johnnysswlab.com

After 1+ year of development, a new workshop is live: Software Optimizations for the Memory Subsystem Modern CPUs are fast. Your software would be too — if memory kept up. Learn more: johnnysswlab.com/optimizations-… Interested? johnnysswlab.com/contact/

Become C/C++/Rust software performance expert in just 16 hours! Information: johnnysswlab.com/optimizations-… Registration: johnnysswlab.com/contact/

In only 16 hours learn how to improve sw performance by better using the memory subsystem. We learn to improve the performance of many codes that work with matrices, hash maps, trees and others. More info at: johnnysswlab.com/optimizations-… Register: johnnysswlab.com/contact/

Vectorization doesn't always speed things up because of SIMD math. Sometimes it speeds things up because it forces you to overlap independent dependency chains. New post: johnnysswlab.com/exposing-more-…

🚀 New on Johnny’s Software Lab: Handling floating-point errors in C++ without killing performance! NaNs, infinities, sticky bits, and traps — what works and what’s a trap🪤. Read more: johnnysswlab.com/floating-point… #Cpp #Performance #FloatingPoint #NaN #Infinity

📬 Mailing list is live! New articles + workshop dates (AVX, NEON, memory optimizations) straight to your inbox. 👉 Go to johnnysswlab.com ➡️ Enter your email in the box on the right

You always wanted to learn AVX but didn't have time or resources? Learn AVX from beginning to end in only 16 hours! Workshop information: johnnysswlab.com/avx-neon-vecto… Register by mail: info@johnnysswlab.com

The discounted price workshop ends tomorrow. Hurry up!

You could learn AVX on your own… But it’ll take months. Or you can learn everything in 16 hands-on hours — intrinsics, loop vectorization, removing blockers, outer-loop tricks, and hitting peak performance. Register now: info@johnnysswlab.com More info: johnnysswlab.com/avx-neon-vecto…

@forked_franz lol

@johnnysswlab 🤗 to another way of wrapping assembly 😈 🤗

Still thinking #Java is slow? A deep dive into Java vs C++ performance will show what are its strengths and what are its weaknesses. johnnysswlab.com/deep-dive-in-j… #java #javaperformance #garbagecollector #jvm

⚡ Last call for the AVX Vectorization Workshop! 🗓️ Nov 24–25, 9–17h CET 💻 Learn SIMD, AVX, and compiler-level optimization hands-on! 💶 €749 / 💵 $799 📧 info@johnnysswlab.com 👉 johnnysswlab.com/avx-neon-vecto…

In this talk we are going to investigate our codes from the viewpoint of ILP: which code has more ILP, which has less ILP, how does ILP affect software performance and what techniques we can use to speed up our software. meetingcpp.com/mcpp/schedule/…

Summary: Instruction Level Parallelism (ILP) is the magic ingredient of a modern CPU that makes it run fast. But not all programs are equal, some have more ILP, others have less.

@i_bogosavljevic will be giving a talk at Meeting C++ in Berlin, November 6th at 17:15, in Track E [Amethyst Room]. Title of the talk: Instruction Level Parallelism and Software Performance

AVX workshop on Nov 24th and 25th. Hurry up, early bird discount ends in three days! More info: johnnysswlab.com/avx-neon-vecto… Register: info@johnnysswlab.com

A new online AVX workshop coming up! More info: johnnysswlab.com/avx-neon-vecto… Register: info@johnnysswlab.com

9 Things Every Fresh Graduate Should Know About Performance johnnysswlab.com/9-things-every…

I am really thinking about starting a new series of posts for @johnnysswlab, called either "Self-taught C/C++ Engineer" or "Self-taught Embedded Engineer" or something along those lines I have many friends and colleagues who work in the field of embedded systems and they are good at it, but they don't have an official software engineering education. And they told me that occasionally they feel this gap in their knowledge, missing knowledge that would help better understand the system they are currently developing. The topics could be: * Basics of assembly and mental model of how C code corresponds to assembly * Virtual memory * Basics of processes and threads - how is everything managed at OS level and how does it look from above * Memory allocation - stack allocation, heap allocation, allocation from the OS ETC * Static vs dynamic libraries The idea is like this: * Short five minute posts * One topic per posts * Simple explanation without too much details * Questions at the end using real world examples to ensure that you understood the essence of the post. I know most of the people here are experts and know a bunch of this stuff, but I am really curious what y'all think about it. There is a lot of content written on these topic, but again similar to performance it is all over internet.