I'm using a physics library called Chipmunk(version 7.0.1) for a game engine in Pharo Smalltalk(version 6).
I initialize a Space with a call to function "cpSpaceNew()" and i´m correctly getting the pointer to cpSpace structure. Then i want to set the gravity to the space using a call to function "cpSpaceSetGravity(space, gravity)" and i get a Segmentation Fault.
I found a similar question : Segmentation fault on library call
but it is old question and did not find solution there.
EDIT:
Getting cpSpace pointer with:
^ self ffiCall: #( cpSpace* cpSpaceNew(void) )
"ffical:" is native call from Smalltalk.
Trying to set the gravity with:
^ (self ffiCalloutIn: thisContext sender)
convention: self ffiCallingConvention;
options: #( optIndirectCall );
function: #(void cpSpaceSetGravity (self , cpVect aVector)) module: self ffiLibraryName
This way of function call is because it is an "inline function" and it is not load in Library.
Which gcc version are you using? I recommend using older or maybe ancient one.
Excerpt from Pharo mailing list:
My understanding is that gcc 4.8(.4) is the default version in trusty,
so if anyone else wants to build the VM with the same compiler,
they'll need gcc 4.8.
At the moment it is needed because a few people have found that the VM
doesn't work properly if compiled with gcc 5.4
Related
I have a custom set of passes created using LLVM to run on some bitcode.
I've managed to get it to compile, but whenever I try to run it with a pass that calls getAnalysis() on another pass type it fails with:
Assertion `ResultPass && "getAnalysis*() called on an analysis that was not " "'required' by pass!"' failed.
The custom pass that is calling getAnalysis() requires its type, specifically;
bool Operators::doInitialization(){
ParseConfig &parseConfig = getAnalysis<ParseConfig>(); // Fails here.
}
.
.
.
void Operators::getAnalysisUsage(AnalysisUsage &AU) const{
AU.addRequired<ParseConfig>();
return;
}
I've spent a few days on this and am quite lost. I know the following is true:
ParseConfig is registered successfully via the RegisterPass<> template, and I have stepped through it in GDB to find that it does get registered.
Also using GDB I have found that when looking into getAnalysis() that the list of registered passes is always empty (which causes the assertion).
Important Note: I will eventually be using this on a Fortran project which is compiled with Flang, thus the LLVM library version I'm using is the Flang fork (found here). That fork is right around LLVM 7.1, but the specific files associated with registering passes seems to not be different from the current LLVM library.
Move getAnalysis function from doInitialization to runOnFunction would make it work.
From LLVM page
This method call getAnalysis* returns a reference to the pass desired. You may get a runtime assertion failure if you attempt to get an analysis that you did not declare as required in your getAnalysisUsage implementation. This method can be called by your run* method implementation, or by any other local method invoked by your run* method.
I'm new to LLVM. I am using the clang c++ API to compile multiple stub files (in c) to IR, and then stick them together using IR builder (after linking them) to eventually run via JIT.
All this works great, unless I add a functionInlining pass to my optimizations, at which point one of these function calls made in IR builder will trigger the following exception when the pass manager is run:
Assertion failed: (New->getType() == getType() && "replaceAllUses of value with new value of different type!"), function replaceAllUsesWith, file /Users/mike/Development/llvm/llvm/lib/IR/Value.cpp, line 356.
This is how I make the call instruction (pretty straight forward):
Function *kernelFunc = mModule->getFunction( (kernel->Name() + StringRef("_") + StringRef(funcName)).str());
if (kernelFunc){
CallInst* newInst = builder.CreateCall(kernelFunc, args);
}
Later the module is optimized:
legacy::PassManager passMan;
PassManagerBuilder Builder;
Builder.OptLevel = 3;
//Builder.Inliner = llvm::createFunctionInliningPass(); //commenting this back in trigger the exception
Builder.populateModulePassManager(passMan);
passMan.run( *mModule ); //exception occurs before this call returns
Any ideas what to look for?
Try running llvm::verifyModule on your module to see if it's correct. You might have an error and have been getting lucky beforehand but it tripped something up in the inliner.
In general assertions check a subset of things that can be wrong with your module but verify checks a lot more.
It could be a bug in LLVM but more than likely it's a bad module, it's easy to happen.
So I finally setup my dev environment so I could inspect the assertion call in the debugger. I turns out the basic block being replaced had a different context set than the one it was being replaced with. going back and making sure IRBuilder was using the same context as the IR parsers solved the problem.
I am having some trouble calling Lua functions from C++ using LuaBridge. The idea is that I want to call "Update" on the script on every game update in C++. The following code is what I have found online:
LuaRef sumNumbers = getGlobal(L, "sumNumbers");
int result = sumNumbers(5, 4);
So in my case this would be:
LuaRef updateFunction = getGlobal(L, "Update");
updateFunction();
However, the getGlobal does not seem to exist in LuaBridge 2.0 (luabridge namespace). This is different from the lua_getglobal
I am wondering if this has been replaced by a different function call or if it has been deprecated out of the 2.0 version. I can of course use the normal C Lua approach, but I was wondering if this has been abstracted in LuaBridge (to make things easier)
Make sure you're using the latest source from the github project repository. When I tested this from luabridge's master branch, luabridge::getGlobal is present and working.
I am new to Lua so I am sorry if this is rather an easy question but it is driving me nuts. In my previous thread A simple query on calling Lua 5.2 from C++ I have used an easy C++ code to embed Lua in it. Things worked well but when I transfer that concept to a more complex project, the same code does not work. I have checked many sources but could not find a solution.
Here is my code:
#include "External/include/lua.hpp"
lua_State *luastate =NULL;
IMPLEMENT_APP(ScienceSuitApp);
bool ScienceSuitApp::OnInit()
{
luastate=luaL_newstate();
luaL_openlibs(luastate);
ScienceSuitFrame* frame = new ScienceSuitFrame(0L);
frame->Show();
return true;
}
Now when I try to compile the code, I am getting invalid arguments error for the luaL_openlibs(luastate) line. The error that the compiler gives:
Invalid arguments '
Candidates are:
void luaL_openlibs(*)
' ScienceSuitApp.cpp /ScienceLab line 33 Semantic Error
This is actually happening whenever I call a Lua function such as luaL_dostring etc.. which takes lua_State as parameter. By the way, I am using Eclipse as IDE and wxWidgets as GUI if that should give a clue. Any my configuration for this set up is:
I "think" I have solved the problem. It stems from the fact that Eclipse throws a "semantic error" and not a "syntax error". Therefore, I followed the advice from Eclipse CDT shows semantic errors, but compilation is ok. But now my question is what does it have to do with Indexing in Eclipse IDT? It feels like I am using Eclipse without knowing any of its internals.
I am using the Aria C++ programming libs for mobile robots (http://robots.mobilerobots.com/wiki/ARIA). I am new to this API so I wanted to start with a simple action class derived from ArAction. Now I tried to develop a small test program (an ArAction) in order to
control a simulated p3dx robot via MobileSim. Development takes place under Ubuntu 10.10, using gcc 4.4.5. Making (compiling) my code works fine, without errors. I can also set the desired speed for example in my ArAction's fire() method, and the simulation is also working as desired.
But, unfortunately, I can't use the ArRobot object attached to the ArAction I am overriding. The problem is that none of the member functions of the ArRobot object seems to work. For example, calling getVel() or getCompass() always returns a zero value. And when I call the hasFrontBumpers() method the program even crashes with the error message "Aria: Received signal 'SIGSEGV'. Exiting.". As soon as I remove this method call and recompile the error is also gone again...
Here is the relevant code that leads to the crash:
ArActionDesired * forward::fire(ArActionDesired d)
{
desiredState.reset();
ArRobot *r = getRobot();
if(r == NULL)
{
printf("ArRobot = NULL\n");
deactivate();
return &desiredState;
}
printf("ok, ArRobot is not NULL, check for bumpers...\n");
r->hasFrontBumpers(); // <-- this leads to the SIGSEV-based "crash"
return &desiredState;
}
Any ideas what I am missing here -- is it a problem with my coding, or with the simulation environment? Thanks in advance for your help!
Kind regards, Matthias
ok, found it out now -- for the records: the Aria libs in version 2.7.2 are based on gcc-3 and libstdc++ 5, but Ubuntu 10.10 (which I am using) is shipped with gcc-4 and libstdc++ 6 per default. So I had to manually install the older versions of both packages, now my code is running fine...
cheers!
Calling hasFrontBumpers() for a p3dx from the fire() works fine for me on a similar Linux platform. If something is wrong, it is not in this method but in the initialization of the system. A reason for the non-moving robot could be that robot.enableMotors() hasn't been called.