1. No more End-Of-Arguments.
2. Daemons close the control socket after they are initialized.
3. No more system properties.
4. ip-up-vpn now creates state to pass the configuration.
5. JNI methods are split again for legacy VPN.
Change-Id: I02fafdf01d425c965345ef712b2bd5fdee3a0cab
Changes in ConnectivityService in hc-LTE when merged with changes
happening in Master caused the build to break.
Change-Id: I92a0b782ae58e9789b1e950c94ef966234fa94af
Now classify background processes into a set of bins of how much
memory they should try to clear. The last bin also involves
destroying all activities in that process.
Removed the old code for the simulator that is no longer needed
(yay). The debugging features it had are now integrated into the
regular oom adj code.
Small fixes to load average service.
Change-Id: Ic8df401714b188c73b50dbc8f8e6345b58f1f3a0
This change moves the cached window and application input state
into the handle objects themselves. It simplifies the dispatcher
somewhat because it no longer needs to fix up references to
transient InputWindow objects each time the window list is updated.
This change will also make it easier to optimize setInputWindows
to avoid doing a lot of redundant data copying. In principle, only
the modified fields need to be updated. However, for now we
continue to update all fields in unison as before.
It turns out that the input dispatcher was inappropriately retaining
pointers to InputWindow objects within the mWindows InputWindow
vector. This vector is copy-on-write so it is possible and the
item pointers to change if an editing operation is performed on
the vector when it does not exclusively own the underlying
SharedBuffer. This bug was uncovered by a previous change that
replaced calls to clear() and appendVector() with a simple use
of operator= which caused the buffer to be shared. Consequently
after editItemAt was called (which it shouldn't have, actually)
the buffer was copied and the cached InputWindow pointers became
invalid. Oops. This change fixes the problem.
Change-Id: I0a259339a6015fcf9113dc4081a6875e047fd425
1. Touch exploration start and end events are generated
by the sytstem to provide additional information for
accessibility services. Since such events do not come
from any particular window they whould not change the
id of the window that currently allows exploring its
content.
2. Touch exploration start and end events were lealing the
touch explorer class wich is private.
bug:5026258
Change-Id: Icaf3e2bd9566716f2afb876cf8e0d50813b0c76e
Zlib compression, with a full flush between each application's
data. Encryption will be performed on the already-compressed data
once that's implemented.
On restore, the streamed data is similarly uncompressed on the fly.
Change-Id: I19b65c88e759a66527d10913d18fffa9df0bc011
When NetworkManagementService starts, it determines if qtaguid
should be enabled. The changes pushes the enabled state to a system
property to control socket tagging.
Change-Id: I5697a990dbaad4b220a2fb055931b18077d07a74
If an address is removed we must reset the connection but
only for the connections associated with that address. For
now we're doing the "all" addresses for a type (IPv6 or IPv4)
in the future we only need to reset a particular addresses
connections.
Bug: 4981919
Change-Id: I97f8071a3ed6f827ed22f32216ca5011bfe6c1d9
When requesting historical values with time ranges, always include
full values for buckets that span current time. (It doesn't make
sense to interpolate partial data.) Move getTotalData() to return
full Entry objects to prepare for packet counts.
Bug: 4691901
Change-Id: I717bd721be9f1d4a47c4121e46e07a56cb15bbf1
if the state transform didn't preserve rectangles, we
would still try to use h/w composer hal using the bounds
of the transformed rect, which isn't correct.
now we correctly fall back to composition.
Change-Id: Iff78f4339ece415d4987e95a5717b04934d370ab
Instead of exposing internal storage details, offer getValues()
accessor which populates values into a recycled structure.
Change-Id: I5228d7099a8f61b2018a435a813435a67dae49b4
This patch enables the Zygote to tell the ActivityManager when
it has started a process with a wrapper attached so that the
ActivityManager can allow it extra time to start up or process
events.
This is useful when wrapping an app with Valgrind or other tools
which add significant runtime overhead.
Bug: 4584468
Change-Id: I5db6f2f15cd30b0ec40f547d2fadfa216de2926d
The new link-status netlink monitor is a bit chatty and we were getting race conditions.
The wl0.1 soft-ap interface was getting created, but in the off state. We would report
it available (link-added always reports it) causing wifi to start tethering, but then
we'd sometimes get a link-status-down notification and a link-status-up, causing wifi
to freak out with these two requests.
The simple fix was to not listen to the new link-status notifications. This should be
redesigned in the future.
bug:4983942
Change-Id: Ica4f34d8bf2a07fc19b35a1e15283a7f8022165e
This makes it more robust when recovering from runtime restarts
Bug: 4986841
Change-Id: I54b94213447130ca881c66da2d0ce490242f0c96
Signed-off-by: Mike Lockwood <lockwood@android.com>
When an iface goes inactive, xt_qtaguid maintains historical counters
instead of discarding. Now reading and returning those values in
NetworkManagementService summary calls. Tests to verify stats are
returned.
Modify NetworkStats to include rxPackets and txPackets values, and
move to Entry to eventually hide internal storage details.
Bug: 4984539
Change-Id: I1ba0bb5580c62e946d9dff2cc2a6f64aff403efc
