Re: CPU usage since 8.7.1

> Am 28.10.2025 um 23:15 schrieb Gleb Smirnoff <[email protected]>:
> 
> On Tue, Oct 28, 2025 at 11:06:05PM +0100, Daniel Stenberg wrote:
> D> > This may actually help our case!
> D> > I'm going to give a try to unmodified 8.16.0.
> D> 
> D> It could be worth exploring more exactly what we're talking about. I mean
> D> what measurements to do and how to tell what's "good" and what's "bad" in
> D> this context.
> 
> The microbenchmark David used is desribed in his email:
> 
> https://www.mail-archive.com/[email protected]/msg02593.html
> 
> The regression was discovered by a huge A/B test of a system overall,
> that involves many machines, many hours and a ton of traffic.  What
> I'm going to do now is to re-run this A/B test with 8.16.0, to avoid
> relying on year old data.

Waiting for your results then. The "limit-rate revisisted" PR is https://github.com/curl/curl/pull/18454

I noticed "busy" loops where we expired transfers that are on a speed limits. Hope that addresses the cpu issue for you. As to the accuracy, we'll see what you find in your setup.

Cheers,
Stefan

PS. This would be a good opportunity to sponsor the project. You seem to get quite some mileage out of it. =)

> 
> -- 
> Gleb Smirnoff

-- 
Unsubscribe: https://lists.haxx.se/mailman/listinfo/curl-library
Etiquette:   https://curl.se/mail/etiquette.html