Arch Linux

cmskipsey

Member

Registered: 2020-04-14

Posts: 14

Slow boot time for kernel, or is it...?

As the title implies, I get a slow boot time of about 24s.  This is a used Lenovo X240 I just picked up, so I don't know if its this SSD (480gb Patriot Burst) or just the system being a bit older.  Just after 'initramfs' it seems to kinda hang for about 10 sec... hence why I think its some issue that needs to be resolved

I've run

systemd-analyze

Startup finished in 3.668s (firmware) + 5.218s (loader) + 11.358s (kernel) + 3.774s (userspace) = 24.019s 
graphical.target reached after 3.774s in userspace.

So I guess my question is - is 11s for Kernel (or any of the other times) considered 'normal'?  My T470 (better specs of course) loads in about 11s total (about 1.5s for kernel)

and

systemd-analyze blame

1.034s dev-sda2.device
 901ms systemd-journal-flush.service
 659ms reflector.service
 508ms systemd-udevd.service
 273ms user@1000.service
 197ms upower.service
 173ms systemd-udev-trigger.service
 171ms systemd-logind.service
 146ms systemd-rfkill.service
 140ms NetworkManager.service
 135ms boot.mount
 130ms systemd-fsck@dev-disk-by\x2duuid-7307\x2dE6B6.service
 128ms systemd-tmpfiles-setup.service
 107ms lvm2-monitor.service
 106ms systemd-journald.service
  99ms systemd-tmpfiles-setup-dev.service
  59ms dbus.service
  54ms systemd-random-seed.service
  53ms modprobe@fuse.service
  46ms systemd-modules-load.service
  34ms wpa_supplicant.service
  32ms dev-hugepages.mount
  31ms dev-mqueue.mount
  29ms sys-kernel-debug.mount
  28ms sys-kernel-tracing.mount
  26ms systemd-sysctl.service
  26ms tmp.mount
  26ms systemd-backlight@backlight:intel_backlight.service
  25ms kmod-static-nodes.service
  23ms systemd-update-utmp.service
  23ms modprobe@configfs.service
  22ms modprobe@drm.service
  22ms polkit.service
  21ms systemd-remount-fs.service

And

systemd-analyze critical-chain

graphical.target @3.774s
└─multi-user.target @3.774s
  └─getty.target @1.991s
    └─getty@tty1.service @1.989s
      └─systemd-user-sessions.service @1.969s +16ms
        └─network.target @1.950s
          └─NetworkManager.service @1.809s +140ms
            └─dbus.service @1.738s +59ms
              └─basic.target @1.735s
                └─sockets.target @1.735s
                  └─dbus.socket @1.735s
                    └─sysinit.target @1.733s
                      └─systemd-update-utmp.service @1.710s +23ms
                        └─systemd-tmpfiles-setup.service @1.579s +128ms
                          └─local-fs.target @1.576s
                            └─boot.mount @1.441s +135ms
                              └─systemd-fsck@dev-disk-by\x2duuid-7307\x2dE6B6.service @1.309s +130ms
                                └─dev-disk-by\x2duuid-7307\x2dE6B6.device @1.307s

ewaller

Administrator

From: Pasadena, CA

Registered: 2009-07-13

Posts: 20,662

Re: Slow boot time for kernel, or is it...?

Seems a tad long for an SSD system.

ewaller@odin/home/ewaller % systemd-analyze
Startup finished in 3.451s (firmware) + 20ms (loader) + 1.385s (kernel) + 636ms (userspace) = 5.493s 
graphical.target reached after 615ms in userspace.
ewaller@odin/home/ewaller % 

You might look in the journal for clues, but it does not start until after process 1.  Maybe check dmesg.

OTOH, look at the bright side; I booted my company issued pig of a Windows laptop this morning.  It only takes 12 minutes to boot and be ready to work.

---

Nothing is too wonderful to be true, if it be consistent with the laws of nature -- Michael Faraday
The shortest way to ruin a country is to give power to demagogues.— Dionysius of Halicarnassus
---
How to Ask Questions the Smart Way

twelveeighty

Member

Registered: 2011-09-04

Posts: 1,453

Re: Slow boot time for kernel, or is it...?

Your systemd-analyze blame only adds up to 4.5-ish seconds. Turn off 'quiet' from your boot line and watch the boot process itself for any weirdness. Also check your startup journal for messages.

cmskipsey

Member

Registered: 2020-04-14

Posts: 14

Re: Slow boot time for kernel, or is it...?

Thanks for your replies folks - ok - turned off 'quiet' and watched boot.  Seems to hang for the longest time after "Loading initial ramdisk..." then after that pause, everything seems to load up quick.  So what is going on behind the "Loading initial ramdisk..." at startup.  dmesg --level=err,warn gives me this:

[    0.000000] efi: [Firmware Bug]: Invalid EFI memory map entries:
[    0.000000] efi: mem41: [Reserved    |   |  |  |  |  |  |  |  |  |   |  |  |  |  ] range=[0x0000000000000000-0x0000000000000000] (invalid)
[    0.000000] efi: mem42: [Reserved    |   |  |  |  |  |  |  |  |  |   |  |  |  |  ] range=[0x0000000000000000-0x0000000000000000] (invalid)
[    0.000000] efi: mem43: [Reserved    |   |  |  |  |  |  |  |  |  |   |  |  |  |  ] range=[0x0000000000000000-0x0000000000000000] (invalid)
[    0.000000] efi: mem44: [Reserved    |   |  |  |  |  |  |  |  |  |   |  |  |  |  ] range=[0x0000000000000000-0x0000000000000000] (invalid)
[    0.000000] efi: mem45: [Reserved    |   |  |  |  |  |  |  |  |  |   |  |  |  |  ] range=[0x0000000000000000-0x0000000000000000] (invalid)
[    0.000000] efi: mem46: [Reserved    |   |  |  |  |  |  |  |  |  |   |  |  |  |  ] range=[0x0000000000000000-0x0000000000000000] (invalid)
[    0.000000] efi: mem47: [Reserved    |   |  |  |  |  |  |  |  |  |   |  |  |  |  ] range=[0x0000000000000000-0x0000000000000000] (invalid)
[    0.000000] efi: Removing 7 invalid memory map entries.
[    0.106997] MDS CPU bug present and SMT on, data leak possible. See https://www.kernel.org/doc/html/latest/admin-guide/hw-vuln/mds.html for more details.
[    0.106997]  #2 #3
[    0.113774] ENERGY_PERF_BIAS: Set to 'normal', was 'performance'
[    0.300440] hpet_acpi_add: no address or irqs in _CRS
[   12.057919] acpi PNP0C14:01: duplicate WMI GUID 05901221-D566-11D1-B2F0-00A0C9062910 (first instance was on PNP0C14:00)
[   12.058132] acpi PNP0C14:02: duplicate WMI GUID 05901221-D566-11D1-B2F0-00A0C9062910 (first instance was on PNP0C14:00)
[   12.491058] platform regulatory.0: Direct firmware load for regulatory.db failed with error -2
[   12.709960] at24 0-0050: supply vcc not found, using dummy regulator
[   13.733514] ACPI: \: failed to evaluate _DSM (0x1001)
[   13.733632] ACPI: \: failed to evaluate _DSM (0x1001)
[   13.733756] ACPI: \: failed to evaluate _DSM (0x1001)
[   13.733793] ACPI: \: failed to evaluate _DSM (0x1001)
[   13.733803] ACPI: \: failed to evaluate _DSM (0x1001)
[   13.733813] ACPI: \: failed to evaluate _DSM (0x1001)
[   13.734075] ACPI: \: failed to evaluate _DSM (0x1001)
[   13.993625] ACPI: \: failed to evaluate _DSM (0x1001)
[   13.993897] ACPI: \: failed to evaluate _DSM (0x1001)
[   13.993909] ACPI: \: failed to evaluate _DSM (0x1001)
[   13.994190] ACPI: \: failed to evaluate _DSM (0x1001)
[   13.994251] ACPI: \: failed to evaluate _DSM (0x1001)
[   13.994539] ACPI: \: failed to evaluate _DSM (0x1001)
[   13.994564] ACPI: \: failed to evaluate _DSM (0x1001)
[   17.370741] kauditd_printk_skb: 42 callbacks suppressed
[  201.836565] kauditd_printk_skb: 3 callbacks suppressed

How much of this is actual errors vs other stuff that is harmless.  Also, the number at the start, is that time?  If so, the ACPI stuff seems to be creating a lot of delay...?

threefour

Member

Registered: 2022-07-25

Posts: 21

Re: Slow boot time for kernel, or is it...?

Are there any packages you installed recently that rebuilt your initramfs, that thereafter caused this slow boot? Or has this been from the beginning?

Last edited by threefour (2022-08-08 14:32:23)

cmskipsey

Member

Registered: 2020-04-14

Posts: 14

Re: Slow boot time for kernel, or is it...?

Nup, fresh install, the delay has been there from the beginning

seth

Member

From: Won't reply 2 private help req

Registered: 2012-09-03

Posts: 76,497

Re: Slow boot time for kernel, or is it...?

What size are the initramfs and kernel image?
What FS is the boot partition? Is it the same device as the root partition? Is it encrypted?

---

How to upload text · How to boot w/o GUI · Disable Windows Fast-Start! · Fix your xinitrc