Making the Tokio scheduler 10x faster

Re:Making the Tokio scheduler 10x faster by 15155 on Monday October 14, 2019 @01:44PM (#21250830)

IIRC: one of the issues with the original Rust stdlib greenthread implementation was that it's extremely difficult to preempt execution in a systems language. It's also quite difficult to add userland execution preemption without incurring significant performance penalties.

How do you preempt syscalls or FFI calls? (Go will almost certainly also have this issue with cgo modules)

I would be very surprised if preemptible Futures are something Tokio can or would implement.

[ Parent | on HN ]

Re:Making the Tokio scheduler 10x faster by wbl on Monday October 14, 2019 @04:48PM (#21252719)

What does systems mean here? There is a dirty trick with a NOP and self rewriting code that's quite cheap when not triggered. I think java does it that way.

[ Parent | on HN ]

Re:Making the Tokio scheduler 10x faster by sansnomme on Monday October 14, 2019 @05:24PM (#21253062)

Elaborate?

[ Parent | on HN ]

Re:Making the Tokio scheduler 10x faster by wbl on Monday October 14, 2019 @06:16PM (#21253469)

So you put in a NOP before each backward jump. To preempt you overwrite it with a jump to a handler that saves the state of the thread.

[ Parent | on HN ]

Re:Making the Tokio scheduler 10x faster by jedisct1 on Monday October 14, 2019 @06:51PM (#21253708)

How multitasking used to work on Atari :)

But this requires code pages to be writable. Which, from a security perspective is awful.

[ Parent | on HN ]

Re:Making the Tokio scheduler 10x faster by nine_k on Monday October 14, 2019 @07:56PM (#21254175)

This presumes writable code segments, which many OSes shun.

[ Parent | on HN ]

Re:Making the Tokio scheduler 10x faster by hyperman1 on Tuesday October 15, 2019 @07:09AM (#21257337)

It's java. Dynamic code generated by a jit.

[ Parent | on HN ]

Re:Making the Tokio scheduler 10x faster by adwn on Tuesday October 15, 2019 @12:15AM (#21255516)

Which NOP do you overwrite? Do you have to do this for all backward-jump-NOPs in all loaded executable code, or do you have to determine, which loop the thread in question is executing right now?

[ Parent | on HN ]

Re:Making the Tokio scheduler 10x faster by heavenlyblue on Monday October 14, 2019 @07:17PM (#21253914)

How does that resolve syscall question?

[ Parent | on HN ]

Re:Making the Tokio scheduler 10x faster by wbl on Monday October 14, 2019 @09:56PM (#21254886)

Syscall you have to do with a signal

[ Parent | on HN ]