Some ANR bugs are very difficult to localize because by the time
the bug report has been captured, the relevant information is
gone. Work around this by capturing a log of the input dispatcher's
state at the exact time of the ANR before anything has changed
and include this information in the bug report.
Also fixed a nit related to some format strings that had
improper field widths specified.
Bug: 6680398
Change-Id: I5323bf18ec5e47a767cd053209753cc46852bf4c
...in Developer options is on
Don't respect stay awake while on as long as a time to lock limit
is being enforced. When we start enforcing one, make sure the
setting is off (since we won't be respecting it anyway).
Bug: 6664140
Change-Id: Id07cb528afa0c64c7766341841c51771f507121d
Call noteWifiScanStartedFromSource() when a scan is started.
Call noteWifiScanStoppedFromSource() when a scan is finished.
The current implementation tracks to UID that requested the scan, and
correctly tracks the duration of the scan. It ignores scan requests
that occur when a scan is already in progress. It does not distinguish
between active and passive scans.
Repurpose all the noteScanWifiLockAcquired/Released() plumbing
for WIFI scan tracking. The WIFI scan locks were never reported
to the user. Use noteFullWifiLock() when WIFI scan locks are used -
this makes sense because the power draw for a WIFI scan lock
should be about the same as for a full WIFI lock.
Bug: 6642581
Change-Id: Ida6e87992853698545b89f875c973a239218317d
Weren't cleaning out any ActivityOptions that are still attached
to a finishing activity.
Bug: 6636731
Change-Id: If0520bbcbf1d4ce19d46ff769918893cefda9c87
People generally expect, if they are using FLAG_KEEP_SCREEN_ON,
that the screen won't immediately dim after it is cleared, even
if it has been passed the user activity timeout since the last
user interaction. So include the flag to reset the user activity
timeout when releasing its wake lock.
Change-Id: If7a8fea8faef3edbf13dff10a2f248adc9e3ff0b
The foreground activity is stopped when the device goes to sleep,
and started again when the device is unlocked. We now distinguish
this case from a "normal" stop, and do not finish() a foreground
noHistory="true" activity inappropriately when the device sleeps.
We also detect the case where an activity is started while the
device is still asleep, in which case the foreground noHistory
activity is cleaned up as part of bringing the new activity to
the foreground.
Bug 6657549
Change-Id: I9c6a0830aed0e47e4207b62803b90067c8486112
Tethering was re-evaluating everything on any connectivity change.
This was causing problems when mobile data was not available as each
re-eval was again requesting a connection which would fail and loop.
Ignore FAILED states to break this cycle in a minimal change.
bug:6668233
Change-Id: Ia1dfc6abff1fe464e9501b50185cdb8d0fbb7ff6
Restricting to pkg="android" didn't filter out things like
open wifi networks, etc. So now we have a whitelist:
notifications must be sent the "android" pseudo-package,
*and* they must have one of these "kind" tags:
- android.system.imeswitcher (IME switcher, needed by SUW)
- android.system.update (OTAs)
Note that OTAs currently use a fullScreenIntent, so they
bypass this logic anyway, but for consistency's sake we now
allow OTA icons in the status bar explicitly.
Bug: 6645469
Change-Id: Ib2e2f22d7a0817a1acaf8137ed4f3c7d3ddf8af5
- Sometimes the black background would flash; changing
animation durations to make this much less likely
- Fixing issue in Recents where we sometimes forgot
to disable drawing caches on views after enabling them
1. Accessibility allows querying only of the active window.
The active window is the one that has input focus or the
one the user is touching. Hence, if the user is touching
a window that does not have input focus this window is
the active one and as soon as the user stops touching
it the active window becomes the one that has input
focus. Currently the active window is not updated properly
when the user lifts his finger. This leads to a scenario
of traversal actions sent to the wrong window and the user
being stuck.
The reason is that the last touch explored event that is
used to determine where to click is cleared when accessibility
focus moves but this event is also used to determine when to
send the hover exit and touch exploration gesture end events.
The problem is that the last hover event is cleared before
it is used for sending the right exit events, thus the event
stream is inconsistent and the accessibility manager service
relies on this stream to update the active window. Now we
are keeping separate copies of the last touch event - one
for clicking and one for determining the which events to
inject to ensure consistent stream.
bug:6666041
Change-Id: Ie9961e562a42ef8a9463afacfff2246adcb66303
1. We are deciding whether the user is performing a gesture or an exploration based
on the gesture velocity. If we are detecting gesture we do the recognition at the
gesture end which is when the finger goes up. This is better than having a mode
toggle gesture for exploring and gestures detection. However, it is possible that
the user really wanted to perform an exploration but was moving too fast and
unless he lifts his finger the device is in gesture detection mode. This is
frustrating since the user has no feedback and assumes exploration does not
work.
We want to perform gesture detection only for a maximal time frame and if the
user did not lift his finger we transition into touch exploration state.
bug:6663173
Change-Id: I954ff937cca902e31b51325d1e1dfce84d239624
