similar to: [LLVMdev] Why code doesn't speed up much with optimization level increase?

I am trying to JIT some code containing an indirect branch (and the corresponding store i8* blockaddress(@label)). I am using LLVM 2.7 code base. I build the ExecutionEngine using EngineBuilder, and call engine->getPointerToFunction(func). When I use setOptLevel(llvm::CodeGenOpt::None), the JITing fails with the following message : JIT.h:131: virtual void*

Hi, I'm seeing rather high memory usage from LLVM and I'd like to track down what I'm doing to cause it. My application is a simple web application server that compiles web pages with embedded script to bitcode and compiles them with the JIT on demand. I've taken tools/lli.cpp as a starting point and extended it to load additional modules. However, if I load successive pages and

On Thu, Feb 11, 2010 at 6:53 PM, James Williams <junk at giantblob.com> wrote: > Hi, > > I'm seeing rather high memory usage from LLVM and I'd like to track down > what I'm doing to cause it. My application is a simple web application > server that compiles web pages with embedded script to bitcode and compiles > them with the JIT on demand. I've taken

I've pretty much finished exposing all I wanted to llvm-c and the ocaml bindings for the soon to be released 2.7. Does anyone need any other functions exposed before the code freeze on the 7th?

With LLVM 3.8 the JIT compiler engine generates an AVX512 instruction although I target an 'avx2' CPU (intel Core I7). I just downloaded the most recent 3.8 and still it happens. It happens with this input module: target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128" define void @module_cFFEMJ(i64 %lo, i64 %hi, i64 %myId, i1 %ordered, i64 %start, i32* noalias align 32

On Fri, Mar 5, 2010 at 5:53 AM, George Giorgidze <giorgidze at gmail.com> wrote: > > Hi Erick, > > Can you make the following functions available in llvm-c. > > createStandardFunctionPasses > createStandardModulePasses > createStandardLTOPasses > > Thanks in advance, George This is a little tricky, so I need some advice from the community. First off, I'm

Erick Tryzelaar <erick.tryzelaar <at> gmail.com> writes: > > I've pretty much finished exposing all I wanted to llvm-c and the > ocaml bindings for the soon to be released 2.7. Does anyone need any > other functions exposed before the code freeze on the 7th? > Hi Erick, Can you make the following functions available in llvm-c. createStandardFunctionPasses

[I assume you meant to send this to the list as well, not just me.] Begin forwarded message: Date: Wed, 3 Nov 2010 14:43:54 +0000 From: Salomon Brys <salomon.brys at gmail.com> To: Török Edwin <edwintorok at gmail.com> Subject: Re: [LLVMdev] Forcing the Interpreter segfaults I have build LLVM in debug mode. Here are the informations of the segfault : memcpy() at 0x7ffff6f6581e

On 06/23/2016 12:56 PM, Craig Topper wrote: > Can you check what value "getHostCPUName" returned? getHostCPUName() = skylake > > On Thu, Jun 23, 2016 at 9:53 AM, Frank Winter via llvm-dev > <llvm-dev at lists.llvm.org <mailto:llvm-dev at lists.llvm.org>> wrote: > > With LLVM 3.8 the JIT compiler engine generates an AVX512 > instruction although I

Hi Salomon, please don't forget to reply to the list too (I've CC'd the list). > I don't think my code is doing anything worng... No, it looks fine to me, and the interpreter certainly supports this. That suggests that the value of %str is not being transmitted to the function right. If it is getting the wrong pointer value, that would explain why it barfs. Ciao, Duncan.

Hi Morten, Something else just occurred to me: can you share your EngineBuilder configuration lines? (http://llvm.org/docs/doxygen/html/classllvm_1_1EngineBuilder.html) In particular - are you explicitly setting the optimization level? The old JIT may have had a different default. - Lang. Sent from my iPad > On Feb 4, 2016, at 10:54 PM, Jim Grosbach via llvm-dev <llvm-dev at

On 12 Eki 2012, at 20:00, Jim Grosbach wrote: > > On Oct 12, 2012, at 7:07 AM, Baris Aktemur <baris.aktemur at ozyegin.edu.tr> wrote: > >> Dear Tim, >> >>> >>> The JIT sounds like it does almost exactly what you want. LLVM's JIT >>> isn't a classical lightweight, dynamic one like you'd see for >>> JavaScript or Java.

Hi, Our DSL LLVM IR emitted code (optimized with -O3 kind of IR ==> IR passes) runs slower when executed with the LLVM 3.3 JIT, compared to what we had with LLVM 3.1. What could be the reason? I tried to play with TargetOptions without any success… Here is the kind of code we use to allocate the JIT: EngineBuilder builder(fResult->fModule);

I get a segfault with this code when setting the data layout: int main(int argc, char** argv) { llvm::InitializeNativeTarget(); llvm::LLVMContext TheContext; unique_ptr<Module> Mod(new Module("A",TheContext)); llvm::EngineBuilder engineBuilder(std::move(Mod)); std::string mcjit_error; engineBuilder.setMCPU(llvm::sys::getHostCPUName());

On Nov 20, 2010, at 2:41 PM, Sdadsda Sdasdaas wrote: > And also the resulting assembly code is very poor: > > 00460013 movss xmm0,dword ptr [esp+8] > 00460019 movaps xmm1,xmm0 > 0046001C addss xmm1,xmm1 > 00460020 pxor xmm2,xmm2 > 00460024 addss xmm2,xmm1 > 00460028 addss xmm2,xmm0 > 0046002C movss dword ptr

Hello, I am trying to modify my LLVM-based compiler to perform an initial, no-optimization compilation synchronously on startup and then perform an asynchronous, optimized recompilation in the background, and I am getting in one of the optimization passes. - I am using the official release of LLVM 2.8 - I have compiled LLVM with threading enabled; I am running llvm::llvm_start_multithreaded() on

> Date: Wed, 26 Oct 2011 11:52:50 -0700 > From: Tanya Lattner <lattner at apple.com> > Subject: [LLVMdev] Release Notes: Volunteers needed > We need some volunteers to help with the 3.0 release notes. Traditionally, Chris has been the one to go > through all the commits (6 months worth!) and come up with a concrete list of things that have changed in 3.0. > Ideally,

Thanks for replying so fast. This UnsafeFPMath trick in fact solves "pxor adds" case, but the resulting code is still not as good as I expected from LLVM. For example expressions like "1+x+1+x+1+x+1+x" (basically adding a lot of constants and variables) are complied to a long series off <add>s both in IR and assembly code. Both GCC and MSVC generates C1*x +C2 (mov +

On Tue, Mar 22, 2011 at 11:51 AM, Peter Zion <peter.zion at fabric-engine.com> wrote: > Hello, > > I am trying to modify my LLVM-based compiler to perform an initial, no-optimization compilation synchronously on startup and then perform an asynchronous, optimized recompilation in the background, and I am getting in one of the optimization passes. > > - I am using the official

I have some LLVM IR after the optimization passes defined in createStandardModulePasses with the optimization level set to 3. It contains what appears to me to be an easily optimizable branch statement. In particular, note in the code below that at the end of the "loop" BasicBlock that there is a conditional branch where in the false case, it branches to the label