ENABLE_REMOTE_AGENT to be inappropriately set on non-trunk branches.
Differential Revision: https://phabricator.services.mozilla.com/D58767
--HG--
extra : moz-landing-system : lando
Upgrading to http 0.2 will give us bytes ^0.5 which removes
a dependency on iovec. iovec is problematic for us because it
depends on a version of winapi that does not compile on Windows AArch64.
The effect of this should be that the remote agent is now possible
to compile on Windows AArch64.
Differential Revision: https://phabricator.services.mozilla.com/D58765
--HG--
extra : moz-landing-system : lando
Most of the code was copied from the String nursery allocation code paths and
then adjusted to work with BigInts. Where applicable, the newly added BigInt
functions were placed right next to the String functions, so it's easier to
compare them against each other. In a follow-up bug, we should try to reduce
these code duplications.
In contrast to Strings, nursery allocated BigInts will also try to nursery
allocate their heap digits, when needed, in the nursery. This works similar to
the slots and elements storage for objects, which can also be nursery
allocated. Supporting nursery allocation for BigInt digits is important when we
want to support nursery allocation of BigInts using heap digits, because it
avoids the extra overhead of registering malloc memory with nursery cells.
Differential Revision: https://phabricator.services.mozilla.com/D54772
--HG--
extra : moz-landing-system : lando
The trace-kind of forwarded cells must match the trace-kind of their target. Add
an assertion to ensure this invariant isn't violated.
Differential Revision: https://phabricator.services.mozilla.com/D56200
--HG--
extra : moz-landing-system : lando
Removes the static_assert because it's too restrictive, only nursery cells are
required to reserve `Cell::ReservedBits` bits in the first word of a cell. For
tenured cells, we only need to use a single bit in the first word. And the
static_assert will also start to fail on 32-bit when `Cell::ReservedBits` is
changed to 3 to account for nursery BigInts.
Also added an extra assertion to `RelocationOverlay::forwardTo` so it's easier
to see that storing the gc-flags doesn't overwrite any set bits.
Differential Revision: https://phabricator.services.mozilla.com/D56199
--HG--
extra : moz-landing-system : lando
BigInts need to use `0x01` as their TraceKind value, but that value is
currently already used for Scripts. Therefore move `Script` to an
out-of-line trace kind to free up `0x01` for BigInts.
Differential Revision: https://phabricator.services.mozilla.com/D56198
--HG--
extra : moz-landing-system : lando
If we are setting up a new permission prompt from iframe navigation, the popup
notification may reuse "checked" value from previous notification, even if the
checkbox is hidden. We should clear checkbox.checked (set to false) if the
checkbox is hidden
Differential Revision: https://phabricator.services.mozilla.com/D58516
--HG--
extra : moz-landing-system : lando
nsIMutable is used only by DOM Blob/File. But Blobs are immutable by spec.
FileBlobImpl has a couple of lazy member values, but those are called when
the object is cloned in order to be sent to a different thread/process.
Differential Revision: https://phabricator.services.mozilla.com/D58716
--HG--
extra : moz-landing-system : lando
When a page navigation happens, we update the state of the toolbox toolbar buttons
as some of them need to disable themselves then (rulers and measurement tool).
In bug 1500142 the logic to do this was changed to retrieve the right inspector front
instead of always the top-level one, stored at toolbox level.
This is good. However a tiny mistake made its way into the code.
Instead of calling getCachedFront("inspector") the code called
getCachedFront("inspectorFront")
This made it return null, and therefore prevented the rest of the button update logic
to run.
Differential Revision: https://phabricator.services.mozilla.com/D57997
--HG--
extra : moz-landing-system : lando
On windows 10-64 asan, we get a failure on the network io test that
checks the sqlite DB is touched. According to the logs, this failure
happens because when the test occurs the write happens in the wal
file. Adapted the test to also check for that location.
Differential Revision: https://phabricator.services.mozilla.com/D58908
--HG--
extra : moz-landing-system : lando
