similar to: [LLVMdev] Structure Types and ABI sizes

Ok, here is the patch again... I also included fixes for the bits that originally gave my mailer fits... Two votes for orange, so I went with orange... Doing diffs in .: --- ./lib/AsmParser/LLLexer.cpp.~1~ 2007-12-14 22:09:06.000000000 -0800 +++ ./lib/AsmParser/LLLexer.cpp 2007-12-15 13:02:47.000000000 -0800 @@ -54,7 +54,7 @@ static uint64_t HexIntToVal(const char * Result +=

Hi, I have a Type * which may come from an IntegerType as shown below: Type.getIntegerBitWidth() tells me numBits. But how to extract the IntegerType.getSignBit? If pType isIntegerType, I need to know if it is signed or unsigned... How to achieve this? Thx Alex llvm::Type * getRandomValid_IntegerType(llvm::LLVMContext &C) { using namespace llvm; //--- determine num of bits

Hi all, I'm fiddling around a bit with programmatically created GEP instructions, which is failing when using i64 for any indix, except the first. The reason behind this, is that StructureType::indexValid only accepts Value* that are constant ints of width 32. I can't really see why this limitation is here. SequentialType solves this slightly differently, it allows for both 32 and 64

Hi all, I just found that TargetData also contains a similar assert, and that I forgot a "return" in my previous patch. Here's an updated patch, which allows to use any integer to index a CompositeType. It seems this enables GEP to work with any integer value as well, I tested indexing a struct with a i16 an i64 (also through llc). This should still be adressed somewhere, unless we

Hi all, any comments about this patch? I'd like to get it out of my working copy :-) Gr. Matthijs > Index: lib/VMCore/Type.cpp > =================================================================== > --- lib/VMCore/Type.cpp (revision 53136) > +++ lib/VMCore/Type.cpp (working copy) > @@ -394,9 +394,8 @@ > > bool StructType::indexValid(const Value *V) const { >

Helper function: template<typename InsnType> static InsnType fieldFromInstruction(InsnType insn, unsigned startBit, unsigned numBits) { assert(startBit + numBits <= (sizeof(InsnType)*8) && "Instruction field out of bounds!"); InsnType fieldMask; if (numBits == sizeof(InsnType)*8) fieldMask =

Hello. Could you please tell me why am I not able to recognize (with LLVM built from the SVN code in Apr 25, 2018) the LLVM IR intrinsic ctpop (described at https://llvm.org/docs/LangRef.html#llvm-ctpop-intrinsic) in the following program: int PopCnt_Simple(int x) { int numBits = 0; int i; //for (i = 0; i < 32; i++) { for (i = 0; x !=

Hi Sunil, I had filed a bug and saw your response stating that it would be fixed in version 14. In the meanwhile , what we want to know is whether this bug is a minor bug and can be ignored for now. Does reporting 100% inodes cause any problem for the OCFS file system or can we ignore this bug and go into production. Also can you tell us by when version 14 would be released. R'gds

Hello, I noticed a bit of a gap in the current code for unions: a ConstantUnion cannot be written out to .ll. Hopefully I'm not stepping on Talin's toes by posting this, it's a fairly straightforward adaptation of the code for structs just above. Tim. -- The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336. --------------

Output df -i ------------------ Filesystem Inodes IUsed IFree IUse% Mounted on / /dev/sde 348548 348195 353 100% /ocfsa01 /dev/sdf 348548 348195 353 100% /ocfsa02 /dev/sdg 348548 348195 353 100% /ocfsa03 /dev/sdk 139410 138073 1337 100% /ocfsq01 Output df -kP ----------------------- Filesystem

I think I might have found a bug in the exection engine's 'constants folding'. Basically APInt's parameters are the wrong way round. Line 577 for lib/ExecutionEngine/ExecutionEngine.cpp :- case Instruction::PtrToInt: { GenericValue GV = getConstantValue(Op0); uint32_t PtrWidth = TD->getPointerSizeInBits(); GV.IntVal = APInt(PtrWidth,

Hello, I'm compiling LLVM with MSVC at the moment, I've got the following warning: BitstreamReader.h(327) : warning C4334: '<<' : result of 32-bit shift implicitly converted to 64 bits (was 64-bit shift intended?) in line uint64_t Piece = Read(NumBits); if ((Piece & (1U << (NumBits-1))) == 0) Is this warning relevant? Is also Read64()

On Tue, Dec 05, 2017 at 11:59:17AM +0200, achiad shochat wrote: > Then we'll have a single solution for both netvsc and virtio (and any > other PV device). > And we could handle the VF DMA dirt issue agnostically. For the record, I won't block patches adding this kist to virtio on the basis that they must be generic. It's not a lot of code, implementation can come first,

On Tue, Dec 05, 2017 at 11:59:17AM +0200, achiad shochat wrote: > Then we'll have a single solution for both netvsc and virtio (and any > other PV device). > And we could handle the VF DMA dirt issue agnostically. For the record, I won't block patches adding this kist to virtio on the basis that they must be generic. It's not a lot of code, implementation can come first,

Hi,everyone. I'm writing a pass in which a CallInst to an external function will be inserted. The function is declared like this: void func(int a, unsigned chat *p); and in the Pass(a Function Pass ,and using the InstVistor template ), I wrote like this: void visitStoreInst(StoreInst &SI) { //Get the refference of the types Module *M =

I tried the following: - ConstantInt::get(Type::getInt64Ty(Ctx), APInt(0xFFFFFFFFFFFFFFFF, 64, false)) - ConstantInt::get(Type::getInt64Ty(Ctx), APInt(-1, 64)) I am receiving the following error: Assertion `NumBits <= MAX_INT_BITS && "bitwidth too large" failed -- Thanks & Regards, Dipanjan -------------- next part -------------- An HTML attachment was scrubbed...

Akshay Jain <jivan.molu at gmail.com> writes: > I have tried it, but I always end up with some kind of error. Can you > explain how can I get a function type for function which returns void > (nothing) and it's arguments are (int, int, int, void *, void *, ...) ?? Instead of getting something to cut&paste, you would be much more enriched if the problematic code and the

I'm using LLVM 3.3 Release. From the Class Reference Page. CallInst has a member function : static CallInst * Create (Value *Func, ArrayRef< Value * > Args, const Twine &NameStr="", Instruction *InsertBefore=0) This is what I want to use! the default name is "" and I also use that . Thanks! ----- Kind Regards! -JinHuang -- View this message in context:

I have tried it, but I always end up with some kind of error. Can you explain how can I get a function type for function which returns void (nothing) and it's arguments are (int, int, int, void *, void *, ...) ?? Thanks in advance. On Tue, Apr 16, 2013 at 7:22 AM, Óscar Fuentes <ofv at wanadoo.es> wrote: > Akshay Jain <jivan.molu at gmail.com> writes: > > > I am

Duncan, Thanks for the quick review! Here is a short description (design) of where I am going with this patch: 1. Motivation: Vectors-of-pointers is the first step in supporting scatter/gather instructions (available in AVX2, for example). I believe that this feature was requested on the mailing list before. As mentioned by Hal Finkel earlier today, this feature is desired by autovectorizers as