[dwenzek]

It's refreshing to see new approaches for memory management exploring notably the power of static analysis at compile time. I will take the time to read this dissertation!

Just a question. It reminds me previous works on static inference of stack-allocated regions. What are the relationships, if any?

* [A Simplified Account of Region Inference](https://hal.inria.fr/file/index/docid/72527/filename/RR-4104...)

* [MLton regions](http://mlton.org/Regions)

[zozbot234]

The GitHub readme links to a scholarly thesis discussing the general approach[0], as well as to the author's own dissertation[1] on this specific project (which "aims to investigate the practical viability" of [0] "by building the technology into a real compiler" for empirical evaluation on real-world hardware). Region inference is discussed throughout [0], and extensively in Chapter 9.

[0] https://www.cl.cam.ac.uk/techreports/UCAM-CL-TR-908.pdf

[1] http://nathancorbyn.com/nc513.pdf

[sitkack]

If the paper is anything like the abstract, this will be wonderful!

[tayistay]

Interesting as well. I see the MLTon regions have issues with bounding space complexity. I wonder if a language could impose some reasonable restrictions to bound region space complexity. That would seem to make regions quite attractive.