I'm trying to use jnim to compile a shared library which can be used from Java
via JNI. I have a minimal example in this [debug
branch](https://github.com/bluenote10/JniScalaNimExample/tree/debug).
On the Java side the code is:
public class NativeTest {
public native static void run();
public static void main(String[] args) {
System.loadLibrary("nativetest");
NativeTest.run();
}
}
On the Nim side it is just:
import jnim
proc Java_NativeTest_run*(env: JNIEnvPtr, obj: jobject) {. cdecl, exportc,
dynlib .} =
system.setupForeignThreadGc()
echo "Printing from JNI..."
In general this works nicely, but unfortunately I'm getting random segfaults,
with a <5% probability in this example.
I'm compiling it using `nim --app:lib -o:libnativetest.so --threads:on c
native.nim`. In the first attempt I was omitting the
`system.setupForeignThreadGc()` (and not enabling threads). @yglukhov suggested
that there might be something wrong related to initializing TLS, so this was
the first thing that came to my mind.
To rule out that it is a general JVM bug, I was doing the same using a C
implementation. This does not seem to cause segfaults, so I think I might be
doing something wrong here. Any ideas how to fix this?
The gdb stacktrace isn't particularly helpful to me either:
Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x7ffff7fc2700 (LWP 14740)]
0x00007fffe10002b4 in ?? ()
(gdb) where
#0 0x00007fffe10002b4 in ?? ()
#1 0x0000000000000246 in ?? ()
#2 0x00007fffe1000160 in ?? ()
#3 0x00007ffff7392bf0 in VM_Operation::_names () from
/home/fabian/bin/jdk1.8.0_74/jre/lib/amd64/server/libjvm.so
#4 0x00007ffff7fc1980 in ?? ()
#5 0x00007ffff6ec5d2d in VM_Version::get_processor_features() () from
/home/fabian/bin/jdk1.8.0_74/jre/lib/amd64/server/libjvm.so
Backtrace stopped: previous frame inner to this frame (corrupt stack?)
(gdb) list
1 events.c: No such file or directory.
