[Bug 57399] Many unnecessary CPU wakeups per second

https://bz.apache.org/bugzilla/show_bug.cgi?id=57399 Yann Ylavic changed: What|Removed |Added Status|NEW |RESOLVED

[Bug 57399] Many unnecessary CPU wakeups per second

https://bz.apache.org/bugzilla/show_bug.cgi?id=57399 Szőgyényi Gábor changed: What|Removed |Added CC|

[Bug 57399] Many unnecessary CPU wakeups per second

https://bz.apache.org/bugzilla/show_bug.cgi?id=57399 Yann Ylavic changed: What|Removed |Added Attachment #34557|0 |1 is

[Bug 57399] Many unnecessary CPU wakeups per second

https://bz.apache.org/bugzilla/show_bug.cgi?id=57399 --- Comment #16 from Stefan Priebe --- Hi Yann, while testing V6 i'm experiencing segfaults. exit signal Segmentation server-error.log: AH00052: child pid 14110 exit signal Segmentation fault (11) currently i'm trying to

[Bug 57399] Many unnecessary CPU wakeups per second

https://bz.apache.org/bugzilla/show_bug.cgi?id=57399 Yann Ylavic changed: What|Removed |Added Attachment #34300|0 |1 is

[Bug 57399] Many unnecessary CPU wakeups per second

https://bz.apache.org/bugzilla/show_bug.cgi?id=57399 --- Comment #14 from Stefan Priebe --- This one no longer applies to httpd 2.4.25. -- You are receiving this mail because: You are the assignee for the bug.

[Bug 57399] Many unnecessary CPU wakeups per second

https://bz.apache.org/bugzilla/show_bug.cgi?id=57399 --- Comment #13 from Luca Toscano --- Anybody that wants to test this patch and provide feedback is more than welcome. Eventually it will become a 2.4.x backport proposal but more testing is probably needed (especially

[Bug 57399] Many unnecessary CPU wakeups per second

https://bz.apache.org/bugzilla/show_bug.cgi?id=57399 Yann Ylavic changed: What|Removed |Added Keywords||FixedInTrunk ---

[Bug 57399] Many unnecessary CPU wakeups per second

https://bz.apache.org/bugzilla/show_bug.cgi?id=57399 Yann Ylavic changed: What|Removed |Added Attachment #34264|0 |1 is

[Bug 57399] Many unnecessary CPU wakeups per second

https://bz.apache.org/bugzilla/show_bug.cgi?id=57399 Yann Ylavic changed: What|Removed |Added Attachment #34263|0 |1 is

[Bug 57399] Many unnecessary CPU wakeups per second

https://bz.apache.org/bugzilla/show_bug.cgi?id=57399 --- Comment #9 from Yann Ylavic --- Created attachment 34264 --> https://bz.apache.org/bugzilla/attachment.cgi?id=34264=edit mpm_event's listener wakeup (with fudge factor) for 2.4.x -- You are receiving this mail

[Bug 57399] Many unnecessary CPU wakeups per second

https://bz.apache.org/bugzilla/show_bug.cgi?id=57399 Yann Ylavic changed: What|Removed |Added Attachment #34167|0 |1 is

[Bug 57399] Many unnecessary CPU wakeups per second

https://bz.apache.org/bugzilla/show_bug.cgi?id=57399 Stefan Priebe changed: What|Removed |Added CC||ste...@priebe.ws --

[Bug 57399] Many unnecessary CPU wakeups per second

https://bz.apache.org/bugzilla/show_bug.cgi?id=57399 --- Comment #7 from Yann Ylavic --- Thanks for testing. Regarding ap_wait_or_timeout(), waking up waitpid() is another beast, possibly with a signal (but signals and threads...). This PR looks fine (generic enough) to

[Bug 57399] Many unnecessary CPU wakeups per second

https://bz.apache.org/bugzilla/show_bug.cgi?id=57399 --- Comment #6 from Anders Kaseorg --- (In reply to Yann Ylavic from comment #5) > Would you please test it Anders? Thanks, that’s a massive improvement. Wakeups are reduced from ~30 per second to 1 per second. That

[Bug 57399] Many unnecessary CPU wakeups per second

https://bz.apache.org/bugzilla/show_bug.cgi?id=57399 Yann Ylavic changed: What|Removed |Added Attachment #34142|0 |1 is

[Bug 57399] Many unnecessary CPU wakeups per second

https://bz.apache.org/bugzilla/show_bug.cgi?id=57399 --- Comment #4 from Luca Toscano --- Something that may be an issue: the calls to process_timeout_queue to manage Keep Alives, Lingering close and Write completion sockets. The function seems to be called after each

[Bug 57399] Many unnecessary CPU wakeups per second

https://bz.apache.org/bugzilla/show_bug.cgi?id=57399 Luca Toscano changed: What|Removed |Added CC|

[Bug 57399] Many unnecessary CPU wakeups per second

https://issues.apache.org/bugzilla/show_bug.cgi?id=57399 --- Comment #1 from Eric Covener cove...@gmail.com --- which says if there’s no reason to wake up, wake up every 100 ms anyway. apr_time_from_msec(100) should be replaced with -1. Wouldn't the server then miss timer events registered

[Bug 57399] Many unnecessary CPU wakeups per second

https://issues.apache.org/bugzilla/show_bug.cgi?id=57399 --- Comment #2 from Anders Kaseorg ande...@mit.edu --- Looking at that code, if replacing 100 ms with -1 would cause the server to miss timers registered when no timers are in progress, then it must already be the case that the server will