https://bugs.kde.org/show_bug.cgi?id=523054

--- Comment #7 from [email protected] ---
(In reply to tomashnyk from comment #6)
> Is there a way to know which hardware did not settle? Some udev logs or
> something?

To answer myself, one needs to add udev.log_level=debug to kernel parameters (I
use systemd-boot, so I can just edit the file corresponding to my current
kernel here: /boot/efi/loader/entries/ with Grub it needs some magic commands).
It turned out the culprit was my badly written custom udev rule. When I fixed
it, the delay is now not 10 seconds, but about 700 ms:

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 @5.292s
└─plasmalogin.service @4.549s +742ms
  └─plymouth-quit.service @4.515s +31ms
    └─systemd-user-sessions.service @4.492s +16ms
      └─remote-fs.target @4.490s
        └─remote-fs-pre.target @4.489s
          └─nfs-client.target @4.489s
            └─gssproxy.service @4.444s +38ms
              └─network.target @4.441s
                └─wpa_supplicant.service @4.419s +21ms
                  └─basic.target @3.230s
                    └─dbus-broker.service @3.045s +86ms
                      └─dbus.socket @3.009s +4ms
                        └─sysinit.target @2.990s
                          └─systemd-binfmt.service @2.174s +815ms
                            └─proc-sys-fs-binfmt_misc.mount @2.952s +26ms
                              └─proc-sys-fs-binfmt_misc.automount @941ms

I would be curious if others see this as well on their installs. It is a lot
smaller, but it is still almost a seventh of my userspace boot time.

-- 
You are receiving this mail because:
You are watching all bug changes.

Reply via email to