We now use Mapped to not double-count cached pages that are
mapped in to app processes. Also read in some more values that
count towards kernel RAM use, and count buffers as free rather
than used RAM.
It also looks like the zram accounting is broken -- it was never
collecting the total zram reserved space. I need to figure out
how I was finding that before.
Change-Id: I883f6fc93966774b5c7d2801d1801666dd11ed41
- Add docs to Binder, Messenger, ResultReceier to explain their
relation (or lack there-of) to process lifecycle.
- Clarify some aspects of process lifecycle for services.
- Fix help text of am command.
- Fix per-package dumping of battery stats to not include history.
- Fix per-package dumping of proc stats to only include aggregated
and current stats and fix some formatting.
- Fix per-process dumping of meminfo to have an option to interpret
the input as a package, so including all processes that are
running code of that package.
- Fix top-level per-package debug output to correctly include all
of these improvements and give them a little more time (10s) to
complete for timing out.
Change-Id: I2a04c0f862bd47b08329443d722345a13ad9b6e2
When creating a user via the UI, disallow phone calls and SMS by
default. Primary user must explicitly enable it via Settings.
Bug: 17832802
Change-Id: I18cad4be8493ddc8890b5d90da2df256cb3f1ec9
Fix Slog.wtf to not acquire the activity manager lock in its code
path, so that it can never deadlock. This was the original intention
of it, but part was missed.
Now we can put back in the code to detect when strict mode data is
getting large (a little more targeted now to the actual problem),
and use Slog.wtf to report it. And as a bonus, when this happens
we will now clear all of the collected violations, to avoid getting
in to the bad case where IPCs start failing. So this should be
good enough for L to fix the problem, with wtf reports for us to
see if the underlying issue is still happening.
Finally, switch a butch of stuff in the system process from Log.wtf
to Slog.wtf, since many of those are deadlocks waiting to happen.
Oh and fix a crash in the settings provider I noticed in APR.
Change-Id: I307d51b7a4db238fd1e5fe2f3f9bf1b9c6f1c041
Drop down the limit on when we log, since under normal operation we
will never get more than a few K of data due to strict mode.
Try to clean up the code paths coming in and out of binder IPCs to
plug any places where we could disrupt the gather flag of a thread,
causing it to keep gathering stack crawls (which is the thing that
is causing our strict mode data to become so large).
Change-Id: I9a46512283d33e863c429840b465855d1fabb74e
32 bit processes will always see values from SUPPORTED_32_BIT_ABIS
and 64 bit processes will always see values from SUPPORTED_64_BIT_ABIS.
bug: 17367265
Change-Id: I9996034c190f9213986b9b1a4ee0e38eca91c7ad
Remove the checks for large parcel sizes. Those were triggering,
and identifyng the area of the problem, but also cause a lot of
trouble by making the unsafe deadlocky code there much more likely
to deadlock.
Add logging for strict mode IPCs, since those seem to be the
problem. Only log when things look bad.
Also add a log when battery stats are reset, to diagnose why they
are getting reset when they shouldn't be.
Change-Id: I588c858fb8d8c45f3c9c164ae2de9ae01547b304
Add the app directory to the arguments for starting a process.
Add a check for NeedsNativeBridge and a call to PreInitializeBridge
in the native fork code.
(cherry picked from commit 2eacd06bfb)
Bug: 17671501
Change-Id: I970db5b284b0c12e2d8a45df3950c1fff2927a4e
Now that we support unreachable routes, use those to block
address families on VPNs. This is a much more elegant solution.
Also update LinkProperties when IP addresses are added and
removed, fixing a TODO.
Bug: 17462989
Change-Id: Ib749d84710dca70d672350b9f129bb91419ec77e
In addition to device owners, profile owners on the primary user
can also set user restrictions that are necessary to lock down the
user.
This is to enable the case of a profile owner registered after setup
wizard is completed, on the primary user.
Also make managed profile vs. profile wording consistent in the
DevicePolicyManager docs.
Bug: 17555025
Change-Id: Ib9d08b8af34a99b25e11757fa7dc83673a7deb32
- Add an explicit power manager call to set the low power mode state,
instead of trying manage everything around a single setting.
- When low-power mode is triggered by falling below the configured
threshold, it does not update the setting.
- The "is-enabled" api returns setting || below configured trigger.
- Move the snooze management into the new api call.
- Callers (sysui + settings) updated to use the api instead of the
setting.
- Handles the case where the level does an unpowered leap out of the
low battery level. (Possible if powered in-between while the device
is off)
Bug:17460535
Change-Id: Ic030504c9cad9868a7137abbe837b170da37852b
Right now different code in System UI, Settings app and other places replace the
user icon with their own default. This tries to make it consistent by moving the
mechanism used in Settings in a helper class.
Bug: 17311038
Change-Id: Ic858c65bf82a98b9806dbba029e7cdcf441f372e
Even though Shell user is allowed to perform cross-user actions,
lock that path down if the target user has restrictions imposed by
the profile owner device admin that prevents access via adb.
If the profile owner has imposed DISALLOW_DEBUGGING_FEATURES, don't
allow the shell user to make the following types of calls:
start activities, make service calls, access content providers,
send broadcasts, block/unblock packages, clear user data, etc.
Bug: 15086577
Change-Id: I9669fc165953076f786ed51cbc17d20d6fa995c3
This can be controlled by MDMs via DPM.
Also fixes:
- javadoc for restrictions
- persisting of cross profile copy/paste restriction
Bug: 17387303
Change-Id: Ie148f56189181d2a4c6345c0823d417ab13a94a3
For restore use-case, session creation needs to complete quickly, so
delay ASEC allocation until session is opened. When preflighting
size checks, only consider external when we have a known size for the
container. Also relax size checks when using MODE_INHERIT_EXISTING
on external, since we don't know how much of existing app will be
copied over.
Consider session as "active" while commit is ongoing, until we're
either finished or pending user interaction.
Always publish first client needle movement away from 0. Use 25% of
internal progress to reflect ASEC allocation.
Avoid CloseGuard messages about leaking PFDs.
Bug: 17405741, 17402982
Change-Id: I6247a1d335d26621549c701c4c4575a8d16ef8c2
Dumb, dumb, dumb mistake.
Also fix battery stats wakeup reason tracking to use a SamplingTimer
(like kernel wake locks) so we can track both the duration and count
for each wakeup reason.
Change-Id: I89d69661006dc533622b1b7e68a139166d3a6975
