search for: aixer77

Hi, all I'd like to write a code that can build different codes based on LLVM version. Like code snippets in the below. #ifdef __LLVM_29__ PHINode *PN = Builder.CreatePHI (Type::getDoubleTy(getGlobalContext()), 2, "iftmp"); #elif __LLVM_28__ PHINode *PN = Builder.CreatePHI (Type::getDoubleTy(getGlobalContext()), "iftmp"); #else assert ("wrong

Kangkook Jee <aixer77 at gmail.com> writes: > Hi, all > > I'd like to write a code that can build different codes based on LLVM version. > Like code snippets in the below. > > #ifdef __LLVM_29__ > PHINode *PN = Builder.CreatePHI (Type::getDoubleTy(getGlobalContext()), 2, "iftmp&quo...

Hi, all I wonder whether I can preserve basic blocks of llvm-bc when it is compiled to the target(x86) binary. for instance, #llc --march=x86 test1.bc -o test1.s #llvm-gcc -O0 -o test1 test1.s then, can BBs from output binary of 'test1' find their correspondents from 'test1.bc' and vice versa? What I eventually want to find out is 1-1 mapping of BBs between llvm-bc and

Hi, all I want to know how a llvm pass output constant char array defined from the input source. Here's an example that I want to do. == test input source == char* msg = "hello, world\n"; void msg_out(char * in) { printf("msg: %s \n", in); } main () { ... msg_out(msg); ... } == llvm pass snippet == ... const CallInst* ci =