Since the clinits may execute methods, we should avoid preloading
classes to avoid executing methods that may not be required. These
samples could cause inaccuracy in the boot image profile.
Test: manual: adb logcat | grep preloadClasses
Bug: 139883463
Change-Id: I5b4568a477724606105196cba010109f80eecec1
The service would query package manager with the provided userId, and
will return true if the package is not visible.
Test: locally, b/143129258
Change-Id: Iab8ddaa76358d5fad63e10d7c4c3f92e2a0f51a2
Merged-In: I92ea650b49743c388bff9943a7ec620e3d61a5d6
Separate the decision on whether to log to two separate ones - logcat
and statsLog to allow separate policies.
Also add startLogAll and endLogAll in preparation for an adb command
that will enable/disable logging to logcat every time.
Test: atest ChangeReporterTest
Bug: 138374585
Bug: 140910281
Change-Id: Ie49368b838a19845f51a2670035f611d3e4b9a1b
This CL adds two new methods to the IPlatformCompat binder interface for
toggling multiple compat changes at once, and for clearing all the
overrides for a given package name.
Bug: 140367850
Test: http://aosp/1113771
Change-Id: I02d08d76e42da308856408dadc2c8c73c7ff20d1
Store the reported changes by app and clear the app info on app launch.
This means that changes reported to the system server would be logged
once per app launch instead of once per device boot.
Bug: 138374585
Test: flash device, run test app
Change-Id: I24f5dc125488c6af3164b6503f236bc3a24e6292
Merged-In: I24f5dc125488c6af3164b6503f236bc3a24e6292
A side effect of CL topic r.android.com/q/topic:bug136256059_attempt2
was that the default MimeMap instance is now loaded eagerly (pre-fork)
regardless of whether any app process ends up using it, whereas
previously it was loaded lazily (post-fork) the first time an app
process used it.
This traded off CPU and memory use post-fork vs. pre-fork. Because it
isn't known / hasn't been proven whether and how many app processes
use that instance, the net CPU and memory impact is not known: if at
least one process makes use of a relevant API, the net impact will be
neutral or positive, or else negative.
Until this can be investigated and undergo system health review, this
CL topic changes the behavior back to post-fork lazy loading of the
default MimeMap instance.
Bug: 142019040
Test: Treehugger
Change-Id: I2a3f244240c51e7ada1f41a8abadb946ebcab40b
Another CL in this topic moves libcore.net.MimeMap to package
libcore.content.type. This CL updates associated references.
Bug: 136256059
Bug: 124232146
Test: Treehugger
Test: CtsMimeMapTestCases
Exempt-From-Owner-Approval: move affecting imports only
Change-Id: I30aac8b28a1740180ea01834a2bb66b61cb255d4
The zygote handles requests by polling on its sockets, and then handling
one request for each active socket. However, it does so by reading from
a socket via a BufferedReader, which means that if multiple requests are
written into the socket before the zygote gets a chance to read them,
the zygote reads multiple requests into its BufferedReader, it handles
one request, and then never responds to the request that's buffered,
leaving its client stuck waiting for a response that will never happen.
For most requests, this can't happen, because the client will wait for a
response to be sent from the zygote before sending another request, but
this isn't true for --boot-completed until this patch.
Bug: http://b/141767463
Test: forrest runs of apct/text/text_native_test-cloud-tf
Change-Id: I8b7a80abfd9443d98f8cf5aedb7669b82c0cb84a
