search for: errorv

...-------------===// // Code Generation //===----------------------------------------------------------------------===// static Module *TheModule; static IRBuilder<> Builder(getGlobalContext()); static std::map<std::string, Value*> NamedValues; static FunctionPassManager *TheFPM; Value *ErrorV(const char *Str) { Error(Str); return 0; } Value *NumberExprAST::Codegen() { return ConstantFP::get(getGlobalContext(), APFloat(Val)); } Value *VariableExprAST::Codegen() { // Look this variable up in the function. Value *V = NamedValues[Name]; return V ? V : ErrorV("Unknown variable...

...> //===----------------------------------------------------------------------===// > > static Module *TheModule; > static IRBuilder<> Builder(getGlobalContext()); > static std::map<std::string, Value*> NamedValues; > static FunctionPassManager *TheFPM; > > Value *ErrorV(const char *Str) { Error(Str); return 0; } > > Value *NumberExprAST::Codegen() { > return ConstantFP::get(getGlobalContext(), APFloat(Val)); > } > > Value *VariableExprAST::Codegen() { > // Look this variable up in the function. > Value *V = NamedValues[Name]; > retur...

...= Args.size(); ++AI, ++Idx) { AI->setName(Args[Idx]); // NamedValues is map<string, Value*> NamedValues[Args[Idx]] = AI; and when it try to get the value of arguement, it simply does: Value *VariableExprAST::Codegen() { Value *V = NamedValues[Name]; return V ? V : ErrorV("Unknown variable name"); } It means that we can access value of arguement directly. However, in llvm.org/demo and the chapter 6 of tutorial, it deals with arguemnt: 1. allocate memory for arguement 2. then store the arguemnt into the memory. 3. CreateLoad() to load the arguement when a...

...----------------------------------------------------===// >> >> static Module *TheModule; >> static IRBuilder<> Builder(getGlobalContext()); >> static std::map<std::string, Value*> NamedValues; >> static FunctionPassManager *TheFPM; >> >> Value *ErrorV(const char *Str) { Error(Str); return 0; } >> >> Value *NumberExprAST::Codegen() { >>  return ConstantFP::get(getGlobalContext(), APFloat(Val)); >> } >> >> Value *VariableExprAST::Codegen() { >>  // Look this variable up in the function. >>  Value *V...

...I->setName(Args[Idx]); > >     // NamedValues is map<string, Value*> >     NamedValues[Args[Idx]] = AI; > > and when it try to get the value of arguement, it simply does: > > Value *VariableExprAST::Codegen() { >   Value *V = NamedValues[Name]; >   return V ? V : ErrorV("Unknown variable name"); > } > > It means that we can access value of arguement directly. Yes, it is possible to access the values (virtual registers) direclty, but you may need to construct SSA form by yourself. > > However, in llvm.org/demo and the chapter 6 of tutorial,...

...// NamedValues is map<string, Value*> > > NamedValues[Args[Idx]] = AI; > > > > and when it try to get the value of arguement, it simply does: > > > > Value *VariableExprAST::Codegen() { > > Value *V = NamedValues[Name]; > > return V ? V : ErrorV("Unknown variable name"); > > } > > > > It means that we can access value of arguement directly. > Yes, it is possible to access the values (virtual registers) direclty, > but you may need to construct SSA form by yourself. > > > > However, in llvm.org/...

quite some speed improvement indeed. attached the updated patch to apply to svn/trunk. j -------------- next part -------------- A non-text attachment was scrubbed... Name: theora-mmx.patch.gz Type: application/x-gzip Size: 8648 bytes Desc: not available Url : http://lists.xiph.org/pipermail/theora-dev/attachments/20040824/5a5f2731/theora-mmx.patch-0001.bin