similar to: [LLVMdev] A question of Sparc assembly generated by llc

Hi, Chris > That is target independent code, so you should not put sparc specific changes there.  It sounds like one of the sparc-specific target hooks is wrong. Since sparc does not provide any hooks for operation of branches (e.g. AnalyzeBranch and friends) it might be possible that generic codegen code is broken in absence of these hooks. -- With best regards, Anton Korobeynikov Faculty

On 11/12/2009, at 10:43 AM, Anton Korobeynikov wrote: > Hi, Chris > >> That is target independent code, so you should not put sparc specific changes there. It sounds like one of the sparc-specific target hooks is wrong. > Since sparc does not provide any hooks for operation of branches (e.g. > AnalyzeBranch and friends) it might be possible that generic codegen > code is

I have the following FORTRAN code converted to a DLL: ! my_xmean.f90 ! ! FUNCTIONS/SUBROUTINES exported from my_function.dll: ! my_function - subroutine ! subroutine my_xmean(X,N,XMEAN) ! Expose subroutine my_function to users of this DLL ! !DEC$ ATTRIBUTES DLLEXPORT,C,REFERENCE,ALIAS:'my_xmean_'::my_xmean ! Body of my_function DOUBLE PRECISION X(N) XMEAN=0D0 DO J=1,N

Dear all, let say I have following list: Dat <- vector("list", length = 26) names(Dat) <- LETTERS My_Function <- function(x) return(rnorm(5)) Dat1 <- lapply(Dat, My_Function) However I want to apply my function 'My_Function' for all elements of 'Dat' except the elements having 'names(Dat) == "P"'. Here I have specified the name

Hi Reid, Sure. Here I attach the trimmed IR(sim_0.ll) which llc.exe takes and produces a 0 storage class symbol. The IR is fairly long but there is only 1 usage of real constant as actual argument passing to function iki_vlog_time() at line 137. The version of llc.exe is $ llc.exe --version LLVM (http://llvm.org/): LLVM version 7.0.1 DEBUG build. Default target: x86_64-pc-win32 Host

There is format in sparc <pre> subcc %r1, %r2, %r3 bne .label </pre> I want instruction like this <pre> bne %r1, %r2, .label </pre> </BR> I am referring to sparc code for this implementation (llvm-3.1).</BR></BR> 1. Should I change the SPBranch and BCOND structs in .td file.</BR> 2. Should I change BuildMI constructs in

Hi Kirill, Don't forget to add X86TargetMachine.obj (add to Additional Dependencies in Linker options, if you are using MSVS) otherwise LLVM will try and use Interpreter instead of JIT. Hope this helps, Rob. > -----Original Message----- > From: llvmdev-bounces at cs.uiuc.edu [mailto:llvmdev-bounces at cs.uiuc.edu] > On Behalf Of kirill havok > Sent: Wednesday, August 20, 2008

kirill havok wrote: > Hi Gordon, > I wrote a small example, but while running I get an error("Tied to > execute an unknown external function"), where I am wrong? > I think the problem is that some_test_func() is a C++ function, so its name is being mangled during compilation. To fix it, I think you want to add a declaration telling the compiler to treat some_test_func()

Hi, I am using the llvm codegen facility (version 7.0.1) to translate LLVM IR for different platforms. I have this error particularly in win64 platform. In my IR code I have such code snippet: %50 = call i8* @my_function(i8* %48, double 2.000000e+00, double 2.000000e+00) ... declare dllimport i8* @my_function(i8*, double, double) By passing it to llc.exe, I find following symbol is declared in

Hi, I'd like to divide each element of a matrix by a specific value per column. These specific values are stored in a list. For example: > x <- c(1,2,3,4,5) > y <- matrix(c(1:30), nrow = 6) Now I want to divide each element in y[,1] by x[1], y[,2] by x[2] etc. I have tried this > my_function <- function(data, ind) data/ind > apply(y, 2, my_function, x) [,1] [,2]

Hi Gordon, I wrote a small example, but while running I get an error("Tied to execute an unknown external function"), where I am wrong? Thanks in advance. Kirill. int some_test_func( int ){ std::cout << "!!!!" << std::endl; return 8848; } int _tmain(int argc, _TCHAR* argv[]){ Module *M = new Module("test"); ExistingModuleProvider* MP = new

Hi Christine, > I am using LLVM 2.8 and llvm-gcc 4.2. Could you please try svn top-of-tree? Clang is also a better choice here. > The assembly files are attached. In the assembly file, the erroneous result > is associated with 'subcc', while the correct ones are associated with 'or'. -- Bruno Cardoso Lopes http://www.brunocardoso.cc

hi all, We are writing a loop transformation using the llvm loop pass class, but we do not want to run the pass on all loops within a program. We are looking for suggestions on how we can uniquely identify certain loops that we want to transform. So far, we are working on a suggestion to transform the loops depending on which function they're run from in the program. Which will be something

Hello World, I have a function that makes pairwise comparisons between two strings. I would like to apply this function to my data (which consists of columns with different strings) in the way that it compares the first with the second entry, and then the third with the fourth, and then the fifth with the sixth, and so on down each column... So (2x-1) and (2x) would be the different entries to be

Hi, I have the following test code print.c: #include <stdio.h> int main(){ fprintf(stderr, "Error!\n"); return 0; } I did the following on linux to cross-compile it for Sparc architecture: clang -m32 -c print.c -emit-llvm -o print.bc (or llvm-gcc -m32 -c print.c -emit-llvm -o print.bc) llvm-dis < print.bc -o print2.bc llc -march=sparc -mcpu=v8 print2.bc -o print_s.s if

I can't reproduce this problem with the recent svn trunk. LLVM 2.9 has lots of fixes for Sparc back-end. So, please at least try with LLVM-2.9. Thanks, On Fri, Sep 16, 2011 at 6:30 PM, Bruno Cardoso Lopes <bruno.cardoso at gmail.com> wrote: > Hi Christine, > >> I am using LLVM 2.8 and llvm-gcc 4.2. > > Could you please try svn top-of-tree? Clang is also a better

On Sun, 1 Oct 2006, Roman Levenstein wrote: > I'm trying to implement a new backend for an embedded CISC processor. > Therefore I thought that it makes sense to take X86 target as a basis, > to save some time. Ok. Note that the X86 backend is one of the most complex though, because it supports several subtargets and ABIs, which makes it more complex than some other targets. >

Hello guys, I'm writing a pass that must check every machine instruction that will go to assembly. I've put my pass registration just before the "addAsmEmmiter" line in LLVMTargetMachine.cpp But, iterating trought machine functions and basic blocks inside the functions, there are still machine instructions that I cant reach. This is the beginnig of the asm output of my example

Hi Venkatraman, I am using LLVM 2.8 and llvm-gcc 4.2. The assembly files are attached. In the assembly file, the erroneous result is associated with 'subcc', while the correct ones are associated with 'or'. Thanks a lot! Christine On Fri, Sep 16, 2011 at 2:29 PM, Venkatraman Govindaraju < venkatra at cs.wisc.edu> wrote: > Hello, > > What is your LLVM version?

Hi, I'd like to know if there is an equivalent for #ifdef (Clangage) in R. I 'd like to do something like: useDebug = defmacro(DEBUG, expr=(DEBUG==1)) if(useDebug(0)){ #Here I do not use debug mode make_addition = function(a, b) { c=a+b plot(c) return(c) } }else{ #here I use debug mode c=a+b } I assume this would work... The problem is : if