setThreadGroup(TOP_APP) will place a thread in the fg stune group but
not the top_app cpuset. This allows android.ui to get more runtime to
render transitions between activities without impacting its CPU
placement.
bug 29512132
Change-Id: I95e031975f83290f88ac6e538abf1062aea27ed4
Add a new mode, controlled by sys.use_fifo_ui property, that enables the
top app's UI and RenderThread to be SCHED_FIFO. This eliminates almost
all jank due to scheduling competition with non-UI critical
threads. This mode may not be suitable for all devices.
bug 24503801
Change-Id: I7b8a31830ad80f7efa00236928d5476998ed4e00
This patch adds in IpReachabilityMonitor a timestamp variable set
everytime that probeAll() send NUD probe requests to RTNETLINK.
This allows to distinguish between:
1) NUD_FAILED events resulting from such a forced NUD probe
2) "organic" NUD_FAILED notifications from the kernel
This distinction is added to IpReachabilityEvent as a one-bit flag.
This patch also changes the formatting of ApfProgramEvent flags to use
'|' as a joining character, similarly to other flags formatting.
Bug: 21859053
Change-Id: I24c64a3f17fa283eace5bd0a05c21a90a2305359
This patch partially undoes ag/869831 (Change-Id:
Ia42ed7aefaebd8caf3eada8e42b6cb7a940d7647) so that ConnectivityManager
does not remove callbacks from its internal request-to-callback map at
unregistration, but instead let the singleton CallbackHandler do it when
receiving a CALLBACK_RELEASED from ConnectivityService.
ag/869831 was thought to fix b/26749700 that reported a callback leak
from sNetworkCallback, but a finer analysis of the code shows that
callbacks were correctly removed by the CallbackHandler before
ag/869831. There was therefore no callback leak.
Bug: 26749700
Bug: 28537383
Change-Id: I421d889d0e225c0e3d1eebea664f44a1cc0f3191
The older Time class is deprecated and doesn't handle edge cases
very well. So migrate the cycle calculation logic to use the
long-standing and well-supported Calendar API.
Bug: 28689087
Change-Id: Ic1802b3f8556402f99bfea4cd625c35dfed81ac0
Use this to make sure that the VR thread belongs to a given process when
it is assigned.
bug 28715706
Change-Id: I4e5b0d8493e69e443eb907f0a6f2a9482fb64bac
We use 'this' for synchronization in NameValueCache but some code
that accesses the generation registry uses 'this' in a different
context ending up syncing on the wrong instance. This is why
sync on this is just a bad idea.
bug:29956424
Change-Id: Ide2d4f07a5f40cb3f0e8f50e4c8de216d15a31ee
Refactor slightly to forbid using myUserId in the zygote (uid == 0).
Also factor Environment to put user data (vs shared data) into its
own class, which should keep the page the shared data is on actually
shared between processes.
Bug: 29338430
Change-Id: I05d1306b57658a83299e38076171f56cb364ea80
This allows the MIME type icons to be used in other system apps.
In this instance, the StorageManager app needs to show file icons
for a given file. By moving the icon mapping into a static library,
we can avoid duplicating the mapping and resources.
Bug: 29826961
Change-Id: Ic2dc7970cc5a5ca430e33425eb9f1ecbd733fe9a
This patch adds lifetime durations of DhcpClient states to
DhcpClientEvents.
To record the duration of a state, the event is now recorded when the
DhcpClient state machine exits that state.
In addition this patch removes event logging of StoppedState,
DhcpState and DhcpHaveLeaseState.
Change-Id: Ibd37b5e3070f35113b6b45942b1e1ff19c27a90b
This patch
- adds a Builder class for RaEvent.
- uses this Builder class for correctly recording the minimum
lifetime seen for every ICMP6 options tracked, instead of
recording the last lifetime seen.
- adds unit test coverage for RaEvent logging.
Change-Id: I6443932f5cf7a613a5c695c65a60eab01e60602a
http://ag/1194313 broke unregisterNetworkCallback because the
system does not parcel the type of the request back to the app.
So when the app calls unregisterNetworkCallback, the
NetworkRequest that's passed in does not have a type and thus
doesn't match the request in mNetworkRequests.
Fix this by parceling over the type as well.
This was not caught by the unit test because the unit test all
runs in the same process with no parceling.
Bug: 23113288
Change-Id: I58b2ed651b9bf5cbdcca5b25c3ca24db53cffdf1
* changes:
Fix synchronization error for Seamless rotation.
Disable seamless rotation while animating.
Force CROSSFADE rotation when launching from double tap gesture.
When activity transition triggers a rotation change, the starting
window will normally be the top window at the time we try
to select the window animation. However, these layout params won't
have the apps rotation animation set (as the client code will set that
on the real window, not the starting window). Eventually we would
like to add API to specify rotation animation via manifest to solve
this problem cleanly. In the mean time, we can force a specific rotation
animation from the double tap gesture, and clean up some camera
ugliness. We accomplish this by attaching an animation hint to
ActivityOptions.
Bug: 28838855
Change-Id: If052cd8cbae76651da43f3b4c590cd9dcc1afc0f
This will allow us to simplify code that deals with
NetworkRequests outside ConnectivityService.
Bug: 23113288
Change-Id: I9b3a859d0c68cad73d7f6baa4b584d13ffd2ae36
With this CL, the system automatically grants a temporary URI permission
to the target application when the IME calls
InputConnection#commitContent() with
InputConnection#INPUT_CONTENT_GRANT_READ_URI_PERMISSION. The temporary
permission will be revoked by any of the following events:
- InputContentInfo#releasePermission() is explicitly called by the
target application.
- The target application returned false in
InputConnection#commitContent().
- All the InputContentInfo instances copied from the original one are
GC-ed.
If we do not do this and there is an application that forgot to call
that method then there is no way for IME developers to prevent
permission denial from happening in the application except for relaxing
the default permission of the ContentProvider just because of such an
application.
Although application developers are still expected to explicitly call
InputContentInfo#{request,release}Permission(), forgetting to call
InputContentInfo#requestPermission() does not hurt the user anymore.
With this CL, calling InputContentInfo#requestPermission() after calling
InputContentInfo#releasePermission() is also allowed.
Bug: 29892936
Change-Id: Id955435dd2e72549ee7134f46b3c6951581694ad
append() is used to optimized insertions in the array, but it must
preserve the order of the hashcode array; when it doesn't, it falls back
to append(), but it should not log a warning message
In particular, PendingIntentRecords might have different hashcodes
across different processes.
Fixes: 29912192
Change-Id: I0ab566249829ddb934fd51cf21399b68cb286bd5
Fixes a case where notification header text could go missing
if a notification view was recycled and previously had a header text.
Reapplying only hid the text without clearing it, so the extraction
logic thought it was still there and hid the text for the children
even though it was not showing for the parent.
Change-Id: I3f96e1e7bebb2f815020d278ad13b2b5d948e63c
Fixes: 29915184
