Forward Progress Guarantees in C++ - Olivier Giroux - CppNow 2023

www.cppnow.org​
/ cppnow
---
Forward Progress Guarantees in C++ - Olivier Giroux - CppNow 2023
Slides: github.com/boostcon
---
If you are coming to a C++ conference, then you probably have heard of the memory model before now. But what have you heard about forward progress guarantees in C++? Without forward progress, most synchronization is meaningless, making the topic a pre-requisite.
Unlike the memory model, however, it is not really an option to leave the topic to experts. That’s because assumptions of progress are implicitly part of the contract for calling libraries that synchronize inside. So, unless you are writing purely serial code, you are making assumptions about forward progress in your C++ code.
This talk will provide a whirlwind tour of the subject, but spend a good chunk of time on where the C++ specification lacks clarity and could be improved.
---
Olivier Giroux
I am a veteran GPU architect of ＞20 years, an ISO C++ committee member of ＞10 years, and its chair of concurrency and parallelism for the past 5 years. I'm the only hardware engineer on the committee. I like to talk about formalizations of parallelism and memory models.
---
Video Sponsors: think-cell and Bloomberg Engineering
Audience Audio Sponsors: Innoplex and Maryland Research Institute
---
Videos Filmed & Edited By Bash Films: bashfilms.com/
CZcams Channel Managed & Optimized By Digital Medium Ltd: events.digital-medium.co.uk
---
CppNow 2024
www.cppnow.org​
/ cppnow
---
#boost #cpp #cppprogramming

[@russCoding]

Thank you Olivier Giroux for the great talk. I and others really appreciate your time helping others to understand the intrinsics of how concurrency and parallelism work under the hood. Representing the code as graphs really help to illustrate any given point.

[@RishabhRD]

so much thanks for the great talks

[@BoostCon]

So pleased to hear that you enjoy them!

[@afigegoznaet]

Amazing talk

[@chuanweizhang3164]

I don't quite understand 34:05, why are we able to enter the loop. For example, when 11 tasks with 2 threads, when dealing with the last task, there would be a problem. But I don't see why we would enter a livelock when numbers can be evenly divided.