On modern versions of Android running in AOT mode
FloatMath is slower than Math. Calls to Math.sqrt(),
etc. are replaced by intrinsics which can be as small
as a single CPU opcode.
When running in interpreted mode the new
implementation is unfortunately slower, but I'm
judging this acceptable and likely to be improved
over time. This change saves a small amount of native
code.
Example timings:
Mako AOSP AOT:
Method: Original / New / Direct call to Math
ceil: 596ns / 146.ns / 111ns
sqrt: 694ns / 56ns / 25ns
Mako AOSP interpreted:
Method: Original / New / Direct call to Math
ceil: 1900ns / 2307ns / 1485ns
sqrt: 1998ns / 2603ns / 1788ns
Other calls Mako AOT:
Method: Original / New
cos: 635ns / 270ns
exp: 566ns / 324ns
floor: 604ns / 150ns
hypot: 631ns / 232ns
pow: 936ns / 643ns
sin: 641ns / 299ns
The advice to use Math directly, in preference to
FloatMath, is still good. FloatMath will be deprecated
separately.
Bug: https://code.google.com/p/android/issues/detail?id=36199
Change-Id: If07fcbd78543d13bc6d75f9743f999860e8d58d7
This enables a change in Settings that allows device
encryption to be disabled by the user.
Fixes bug 17881324
Change-Id: I34dfc586df1a598bf969dc82adf8537a2730d345
Updating the accessibility layer behavior to reflect the new
model where accessibility no longer overrides strong encryption.
Now enabling an accessibility service lowers the encryption
level but the user can bump it up in settings if desired.
bug:17881324
Change-Id: Ic60d760c267d3f934040a42e1963b179bd8b9f5f
DPM's method will return false if encrypted by default password,
preventing the changing of encryption password to lockscreen password.
Check if the device is encrypted by some means, instead.
Also fix a SecurityException when Device Admin queries encryption state
(recent regression)
Bug: 17881324
Change-Id: Id897e61c5e254ab3f8dc569285428a73005303ea
To clarify that BluetoothLeAdvertiser object will return null
when BT is off OR if the hw doesn't support these capabilities
bug: 18006072
Change-Id: I635d7971711a3cae7c58f7a0636faf9a03f19970
Some of the admin policies are throwing security exceptions in
a managed profile without being documented correctly and others
shouldn't be throwing security exceptions.
Changed setCameraDisabled() to not throw an exception. It now just
prevents work profile apps from using the camera.
Changed wipeData() to allow passing in ERASE_EXTERNAL_STORAGE. In
secondary users/profiles, this is just going to remove the user, so
the flag is harmless.
Updated documentation for setKeyguardDisabledFeatures() and resetPassword()
to indicate that they cannot be called in a managed profile.
Bug: 17987913
Change-Id: I8060be4c2d32bdd4edb46ce543551fabb9c8c983
When restoring hundreds of apps on low-DPI devices, we end up sending
icon Bitmaps inline in the response instead of splitting into ashmem
regions. To avoid triggering TransactionTooLargeException, switch to
using ParceledListSlice under the hood.
Bug: 17926122
Change-Id: Ib4da6775e79d2fcb4aaea15f58ed998df203a5f9
On modern versions of Android running in AOT mode
FloatMath is slower than Math. Calls to Math.sqrt(),
etc. are replaced by intrinsics which can be as small
as a single CPU opcode.
When running in interpreted mode the new
implementation is unfortunately slower, but I'm
judging this acceptable and likely to be improved
over time. This change saves a small amount of native
code.
Example timings:
Mako AOSP AOT:
Method: Original / New / Direct call to Math
ceil: 596ns / 146.ns / 111ns
sqrt: 694ns / 56ns / 25ns
Mako AOSP interpreted:
Method: Original / New / Direct call to Math
ceil: 1900ns / 2307ns / 1485ns
sqrt: 1998ns / 2603ns / 1788ns
Other calls Mako AOT:
Method: Original / New
cos: 635ns / 270ns
exp: 566ns / 324ns
floor: 604ns / 150ns
hypot: 631ns / 232ns
pow: 936ns / 643ns
sin: 641ns / 299ns
The advice to use Math directly, in preference to
FloatMath, is still good. FloatMath will be deprecated
separately.
Bug: https://code.google.com/p/android/issues/detail?id=36199
Change-Id: I8d1947d88b3c576643138b1df589fb9da7c1ab88
We used to set this from LoadedApk.getClassLoader. This had
the unfortunate side effect of clobbering the threads default
classloader when Context.createPackageContext().getClassLoader()
was called. This happens, for example, to every app that uses
a updatable webview.
bug: 17986934
Change-Id: If2b5938a19f836ea5fe60c94ea53a7125c2098c0
The system should always be using new startActivityAsCaller() when
starting activities on behalf of someone else, to ensure that
security checks are enforced as the original caller.
Bug: 17983737
Change-Id: Ic40816a797cfdb13c0adb48b86ed4ed7d6aae8eb
