Re: Slow (higher scheduling latency?) system after S3 - regression 4.17 -> 4.19 ?

[Marek Marczykowski-Górecki <marmarek@xxxxxxxxxxxxxxxxxxxxxx>]

On Fri, Apr 03, 2026 at 03:55:07PM +0200, Marek Marczykowski-Górecki wrote:
> On Thu, Apr 02, 2026 at 03:02:15PM +0200, Marek Marczykowski-Górecki wrote:
> > On Fri, Mar 27, 2026 at 01:47:55PM +0100, Marek Marczykowski-Górecki wrote:
> > > Some further observation, this time regarding timers:
> > 
> > In the meantime Roger suggested it might be about C-states. So, I just
> > tried booting with max_cstate=0. I got this:
> 
> We had a design session on this issue, and managed to find (thanks Roger!)
> that adding:
> 
>     tsc_mode='always_emulate'
> 
> to the domU config fixes the issue. Now, the question is how to fix it
> properly...

I also did another test with NTP inside domU (after host S3):

$ ntpdate -q pool.ntp.org
2026-04-06 00:07:01.559218 (+0200) -0.216281 +/- 0.012486 pool.ntp.org 
141.144.241.16 s2 no-leap
$ ntpdate -q pool.ntp.org
2026-04-06 00:57:40.852612 (+0200) -0.255782 +/- 0.013022 pool.ntp.org 
185.252.140.125 s2 no-leap
$ ntpdate -q pool.ntp.org
2026-04-06 13:07:57.647126 (+0200) -0.791802 +/- 0.056578 pool.ntp.org 
185.248.188.98 s1 no-leap
$ ntpdate -q pool.ntp.org
2026-04-06 13:08:01.844118 (+0200) -0.832599 +/- 0.012563 pool.ntp.org 
185.248.188.98 s1 no-leap
$ ntpdate -q 141.144.241.16
2026-04-06 13:08:47.835271 (+0200) -0.819725 +/- 0.028655 141.144.241.16 s2 
no-leap
$ ntpdate -q 141.144.241.16
2026-04-06 13:08:51.290526 (+0200) -0.833668 +/- 0.015501 141.144.241.16 s2 
no-leap
$ ntpdate -q 141.144.241.16
2026-04-06 13:08:56.136567 (+0200) -0.832294 +/- 0.014493 141.144.241.16 s2 
no-leap
$ ntpdate -q 141.144.241.16
2026-04-06 17:21:41.017554 (+0200) -1.044511 +/- 0.021817 141.144.241.16 s2 
no-leap

So, domU's clock definitely drifts.

-- 
Best Regards,
Marek Marczykowski-Górecki
Invisible Things Lab

Attachment: signature.asc
Description: PGP signature