Add a test for emptiness before accessing either mTaskHistory[0] or
TaskRecord.mActivities[0]. This will keep us from hitting
IndexOutOfBoundsException.
Fixes bug 10789624.
Change-Id: If726df888a2c8b393788793b6220a6bffe2df883
The InputMethod window was being assigned to the bottommost
window as a fall-through situation when no window could receive
input. This fix changes that so that if the bottommost window
cannot receive IME input then the InputMethod will be unassigned.
In most ANRs associated with bug 10689184 mInputMethodTarget
was assigned to the Wallpaper window and the InputMethod was
placed directly above it. When the ANR occurred the window that
was awaiting focused was then always placed immediately above the
InputMethod. This fix will keep that situation from happening.
Change-Id: Ic247e8132a907f2712a9f8a89e43c099142ec851
NPE at com.android.server.am.ProcessRecord.resetPackageList(ProcessRecord.java:596)
Take care of some more cases now that baseProcessTracker can be null.
Change-Id: I394c0b7802788118c3ad6bcac5dfdd23eeda8d58
...while setting up a new user from settings.
The delayed service start stuff was too aggressive -- it would
allow a process to be killed between the an onReceive() that calls
startService() and that service being started. This means that
apps that set up global state that they expect to remain set up
during that time could be lost.
This is the first part of a fix, which tightens up when we allow
services to be delayed. Now we will immediately start the service
as long as it currently as a process running that is not in the
cached state. (Previously we would delay if the process was in
the receiver state.)
This unfortunately means that our service start delay is much
less effective. To address that, there will be a follow-on change
to tie broadcast delivery into this to see if we can delay the
finish of a broadcast as long as there are background services
starting in that process.
Change-Id: I2bba2295d10699ee3479375bbe87114b2cbb0826
We now expect that the transport service's host process will be
OOM-killed sometimes, but will come back automatically because we
still hold the binding. So, we no longer drop it as the
expected-to-be-current transport when that happens.
Bug 10728767
Change-Id: I5e756e8942e7c4c3567632f10460ee31b9618d75
1. The UI of a printing app was freezing a little when calling the print
method since the print manager service was waiting for it to bind to the
print spooler which generated the print job id (and the initial print
job info really). Now the print manager service is responsible for job
id generation and does not not wait for the print spooler to spin. Hence,
the app UI is not blocked at all. Note that the print manager initiates
the binding to the spooler and as soon as it completes the spooler shows
the print UI which is hosted in its process. It is not possible to show
the print UI before the system is bound to the spooler since during this
binding the system passes a callback to the spooler so the latter can
talk to the system.
2. Changed the print job id to be an opaque class allowing us to vary the
way we generate print job ids in the future.
3. The queued print job state was hidden but the print job returned by the
print method of the print manager is in that state. Now now hidden.
4. We were incorrecly removing print job infos if they are completed or
cancelled. Doing that is problematic since the print job returned by
the print method allows the app to query for the job info after the
job has been say completed. Hence, an app can initiate printing and
get a print job whose state is "created" and hold onto it until after
the job is completed, now if the app asks for the print job info it
will get an info in "created" state even though the job is "completed"
since the spooler was not retaining the completed jobs. Now the spooler
removes the PDF files for the completed and cancelled print jobs but
keeps around the infos (also persisting them to disc) so it can answer
questions about them. On first boot or switch to a user we purge the
persisted print jobs in completed/cancelled state since they
are obsolete - no app can have a handle to them.
5. Removed the print method that takes a file since we have a public
PrintDocumentAdapter implementation for printing files. Once can
instantiate a PrintFileDocumentAdapter and pass it to the print
method. This class also allows overriding of the finish method to
know when the data is spooled and deleted the file if desired, etc.
6. Replaced the wrong code to slice a large list of parcelables to
use ParceledListSlice class.
bug:10748093
Change-Id: I1ebeeb47576e88fce550851cdd3e401fcede6e2b
Haven't found the underlying cause, but this will give us more
information when we get into the bad state.
Change-Id: I9aebd3a025a7c0d931f43098461b64ee3c220746
Bug: 9520957
DevicePolicyManagerService will play dumb if the feature is not installed.
Continue to keep track of failed password attempts for keyguard's use.
Change-Id: I28d258dc09a8b4976b188da6f453d8daabcc4bdd
We now have the activity manager kill long-running processes
during idle maintanence.
This involved adding some more information to the activity manager
about the current memory state, so that it could know if it really
should bother killing anything. While doing this, I also improved
how we determine when memory is getting low by better ignoring cases
where processes are going away for other reasons (such as now idle
maintenance). We now won't raise our memory state if either a process
is going away because we wanted it gone for another reason or the
total number of processes is not decreasing.
The idle maintanence killing also uses new per-process information
about whether the process has ever gone into the cached state since
the last idle maintenance, and the initial pss and current pss size
over its run time.
Change-Id: Iceaa7ffb2ad2015c33a64133a72a272b56dbad53
