similar to: CallSiteBase::getCalledFunction and non-trivial calls

The immediate change I have in mind is in CallGraph; our implementation of LowerCall in AMDGPU currently relies on the the callee arguments being lowered before the call is lowered, and we simply do not support indirect calls. However, we should be able to support these bitcast calls, as they are effectively direct for our purposes, but the CallGraph does not seem to consider them (it uses

Hi, The getCalledFunction on CallInst is returning NULL, although it is (directly) calling a function, which is defined in the same file. ---------------------- if (CallInst* callInst = dyn_cast<CallInst>(&inst)) { std::cerr<<callInst->getCalledFunction(); } ----------------------- When ------------------ std::cerr<<*callInst ---------------- the

I was wondering if someone could explain why CallInst::getCalledFunc behaves the way it does. For simple, direct call instructions in my IR, that method behaves just as one would expect. However, for instructions like this: %25 = call i32 (%struct._IO_FILE*, ...)* bitcast (i32 (...)* @close to i32 > (%struct._IO_FILE*, ...)*)(%struct._IO_FILE* %24), !dbg !695 getCalledFunc returns null. I

i am sorry, i get this error: opt: /home/llvm/src/include/llvm/Support/CallSite.h:87: ValTy* llvm::CallSiteBase<FunTy, ValTy, UserTy, InstrTy, CallTy, InvokeTy, IterTy>::getCalledValue() const [with FunTy = const llvm::Function, ValTy = const llvm::Value, UserTy = const llvm::User, InstrTy = const llvm::Instruction, CallTy = const llvm::CallInst, InvokeTy = const llvm::InvokeInst, IterTy =

I have the following list of strings: x <- c("foo, foo2, foo3", "bar", "qux, qux1") what I want to do is to obtain foo, bar qux Namely for each element in the vector obtain only string before the first comma. What's the way to do it? - G.V. [[alternative HTML version deleted]]

Dear all, Suppose I have this data frame: > data_main V1 V2 foo 13.1 bar 12.0 qux 10.4 cho 20.33 pox 8.21 And I want to split the data into two parts first part are the one contain in the source array: > src [1] "bar" "pox" and the other one the complement. In the end we hope to get this two dataframes: > data_child1 V1 V2 bar 13.1 pox

Hi, I'm trying to write a pass to detect all free()/delete() call instructions in LLVM IR.The method is as follows. First I find Call Instructions: CallInst *CI=dyn_cast<CallInst>(&*i); then see if the Function name matches: name=CI->getCalledFunction()->getName(); if(name=="_ZdlPv"||name=="_ZdaPv"||name=="free")

Dear all, I found that printing with 'cat' is very slow. For example in my machine this snippet __BEGIN__ # I need to resolve to use this type of loop. # because using write(), I need to create a matrix which # consumes so much memory. Note that "foo, bar, qux" object # is already very large (>2Gb) for ( s in 1:length(x) ) {

Dear all, I would like to keep track of all the indirect calls that may caused from function pointers inside a program. I need this in order to be able to construct the control flow graph of all the indirect calls, that is which function is legal to call another function. Is there a module that implements this functionality in llvm? If not, is there a way to do it? Maybe through implementing a

Hello, This is a proposal for implementing "module flags". Please take a look at this and give any feedback you may have. Thanks! -bw Module Flags Metadata Information about the module as a whole is difficult to convey to LLVM's subsystems. The LLVM IR isn't sufficient to transmit this information. One should instead use the llvm.module.flags named

Dear R community, I have a question concerning recoding of a variable. I have a data set in which there is a variable devoted to the ISCO code describing the occupation of this certain individual (http://www.ilo.org/public/english/bureau/stat/isco/isco88/major.htm). Every type of occupation begins with a number and every number added to this number describes th occupation more detailed. Now my

Hi all, The patch below is to fix a problem with unaligned memcpys. This program: void Bork() { int Qux[33] = {0}; } will currently produce this LLVM code on PPC64: @C.0.937 = internal constant [33 x i8] zeroinitializer define void @Bork() { entry: %Qux = alloca [33 x i8] %Qux1 = bitcast [33 x i8]* %Qux to i8* call void @llvm.memcpy.i64( i8* %Qux1, i8* getelementptr ([33 x i8]*

My friends, I ran a function pass on a .bc file, intending to insert a CallInst to my self-made check function. The compilation is successful. BUT after I ran that pass on the .bc file, the size of the file didn't get any bigger!! Does this mean my instrumentation work failed?? BTW the opt command I use is "opt -load ../../../Debug+Asserts/lib/Hello.so -hello <hello.bc> -o

Hi all, Usually if we want to get the called Function we can directly use CallInst->getCalledFunction(), however, today i encounter an unusual CallInst as follows: %call11 = call double (...)* bitcast (double ()* @quantum_frand to double (...)*)() the original C source involve type cast: float u,v; extern double quantum_frand(); u = 2 * quantum_frand() - 1; v = 2 * quantum_frand() -

1. Same question as David, why &c - 8 is invalid? Is it related to below statements In C99 standard? 6.5.3.3: "Among the invalid values for dereferencing a pointer by the unary * operator are a null pointer, an address inappropriately aligned for the type of object pointed to, and the address of an object after the end of its lifetime." 2. We are trying to preserve 1st load and

On 05/12/2019 11:00 a.m., Johann R. Kleinbub wrote: > It's been three months without an answer, is it ok to thread bump? > Would someone provide a pointer? I agree it's a bug, and agree with your analysis. You should report it on bugs.r-project.org. (If you don't have an account there, let us know, and either someone will give you one, or someone will report it for you.)

Hi all, This patch is to fix a problem on PPC64 where an unaligned memcpy is generated. The testcase is this: $ cat testcase.c void Qux() { char Bar[11] = {0}; } What happens is that we produce LLVM code like this: call void @llvm.memcpy.i64( i8* %event_list2, i8* getelementptr ([11 x i8]* @C.103.30698, i32 0, i32 0), i64 11, i32 8 ) Notice that it has an 8-byte alignment. However, the Bar

Realized I didn't document the original reviews that motivated this thread: A stack of reviews, split off from here: https://reviews.llvm.org/D69787 Alok's posted a new patch (with smaller patches split off from the monolithic one) here: https://reviews.llvm.org/D84112 I haven't had a chance to page in all the old context, nor look at the new ones in detail yet. But probably worth

Could you expand on likely uses other than ObjC? For example, should float ABI (soft/softfp/hard) be conveyed via this mechanism? deep On Wed, Jan 18, 2012 at 9:36 PM, Bill Wendling <wendling at apple.com> wrote: > Hello, > > This is a proposal for implementing "module flags". Please take a look at this and give any feedback you may have. > > Thanks! > -bw >

Thanks for the response. I looked and I cannot see what exactly I need. I saw getCalledFunction() so I need CallSite CS(cast<Value>(II)) where II is a basic block iterator, so an instruction. It seems not easier than the "unelegant" version....if I am still at the Instruction level... I need a method that takes from a "leaf" basic block from a function (Maybe there is