If a component was a device with MUST_ISOLATE set on it, we would try to
bind to the non-proxied side of it, which doesn't speak the device
protocols.

Change-Id: I64eaccec5d51997f4374d0277b040a952f681667

This adds the configuration that specifies if stories should use new runners or
reuse their parents' runners.

MF-353 #comment Add story_realm_inherit_parent_runner config and update docs
TEST=fx run-test peridot-tests

Change-Id: Ifed85a9dacce34db614a5e7fda82e5da3d23630d

replace "two" with "three" prior to list of three items.

Test: none.
Change-Id: I947a4dfa7cd56af8a9abca43b10273984211b460

Failure to communicate with a test server peer caused both the test
runner and the go and c++ test servers to just hang. They should (and
now do) fail loudly.

Change-Id: Ic8116045935a45c1fca19cffdf80a39abf343531

When a memfs is created without an explicit max page limit, the fs
assumes UINT64_MAX, which is confusing when manifested in
QueryFilesystem calls. This patch refines the cap to the amount of
available physical memory.

Test: /boot/test/fs/memfs-test
Change-Id: I10fada0b8b2eb7b33f270cd31c9d859157475c1d

This change creates the new libiquery library and implements the first
part of it: Discovery.

Follow ups will include: Reading, Formatting

CF-714: #progress

Change-Id: I4bc6af3367ef8030479bd722c8b4462cef981764

During end to end testing, the system load on some devices may be
high enough to cause the scheduler to not give audio_core another
time slice, leading to UNDERFLOW conditions.  This is causing
failures in CI.

BUG: MTWN-269
TEST: Ran end to end test repeatedly with no failures.
Change-Id: I3b0cc24140b88bf9315abddf59a461491d504d53

missing. Added a unit test that failed without the changes.

Change-Id: Idaf9913bf9006384e021f39047df1e8e8a855369

* it turns out async::PostDelayedTask returns a status so we might as well log an error if we couldn't post the timeout!

TESTED=I don't know an easy way to make the async loop reject the task :-(

Change-Id: I99b3a685f8ac2176aa0274890bc2a51e6661ce91

Test: CQ

Change-Id: I5e8322db78c313bad9af2a71759cdf6e71dbd0c9

Following earlier discussions with Satsuki the precedent we set with
these packages didn't make sense when applied to token manager. We're
probably stomping on a wider namespace than we'd ideally like but thats
a common theme across rust package names at the moment plus its
possible we'll roll both these into token manager at some point anyway.

Change-Id: I825ca9e0a853cdafde845e96077a64d1d5729bc4

Implements ZX_JOB_TERMINATED, which is signaled when a job is killed,
after all its children have finished cleaning up.

This also changes job tracking behavior. Previously, if a job was killed
it would stay in the job trees until the last handle was closed. With
this patch, jobs now behave like threads or processes and remove
themselves from the task trees as soon as they (and all children) have
been killed. This provides a more consistent behavior among all tasks.

As a nice side-effect, now if a job fails to create (e.g. if the parent
is already dead) we never add it to the all-jobs tree; previously it
would be briefly added and then removed.

ZX-3919

Test: runtests
Test: run-zircon-x64 -C -b -k -z core-tests-x64.zbi
Change-Id: Ic367b7df9704e5012a1c73636d76784c527a4636

It has been replaced by the //sdk/bundles:tools set of packages.

Bug: DX-1269
Change-Id: I49e1cf26ede67ac15da06b813b90968b25e7e010

This reduces the externally and internally mutable surface of the datastructure

Change-Id: I26f045a62cfa2bd78c3c02a65ee0c4b062f9c6f3

Component manager will not always be able to rely on
fuchsia.logger.LogSink being present in its namespace, so this commit
adds new logging macros to src/log.rs to log to zircon's debuglog. If
debuglog creation fails, logs are sent to stderr.

CF-707 #done

Change-Id: I5893719e6bb72aaba481ac0a0cbe43911cfea29c

...reading fuchsia.netstack.NetInterface{,2}.Ipv6addrs from upstream and
removing netiface.NIC.Ipv6addrs, which is no longer referenced.

NET-1181 #done
NET-1223 #progress

Change-Id: I4fee8d2d144b705e8643b3f9d73075a00d5e1fb1

This method has been deprecated for a while and is now unused.

Bug: ZX-2620 #done
Change-Id: I742ca8299a598000b34aaee17c2493a441db6076

This CL moves the CopyArrayToVector from sandbox.cc's anonymous
namespace to a helper method in json_parser.cc, and updates the
callers to use it as appropriate.

Bugs: CF-723

Tests: fx run-test json_unittests
Change-Id: I8546d76eeec621c0c5539521b7c000f534e95e15

Change-Id: I706db37f1aba1d62d56bc649cb4a69f149a59c76

Change-Id: I907aeffc7303998b136a59c63327badf42fc08e5

Mozart no longer exists, as a project or a team.  It's nice to not have
it show up when grepping the code-base.

Tested: CQ; no behavior change
SCN-580 #comment

Change-Id: I5456c927e7fdf63be820a970f04636c9345db64b

Use ZX_INFO_THREAD and thrd_join to eliminate races.

Use test infrastructure's timeout rather than a test specific timeout.

Change-Id: Ib25a5a1c54ea5a358c863202447c38706746207a

FLK-168 #progress

Change-Id: I4b049fb145d88bc3f028e7bb6665a36634418f4d

BUG: FLK-178 #progress

- Disable SandboxTest.DestructorRunsCleanly. Race condition causing
flakiness.

Change-Id: Ia7115f536d3f4a0b25e0deced2ab8f0133c3a69e

Change-Id: I3413b9b5d910e3f695437a8622d61cf821158603

* Update the location of the dart-sdk
* Add more troubleshooting tips

Change-Id: Iba475bd1505e390dd022a163a75ee91e3f1c4365

... and support --detach with spawned apps.

PT-141 #done

Change-Id: If980b4e8227d72e7c94af022ad344b374b09e335

WLAN-1078 #done

Test=Tested Ralink and ath10k. Ralink got a 3x speed boost, ath10k
doesn't show any notable improvement.

Change-Id: I55dbf79cb9dd9885da35260d3311d5a85f1a0cfc

Change-Id: I529d92ecf6d56f3d4abb7109955dbcaea93be340

No driver supports the new SME channel approach yet.
Thus, none of the new code is used yet.

WLAN-927 #comment

Test=Sanity check with broadcom and ath10k

Change-Id: I7cde52becea9ac6b2985ef6a392b7504c04a4fdc

Add a README to the A2DP sink profile outlinig how to run the profile
and the runtime requirements.

Move checks for player support of codecs to the startup of the profile,
so that it will fail early if it cannot play audio.

BT-840 #done
BT-533

Test: run bt-a2dp-sink without and with required components available

Change-Id: I5fd2fdf81d20c95fe46450abf8b9bfe100bd774c

This is a continuation of Doug Evans' CL 266326 but
under new management :-)

Change-Id: If4c4e088bd9d8ae8048ed7d4b8830738098f4be5

Test: CQ

Change-Id: I2f553cb853d2e4c506d1e855648cbe0681834bb5

Change-Id: I166512e6a74d4c8eb25922ed2c6e924e956d36a9

When constructing Pagechange for the PageWatcher notification, correctly
return an entry with an empty value if the entry has the priority LAZY
and the value is not present loacally on the device.

TEST=SyncIntegrationTest.PageChangeLazyEntry
LE-747 #done

Change-Id: I59b28b9f00cc030c784d2fc0265e25a3e60a29db

Test: CQ

Change-Id: I9922fc651b6dde0a6eb8232d501b4211e9e2810b

Ensure that device notifications for a device removal can only be created after the notification for the device being added is dispatched.

LE-739 #done

Change-Id: I8c33ae209916f4a82277782731f1d837b501291b

Change-Id: Ifb0945dd25ebc9e37f6fcb3926d8ef3ffa0f3adb

The t-test CIs are wider and more accurate, especially for
smaller sample sizes.  For our current sample size of 30 and setting
of alpha = 0.01, they are wider by 7%:

>>> scipy.stats.t.ppf(alpha/2, 30-1) / scipy.stats.norm.ppf(alpha/2)
1.0700964927577572

The blocker for using t-test CIs was making the SciPy library
available to perfcompare when run on the bots.  This is achieved by
adding a vpython spec file.  perfcompare.py is already run through
vpython by the fuchsia_perfcompare.py recipe.  vpython will download
the dependencies specified in .vpython and make them available to
perfcompare.py.

The dependencies used here are based on those used by Chromium's main
.vpython file:
https://chromium.googlesource.com/chromium/src/+/33da179d28beb0f03278a04b6ee2d97bbd8daf82/.vpython

The numpy and scipy packages come from here:
https://chrome-infra-packages.appspot.com/p/infra/python/wheels/

Bug: IN-646
Test: "python perfcompare_test.py" + perfcompare trybot
  + manually checked that this makes CIs wider by 7% for a given set of perf results
Change-Id: Ib3a6b1d733ea328180b804c3f2a080eb6330a263

We also rename all MZ-* bugs to SCN-*.

SCN-580 #comment

Change-Id: I82294f7958b529f1deded43a1d6d5210bbe43f6f