similar to: [LLVMdev] LLVM Backend

> On Feb 27, 2015, at 1:59 AM, Ambuj Agrawal <ambujbwt at gmail.com> wrote: > > Is it possible to get a register number to which the value is allocated to in MIPS in DAGToDAG class? > > More Specifically: > SDValue Reg3 = Node->getOperand(3); > if (RegisterSDNode *R = dyn_cast<RegisterSDNode>(Reg3)) >

Where can I find slides and videos for this year's EuroLLVM Developer's meeting? Thanks, Ambuj -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20150506/911e659f/attachment.html>

Thanks for your reply Quentin. I do understand that the registers are allocated much later in the pipeline. I am assuming that the physical registers are allocated before MipsAsmPrinter class. I am doing something like if (MI->getOpcode() == Mips::OPCODE) { unsigned n = MI->getNumOperands(); for(unsigned i=0 ; i < n ; i++) { const MachineOperand &MO =

Is it possible to get a register number to which the value is allocated to in MIPS in DAGToDAG class? More Specifically: SDValue Reg3 = Node->getOperand(3); if (RegisterSDNode *R = dyn_cast<RegisterSDNode>(Reg3)) { op3 = cast<RegisterSDNode>(Reg3)->getReg();

I have created custom Clang flag -no_nop_optimise which can be passes with clang on command line. I am planning to implement someting like: when -no_nop_optimise flag is passed on the command line do not emit a nop instruction. i.e. inside SelLowering class if ( !OPT_no_nop_optimise) { BuildMI(*BB, MI, DL, TII->get(Mips::NOP)); } Any clues on how can I pass the flag to the LLVM backend

I am currently working on DAGToDAGISel class for MIPS and am trying to figure out a way to use INTRINSIC_W_CHAIN for an intrinsic which can return a value. My intrinsic is defined as: Intrinsic<[llvm_i32_ty],[llvm_i32_ty,llvm_i32_ty,llvm_i32_ty,llvm_i32_ty],[IntrReadWriteArgMem]>; i.e. it has four arguments and one return value In DAGToDAGISel when I try to pass it with four arguments and

Hi Harris Bakiras, Thanks for reply. It working now. Actually I wanted to try vmkit VM to run jruby codes. vmkit is able to run Java program, but when I try to run JRuby code then I get following error - root at komal:/home/komal/Desktop/GSOC/programs# jruby hello.rb > > Platform.java:39:in `getPackageName': java.lang.NullPointerException > > from ConstantSet.java:84:in

I am working on a project to port JRuby on Embedded systems. JRuby converts Ruby code to bytecode which is executed by any JVM. For this project I am testing performance of JRuby with various available JVMs. I have chosen ARM architecture. Does vmkit support ARM architecture? On Fri, Jul 19, 2013 at 8:01 PM, Harris BAKIRAS <h.bakiras at gmail.com> wrote: > I don't know how JRuby

Hi,    I have a file containing gene expressions written using the tillingArray package. I used load() and attach() to get the data into R. Both of them works fine. Now I want to see the contents of the file. How can I see the contents of the file? Thankyou for your time. Regards, Ambuj The INTERNET now has a personality. YOURS! See your Yahoo! Homepage. [[alternative HTML version

I don't know how JRuby works, maybe it uses some new feature that GNU Classpath does not provide. VMKit's openJDK version is unstable on 64 bits since package version 6b27. You can still use it for very small programs which does not need GC but that's all. It works fine on 32 bits. So you can try it on 32 bits or revert your java version to a previous one (< than 6b27) to test

Hi,     I want to use segChrom() method in tilingArray package. For that I need to create a probeAnno object. I could not find much much info by ?probeAnno. I need help in creating  probeAnno object. Snap shot of the file(.txt): chr1 2500014 2500038 + 0.232689943122845 chr1 2500039 2500063 + 2.60502410304227 chr1 2500062 2500086 + 0.0756595313279895 chr1 2500080 2500104 + 0.78574617788405 chr1

here <http://vmkit.llvm.org/> its mentioned that its portable on ARM. So simply cross-compiling will work? On Mon, Jul 22, 2013 at 7:37 PM, Harris BAKIRAS <h.bakiras at gmail.com> wrote: > Hello Kumar, > > Unfortunately we never experienced on ARM architecture and we are not > planning to port VMKit on ARM for the moment. > > Regards, > > Harris Bakiras >

Hello Kumar, Unfortunately we never experienced on ARM architecture and we are not planning to port VMKit on ARM for the moment. Regards, Harris Bakiras On 07/19/2013 05:50 PM, Kumar Sukhani wrote: > I am working on a project to port JRuby on Embedded systems. JRuby > converts Ruby code to bytecode which is executed by any JVM. For this > project I am testing performance of JRuby

Hi Kumar, It's a mistake, we will correct that, we haven't tested vmkit on arm. As LLVM supports arm, by cross-compiling vmkit in arm/linux, it could work, but you will probably have to adapt some few parts of the code. If you plan to make the port, feel free to send us a lot of patches :) Gaël Le 22 juil. 2013 17:09, "Kumar Sukhani" <kumarsukhani at gmail.com> a écrit :

I have an audio file compressed by JSpeex that I'm trying to decompress on the iPhone. I've been trying to work from the speexdec example in an attempt to port the core logic to ObjC. I have a class that does the decode and passes the result back to another class. (currently the other class is a unit test.) My other class is writing the converted output to disk. I compare the results of my

Dear LLVM experts! we would like to use BeignetBackend as and LLVM Target It shown in teh Beignet code there is a note "Well, the complete code base is somehow a compiler backend for LLVM. Here, we really speak about the final code generation passes that you may find in `src/backend`." http://www.freedesktop.org/wiki/Software/Beignet/Backend/compiler_backe nd/ But Beignet does not

Hi Kumar, There is an error on your configuration line, you should provide the path to llvm-config binary instead of configure file. Assuming that you compiled llvm in release mode, the llvm-config binary is located in : YOUR_PATH_TO_LLVM/Release+Asserts/bin/llvm-config Try to change the -with-llvm-config-path option and it will compile. Harris Bakiras On 07/19/2013 02:36 PM, Kumar

To compile vmkit on Ubuntu 12.04 64-bit machine, I followed the steps giving here <http://vmkit.llvm.org/get_started.html>[1]. but when I run ./configure I am getting following error- root at komal:/home/komal/Desktop/GSOC/vmkit/vmkit# ./configure >> -with-llvm-config-path=../llvm-3.3.src/configure >> --with-gnu-classpath-glibj=/usr/local/classpath/share/classpath/glibj.zip

This patch just includes the repo name in the subject.

This one incorporates feedback from mburns and apevec to ensure that all multipath devices are excluded in the list of available drives. Only drives with an identified bus are included in the list now.