As per a comment on an earlier code review.
(cherry-picked from commit a9d64733421d6765eab5c2730fa912f068e26047)
Change-Id: I064cffc13c323b721f3a16c83e0e95ee348ef9f6
It's better not to cancel display lists updates. In the emulator, if we use
low precision clock source like refined-jiffies, the display list which
should be updated with a new frame, won't be updated due to the delta time
is 0 . And this makes all the hardware accelerated GUI out of order.
Change-Id: I41dba8965c96b6ad1b6d326ad1d6fa4a0a49173f
Signed-off-by: Tina Zhang <tina.zhang@intel.com>
If one of the ABI list system properties is undefined/empty (as
ro.product.cpu.abilist64 is on a 32-bit only device), getString returns
"unknown", which ends up creating a 1-element array with "unknown" as a
member. Fix this to instead just get the empty string and split that
into a 0-element array.
Change-Id: I0d0a54eb06bb04427bcf0487e2a16d4180b81116
The zygote that's responsible for starting up the system server
now checks if there's another zygote on the system, and waits
for it to start up. Also, a few minor clean ups :
- Address a long standing TODO about zygote retries.
- Have functions throw IOException where appropriate and
wrap them in ZygoteStartFailedEx with a filled in cause.
bug: 14869939
Change-Id: I9e514659b79b3d2c98a4c5f93c0c376843f6c881
As the CursorWindow will be preloaded by zygote, and there is one static
member need use the system resource to initialize. But actually the zygote
do not preload the resource now. (In fact, the zygote will preload classes
first, then preload the resource.) We need ensure the zygote to create the
system resource first.
So change to init this static member as -1, and it will be evaluated when
first used.
Change-Id: Ibccb84e3c410c73788ac682fe76b720306e81dc4
- Pass down the app's instruction set to dexopt so that
it can compile the dex file for the right architecture.
- Also pass down the app's instruction set to rmdex, movedex
and getSize so that they can construct the cache file
location properly.
- Temporarily compile "system" jars such as am,wm etc. for
both architectures. A follow up change will ensure that
they're compiled only for one architecture (the same
arch. as the system server).
- Java "shared" libraries are now compiled for the right
architecture when an app requires them.
- Improve the app native library ABI detection to account
for system apps installed in /system/lib{64}/<packagename>
and also handle sdcard and forward locked apps correctly.
(cherry-picked from commit b4d35dc8e9702f9d0d82d35a105f0eea35672b52)
Partial cherry-pick of changes 4ca728c0 and 21de56a9, which
can't be cherry-picked due to their large surface area.
Change-Id: Ife46e150d360cd5241dea93863141749233c1805
- Differentiate between clean exits and signals.
- Change the logic for the waitpid warning. We should log it
only if the *first* call to waitpid returned ECHILD.
Change-Id: I6bc50e4ffd100fb7f3f526d6b4056bed10b8045f
Originally the api 'getCropAndSetWallpaperIntent' does not check
whether the parsed parameter is a NULL pointer, as leads to the
whole app crash if it call this api with a null parameter accidentally.
A null pointer check logic is added into its original parameter check
logic, so that this API will throw an IllegalArgumentException if it
is parsed with a null pointer
Change-Id: Ib7de40e571419d09e1a744edc969eb7162766b75
System.gc doesn't always do a GC unless to attempt to run the
finalizers after you call System.gc.
Bug: 14325353
Change-Id: Iebed7fad5576d610cea13a86e7d3d46652c67478
- critical issue with using %lld to scanf into a jlong
- subsidiary issues with mix jint, jsize and jlong printing
Change-Id: I77c6b3d3a202fec741cacc540fc39ad6cba25f52
