I just want to refresh that this commit helped to restore the precision of rate throttling. Before it, the speed deviations from the limit were more than 10% in certain cases.
And depending on the network speed, they occurred on relatively big downloads (i.e. 5 MB) as well. I think that CPU usage with this commit should be compared to previous releases, like 7.84.0, which had a quite precise speed limit. Thanks, Dmitry Karpov From: curl-library <curl-library-boun...@lists.haxx.se> On Behalf Of David Pfitzner via curl-library Sent: Wednesday, May 15, 2024 4:04 AM To: Daniel Stenberg <dan...@haxx.se> Cc: David Pfitzner <dpfitz...@netflix.com>; David Pfitzner via curl-library <curl-library@lists.haxx.se> Subject: [EXTERNAL] Re: Increase in CPU usage in 8.7.1 vs 8.6.0 for rate-limited downloads On Wed, May 15, 2024 at 5:31 PM Daniel Stenberg <dan...@haxx.se<mailto:dan...@haxx.se>> wrote: On Wed, 15 May 2024, David Pfitzner via curl-library wrote: > From testing, this change appears to be due to this commit: > > https://github.com/curl/curl/commit/db5c9f4f9e0779b49624752b135281a0717b277b So just reverting this single commit makes it "good" again? Correct. > Perhaps it would be useful for a user of libcurl to be able to (somehow) > control this tradeoff between rate-limiting accuracy and CPU usage? Perhaps getting more data would be a first step. How big difference in rate-limit accuracy does this commit make in your case? I have not looked at that carefully, but casually I don't see much difference if any. But possibly closer inspection may find a systematic difference. I also suspect one may see bigger differences in some regime different to what I'm looking at - eg, smaller files. -- David
-- Unsubscribe: https://lists.haxx.se/mailman/listinfo/curl-library Etiquette: https://curl.se/mail/etiquette.html
