Announcement

Collapse
No announcement yet.

FreeBSD Shortening Its Boot Time, ASLR By Default & Better Intel WiFi Support

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • #11
    Originally posted by er888kh View Post

    Manjaro can definitely be improved. It has a couple of scripts that take the majority of time. have you tried "systemd-analyze plot" ?
    Yeah, but there isn't much I can do.
    • Kernel to systemd is 5 secs
    • systemd to GUI is another 12 secs
    • 17 seconds total

    Not much I can do about the 5 kernel seconds (maybe switch from un. 6 of those 11 systemd ones are importing ZFS pools and mounting datasets. I could turn that off but then I'd have to manually do it every single boot. There's another second used by my Windows disk. From there, everything is minuscule in regards to milliseconds used.

    Wait...right after posting that I realized that I forgot to set /etc/mkinitcpio.conf to from gzip to lz4. That might shave something off the kernel's 5 seconds...

    Annnnd....there was no change when switching to LZ4.
    Last edited by skeevy420; 11 March 2022, 02:18 PM.

    Comment


    • #12
      Originally posted by skeevy420 View Post
      ...
      My advice: try to trim your initramfs and use zstd rather than lz4.

      Comment


      • #13
        Originally posted by stormcrow View Post

        I've used it for exactly that purpose - ZFS mirrors served via Samba. For some reason it always idled at 10W more than on the same machine with Debian (35W on FreeBSD vs. 25W on Debian). I didn't really dig in why beyond trying out powerd and power++. I've read FreeBSD plays ultra safe on not trusting ACPI tables so the CPU didn't idle properly might be the cause. I might dig in deeper sometime in the future and find out if it can be better tuned. Right now everything is packed away for a long distance move.
        Thanks for your take on this, always enjoy hearing about other people's experiences. Obviously with a powered on 24/7 file server, nice to see good efficiency. But yeah, happy to hear your take. Good luck with the move, that can be a lot of work, but once done it is done! That must have been you on that other thread who got rid of the old P4 or whatever you were holding onto

        Comment


        • #14
          Originally posted by Danielsan View Post
          It doesn't bother me the boot's slowness of FreeBSD however still feels weird in 2022...
          It feels weird that OS's have to boot in 2022. Uh huh.

          Comment


          • #15
            Wireless AC support...when?

            edit: Yes this matters for pfSense reasons, and there really isn't anything to replace pfSense in the space its in.

            Comment


            • #16
              Originally posted by GI_Jack View Post
              Wireless AC support...when?
              Maybe by 14.

              Comment


              • #17
                Originally posted by stormcrow View Post
                Windows cheats by default. It uses a hybrid boot mode that's a cross between a cold boot and hibernation. That's why it boots "fast"..
                Please stop exposing Microsoft's tricks to the beginners. They lose out on an important self-learning exercise if they are just told the answer.

                But most importantly, it is more fun to see them run into issues when they realize they don't actually know how to shutdown a computer anymore .
                Last edited by kpedersen; 11 March 2022, 07:36 PM.

                Comment


                • #18
                  Originally posted by Danielsan View Post
                  It doesn't bother me the boot's slowness of FreeBSD however still feels weird in 2022...
                  In the thirty-odd years that I've been using it, boot speed has never been even at the bottom of my concern list: booting isn't something that happens often enough to optimize, for me. I suppose that it's nice that it matters enough to someone else to work on it, as long as doing so doesn't add a bunch of extra complexity or fragility to the system.

                  I always thought that the majority of the boot delay was the BIOS system sitting there, waiting for me to run it, instead of the OS. I expect that's a delay that I could turn off somewhere, if I could be bothered...

                  Comment


                  • #19
                    Anyone know if the ASLR and boot time improvements are coming to 13.1 or if we have to wait years for 14.0 to come out?

                    Comment


                    • #20
                      Originally posted by sinepgib View Post

                      My advice: try to trim your initramfs and use zstd rather than lz4.
                      Without changing from mkinitcpio to something else or trimming it manually, it's as trimmed as possible. I tested lz4, zstd, gzip, and uncompressed.
                      1. cat 5.6
                      2. zstd 5.71
                      3. gzip 5.73
                      4. lz4 5.75
                      I was surprised at those results. While not much different, I expected lz4 to be one of the better performers.

                      Cat is dead on the 5.6 line, the rest are about where they finished between 5.7 and 5.8.

                      Ext4 root and Fat32 EFI on an SSD. Nothing esoteric or complex; no other compression layer to account for. I'm not sure how to infer these results outside of compressing the kernel adds a hair over a second to boot time.
                      Last edited by skeevy420; 12 March 2022, 09:37 AM.

                      Comment

                      Working...
                      X