nihalpasham

CubeCL

#gpu #kernel #rust

High Level Overview:

• GPU kernels in Rust
• Comptime
  • Automatic vectorization
  • Instruction and shape specialization
• Loop unrolling

juanbono

use cranelift::prelude::*;
use cranelift_module::{Linkage, Module};
use cranelift_object::{ObjectBuilder, ObjectModule};

fn main() {
    // create a settings builder to configure the opt level
    let mut settings_builder = settings::builder();
    // disable optimizations
    // TODO: take the opt level in the CLI.
    settings_builder.set("opt_level", "none").unwrap();

mandrean

# Install xpm using npm
$ npm install --global xpm@latest

# Install xPack distribution of Arm toolchain with Apple M1 Silicon support
$ xpm install @xpack-dev-tools/[email protected]

# Add Arm toolchain binaries to PATH in your .bashrc or .zshrc
$ echo 'export PATH="$HOME/Library/xPacks/@xpack-dev-tools/arm-none-eabi-gcc/10.3.1-2.3.1/.content/bin:$PATH"' >> ~/.bashrc
$ source ~/.bashrc

mepcotterell

Implementing a Mutex

Use the following commands to compile and link the examples:

$ gcc -std=c17 -pedantic-errors -O0 -g -S mutex.c
$ as --gstabs -o mutex.o mutex.s 
$ gcc -o mutex mutex.o -lpthread

This implementation makes use of the C11 Atomic Operations Library.