Jim Newsome

It might be worth considering dropping ubuntu 18.04, too. It's not EOL until 2023, but it's holding back our minimum cmake version. #2418 OTOH as code moves into Rust maybe...

Came here to file the same issue for `process_vm_readv`, which has the same problem. IANA expert, but I started digging into this when I was writing a wrapper for `process_vm_readv`,...

Example in the Rust playground that (I think) implements the same behavior as these methods. It works with `unsafe`, but doesn't pass the Miri checks. https://play.rust-lang.org/?version=stable&mode=debug&edition=2018&gist=8cb8a4507c42d86f6efd8a81d31f89d6

> In the meantime I think I may be able to capture the compiler commands by wrapping the compiler with a shell script... That was ultimately our solution in [shadow](https://github.com/shadow/shadow),...

This is partly fixed now. See e.g. #3055 and #3064. There are now alternate modes of operation that don't try to store anything in native TLS, but native TLS is...

This is now implemented

@scanlime thanks! Yeah I agree it's worth taking a closer look at rr's approach. It's definitely nice to have an option more deterministic than the other performance counters.

Another motivating example: arti's way of using the obs4 pluggable transport currently involves forking and execing an obs4 process. Investigating whether it makes sense to add an alternative mechanism to...

> fwiw, using fork/exec to start a pluggable transport such as obfs4 is not an arti thing, it also concerns tor, and is actually a [requirement from the PT spec](https://gitweb.torproject.org/torspec.git/tree/pt-spec.txt#n100)....

Starting to look at this now. My plan is roughly: * Migrate the `clone` handler to Rust * Implement `clone3` (which is a superset of `clone`), and change the `clone`...