Grant Limberg
ab34884e3e
Set -O0
...
anything above that currently goes kaboom on ARM platforms
2015-07-03 18:14:50 -07:00
Grant Limberg
0b7cd2f40a
change some build settings. add a lock
2015-07-01 20:26:14 -07:00
Grant Limberg
f803dd5ff4
...
2015-07-01 18:13:39 -07:00
Grant Limberg
aee8e95d49
logging
2015-06-30 19:21:46 -07:00
Grant Limberg
c9919cc5ba
reflect changes to ZT1_PeerRole in JNI
2015-06-26 18:26:57 -07:00
Grant Limberg
bfb152f53f
configure the NDK to build all supported ABIs and package them up in the jar
2015-06-24 20:31:22 -07:00
Grant Limberg
759d71037e
added Comparable interface to VirtualNetworkConfig so we can sort arrays containing it.
2015-06-23 23:03:02 -07:00
Grant Limberg
adf89901c4
dont consider multicast groups in VirtualNetworkConfig.equals()
2015-06-19 19:00:06 -07:00
Grant Limberg
dfdd46db80
Merge branch 'adamierymenko-dev' into android-jni
2015-06-17 18:42:23 -07:00
Grant Limberg
abbcb0a12c
Modified ant build script so that it can be integrated with Android Studio's build system
2015-06-13 14:38:04 -07:00
Grant Limberg
472206dfb2
Rename JniCache to JniLookup
...
Removed caching capabilities as the cached methods, fields, and objects appears to be broken on Android
2015-06-10 20:16:13 -07:00
Grant Limberg
7e84f5a7db
killing whitespace
2015-06-09 23:24:54 -07:00
Grant Limberg
6889fcfc28
Looks like it was the JNI cash causing the crash.
...
Forcing it to look up classes and methods instead of caching them stopped the crashes in the GC. Will investigate more later.
2015-06-09 23:24:47 -07:00
Grant Limberg
4dc0ff8f13
Replace calls to Get<Type>ArrayElements with GetPrimitiveArrayCritical.
...
This puts code accessing the data in a critical section so that the GC cannot run while JNI has access to the array. This helps with stability somewhat, but I'm still getting some crashes in the GC
2015-06-09 23:12:44 -07:00
Grant Limberg
ced040c503
Logging and adding .equals() methods to MulticastGroup and VirtualNetworkCofnig
2015-06-09 19:38:05 -07:00
Grant Limberg
7cc64c5cb6
Might help to set the enabled field on a VirtualNetworkConfig object :)
2015-06-03 21:29:19 -07:00
Grant Limberg
b84dba3ecb
more logging
2015-06-03 21:29:07 -07:00
Grant Limberg
8fb9df9751
delete dead test code
2015-06-03 18:35:38 -07:00
Grant Limberg
fb0bf52b1b
Fixed sending a port number of 0 to ZT1.
...
Fixed VirtualNetworkConfigFunction (creating array to send to Java)
Fixed the creating InetAddress instead of InetSocketAddress in virtual network config callback
2015-06-01 20:03:28 -07:00
Grant Limberg
74f7d5377d
Enable ZT_TRACE and more verbose build output
2015-06-01 20:01:37 -07:00
Grant Limberg
5015c0ae78
more logging
2015-05-28 21:03:37 -07:00
Grant Limberg
b6196a53de
Fix DataStoreGetFunction.
2015-05-28 20:37:16 -07:00
Grant Limberg
58ca657da6
Use object arrays rather than ArrayList. JNI side
2015-05-28 20:36:54 -07:00
Grant Limberg
7822616966
Replace ArrayList instances with []
2015-05-28 20:01:30 -07:00
Grant Limberg
4c680e3126
LOTS more error checking and logging.
...
Running into a wall attempting to add items to an ArrayList from the JNI side
2015-05-27 20:42:54 -07:00
Grant Limberg
09631ad382
fix datastore get call in JNI
...
Just delete the local reference rather than do an array delete. Array delete was causing a crash on loading network config files. Deleting the local ref works fine, though, as it still lets the JVM garbage collecter know that the native code is done with the array
2015-05-26 21:32:56 -07:00
Grant Limberg
2d67196667
added the android verbose log level and moved most jni debug messages to verbose
2015-05-26 21:31:34 -07:00
Grant Limberg
9755dad7ff
add a missing break from a switch
2015-05-26 21:30:33 -07:00
Grant Limberg
adaf9cf32b
still getting re-used to all the places you have to put access modifiers in Java
2015-05-26 20:03:47 -07:00
Grant Limberg
6e998efd15
fixing up data store stuff
2015-05-23 18:29:16 -07:00
Grant Limberg
1e043a3f66
Remove link desperation from java API
2015-05-21 19:34:19 -07:00
Kees Bos
a425bbc673
Renamed supernode to rootserver
2015-05-06 12:05:20 +02:00
Grant Limberg
e28712555d
build & package libZeroTierOneSDK.jnilib on Mac oS
...
Signed-off-by: Grant Limberg <glimberg@gmail.com>
2015-05-04 20:04:44 -07:00
Grant Limberg
7054c365f2
put a read timeout on the socket
2015-05-02 18:34:47 -07:00
Grant Limberg
06ab338ecc
added naive OneService implementation.
...
Does nothing but send packets to master nodes and listen for packets back.
2015-05-02 18:24:52 -07:00
Grant Limberg
ad6ec22857
DataStore now works
2015-05-02 18:22:56 -07:00
Grant Limberg
a9307693a6
Adding Data Store implementations for Android and normal Java
2015-05-02 17:32:12 -07:00
Grant Limberg
742c59a7c7
Refactor package name from com.zerotierone.sdk to com.zerotier.sdk
2015-05-02 16:54:00 -07:00
Grant Limberg
e79a6915ff
ensure java is compiled for JVM 1.7. Android cant read .class files compiled for 1.8
2015-05-02 12:06:07 -07:00
Grant Limberg
818339f966
replace old direct calls to FindClass, GetMethodID, etc with calls that use the cache
2015-05-02 12:05:40 -07:00
Grant Limberg
079d248eab
added a JniCache objectet for caching jclass, jmethodID, and jfieldID objects
2015-05-02 12:05:02 -07:00
Grant Limberg
9f2732d661
add the basedir to the ant buildfile
2015-05-02 10:16:10 -07:00
Grant Limberg
fb6b56a85f
clean up unused imports
2015-05-01 18:27:34 -07:00
Grant Limberg
02b33e3763
Yeah it helps to byte swap the port back to little-endian before sending the port off to Java
2015-04-30 22:07:14 -07:00
Grant Limberg
0a15eae00f
Setup native build for Windows Java.
...
Still need to do Mac & Linux
2015-04-30 21:50:28 -07:00
Grant Limberg
75d7137025
added a class I found to load JNI code from within JAR files on non-android platforms.
...
That doesn't come for free, unlike android.
2015-04-30 21:50:05 -07:00
Grant Limberg
9464504e4a
DataStoreGet and Put and EventListener appear to work
2015-04-29 21:20:50 -07:00
Grant Limberg
7e0cc28a89
Fixes for bugs pointed out by updating warnings
2015-04-29 19:29:35 -07:00
Grant Limberg
9bf9670ff9
add more warnings to the JNI build
2015-04-29 19:14:52 -07:00
Grant Limberg
5983b4367b
jclass pointers aren't as cacheable as originally thought.
...
There is a way to do it. We can try it later if we determine it's needed for performance reasons. Otherwise, don't use static to cache them
2015-04-27 18:38:13 -07:00
