This removes all GlobalRef allocation as part of building BinderProxys.
Previously these were used to map IBinders to the corresponding
Java object, so the Java objects could be reused. We now keep
that mapping at the Java level.
This means we often need to call into Java to look up or allocate
a BinderProxy. But this replaces a prior call to Java to dereference
a WeakReference. The Java custom Java map-to-WeakReference data
structure is probably not terribly efficient, but the original
attachement mechanism did not seem to be either. And this
avoids potentially even more catastrophic issues when the number
of GlobalRefs approaches its limit.
We decrease GC triggering frequency from 200 to 1000 allocated
references. This now only applies to other kinds of JNI References
allocated by Binder.
I saw a maximum bucket size of 16 for the ProxyMap data structure
while briefly exercising a freshly booted device. That occurred
in system_server.
Bug: 65760710
Test: Built and booted master with some debugging output. Looks sane.
Change-Id: I322c4d8e9c8e198586d591580c2cdbb094906677
Change the Java BinderProxy to only contain a single native pointer,
so that we can get by with a single NativeAllocationRegistry
registration. This adds some indirections and a new allocation. But it
marginally reduces the number of (expensive) JNI field lookups from
native code, and the extra allocation involves significantly less
overhead than registering each object twice. This also cleans up the
code a little by avoiding some explicit reference count adjustments.
Change BinderProxy Binder to use NativeAllocationRegistry instead of
finalize().
Change the mObject field in Binder to hold a non-reference-counted
but owning pointer to JavaBBinderHolder. Have JavaBBinderHolder no
longer inherit from RefBase.
Make it clear that neither Binder.mObject, not BinderProxy.mNativeData
can be null. Remove null checks.
Avoid checking for null returns from C++ new. It would throw
anyway, which would cause the process to abort.
Test: Booted master.
Bug: 65760710
Change-Id: I323d4bdc7e25f8c27b847b6fe2c073eac3f2efe5
Also add appropriate @NonNull and @Nullable annotations.
Test: built
Change-Id: I22de48105ef685baf594cfc004dd3e27e2ba09e9
Merged-In: I22de48105ef685baf594cfc004dd3e27e2ba09e9
(cherry picked from commit 4cd650c008)
getService: wait for service if it is declared in the manifest
tryGetService: only return if the service is immediately available
Bug: 67981006
Test: hidl_test_java
Change-Id: I4485b84f0fde98851cf5f64d198a8c5410795c8c
We now disallow non-UTF8 property values directly in property_service,
so there is no need for this warning.
This partially reverts 33aea8d40e.
Bug: 63177684
Test: make
Change-Id: I61f49f635c6095013f55b906e00925a8cbd1ea8c
Instead of using the whole slab mem for kernel usage, split the
unreclaimable to kernel and reclaimable to cache (since is freed
under mem. pressure).
Test: tested on gobo
Bug:67753120
Change-Id: I0f5a310bb88603ad7bb28e5398ea57c249c04fc2
Cleanup of Cellular/Wifi aggregate statistics in Batterystats dumpsys
Cellular
- Modification of field names to improve clarity
- Removed count "x" values
- Added power values to Rx signal strength
- Added power values to Tx power
- Added "Cellular Sleep time"
- Removed erroneous battery field (This is being fixed in b/67213886,
b/67213967)
- Removed the following fields (Mobile radio active unknown time, Mobile
radio active adjusted time, Signal scanning time)
Wifi
- Modification of field names to improve clarity
- Removed count "x" values
- Added power values to Rx signal strength
- Removed the following field (Wifi on, Wifi running)
BUG:67322291
Change-Id: I9630f154a73c7e3cb4bae4779609302f3674602f
Check whether the guard is null to avoid:
Uncaught exception thrown by finalizer
java.lang.NullPointerException: Attempt to invoke virtual method 'void dalvik.system.CloseGuard.close()' on a null object reference
at android.os.ParcelFileDescriptor.closeWithStatus(ParcelFileDescriptor.java:740)
at android.os.ParcelFileDescriptor.finalize(ParcelFileDescriptor.java:990)
at java.lang.Daemons$FinalizerDaemon.doFinalize(Daemons.java:250)
at java.lang.Daemons$FinalizerDaemon.runInternal(Daemons.java:237)
at java.lang.Daemons$Daemon.run(Daemons.java:103)
at java.lang.Thread.run(Thread.java:764)
Follow-up to commit da5a3e12f4.
Bug: 7426029
Bug: 10330121
Test: m
Change-Id: I903f1545ab784008727ff23bb95fe182bd95b62a
Fixes: 64899521
Test: manual - flash build, reset batteryStats, use device > 1 hr
with alternating pattern between screen on/off/AOD. Dump stats, check
all screen related stats look normal, esp. record matches actual time
spent in each screen state. In raw bugreport:
Search "amount discharged" for % discharge;
Search "Screen on/off/doze discharge" for mAh discharge;
Search "time on battery" for up/real time in each state.
Test: Added two unit tests for note AOD screen state
Merged-In: I7193a36751124dd380818b2b665303c0f0d8c984
Change-Id: I51cead7f92abd9e4c620f7dfde393993cdad494e
This only worked on broadcom devices, and was superseded in
M by a wifi HAL call made by IpManager.
Test: bullhead builds, boots
Change-Id: I711cae7dafe171c2c8b4e84a229adbcad27f3d14
Also fix up imports to make repohooks happy and some whitespace issues.
A very small step towards making this code more understandable.
Bug: 65760710
Test: Builds.
Change-Id: I0396c06bb303e0b06ad0cbbbff4fdc1ac527ac6c
Without this test, someone could trick us into constructing other
shady classes.
Test: builds, boots
Bug: 65281159
Change-Id: If678d0681708d1b0dcf056aa1133830ad3dbce31
