Skip to content

Instantly share code, notes, and snippets.

View samcom12's full-sized avatar
💭
available

Samir Shaikh samcom12

💭
available
  • CDAC
  • Pune
View GitHub Profile

GCC compiler optimization for ARM-based systems

2017-03-03 fm4dd

The gcc compiler can optimize code by taking advantage of CPU specific features. Especially for ARM CPU's, this can have impact on application performance. ARM CPU's, even under the same architecture, could be implemented with different versions of floating point units (FPU). Utilizing full FPU potential improves performance of heavier operating systems such as full Linux distributions.

-mcpu, -march: Defining the CPU type and architecture

These flags can both be used to set the CPU type. Setting one or the other is sufficient.

@samcom12
samcom12 / build-tensorflow-from-source.md
Created June 19, 2021 06:22 — forked from Brainiarc7/build-tensorflow-from-source.md
Build Tensorflow from source, for better performance on Ubuntu.

Building Tensorflow from source on Ubuntu 16.04LTS for maximum performance:

TensorFlow is now distributed under an Apache v2 open source license on GitHub.

On Ubuntu 16.04LTS+:

Step 1. Install NVIDIA CUDA:

To use TensorFlow with NVIDIA GPUs, the first step is to install the CUDA Toolkit as shown:

@samcom12
samcom12 / building_tensorflow.md
Created June 20, 2021 06:45 — forked from kmhofmann/building_tensorflow.md
Building TensorFlow from source

Building TensorFlow from source (TF 2.3.0, Ubuntu 20.04)

Why build from source?

The official instructions on installing TensorFlow are here: https://www.tensorflow.org/install. If you want to install TensorFlow just using pip, you are running a supported Ubuntu LTS distribution, and you're happy to install the respective tested CUDA versions (which often are outdated), by all means go ahead. A good alternative may be to run a Docker image.

I am usually unhappy with installing what in effect are pre-built binaries. These binaries are often not compatible with the Ubuntu version I am running, the CUDA version that I have installed, and so on. Furthermore, they may be slower than binaries optimized for the target architecture, since certain instructions are not being used (e.g. AVX2, FMA).

So installing TensorFlow from source becomes a necessity. The official instructions on building TensorFlow from source are here: ht

@samcom12
samcom12 / biotools_packages.txt
Created July 27, 2021 07:09 — forked from dansondergaard/biotools_packages.txt
Package names from bio.tools and various conda channels (anaconda, conda-forge, bioconda, r).
KNIME - Cellular phenotyping
Galaxy Image Analysis
COMPSs
YEASTRACT
WormBase
Reactome knowledgebase of biological pathways
PRIDE
PANTHER
AnalysisServiceService
Kyoto encyclopedia of genes and genomes KEGG
@samcom12
samcom12 / CUDA_Compilers.md
Created September 1, 2021 07:33 — forked from ax3l/CUDA_Compilers.md
CUDA Compilers