similar to: [LLVMdev] spilling & xmm register usage

On Sep 29, 2010, at 8:35 AMPDT, Ralf Karrenberg wrote: > Hello everybody, > > I have stumbled upon a test case (the attached module is a slightly > reduced version) that shows extremely reduced performance on linux > compared to windows when executed using LLVM's JIT. > > We narrowed the problem down to the actual code being generated, the > source IR on both systems

Dear Ashutosh, Thank you, I can handle some indirect callSites by getFunction() of InvokInst and CallInst. However, when I am handling C++ programs, I found the calls of member functions are converted to some strange indirect calls. For example: table->truncate(sysTransaction); // from mysql are translated to the next complex llvm IR: _ZN8Database20getSystemTransactionEv.exit: ;

So, I won't know the called function statically, Right? -------------------------------------------- Qiuping Yi Institute Of Software Chinese Academy of Sciences On Wed, Jan 20, 2016 at 2:24 PM, Nema, Ashutosh <Ashutosh.Nema at amd.com> wrote: > Typically for C++ virtual function you will see an indirect callSite > (unless not de-virtualized). > > > > Regards, >

Hi all, I want to find all the called functions in each function, thus I iterate the calledFunctions of each CallGraphNode in CallGraph as follow: for (CallGraph::iterator it = CG->begin(); it != CG->end(); ++it) { CallGraphNode* node = it->second; for (CallGraphNode::iterator it2 = node->begin(); it2 != node->end(); ++it2) { Function* calledFunc =

We are seeing some behavior we don't understand. Some of our functions are not showing up in the Call Graph because their calls contain bitcasts. The Call Graph then considers them to be external nodes. The calls look as follows: %tmp35 = call i32 (...)* bitcast (i32 (i32, i8*, i32*)* @convert to i32 (...)*)( i32 %tmp33, i8* %tmp34, i8* %tmp30 ) ; <i32> [#uses=1] After looking

I'm attempting to modify a parser generator to emit LLVM code instead of C. So far the experience has been trivial, but I am now running into an error regarding labels that I can't seem to solve. Situation 1: A label is used immediately after a void function call (l6 in this case): <snip> %tmp26 = load i32* @yybegin, align 4 %tmp27 = load i32* @yyend, align 4 call void

I am compiling something without standard libraries, including no libc. Somehow, this bitcode is being generated by plain array manipulation code: %12 = add i8 %11, -19 %tmp35 = icmp sgt i32 %9, 1 %smax36 = select i1 %tmp35, i32 %9, i32 1 call void @llvm.memset.p0i8.i32(i8* getelementptr inbounds ([100 x i8]* @global_array_char, i32 0, i32 0), i8 %12, i32 %smax36, i32 1, i1 false)

Hi, Here is C function: uint64_t mul(uint32_t x, uint32_t y) { return x * y; } current llvm-gcc-4.0 with -O3 will compile it to: define i64 @mul(i32 %x, i32 %y) nounwind { entry: %tmp3 = mul i32 %y, %x ; <i32> [#uses=1] %tmp34 = zext i32 %tmp3 to i64 ; <i64> [#uses=1] ret i64 %tmp34 } This seems incorrect. I think it should extend %x, %y to i64 first and

Thanks, Sanjay, that makes sense. The opportunity for improving instcombining splat sounds promising. Another question about shuffle simplification. This is a testcase from test/Transforms/InstCombine/vec_shuffle.ll: define <4 x i32> @test10(<4 x i32> %tmp5) nounwind { %tmp6 = shufflevector <4 x i32> %tmp5, <4 x i32> undef, <4 x i32> <i32 1, i32 undef, i32

Hi, I have been trying to output my graphs to a file (jpeg, pdf, ps, it doesnt matter) but i cant seem to be able to get it to output. I tried a few things but none of them worked and am lost as what to do now. I am using the scatter3d function, and it prints out the graphs on tot he screen without any problems, but when it comes to writing them to a file i cant make it work. Is there any

David Meyer wrote: > I am compiling something without standard libraries, including no libc. > > Somehow, this bitcode is being generated by plain array manipulation code: > > %12 = add i8 %11, -19 > %tmp35 = icmp sgt i32 %9, 1 > %smax36 = select i1 %tmp35, i32 %9, i32 1 > call void @llvm.memset.p0i8.i32(i8* getelementptr inbounds ([100 x > i8]*

Hello. I have a problem which is quite basic for array optimization, amd I wonder whether I am missing something, but I could not find the LLVM pass that does it. Consider the following code snippet: int test() { int mat[7][7][7]; int i,j,k,sum=0; for(i=0;i<7;i++){ for(j=0;j<7;j++){ for(k=0;k<7;k++){ sum+=mat[i][j][k]^mat[i][j][k^1]; } } } return

The LLVA and LLVM papers motivate the GetElementPtr instruction by arguing that it abstracts implementation details, in particular pointer size, from the compiler. While it does this fine for pointer addresses, it does not manage it for address offsets. Consider the following code: $ cat test.c int main() { int *x[2]; int **y = &x[1]; return (y - x); } $ llvm-gcc -O3 -c test.c

As Sanjay noted in D31426<https://reviews.llvm.org/D31426#712701>, InstructionSimplify is missing the following simplification: This function: define <4 x i32> @splat_operand(<4 x i32> %x) { %splat = shufflevector <4 x i32> %x, <4 x i32> undef, <4 x i32> zeroinitializer %shuf = shufflevector <4 x i32> %splat, <4 x i32> undef, <4 x i32>

I am testing vectorization on the following test case: float x[1024], y[1024]; void myloop1() { for (long int k = 0; k < 512; k++) { x[2*k] = x[2*k]+y[k]; } } Vectorization failed due to "unsafe dependent memory operation". I traced the LoopAccessAnalysis.cpp and found the reason is the NoWrapFlag for SCEVAddRecExpr is not set and consequently the

I'm attempting to instrument virtual function calls in my code. After each virtual call I'm calling my own registerMethod function, with an integer marking the location of the call and a pointer to the function that was called. However, and this is where I get confused, the function pointer doesn't match any of the functions in my module. I'd hoped to call

Hello, I want to compile a LLVM file into an executable running on ARM platform. I use LLVM 2.0 with the following command lines: llvm-as -f -o test.bc test.ll llc -march=arm -mcpu=arm1136j-s -mattr=+v6 -f -o test.s test.bc arm-linux-gnu-as -mcpu=arm1136j-s test.s With the last command, I obtain the following error: rd and rm should be different in mul The bad instruction is

Hello all, My app is using LLVM JIT as a runtime engine for image analysis (similar to opencl). I'm placing values into a struct to pass byref to a series of functions. After running the createStandardModulePasses(...) set of optimizations, the function calls are all inlined but the struct store/load pairs haven't been optimized away. I've attached the code below. What I'd like to

Hello all, My app is using LLVM JIT as a runtime engine for image analysis (similar to opencl). I'm placing values into a struct to pass byref to a series of functions. After running the createStandardModulePasses(...) set of optimizations, the function calls are all inlined but the struct store/load pairs haven't been optimized away. I've attached the code below. What I'd like to

[+Arnold] > On Jun 10, 2015, at 1:29 PM, Sanjoy Das <sanjoy at playingwithpointers.com> wrote: > > [+CC Andy] > >> Can anyone familiar with ScalarRevolution tell me whether this is an >> expected behavior or a bug? > > Assuming you're talking about 2*k, this is a bug. ScalarEvolution > should be able to prove that {0,+,4} is <nsw> and