Re: PHP performance on Xen domU with mulitple vcpu

[Greg Troxel <gdt%lexort.com@localhost>]

Emmanuel Dreyfus <manu%netbsd.org@localhost> writes:

> On Thu, Apr 03, 2025 at 11:00:50AM -0400, Greg Troxel wrote:
>> Have you written a test program to log and examined the return values?
>
> I ran gettimeofday 1M time, it took 5.12, which is much faster than
> what happens in PHP. Here is the distribution of results:
>
> count  value
> 239726 1743694322.907498
> 119400 1743694324.187487
> 327599 1743694325.467476
> 174425 1743694326.747465
> 138850 1743694328.027453

Wow, that is messed up!

>> Is xen somehow using a timecounter that is incrementing in larger jumps?
>
> Oh, yes, good pick! I use kern.timecounter.hardware=clockinterrupt
> because with xen_system_time the domU's ntpd is unable to keep in sync.
>
> With kern.timecounter.hardware=xen_system_time, the uniqid() test runs
> at normal speed.

It sounds then like the clockinterrupt timecounter is seriously broken.

I guess someone(tm) needs to fix xen_system_time to be ok on MP.