similar to: [LLVMdev] BitCast or GEP?

On Jan 10, 2013, at 22:10 , Sean Silva <silvas at purdue.edu> wrote: > On Thu, Jan 10, 2013 at 8:25 AM, Rick Mann <rmann at latencyzero.com> wrote: >> I've been writing up my experience embedding LLVM in a Cocoa app. If anyone would like to see it and give feedback, I'd appreciate it. It's still not finished, but the first part is there: >> >>

Exciting!!! > On Sep 29, 2015, at 17:03 , Hal Finkel via llvm-dev <llvm-dev at lists.llvm.org> wrote: > > ----- Original Message ----- >> From: "Dylan McKay via llvm-dev" <llvm-dev at lists.llvm.org> >> To: llvm-dev at lists.llvm.org, "David Siegel" <agnat at me.com> >> Sent: Thursday, September 3, 2015 10:35:16 PM >>

On 27 Oct 2015, at 09:48, Senthil Kumar <senthil.thecoder at gmail.com> wrote: > > Haven't worked on this yet, but I work on the gcc AVR backend (and binutils), and I'm very interested in this - I intend to work on it in my spare time. And do you have any comments / objections to Dylan becoming the code owner? David

I've been writing up my experience embedding LLVM in a Cocoa app. If anyone would like to see it and give feedback, I'd appreciate it. It's still not finished, but the first part is there: http://roderickmann.org/stuff/LLVMTut/HowToEmbedLLVMWithXcode.html If it's worthy, hopefully it'll end up in the LLVM docs. -- Rick

On Jan 8, 2013, at 13:55 , "Kaylor, Andrew" <andrew.kaylor at intel.com> wrote: > You need to include 'llvm/ExecutionEngine/JIT.h' (or 'llvm/ExecutionEngine/JIT.h' if you want that engine) from your main file. Including that file forces the JIT static constructor to be linked into your executable. Without it, the JIT static constructor gets optimized out and

My host app calls runFunction() on my JITed code. It, in turn, calls a C function ("decode()") in the host app that then calls dispatch_async(). The runFunction() call returns as expected, but the block passed to dispatch_async() never gets called. The async block is supposed to call a function pointer callback that was passed in to decode(). Everything is being called on the main

On Thu, Jan 10, 2013 at 8:25 AM, Rick Mann <rmann at latencyzero.com> wrote: > I've been writing up my experience embedding LLVM in a Cocoa app. If anyone would like to see it and give feedback, I'd appreciate it. It's still not finished, but the first part is there: > > http://roderickmann.org/stuff/LLVMTut/HowToEmbedLLVMWithXcode.html This is great. People

Hi, In PromoteMemoryToRegister.cpp, it seems to rely on the fact that the only users of bitcast/GEP instruction are lifetime intrinsics (llvm.lifetime.start/end). I did some searching in llvm/test folder, it seems to be true. However, by reading LLVM IR manual, I don't see any restriction stated on the possible user of bitcast/GEP instruction. So my question is who impose the restriction ?

----- Original Message ----- > From: "guoqing zhang" <gqzhang81 at gmail.com> > To: llvmdev at cs.uiuc.edu > Sent: Friday, January 30, 2015 4:29:16 AM > Subject: [LLVMdev] About user of bitcast/GEP instruction > > Hi, > > > In PromoteMemoryToRegister.cpp, it seems to rely on the fact that the > only users of bitcast/GEP instruction are lifetime

Hi, In PromoteMemoryToRegister.cpp, it seems to rely on the fact that the only users of bitcast/GEP instruction are lifetime intrinsics (llvm.lifetime.start/end). I did some searching in llvm/test folder, it seems to be true. However, by reading LLVM IR manual, I don't see any restriction stated on the possible user of bitcast/GEP instruction. So my question is who impose the restriction ?

Hi, If the special handling in the meg2reg pass is to look for lifetime intrinsics, shouldn't it cast to <IntrisicInst> and then use getInstrinsicID to check for lifetime_start and lifetime_end ? The thing that I don't understand is the following piece of code, which finds all the users and cast it to <Instruction> then eraseFromParent(). How can this guarantee that it only

> From: llvmdev-bounces at cs.uiuc.edu [mailto:llvmdev-bounces at cs.uiuc.edu] > On Behalf Of Sean Silva > Subject: Re: [LLVMdev] What would cause instructions to NOT make it into the module? > It is a pass that can be run like any other pass. See createVerifierPass(). You can also just call verifyModule() with the appropriate parameters; it's defined in

I've been working through the getting started and language reference docs. One thing that's not immediately clear if it's possible is to define a global pointer to a specific memory address. I'm looking at ways to use LLVM to target a small ARM embedded processor, and there are a great many hardware registers mapped into the address space. I tried making a little sample C program

On Apr 9, 2010, at 23:32:12, Chris Lattner wrote: > Several people have asked what the MC project is all about, and it is now getting to a point where it is pretty interesting and there are lots of potential projects for people who are interested. I wrote a big post about what it is here: > http://blog.llvm.org/2010/04/intro-to-llvm-mc-project.html Thanks for the post, Chris. It

Hi. I've had moderate success so far with LLVM. I've got an OS X app in Xcode that is able to parse my DSL, generate code into an LLVM Module, and pass it to an ExecutionEngine to execute. But now I'm stuck. That code needs to reference an external function I define in my app. If I define that function in the same source file as the file that invokes the ExecutionEngine, everything

On Dec 12, 2013, at 04:25 , Gasiunas, Vaidas <vaidas.gasiunas at sap.com> wrote: > To address these requirements we have developed a set of patches improving performance and malfunction safety of certain compiler passes and would be interested in contributing them at some point. Before proposing concrete changes, we would like to know what the general interest is with respect to making

Hi list, Does LLVM provide support for Apple-style blocks? Or is that all implemented in clang using LLVM primitives? I'm looking at the IR that clang generates, and it sure seems confusing to me. I'm hoping there's a builder that can help me implement them. Thanks! -- Rick

On Jan 17, 2013, at 19:33 , Eric Christopher <echristo at gmail.com> wrote: > You'll want to look at clang/lib/CodeGen/CGBlocks.[cpp, h]. Thanks. I think proper blocks are way over my head for my little project. I'll try to implement something "block-like" using function pointers. -- Rick

On Jan 23, 2013, at 21:47 , Sean Silva <silvas at purdue.edu> wrote: > Try running the verifier pass (the one that is invoked by `opt -verify`). I'm not sure how to do this. I'm writing a Cocoa app that calls the LLVM API directly; I don't use any of the command-line tools. Is there a way to verify via API? I think I've actually done this for individual types, dunno

Several people have asked what the MC project is all about, and it is now getting to a point where it is pretty interesting and there are lots of potential projects for people who are interested. I wrote a big post about what it is here: http://blog.llvm.org/2010/04/intro-to-llvm-mc-project.html Thoughts and comments welcome, -Chris