Jon Gjengset

Also, for international shipping, the book will be distributed through Penguin Random House, so just like TRPL it should be available at local retailers (whether online or brick-and-mortar) once finished, which will save you the shipping cost :)

Oh yeah, and for those asking about other electronic formats than PDF:https://twitter.com/nostarch/status/1400113023491227648 …

I was also lucky enough to have @davidtolnay agree to be the technical reviewer for the book, and his input so far has been absolutely fantastic and has made the book better than I could have done on my own

I'm still working on the later chapters, hence the early access rather than a full release, but they're all well on their way, and will become available as part of the early access to anyone who gets it now!

It's written so that each chapter is more or less standalone, so you can read up on the subjects you want to know more about, rather than having to read through the whole thing start-to-finish. Though of course you can do that too if you want!

The book is, as its name implies, written for people who are already familiar with Rust. The idea is that you read The Rust Programming Language first, play around with Rust for a bit on your own, maybe start using it "for real", and then pick this up to hone your skills.

I wonder if there's any chance of an in-person @rustlang conference in 2021? I miss those so much. Or will we have to wait until 2022?

For more context on hazard pointers in relation to Rust and epoch-based reclamation, see ticki's post https://ticki.github.io/blog/fearless-concurrency-with-hazard-pointers/ …. And for more context on epoch-based reclamation, see @aaron_turon's post https://aturon.github.io/blog/2015/08/27/epoch/ …. Did I mention rabbit holes?

Why am I thinking about this at 9:30pm the evening before a long stream? Well, because I like rabbit holes, and because I realized that memory reclamation will be a big, obvious gaping hole in what we implement tomorrow, and I don't want to fill it with epoch-based reclamation.

For reference, I'm talking about the code rooted here: https://github.com/facebook/folly/blob/master/folly/synchronization/Hazptr.h …. The original hazard pointer algorithm is described in https://erdani.org/publications/cuj-2004-12.pdf … (by @incomputable and Maged Michael).

It'd be really fun to port Facebook's hazard pointer implementation in Folly to Rust. Would potentially be great for implementing concurrent algorithms in Rust going forward, since concurrent memory reclamation is hard. Maybe that's a good topic for the next stream?

Quick reminder that this is happening in 18 hours!https://twitter.com/jonhoo/status/1394369764273713153 …

Luckily I managed to get in touch with Shahar, and now have a copy of the code that we can use for the stream. I'm also hoping I'll be able to share that code in the eventual repository for easy reference!

Oh, and the delay in getting this off the ground was primarily that I wanted to get a hold of the original (Java) source code for reference. These algorithms can be tricky and have a lot of details that aren't easy to extrapolate from the paper, so having running code is helpful.

For those who want _more_ to read, Shahar was also awarded the @podc_disc dissertation of the year award for his thesis on this exact algorithm in 2016, which is pretty cool! https://www.podc.org/dissertation/2016-dissertation-award/ …