我正在编写一个Android应用程序，通过OpenGL ES在本机端进行图形渲染。当我在调试模式下编译它时，应用程序崩溃，但在发布模式下不会崩溃。我已经检测到它只是在设置渲染器时发生，就在它完成对（In [19]: import numpy as np
In [20]: float_list = np.array([167.233, 95.6242, 181.367, 20.6354, 147.505, 41.9396, 20.3126])
# In [22]: np.split(float_list, np.where(np.diff(float_list) > 0)[0] + 1)
# Out[22]:
# [array([ 167.233 , 95.6242]),
# array([ 181.367 , 20.6354]),
# array([ 147.505 , 41.9396, 20.3126])]
In [23]: map(len, np.split(float_list, np.where(np.diff(float_list) > 0)[0] + 1))
Out[23]: [2, 2, 3]
）函数的调用时。发生的其他事情是当我在onSurfaceCreated行中放置一个断点时，即使我没有进入，只需继续程序的正常流程，然后它就不会崩溃。似乎存在线程同步问题，但我无法弄清楚如何解决它。
这是我的渲染器类
XXXActivity.nativeSurfaceCreated();}
这是我的堆栈：
public class Renderer implements GLSurfaceView.Renderer {
public Renderer(Context ctx)
{
}
public void onSurfaceCreated(GL10 gl, EGLConfig config) {
XXXActivity.nativeSurfaceCreated();
}
public void onSurfaceChanged(GL10 gl, int w, int h) {
XXXActivity.nativeSurfaceChanged(w, h); //HERE is the problem just when it finish the execution of this line.
}
public void onDrawFrame(GL10 gl) {
XXXActivity.nativeDrawFrame();
}
这是后面的跟踪
F/art (26519): art/runtime/mirror/object.cc:190] Invalid state during hashcode ForwardingAddress
F/art (26519): art/runtime/runtime.cc:284] Runtime aborting --- recursively, so no thread-specific detail!
F/art (26519): art/runtime/runtime.cc:284]
--------- beginning of crash
F/libc (26519): Fatal signal 6 (SIGABRT), code -6 in tid 26533 (HeapTrimmerDaem)
I/DEBUG ( 299): [2016-09-28 10:42:59.064]
I/DEBUG ( 299): *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** ***
I/DEBUG ( 299): Build fingerprint: 'lge/w7_tmo_us/w7:5.0.2/LRX22G/150651744a25e:user/release-keys'
I/DEBUG ( 299): Revision: '6'
I/DEBUG ( 299): ABI: 'arm'
I/DEBUG ( 299): pid: 26519, tid: 26533, name: HeapTrimmerDaem >>> com.XXX.ar.XXX <<<
I/DEBUG ( 299): signal 6 (SIGABRT), code -6 (SI_TKILL), fault addr --------
W/NativeCrashListener( 972): Couldn't find ProcessRecord for pid 1530015793
I/DEBUG ( 299): Abort message: 'art/runtime/gc/collector/semi_space-inl.h:37] Check failed: IsAligned<kPageSize>(obj) '
E/DEBUG ( 299): AM write failure (32 / Broken pipe)
I/DEBUG ( 299): r0 00000000 r1 000067a5 r2 00000006 r3 00000000
I/DEBUG ( 299): r4 a452edd8 r5 00000006 r6 00000002 r7 0000010c
I/DEBUG ( 299): r8 00000001 r9 b78f2f48 sl b7eabab0 fp b81c5550
I/DEBUG ( 299): ip 000067a5 sp a452e288 lr b6ec57f5 pc b6eed344 cpsr 60070010
I/DEBUG ( 299):
I/DEBUG ( 299): backtrace:
I/DEBUG ( 299): #00 pc 0003b344 /system/lib/libc.so (tgkill+12)
I/DEBUG ( 299): #01 pc 000137f1 /system/lib/libc.so (pthread_kill+52)
I/DEBUG ( 299): #02 pc 00014403 /system/lib/libc.so (raise+10)
I/DEBUG ( 299): #03 pc 00010c55 /system/lib/libc.so (__libc_android_abort+36)
I/DEBUG ( 299): #04 pc 0000f3f4 /system/lib/libc.so (abort+4)
I/DEBUG ( 299): #05 pc 00226ccf /system/lib/libart.so (art::Runtime::Abort()+170)
I/DEBUG ( 299): #06 pc 000a6c59 /system/lib/libart.so (art::LogMessage::~LogMessage()+1360)
I/DEBUG ( 299): #07 pc 001efbb3 /system/lib/libart.so (art::mirror::Object::IdentityHashCode() const+390)
I/DEBUG ( 299): #08 pc 001f4223 /system/lib/libart.so (art::Monitor::DescribeWait(std::__1::basic_ostream<char, std::__1::char_traits<char> >&, art::Thread const*)+306)
I/DEBUG ( 299): #09 pc 00234627 /system/lib/libart.so (art::StackDumpVisitor::VisitFrame()+358)
I/DEBUG ( 299): #10 pc 0022e8ad /system/lib/libart.so (art::StackVisitor::WalkStack(bool)+272)
I/DEBUG ( 299): #11 pc 00236d69 /system/lib/libart.so (art::Thread::Dump(std::__1::basic_ostream<char, std::__1::char_traits<char> >&) const+224)
I/DEBUG ( 299): #12 pc 0023fc1d /system/lib/libart.so (art::ThreadList::DumpLocked(std::__1::basic_ostream<char, std::__1::char_traits<char> >&)+120)
I/DEBUG ( 299): #13 pc 00226a2d /system/lib/libart.so (art::AbortState::Dump(std::__1::basic_ostream<char, std::__1::char_traits<char> >&)+272)
I/DEBUG ( 299): #14 pc 00226c77 /system/lib/libart.so (art::Runtime::Abort()+82)
I/DEBUG ( 299): #15 pc 000a6c59 /system/lib/libart.so (art::LogMessage::~LogMessage()+1360)
I/DEBUG ( 299): #16 pc 0012d2f1 /system/lib/libart.so (art::gc::collector::BitmapSetSlowPathVisitor::operator()(art::mirror::Object const*) const+160)
I/DEBUG ( 299): #17 pc 0012f5cf /system/lib/libart.so (void art::mirror::Object::VisitReferences<true, (art::VerifyObjectFlags)0, art::gc::collector::SemiSpaceMarkObjectVisitor, art::gc::collector::SemiSpaceMarkObjectVisitor>(art::gc::collector::SemiSpaceMarkObjectVisitor const&, art::gc::collector::SemiSpaceMarkObjectVisitor const&)+1122)
I/DEBUG ( 299): #18 pc 0013061b /system/lib/libart.so (art::gc::collector::SemiSpace::ProcessMarkStack()+126)
I/DEBUG ( 299): #19 pc 001307a9 /system/lib/libart.so (art::gc::collector::SemiSpace::MarkReachableObjects()+324)
I/DEBUG ( 299): #20 pc 00130beb /system/lib/libart.so (art::gc::collector::SemiSpace::MarkingPhase()+370)
I/DEBUG ( 299): #21 pc 0012e479 /system/lib/libart.so (art::gc::collector::SemiSpace::RunPhases()+480)
I/DEBUG ( 299): #22 pc 00121e37 /system/lib/libart.so (art::gc::collector::GarbageCollector::Run(art::gc::GcCause, bool)+246)
I/DEBUG ( 299): #23 pc 00134657 /system/lib/libart.so (art::gc::Heap::Compact(art::gc::space::ContinuousMemMapAllocSpace*, art::gc::space::ContinuousMemMapAllocSpace*, art::gc::GcCause)+58)
I/DEBUG ( 299): #24 pc 00138fa3 /system/lib/libart.so (art::gc::Heap::PerformHomogeneousSpaceCompact()+1818)
I/DEBUG ( 299): #25 pc 0013fcc7 /system/lib/libart.so (art::gc::Heap::DoPendingTransitionOrTrim()+818)
I/DEBUG ( 299): #26 pc 0001bd93 /system/framework/arm/boot.oat
I/DEBUG ( 299):
I/DEBUG ( 299): Tombstone written to: /data/tombstones/tombstone_05
I/BootReceiver( 972): Copying /data/tombstones/tombstone_05 to DropBox (SYSTEM_TOMBSTONE)
I/Zygote ( 325): Process 26519 exited due to signal (6)
答案 0 :(得分：0)
我设法解决了自己的问题。碰巧我在XXXActivity.nativeSurfaceCreated();中初始化了一个与纹理相关的代码，但不知怎的XXXActivity.nativeDrawFrame();我在XXXActivity.nativeSurfaceCreated();之前绘制了纹理，并且由于没有创建该资源，它引发了崩溃，这就是OpenGL线程崩溃的原因。我将代码移到nativeDrawFrame中进行初始化一次。它可能不是必须美丽的解决方案，但它的工作原理。如果有人有其他解决方案，请不要犹豫在这里发布。