Title: P4: I have compilled PyTorch with CUDA and CUDNN. [2025-06-03 Tue]
current recent process, which I share at my gentoo
overlay as a package. #dailyreport #deeplearning #gentoo #llvm #clang #programming #toolchain #pytorch #caffe2
Title: P4: I have compilled PyTorch with CUDA and CUDNN. [2025-06-03 Tue]
current recent process, which I share at my gentoo
overlay as a package. #dailyreport #deeplearning #gentoo #llvm #clang #programming #toolchain #pytorch #caffe2
Title: P3: I have compilled PyTorch with CUDA and CUDNN. [2025-06-03 Tue]
(GCC:binutils, LLVM:lld) and ABI. Between “toolchain”
and “build pipeline”.
Gentoo STL:
- libc++: sys-devel/gcc
- libstdc++: llvm-runtimes/libcxx
Gentoo libc: sys-libs/glibc and sys-libs/musl
I learned how Nvidia CUDA and CUDNN distribud and what
tools PyTorch have.
Also, I updated my daemon+script to get most heavy #dailyreport #deeplearning #gentoo #llvm #clang #programming #toolchain #pytorch #caffe2
Title: P0: I have compilled PyTorch with CUDA and CUDNN. [2025-06-03 Tue]
PyTorch is mainly a Python library with main part of
Caffe2 C++ library.
Main dependency of Caffe2 with CUDA support is
NVIDIA "cutlass" library (collection of CUDA C++
template abstractions). This library have "CUDA code"
that may be compiled with nvcc NVIDIA CUDA compiler,
distributed with nvidia-cuda-toolkit, or with LLMV #dailyreport #deeplearning #gentoo #llvm #clang #programming #toolchain #pytorch #caffe2
Title: P2: I have compilled PyTorch with CUDA and CUDNN. [2025-06-03 Tue]
compile PyTorch CUDA code with Clang++ compiler.
I learned cmake config files and difference between
Compiler Runtime Library (libgcc and libatomic,
LLVM/Clang: compiler-rt, MSVC:vcruntime.lib) and C
standard library (glibc, musl) and C++ Standard Library
(GCC: libstdc++, LLVM: libc++, MSVC STL) and linker #dailyreport #deeplearning #gentoo #llvm #clang #programming #toolchain #pytorch #caffe2
Title: P1: I have compilled PyTorch with CUDA and CUDNN. [2025-06-03 Tue]
Clang++ compiler. But llvm support CUDA only up to 12.1
version, but may be used to compile CUDA for sm_52
architecture. Looks like kneeling before NVIDIA. :)
Before installing dev-libs/cutlass you should do:
export CUDAARCHS=75
I sucessfully compiled cutlass, now I am trying to #dailyreport #deeplearning #gentoo #llvm #clang #programming #toolchain #pytorch #caffe2
Facebook hat eine neue Version seines Deep-Learning-Frameworks PyTorch angekündigt, die den Wechsel von Forschung und Training der Modelle zum Produktionseinsatz vereinfachen soll. Eine erste Beta ist im Laufe des Jahres geplant.
Das Open-Source-Projekt Detectron soll besonders der Forschung im Bereich der Objekterkennung dienen. Die in Python verfasste Software baut auf Caffe2 und ergänzt das Portfolio der APIs und Open-Source-Projekte für die Bildanalyse.