All of lore.kernel.org
 help / color / mirror / Atom feed
From: Greg KH <gregkh@linuxfoundation.org>
To: Alvaro Kuolas <kuolas@gmail.com>
Cc: regressions@lists.linux.dev
Subject: Re: Longer boot times with kernel 6.11.7 and newer (including 6.12.6).
Date: Fri, 27 Dec 2024 08:58:53 +0100	[thread overview]
Message-ID: <2024122756-ravioli-rabid-5692@gregkh> (raw)
In-Reply-To: <CAKUwrzBKjtxYkGoL-=mrJ9PoN_PACW0VO1Si1rUdLXfvW_h1cA@mail.gmail.com>

On Thu, Dec 26, 2024 at 02:51:13PM -0300, Alvaro Kuolas wrote:
> Hello!
> 
> I have experienced a delay booting Gnome graphical user interface under
> Fedora 41. This occurred when upgrading to kernel 6.12.4, but I have tested
> 6.13.0-rc3 and the slowdown is still present.
> 
> I have compiled versions 6.11.4 to 6.11.11, 6.12.6 and 6.13.0-rc3 and
> traced the regression on version 6.11.7.

6.11.y is long end-of-life, we can't do anything there, sorry.  But for
6.12.y we can.

> Using systemd-analyze I gathered the following information:
> (attached are more output from the different kernels that I have tested)
> 
> -------------------------------------------------------------------------------------------------------------------------------
> 
> Fedora Kernel 6.11.4-301.fc41.x86_64
> 
> [alvarok: boot]$ systemd-analyze time
> Startup finished in 11.270s (firmware) + 4.419s (loader) + 1.867s (kernel)
> + 3.465s (initrd) + 10.087s (userspace) = 31.110s
> graphical.target reached after 10.047s in userspace.
> 
> [alvarok: ~]$ systemd-analyze critical-chain
> The time when unit became active or started is printed after the "@"
> character.
> The time the unit took to start is printed after the "+" character.
> 
> graphical.target @10.047s
> └─multi-user.target @10.047s
>   └─plymouth-quit-wait.service @8.125s +1.920s
>     └─systemd-user-sessions.service @8.094s +29ms
>       └─remote-fs.target @8.092s
>         └─remote-fs-pre.target @2.524s
>           └─nfs-client.target @2.524s
>             └─gssproxy.service @2.093s +22ms
>               └─network.target @2.091s
>                 └─wpa_supplicant.service @3.284s +22ms
>                   └─basic.target @1.715s
>                     └─dbus-broker.service @1.686s +27ms
>                       └─dbus.socket @1.676s
>                         └─sysinit.target @1.673s
>                           └─systemd-resolved.service @1.628s +44ms
> 
> └─run-credentials-systemd\x2dresolved.service.mount @2.471s
> 
> [alvarok: ~]$ systemd-analyze blame
> 13.401s fwupd.service
>  5.999s NetworkManager-wait-online.service
>  4.051s sys-module-fuse.device
>  3.970s
> sys-devices-platform-serial8250-serial8250:0-serial8250:0.3-tty-ttyS3.device
>  3.970s dev-ttyS3.device
>  3.968s
> sys-devices-platform-serial8250-serial8250:0-serial8250:0.1-tty-ttyS1.device
>  3.968s dev-ttyS1.device
>  3.967s dev-tpmrm0.device
>  3.967s sys-devices-LNXSYSTM:00-LNXSYBUS:00-MSFT0101:00-tpmrm-tpmrm0.device
> 
> -------------------------------------------------------------------------------------------------------------------------------
> 
> Vanilla Linux 6.11.7 #1 SMP PREEMPT_DYNAMIC Thu Dec 26 14:11:07 -03 2024
> x86_64 GNU/Linux
> 
> [alvarok: ~]$ systemd-analyze time
> Startup finished in 11.278s (firmware) + 6.574s (loader) + 2.310s (kernel)
> + 3.542s (initrd) + 28.050s (userspace) = 51.755s
> graphical.target reached after 28.038s in userspace.
> 
> [alvarok: ~]$ systemd-analyze critical-chain
> The time when unit became active or started is printed after the "@"
> character.
> The time the unit took to start is printed after the "+" character.
> 
> graphical.target @28.038s
> └─upower.service @1.824s +26.214s
>   └─basic.target @1.783s
>     └─dbus-broker.service @1.754s +27ms
>       └─dbus.socket @1.744s
>         └─sysinit.target @1.741s
>           └─systemd-resolved.service @1.694s +46ms
>             └─run-credentials-systemd\x2dresolved.service.mount @2.493s
> 
> [alvarok: ~]$ systemd-analyze blame
> 26.214s upower.service
> 25.112s tuned-ppd.service
> 20.057s cups.service

So upower is taking a long time now, makes sense, as we fixed a
regression there, as you found in your next email.

thanks,

greg k-h

      parent reply	other threads:[~2024-12-27  7:58 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-12-26 17:51 Longer boot times with kernel 6.11.7 and newer (including 6.12.6) Alvaro Kuolas
2024-12-26 21:33 ` Alvaro Kuolas
2024-12-27  8:01   ` Greg KH
2024-12-27 12:37     ` Alvaro Kuolas
     [not found]     ` <CAKUwrzA2ue3g3Ys=Vg=eeunE1mw-RSockLPoqnU=F6jB1jjVBw@mail.gmail.com>
2024-12-27 12:40       ` Greg KH
2024-12-27 15:28         ` Alvaro Kuolas
2024-12-27  7:58 ` Greg KH [this message]

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=2024122756-ravioli-rabid-5692@gregkh \
    --to=gregkh@linuxfoundation.org \
    --cc=kuolas@gmail.com \
    --cc=regressions@lists.linux.dev \
    /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.