1. Send PROXY_CHANGE_ACTION broadcast when any network's proxy changes,
not just the default network.
2. When a process is bound to a particular Network, update the proxy
system properties to those for the bound Network, and keep them
updated when PROXY_CHANGE_ACTION broadcasts are received.
3. Make Network.openConnection() use the proxy for the Network.
bug:17905627
bug:17420465
bug:18144582
(cherry-pick of https://android-review.googlesource.com/#/c/115170)
Change-Id: Ia2819985e6108a8c121e74c683a5646becfd0a97
Tune sampling periods:
- Lower minimum sample interval when in test mode.
- Add a check when sampling pss to make sure this is at least 1
second since the pss changed, to reduce the amount of bad
data (very small pss samples) that we see.
Change-Id: I624a7d0480f3f6b22d229744e028d13a2c7b0e89
This causes crashes on all devices that disable the BluetoothHeadset
service when disabling bluetooth and thereby trying to unbind all
services.
Bug: 18669882
Change-Id: I48978a1556ead967c848c5bd9f6e001d38754b8d
http://ag/261732 introduced an optimization where we only take
the activity screenshot if is is not the activity we took a
screenshot for last. This causes us to display a stale screenshot
for activities that change their display content (using fragments)
without going through the pause/resume cycle. It should be safe
to always take a screenshot when we are pausing since the code
path is only called once per pause/resume cycle.
Bug: 18682160
Change-Id: Ie5c43cfd806286808af4233c5917ae7071908ebb
(cherry picked from commit eacdf2ce04899b0a3abef55e4e7e1fe9c1649932)
Previously the policy change would always re-enable the keyguard,
but that would break the current state. Instead, when the keyguard
is disabled, we need to recalculate the allow state immediately
and only re-enable it if necessary.
Bug: 18430385
Change-Id: Id9e701df2db765031e177c8bc8d3730cfb7f162a
Fixes a regression that prevents the user from being fully shutdown
after being stopped. Wasn't killing all user processes.
Bug: 7462778
Change-Id: Iee7231f7c8a167e8a61549b149294b2dbef071c1
If an accessibility service opts-in for the new window introspection APIs
when calling into a window for accessibility node infos we compute the
window's interactive region, i.e. the portion that is touchable and not
covered by other touchable windows, in roder to deterine whether these
nodes are visible, i.e. the user can interact with them. There was a bug
in the interactive region computation what we were not ignoring accessibility
overlay windows which are intended to not change what an accessibility
service can "see" and are put by such a service.
bug:18652374
Change-Id: I24ce54069087af9bc308e0c6a7b2aa6b952a6136
You can now do "adb shell dumpsys procstats --start-testing" to
enable high frequency pss sampling.
Also improved the low on RAM mem reporting to separate out RAM
from memtrack, in case the data we are getting from that is bad.
And fixed meminfo --oom to work correctly again.
Change-Id: I7af17eab110a82298bd7b0ce381f8fa5c96c1f6a
INSTALL_FAILED_UPDATE_INCOMPATIBLE is a better description of the
real underlying cause when signatures don't match.
Bug: 18413784
Change-Id: Ib81591df7d0d79f6725a4fa7300ae4e20958f004
Connectivity broadcasts recently changed and are no longer sent for
certain types of network changes. For example, when stacked network
interfaces change for a mobile network. To ensure that we pick up
all these details, directly wire the two services together.
Also remove some unused code for split network types.
Bug: 18666753
Change-Id: I0467bd5b330c0e0cb51af2306d821b41ad16337a
There are some cases where multiple subscriber identities (IMSI)
should be treated as "merged together" from a data usage
perspective. This is done by extending the template used for
matching purposes to support multiple subscribers.
Then, when we query historical usage or set network policies, we
normalize the matching template to merge to any other identities
that should be included. When normalizing, the "lowest" identity
is always used for equality and storage purposes, which allows
identities to come and go over time.
This change also fixes data usage recording for multi-SIM devices
by passing along the concrete subscriber identity for each network
interface. Also correctly create default policies for multi-SIM
devices. This change also drops setPolicyDataEnable() until it can
be wired up to the right underlying NetworkAgent. (This means we
still bring up the network, and then rely on iptables rules to block
traffic when over the limit, instead of proactively disabling the
connection.)
Bug: 18012787
Change-Id: If6acf32009fdfea2b836f5aff8e2f3e5e0248b4a
- Downtime: Allow user to enter downtime early, offer as an end
condition four hours before downtime starts. Available in
either none or priority, regardless of settings configuration.
- Downtime: Always exit before next alarm if zen=none.
- Downtime: Make more like any other condition provider, remove
special status (mostly).
- Downtime: New auto-triggering rules, allow triggering after a
manual condition ends, once.
- Decouple NextAlarm + Downtime providers, allow them to offer
their conditions at the same time.
- Downtime/NextAlarm: Update conditions if they change while being
requested, even if unsubscribed.
- Make all three built-in condition providers optional, via config.
- New internal helper for runtime config.
- Don't follow changes to next alarm, consider the condition false.
- Isolate downtime calendar logic into separate class (for testing).
- Allow a:bb -> a:bb as a valid downtime range (all day).
- Volume dialog: configuration establishes maximum number of visible
conditions, including built-ins.
- Zen mode panel: avoid widget updates during layout transition.
- Zen mode panel: move controller callers to background thread.
- Zen mode panel: hide/show/rebind rows instead of adding/removing.
- ZenLog: Add downtime autotrigger results.
- Volume panel: Smarter refresh on ringer/zen changes.
Bug: 16373455
Change-Id: I4f801018ddb0beb6eb9fa03a81c79f7949888a3f
CEC device information cannot be updated/notified properly if
the device was plugged to other port. This change makes
sure the updated information is reflected by device event
callback listeners.
Bug: 18513567
Change-Id: I62bbb741d3f89e225a50f86e315da014832d8118
Since optimistic addresses are useable upon kernel notification
there is no need for this extra connectivity delay.
---
This functionality was originally submitted in ag/572619. Owing
to issues with bind()ing to optimistic addresses (see b/18609055)
this was reverted in ag/598673.
This reverts the revert. :-)
Bug: 17769720
Change-Id: Ibee490b2af72050693b6bd748193f51e312ca527
