similar to: [LLVMdev] extract thread form sequential program

I want automatically parallelize sequential program in thread level to run on multi-core processors with software pipelining and use llvm. is it a suitable tools in this project ? can you help me in this topic? -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20110128/b84ebf3d/attachment.html>

Hi neda 8664, > how i can assign instruction of program to a thread? LLVM doesn't have special support for threads, you need to call library functions the same as you would from C. I suggest you write some C code that does what you want, then run it through http://llvm.org/demo/ to see the corresponding LLVM IR. Ciao, Duncan.

LLVM has no facility for dealing with threads explicitly. If you're implementing a research tool to automatically pipeline software, LLVM could be of use because it is also a JIT with a good API. On Fri, Jan 28, 2011 at 8:35 AM, neda 8664 <neda8664 at gmail.com> wrote: > I want automatically parallelize sequential program in thread level to run > on multi-core > processors

Thank you for your reply Yes, I change them, so what should I do for another loops? On Mon, Dec 12, 2011 at 7:54 PM, neda 8664 <neda8664 at gmail.com> wrote: > Thank you for your reply > > Yes, I change them, so what should I do for another loops? > > On Mon, Dec 12, 2011 at 7:42 PM, John Criswell <criswell at illinois.edu>wrote: > >> On 12/12/11 9:59 AM,

hi all, I want access to all basic blocks of function in a loop, so I used the following code: *bool parallel::runOnLoop(Loop *L, LPPassManager &LPM) { for (Function::iterator bi= func->begin(); bi != func->end(); bi++){ // } }* First loop run without problem, but for second loop I get the following error: *0 libLLVM-2.9.so 0x0137d530 1 libLLVM-2.9.so 0x0137fa6c 2

On 12/12/11 10:25 AM, neda 8664 wrote: > > Thank you for your reply > > Yes, I change them, so what should I do for another loops? I don't really know what you should do since I don't know what your code does. All I know is that it's changing the function's control-flow graph. If you're not modifying the structure of the loops in the function, then you can

Hi, All I want to implement DSWP Which is used for parallelization of loops. For this purpose, the loop was replaced with a new basic block in main function. And new functions were created and basic blocks of Loop assigned to them.I have checked blocks and branches for Succ and Pred relation and I have not found any problems. However I get the following error: * **opt:

I am changing structure of loops, I used std::vector<> but i get same error, I can’t use FunctionPass :( On Mon, Dec 12, 2011 at 8:03 PM, John Criswell <criswell at illinois.edu>wrote: > On 12/12/11 10:25 AM, neda 8664 wrote: > > Thank you for your reply > Yes, I change them, so what should I do for another loops? > > > I don't really know what you should

I've study their work carefully. My problem is the implementation of threads in llvm after partitionning. I want to have information about how to implement producer/consumer thread in llvm. I do not know where I should start in llvm for code generation and create thread and insert produce and consume statement . -------------- next part -------------- An HTML attachment was scrubbed... URL:

i could not use profile.pl, however i did reinstall the llvm but didn't solve my problem. is there is another way to extract profilling information in llvm? i have another question, i want know how can estimate execution time for each operand such as mul ,add, ... ? -------------- next part -------------- An HTML attachment was scrubbed... URL:

hi I changed llvm-prof and make it, but when use profile.pl ,i could not see any change in result. -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20110907/fc9be189/attachment.html>

I am MSc student and my thesis subject is "parallelization of sequential program in thread level to run on multi-core processors". I want automatically parallelize sequential program in thread level to run on multi-core processors with software pipelining. I've use LLVM to code analysis and translate code to IR, I could extract PDG graph form IR code and convert it to SCCDAG and

hi i have only one week to do this project.i know its impossible.can you help me? -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20120216/aae3954d/attachment.html>

John Criswell <criswell <at> illinois.edu> writes: > > > On 12/12/11 9:59 AM, neda 8664 wrote: > > > hi all, > I want access to all basic blocks of function in a loop, so I used > the following code:bool parallel::runOnLoop(Loop *L, LPPassManager > &LPM) > { >     for

On 12/12/11 9:59 AM, neda 8664 wrote: > hi all, > > I want access to all basic blocks of function in a loop, so I used the > following code: > > /bool parallel::runOnLoop(Loop *L, LPPassManager &LPM) > { > for (Function::iterator bi= func->begin(); bi != func->end(); bi++){ > // > } > }/ Are you modifying anything within this code

Hi all, I am looking appropriate Benchmarking for the assessment of automatic parallelization project. What Benchmarking do you suggest me? regards neda -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20111201/95e7779c/attachment.html>

I use "opt -dot-cfg sample.bc" command to view cfg, but I encountered the following warning. WARNING: You're attempting to print out a bitcode file. This is inadvisable as it may cause display problems. If you REALLY want to taste LLVM bitcode first-hand, you can force output with the `-f' option. I see unmeaning thing when use -f , in additinal I've installed graphvis

Hi All I extract Def_Use chain by following code: for (Value::use_iterator i = F->use_begin(), e = F->use_end(); i != e; ++i) if (Instruction *Inst = dyn_cast<Instruction>(*i)) { errs() << "F is used in instruction:\n"; errs() << *Inst << "\n"; } Now, I want to distinguish the register name or memory variable that lead to

Hi all I want to use "'/home/llvm/src/utils/profile.pl' -edge sort.bc" command, but i face this error: Error opening '/usr/local/lib/profile_rt.so': /usr/local/lib/profile_rt.so: cannot open shared object file: No such file or directory -load request ignored. LLVM ERROR: Program used external function 'llvm_start_edge_profiling' which could not be resolved!

hi i have another question, how can change the display of result on screen?For this purpose, i have modified the llvm-prof file and make it again. But received no results