search for: nvptx

Hi all, I'm trying to use the newly added (in LLVM 3.2) NVPTX backend for producing PTX (Parallel Thread eXecution) assembly from simple C programs. While using llc with -march for mips and x86 works, -march=nvptx doesn't work. This seems reasonable since I can see that the libLLVMNVPTXAsmPrinter.a library is about 500 bytes (thus empty). Howev...

On Mon, Feb 4, 2013 at 10:46 AM, <nkavv at physics.auth.gr> wrote: > Hi all, > > I'm trying to use the newly added (in LLVM 3.2) NVPTX backend for > producing PTX (Parallel Thread eXecution) assembly from simple C programs. > > While using llc with -march for mips and x86 works, -march=nvptx doesn't > work. This seems reasonable since I can see that the > libLLVMNVPTXAsmPrinter.a library is about 500 bytes (thus...

Dear all, Looks like the NVPTX backend cannot handle array-of-arrays contant (please see the reporocase below). Is it supposed to work? Any ideas how to get it working? Important for our target applications. Thanks, - Dima. $ cat test.ll ; ModuleID = '__kernelgen_main_module' target datalayout = "e-p:64:64:64-i1:8...

I think our test case demonstrates that requiring the array item being initialized to be constant is incorrect. NVPTX does not crash anymore and produces correct result with the following change: --- NVPTXAsmPrinter.cpp 2012-09-03 15:14:00.000000000 +0200 +++ NVPTXAsmPrinter.cpp 2012-09-04 15:47:17.859398193 +0200 @@ -1890,17 +1890,15 @@ case Type::ArrayTyID: case Type::VectorTyID: case Type::StructTyID:...

NVCC successfully handles the same IR, if we try to process the same .cu file with clang+nvptx and nvcc: CLANG/NVPTX: ============= $ cat dayofweek.cu __attribute__((device)) char yweek[7][4] = { "MON", "TUE", "WED", "THU", "FRI", "SAT", "SUN" }; $ clang -cc1 -emit-llvm -fcuda-is-device dayofweek.cu -o dayofweek.ll $ cat...

On 09/04/2012 09:57 AM, Dmitry N. Mikushin wrote: > I think our test case demonstrates that requiring the array item being > initialized to be constant is incorrect. NVPTX does not crash anymore > and produces correct result with the following change: > > --- NVPTXAsmPrinter.cpp 2012-09-03 15:14:00.000000000 +0200 > +++ NVPTXAsmPrinter.cpp 2012-09-04 15:47:17.859398193 +0200 > @@ -1890,17 +1890,15 @@ > case Type::ArrayTyID: > case Type::V...

Hi Nikolaos, Following commands work great for me. $ clang -S -emit-llvm -target nvptx -x cl -include clc/clctypes.h ../data-types/scalar.cl $ llc -mcpu=sm_30 scalar.s You can follow Justin's blog [1]. It helped me a lot to understand where to start. [1] http://jholewinski.org/blog/llvm-3-0-ptx-backend/ Best, Ankur On Mon, Feb 4, 2013 at 11:40 PM, Justin Holewinski < just...

Hi Justin, >> Has anyone had similar problems with the NVPTX backend? Shouldn't this >> code be linked to the AsmPrinter library for NVPTX (already)? > > What do you mean by "doesn't work"? The AsmPrinter library really houses > the MCInst printer, which isn't implemented for NVPTX yet. The older > assembly printer w...

Dear LLVM NVPTX maintainers, Just to have the issue recorded, I don't know how important it is: clang generates linkonce_odr out of __inline__, and NVPTX generates .weak out of linkonce_odr (how it happens - a big question, btw, because I can't find anything related in NVPTX asm printer - does it chain t...

On Tue, Jun 12, 2012 at 6:11 PM, Dmitry N. Mikushin <maemarcus at gmail.com>wrote: > Dear LLVM NVPTX maintainers, > > Just to have the issue recorded, I don't know how important it is: > > clang generates linkonce_odr out of __inline__, and NVPTX generates .weak > out of linkonce_odr (how it happens - a big question, btw, because I can't > find anything related in NVPTX a...

On Mon, Feb 4, 2013 at 1:09 PM, <nkavv at physics.auth.gr> wrote: > Hi Justin, > > > Has anyone had similar problems with the NVPTX backend? Shouldn't this >>> code be linked to the AsmPrinter library for NVPTX (already)? >>> >> >> What do you mean by "doesn't work"? The AsmPrinter library really houses >> the MCInst printer, which isn't implemented for NVPTX yet. The...

In ptx, variables need to be defined before referenced. NVPTX emits the global variables in the order as in the LLVM IR and does not sort them. It is a bug in the NVPTX backend. Thanks. Yuan From: Dmitry N. Mikushin [mailto:maemarcus at gmail.com] Sent: Wednesday, July 18, 2012 7:44 AM To: LLVM-Dev Cc: Justin Holewinski; Yuan Lin Subject: [NVPTX] PTXAS -...

Dear NVPTX community, PTXAS fails to compile the ptx code generated by NVPTX. Is it an issue of backend or an issue of PTXAS or a known reasonable restriction? Thanks, - Dima. > cat test.ll ; ModuleID = '__kernelgen_main_module' target datalayout = "e-p:64:64-i64:64:64-f64:64:64-n1:8:16:32:...

Hi LLVMers, We at NVIDIA would like to contribute back to the LLVM open-source community by up-streaming the NVPTX back-end for LLVM. This back-end is based on the sources used by NVIDIA, and currently provides significantly more functionality than the current PTX back-end. Some functionality is currently disabled due to dependencies on LLVM core changes that we are also in the process of up-streaming, but th...

Hi everybody, I have noticed that the new NVPTX backend requires new metadata to identify the kernels in the module: define void @metadata_kernel(float* %a) { ret void } !nvvm.annotations = !{!1} !1 = metadata !{void (float*)* @metadata_kernel, metadata !"kernel", i32 1} Is clang going to support the emission of this metadata soon...

Dear colleagues, I'm looking if we can replace nvopencc with LLVM NVPTX in our project. It turns NVPTX won't work with the code nvopencc can handle (please see the log below). So are atomic intrinsics not supported or am I doing call in a wrong way? Thanks, - Dima. SOURCE ======== dmikushin at hp2:~> cat kernelgen_monitor.ll ; ModuleID = '/opt/kernelgen/i...

This new metadata format is currently optional. The old ptx_kernel calling convention should still work. The only thing you should have to change when converting from PTX -> NVPTX is the address space map. The calling conventions and intrinsics should be compatible with both. > -----Original Message----- > From: llvmdev-bounces at cs.uiuc.edu [mailto:llvmdev-bounces at cs.uiuc.edu] > On Behalf Of Alberto Magni > Sent: Monday, May 07, 2012 6:09 AM > To: LLVM...

Dear LLVM, FWIW, attached presentation provides a simple jump-start tutorial for newbies to start doing something useful with LLVM, DragonEgg, NVPTX backend and some custom high-level language. Along with short software overview from the user perspective it contains instructions on building components necessary to experiment with NVPTX and an example usecase of generating PTX assembler out of the Ada source code with help of DragonEgg. Best, -...

> -----Original Message----- > From: Dmitry N. Mikushin [mailto:maemarcus at gmail.com] > Sent: Wednesday, May 16, 2012 5:44 AM > To: LLVM-Dev > Cc: Justin Holewinski > Subject: NVPTX: __iAtomicCAS support ? > > Dear colleagues, > > I'm looking if we can replace nvopencc with LLVM NVPTX in our project. > It turns NVPTX won't work with the code nvopencc can handle (please > see the log below). So are atomic intrinsics not supported or am I > doing c...

...ement, container.end() being evaluated every iterations of loops, etc, etc.. -Jim On Apr 24, 2012, at 11:50 AM, Justin Holewinski <jholewinski at nvidia.com> wrote: > Hi LLVMers, > > We at NVIDIA would like to contribute back to the LLVM open-source community by up-streaming the NVPTX back-end for LLVM. This back-end is based on the sources used by NVIDIA, and currently provides significantly more functionality than the current PTX back-end. Some functionality is currently disabled due to dependencies on LLVM core changes that we are also in the process of up-streaming, but th...