Re: [PATCH experiment 00/35] stackless coroutine backend

[Stefan Hajnoczi]

On Thu, Mar 10, 2022 at 09:14:07PM +0100, Paolo Bonzini wrote:
> On 3/10/22 18:42, Stefan Hajnoczi wrote:
> > There are a lot of details to decide on in the translator tool and
> > runtime to optimize the code. I think the way the stack frames are
> > organized in this patch series is probably for convenience rather than
> > performance.
> 
> Yes, sometimes the optimizations are there but mostly because they made my
> job easier.
> 
> > Out of curiousity, did you run the perf tests and compare against
> > ucontext?
> 
> Not quite voluntarily, but I noticed I had to add one 0 to make them run for
> a decent amount of time.  So yeah, it's much faster than siglongjmp.

That's a nice first indication that performance will be good. I guess
that deep coroutine_fn stacks could be less efficient with stackless
coroutines compared to ucontext, but the cost of switching between
coroutines (enter/yield) will be lower with stackless coroutines.

Stefan

signature.asc
Description: PGP signature