similar to: [LLVMdev] llvm::Value

yes - i checked that the src->getType()->isFloatTy() is true Yakov On Mon, Nov 14, 2011 at 8:24 PM, Duncan Sands <baldrick at free.fr> wrote: > On 14/11/11 19:20, Yakov Malinkovich wrote: > >> I sure that is. >> > > Did you test it? Can you do: src->getType()->isFloatTy() > > > What could be other reasons for such error? > > The only

I sure that is.What could be other reasons for such error? Yakov On Mon, Nov 14, 2011 at 5:44 PM, Duncan Sands <baldrick at free.fr> wrote: > On 14/11/11 16:39, Yakov Malinkovich wrote: > >> It doesnt work it fails with assertation that cast is invalid .What >> could be done? >> > > Maybe src doesn't have Float type? > > Ciao, Duncan. > > >

You'll probably need to dump both the source and the dest and show the code that's being generated. A lot of guessing here that's not getting us very far very fast. -eric On Nov 14, 2011, at 10:56 AM, Yakov Malinkovich wrote: > yes - i checked that the src->getType()->isFloatTy() is true > Yakov > > > On Mon, Nov 14, 2011 at 8:24 PM, Duncan Sands <baldrick

On 14/11/11 19:20, Yakov Malinkovich wrote: > I sure that is. Did you test it? Can you do: src->getType()->isFloatTy() What could be other reasons for such error? The only other possibility I can think of is that src was created using a different context. Ciao, Duncan. > Yakov > > > On Mon, Nov 14, 2011 at 5:44 PM, Duncan Sands <baldrick at free.fr >

On 14/11/11 21:11, Yakov Malinkovich wrote: > Here is the error that I get: > Assertion failed: (i >= FTy->getNumParams() || FTy->getParamType(i) == > Params[i]->getType()) && "Calling a function with a bad signature!" That's not being generated by the CreateCast, so it looks like your description of the problem was quite misleading. Ciao, Duncan. >

Here is the error that I get: Assertion failed: (i >= FTy->getNumParams() || FTy->getParamType(i) == Params[i]->getType()) && "Calling a function with a bad signature!" Yakov On Mon, Nov 14, 2011 at 9:05 PM, Eric Christopher <echristo at apple.com>wrote: > You'll probably need to dump both the source and the dest and show the > code that's being

So what do you think the problem is? Thank you. Yakov On Mon, Nov 14, 2011 at 10:20 PM, Duncan Sands <baldrick at free.fr> wrote: > On 14/11/11 21:11, Yakov Malinkovich wrote: > >> Here is the error that I get: >> Assertion failed: (i >= FTy->getNumParams() || FTy->getParamType(i) == >> Params[i]->getType()) && "Calling a function with a

On 14/11/11 16:39, Yakov Malinkovich wrote: > It doesnt work it fails with assertation that cast is invalid .What > could be done? Maybe src doesn't have Float type? Ciao, Duncan. > > > On 11/14/11, Duncan Sands<baldrick at free.fr> wrote: >> Hi Yakov, that looks correct to me. You can also use CreateFPExt which is >> slightly simpler. >> >>

It doesnt work it fails with assertation that cast is invalid .What could be done? On 11/14/11, Duncan Sands <baldrick at free.fr> wrote: > Hi Yakov, that looks correct to me. You can also use CreateFPExt which is > slightly simpler. > > Ciao, Duncan. > > >> I want to transfer value (Value* src) of the type `FloatTyID` to >> `DoubleTyID`(I >> need all

> %temp = shufflevector <3 x i16> %incoming, <3 x i16> undef, <4 x i32> > <i32 0, i32 1, i32 2, i32 undef> > %out = bitcast <4 x i16> %temp to i64 I seem to have misread the destination type, you'd obviously want "<1 x i64>" instead of "i64" in everything I've written. Tim.

I have a function that should work only with vector types having element type as a power of 2. So if I get total 48 bit vector I should force it to be rounded to total 64 bit vector Yakov On Fri, May 4, 2012 at 10:37 PM, Tim Northover <t.p.northover at gmail.com> wrote: > > > %temp = shufflevector <3 x i16> %incoming, <3 x i16> undef, <4 x i32> > >

I implement the following function,which gets CallInst * and should perform the following: 1. Change the value of the argument if condition1 takes place 2. Change the type of the argument if condition2 takes place 3. Add addition argument/s if condition3 takes place void argChange(CallInst * I) { for (unsigned index = 0; index < I->getNumOperands(); ++index) {

Hello I have a pointer to llvm::Value *,I know its a pointer to some llvm::CompositeType (vector) Could you please tell me how can I get a value of specific index in the vector. Thank you. Yakov -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20111107/0b475afd/attachment.html>

Hello I want to transfer value (Value* src) of the type `FloatTyID` to `DoubleTyID`(I need all floats to be saved as 8 bytes).I have tried: builder.CreateCast(llvm::Instruction::FPExt, src, llvm::Type::getDoubleTy(llvmContext),””) I don`t know if I use CreateCast correctly and if this approach is correct at all. Thank you. Yakov -------------- next part -------------- An HTML attachment

Hello I write functionPath,for function which get floats and makes some calculations for them ,can I add in the functionPath check if this float values are infinite or NAN if yes what is the best way to do it. Thank you Yakov -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20111212/4ca544ab/attachment.html>

Hi Yakov, that looks correct to me. You can also use CreateFPExt which is slightly simpler. Ciao, Duncan. > I want to transfer value (Value* src) of the type `FloatTyID` to `DoubleTyID`(I > need all floats to be saved as 8 bytes).I have tried: > > builder.CreateCast(llvm::Instruction::FPExt, src, > llvm::Type::getDoubleTy(llvmContext),””) > > I don`t know if I use

You are right some cases would definitely not be right like undef + Nan -> undef. For 2.0f case I'm not sure either if any bits could be known. It seems that in general fadd( float undef, float %1) -> float %1 should always be safe and I just checked with latest code this doesn't happen. Do you think the right solution would be to add such optimization? Is there any reason why we

----- Original Message ----- > From: "Philip Reames" <listmail at philipreames.com> > To: llvmdev at cs.uiuc.edu > Sent: Tuesday, December 10, 2013 2:55:36 PM > Subject: Re: [LLVMdev] Float undef value propagation > > > > On 12/9/13 2:13 PM, Raoux, Thomas F wrote: > > > > > > Constant propagation pass generates constant expression

I'm trying to return an aggregate structure (a complex number, really) containing two 32-bit floats or two 64-bit doubles, by value. LLVM 2.6 compiled as 64-bit on OSX 10.6/x86-64. The IR I generate for a simple test case looks like this (float case): %0 = type { float, float } define %0 @test600() { entry: ret %0 { float 4.200000e+01, float 3.500000e+01 } } Running that through

On 12/9/13 2:13 PM, Raoux, Thomas F wrote: > > Constant propagation pass generates constant expression when undef is > used in float instructions instead of propagating the undef value. > > ; Function Attrs: nounwind > > define float @_Z1fv() #0 { > > entry: > > %add = fadd fast float undef, 2.000000e+00 > > ret float %add > > } > > Becomes: