Re: High CPU usage associated to busy loop in curl_timediff_us()

Daniel Stenberg via curl-library <[email protected]>
Newsgroups gmane.comp.web.curl.library
Message-ID <[email protected]>
On Fri, 12 Sep 2025, Stephane Cruchon wrote:

> I checked the history of Curl_splay function - last modification was :
>
> ------------------------------------------------------------------------------------
> $ git log -L :Curl_splay:lib/splay.c
> commit dcb51bafabf9d0776135de9111c1a3fe437d570c

No it's not the latest modification. There are four commits done since:

$ git log --oneline dcb51bafabf9d0776135de9111c1a3fe437d570c.. -- lib/splay.c
aec28f3ad3 splay: add another assert for detected usage problem
40caca581f splay: rename KEY_NOTUSED TO SPLAY_SUBNODE
255aac56f9 curlx: move into to curlx/
9cc246401e source: avoid use of 'very' in comments

But more importantly: one of the fixes I thought about was in how the splay is 
used, so not a change in lib/splay.c. See commit 019991c25ef0b08.

> Is it this modification that could help us maybe ?

Rather than wasting braincells on even thinking about that, it would just be 
helpful if you tried the latest curl release to see if the problem remains. 
Because if it works, then it doesn't matter which fix that fixed it, and if it 
doesn't work then it doesn't matter which fix we thought would fix it...

-- 

  / daniel.haxx.se || https://rock-solid.curl.dev
-- 
Unsubscribe: https://lists.haxx.se/mailman/listinfo/curl-library
Etiquette:   https://curl.se/mail/etiquette.html
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.