similar to: [LLVMdev] function inlining

On Tue, Feb 2, 2010 at 11:40 AM, Jochen Wilhelmy <j.wilhelmy at arcor.de> wrote: > Hi! > > I have a llvm::Module that references an external function > and a second mdoule that implements the function. > The implementation consists of only one instruction. > How can I explicitly inline this function? > > -Jochen > >

Hi, We want to apply the deepest possible inlining to the target code. Looks like builder.Inliner = createFunctionInliningPass(2000); does the sufficient inlining, but what would be the best value for threshold here? Apparently, builder.Inliner = createFunctionInliningPass(numeric_limits<int>::max()); gives less inlining, that 2000. Thanks, - D.

Hi Dmitry, > We want to apply the deepest possible inlining to the target code. Looks like you could give every function the alwaysinline attribute. Ciao, Duncan.

Hi all, I have an LLVM bitcode file, with several functions in it. I'm trying to inline these functions as much as possible, so that I can obtain a big single function. There is opt -inline, but it is not sufficient. Any idea ? -- Julien Henry http://www-verimag.imag.fr/~jhenry/

Hi Julien, > I have an LLVM bitcode file, with several functions in it. > I'm trying to inline these functions as much as possible, so that I can > obtain a big single function. > There is opt -inline, but it is not sufficient. mark your functions with the alwaysinline attribute. Ciao, Duncan.

Hi! > sin/cos etc should already be handled by lib/Analysis/ConstantFolding.cpp. > Thanks for the hint and it works! Now I have a new Problem: I have this function: float foo(float a, float b) { float x = a * b * 0.0f; return cos(0.5) * sin(0.5) * x; }; after compiling it with clang (cpp mode) and renaming _ZSt3sinf to sin and _ZSt3cosf to cos I get the following: define

Hi! Of course nobody wants to implement unicode support for windows because windows should support an utf8-locale and windows is obsolete anyway ;-) But there is a simple solution: use boost::filesystem::path everywhere you use file names and paths, for example in clang::FileManager::getFile. With version 3 opening a file is easy: std::fstream file(path.c_str()). Internally

On 25.11.2010 23:56, Michael Spencer wrote: > On Nov 25, 2010, at 5:01 PM, Jochen Wilhelmy <j.wilhelmy at arcor.de > <mailto:j.wilhelmy at arcor.de>> wrote: > >> Hi! >> >> Of course nobody wants to implement unicode support for windows >> because windows should support an utf8-locale and windows is obsolete >> anyway ;-) >> >> But

On Nov 25, 2010, at 5:01 PM, Jochen Wilhelmy <j.wilhelmy at arcor.de> wrote: > Hi! > > Of course nobody wants to implement unicode support for windows > because windows should support an utf8-locale and windows is obsolete > anyway ;-) > > But there is a simple solution: use boost::filesystem::path everywhere you > use file names and paths, for example in

Now I have -DLLVM_INCLUDE_EXAMPLES:BOOL=OFF but Kaleidoscope is still there and selected for build (-G "Visual Studio 9 2008") -Jochen

Hi! Is it possible that you add an example how to write a .o with llvm? Just like examples/ModuleMaker but instead of printing to stdout writing a .o file as starting point for the new MC functionality. -Jochen

Hi! In DataTypes.h (generated from DataTypes.cmake) there is an extra code path for Visual C. This can produce macro redefinitions for INT8_MAX, INT8_MIN etc. since other headers may also define them. Therefore please protect the macros like INT8_C etc. Also it would be nice if the auto-generated HAVE_STDINT_H and HAVE_INTTYPES_H would be used also for Visual C since it is possible to add them

On 3 March 2010 11:56, Jochen Wilhelmy <j.wilhelmy at arcor.de> wrote: > > the sin and cos calls are folded, but not the mul by zero. > Is x*0 => 0 true if isnan(x)? And cos(x)*sin(x) makes me desperately want to fold it to sin(2*x)/2, but I suppose that's not allowed either.

>> Hi! >> >> before InstCombine (llvm::createInstructionCombiningPass()) I have >> a trunc from i8 to i1 and then a select: >> >> %45 = load i8* @myGlobal, align 1 >> %tobool = trunc i8 %45 to i1 >> %cond = select i1 %tobool, float 1.000000e+00, float -1.000000e+00 >> >> after instCombine I have: >> >> %29 = load i8*

Am 29.12.2011 19:52, schrieb Reid Kleckner: > I think Chris is saying that the and is necessary because with your i1 > trunc you're ignoring all of the high bits. The and implements that. > If you don't want this behavior, don't generate the trunc in the > first place and just compare the full width to zero. But if a backend sees trunc from i8 to i1 it should know

Hi! In the llvm-doc is the following: Also, since many platforms define their ABIs in terms of C, and since LLVM is lower-level than C, front-ends currently must emit platform-specific IR in order to have the result conform to the platform ABI. Can you give an example for this except system calls (which e.g. on windows use a different calling convention)? If you write your own operating

Hi! when compiling projects using llvm 2.9 and ms vc 10 I get these warnings: 1>e:\Jochen\Lib\lib\include\llvm/Use.h(218): warning C4624: 'llvm::AugmentedUse' : destructor could not be generated because a base class destructor is inaccessible 1>C:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\stdint.h(72): warning C4005: 'INT8_MIN' : macro redefinition

Hello Jochen- Your expectation is incorrect, I'm afraid. Typically this would be expanded to something like this: … %cmp = … br i1 %cmp, label %ternary.true, label %ternary.false ternary.true: %truevalue = … br label %ternary.end ternary.false: %falsevalue = … br label %ternary.end ternary.end: %value = phi %type [%truevalue, %ternary.true], [%falsevalue,

Hi! is there a way to get the c backend using llvm::TargetRegistry::lookupTarget()? e.g. llvm::TargetRegistry::lookupTarget("c", error)? - Jochen

Hi! Is there a pass that optimizes vector operations? If I have for examle a sequence of shufflevector instructions that optimizes them? (in opencl notation e.g. a.xyzw.wzyx.xxxx -> a.wwww) -Jochen