From: "Marek Marczykowski-Górecki" <marmarek@invisiblethingslab.com>
To: Jan Beulich <jbeulich@suse.com>
Cc: xen-devel <xen-devel@lists.xenproject.org>,
"Jürgen Groß" <jgross@suse.com>,
"Roger Pau Monné" <roger.pau@citrix.com>
Subject: Re: Slow (higher scheduling latency?) system after S3 - regression 4.17 -> 4.19 ?
Date: Thu, 2 Apr 2026 15:02:15 +0200 [thread overview]
Message-ID: <ac5l-HgbDRcVkiWb@mail-itl> (raw)
In-Reply-To: <acZ8ez4dmQay1JSw@mail-itl>
[-- Attachment #1: Type: text/plain, Size: 8571 bytes --]
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:
[2026-04-02 11:41:57] (XEN) [ 4.327519] mwait-idle: disabled
But then xenpm reports C1 is still used:
# xenpm get-cpuidle-states
Max possible C-state: C0
cpu id : 0
total C-states : 2
idle time(ms) : 3727115
C0 : transition [ 0]
residency [ 3769835 ms]
C1 : transition [ 0]
residency [ 0 ms]
cpu id : 1
total C-states : 2
idle time(ms) : 3763654
C0 : transition [ 1]
residency [ 0 ms]
C1 : transition [ 1]
residency [ 3769838 ms]
cpu id : 2
total C-states : 2
idle time(ms) : 3729168
C0 : transition [ 1]
residency [ 0 ms]
C1 : transition [ 1]
residency [ 3769838 ms]
cpu id : 3
total C-states : 2
idle time(ms) : 3750226
C0 : transition [ 1]
residency [ 0 ms]
C1 : transition [ 1]
residency [ 3769840 ms]
cpu id : 4
total C-states : 2
idle time(ms) : 3726369
C0 : transition [ 1]
residency [ 0 ms]
C1 : transition [ 1]
residency [ 3769840 ms]
cpu id : 5
total C-states : 2
idle time(ms) : 3755733
C0 : transition [ 1]
residency [ 0 ms]
C1 : transition [ 1]
residency [ 3769840 ms]
cpu id : 6
total C-states : 2
idle time(ms) : 3725747
C0 : transition [ 1]
residency [ 0 ms]
C1 : transition [ 1]
residency [ 3769841 ms]
cpu id : 7
total C-states : 2
idle time(ms) : 3730190
C0 : transition [ 1]
residency [ 0 ms]
C1 : transition [ 1]
residency [ 3769841 ms]
cpu id : 8
total C-states : 2
idle time(ms) : 3716485
C0 : transition [ 1]
residency [ 0 ms]
C1 : transition [ 1]
residency [ 3769841 ms]
cpu id : 9
total C-states : 2
idle time(ms) : 3710830
C0 : transition [ 1]
residency [ 0 ms]
C1 : transition [ 1]
residency [ 3769842 ms]
cpu id : 10
total C-states : 2
idle time(ms) : 3704881
C0 : transition [ 1]
residency [ 0 ms]
C1 : transition [ 1]
residency [ 3769842 ms]
cpu id : 11
total C-states : 2
idle time(ms) : 3719333
C0 : transition [ 1]
residency [ 0 ms]
C1 : transition [ 1]
residency [ 3769842 ms]
cpu id : 12
total C-states : 2
idle time(ms) : 3715059
C0 : transition [ 1]
residency [ 0 ms]
C1 : transition [ 1]
residency [ 3769842 ms]
cpu id : 13
total C-states : 2
idle time(ms) : 3711681
C0 : transition [ 1]
residency [ 0 ms]
C1 : transition [ 1]
residency [ 3769843 ms]
cpu id : 14
total C-states : 2
idle time(ms) : 3709380
C0 : transition [ 1]
residency [ 0 ms]
C1 : transition [ 1]
residency [ 3769843 ms]
cpu id : 15
total C-states : 2
idle time(ms) : 3720119
C0 : transition [ 1]
residency [ 0 ms]
C1 : transition [ 1]
residency [ 3769843 ms]
cpu id : 16
total C-states : 2
idle time(ms) : 3737637
C0 : transition [ 1]
residency [ 0 ms]
C1 : transition [ 1]
residency [ 3769844 ms]
cpu id : 17
total C-states : 2
idle time(ms) : 3737662
C0 : transition [ 1]
residency [ 0 ms]
C1 : transition [ 1]
residency [ 3769844 ms]
cpu id : 18
total C-states : 2
idle time(ms) : 3729870
C0 : transition [ 1]
residency [ 0 ms]
C1 : transition [ 1]
residency [ 3769844 ms]
cpu id : 19
total C-states : 2
idle time(ms) : 3745564
C0 : transition [ 1]
residency [ 0 ms]
C1 : transition [ 1]
residency [ 3769845 ms]
cpu id : 20
total C-states : 2
idle time(ms) : 3717089
C0 : transition [ 1]
residency [ 0 ms]
C1 : transition [ 1]
residency [ 3769845 ms]
cpu id : 21
total C-states : 2
idle time(ms) : 3590257
C0 : transition [ 1]
residency [ 0 ms]
C1 : transition [ 1]
residency [ 3769845 ms]
And the issue still happens...
Honestly, I'm confused about the above output. The system is mostly
idle, but I wouldn't expect it to be _that_ idle - there is xenstored +
xenconsoled running, plus occasional xl create for testing with
short-lived domU. Does it really all happen only on CPU0 here?
--
Best Regards,
Marek Marczykowski-Górecki
Invisible Things Lab
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]
next prev parent reply other threads:[~2026-04-02 13:02 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-10-25 12:23 Slow (higher scheduling latency?) system after S3 - regression 4.17 -> 4.19 ? Marek Marczykowski-Górecki
2026-03-25 19:13 ` Marek Marczykowski-Górecki
2026-03-26 8:21 ` Jan Beulich
2026-03-26 13:05 ` Marek Marczykowski-Górecki
2026-03-26 8:27 ` Jan Beulich
2026-03-26 12:44 ` Marek Marczykowski-Górecki
2026-03-26 13:20 ` Jan Beulich
2026-03-26 15:53 ` Marek Marczykowski-Górecki
2026-03-26 16:02 ` Jan Beulich
2026-03-27 12:47 ` Marek Marczykowski-Górecki
2026-04-02 8:23 ` Jan Beulich
2026-04-02 8:50 ` Jürgen Groß
2026-04-02 13:02 ` Marek Marczykowski-Górecki [this message]
2026-04-02 13:03 ` Jan Beulich
2026-04-03 13:55 ` Marek Marczykowski-Górecki
2026-04-06 15:29 ` Marek Marczykowski-Górecki
2026-04-07 6:14 ` Jan Beulich
2026-04-07 12:01 ` Marek Marczykowski-Górecki
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=ac5l-HgbDRcVkiWb@mail-itl \
--to=marmarek@invisiblethingslab.com \
--cc=jbeulich@suse.com \
--cc=jgross@suse.com \
--cc=roger.pau@citrix.com \
--cc=xen-devel@lists.xenproject.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.