Valgrind also does stuff like memory tracking and memory-profiling, so this is great also from a performance tracking point of view.

There are definitely some parts that are common between these that it could be nice to propagate to the standard. https://alexanderpetros.com/triptych/ lists some of them but I'm not certain how specific to htmx it is or how much it apply to the different variants.

If nothing else, it is a path to making them hotswappable.

It might have been a good idea to do that communication BEFORE creating all that drama.

No. Getaddrinfo is libc, not the kernel. It is of course possible, but complicated, to implement dns resolution with io_uring, but making it behave the same as glibc is very much a nontrivial piece of work.

To be fair, the same is very true with Wayland - you can't do much without extensions.

Uh, io_uring does that just fine.

That's yet another thing though.

It is part of the kernel and it does async file writes, so I don't really understand your objection.

Not all courts post their judgements online with full access for everyone.

I've seen plenty of times where the problems has been explained to Kent. But he just don't give a shit about the problems of people that isn't himself or that doesn't use his file system experiences.

It seems very clear to me that it's almost always a "you can't argue canon law with the Pope" situation - the rules say no new features, and it doesn't matter what the definition of "feature" is if the definition AND the rule come from the same person, Linus.

You can't win a rules-lawyer argument with the rulemaker.

Well, io_uring came along and removed a lot of the incentive.