• NetBSD

  • Wishlist for NetBSD 10.0

vazub 7. 'doas' is base

Any reason why su won't work? That's in base.

      7 days later

      I'm not sure if this feature is useful or not and I don't think many people may need or want it more than hardware support but do we have any progress for unicode characters support in wscons?

      Coming close to the end of the year, I just wish netbsd source code to be branched (preparing for 10.0 release) within this year (2021).
      If it doesn't, god only knows when it will be released... (when it is ready - yeah I know.. 😆)

        22 days later
        2 months later

        This thread needs splitting by a mod, it's mostly "stuff we want in pkgsrc" now

          15 days later

          Since it's been brought up by the other thread, I'll leave a note here about revising the current FFS discard(ATA TRIM) implementation.

          The second suggestion is bringing zfs(8) in par with current OpenZFS, thus to reflect the changes upstreamed by ZoL and prevent incompatibilities with FreeBSD, Linux and illumos.

          I'm not particularly interested in or concerned about those features, but I'm confident a good portion of the userbase is, at least very least those using (or evaluating the possibility of using) the OS in a professional environment, as well as those enthusiasts who'd pick out NetBSD for storage/database operations.

          6 days later

          2nd OpenZFS tracking, root on ZFS.

          FreeBSD FIBs or similar "N routing tables" would be a great addition.

          my wish for NetBSD 10? Make wireless config as easy as it is on OpenBSD.........

          • nia replied to this.

            This is obviously not for 10.0, but I am curious if there is any interest among NetBSD devs to look into bringing HAMMER2 in, as an alternative solution to ZFS.

            • nia replied to this.

              vazub there doesn't seem to be much interest in HAMMER. Current filesystem developments in the pipeline are improved UDF support and ZFS in the bootloader.

                  nia I see, thanks for the update. Are there any technical considerations against that or mostly lack of devs/capacity? Sorry if this is not the proper topic to ask about this, but I am craving for some tech insights and can't seem to find any at the moment. At least on paper, HAMMER2 boasts kinda same perks ZFS has, while still being permissive-licensed - so my default assumption would be that it might interest BSD-crowd. Yet I don't see much interest in its adoption at the moment, which makes me curious as to why not.

                  • nia replied to this.

                      vazub we had a brief discussion about it at some point. I remember the general conclusion being "too much complex work when there's already a solution we find reliable" (ZFS). a lot is at risk when you add a new filesystem with write support.

                          nia what about CHFS? Is there any practical use for it on flash drives?

                          • nia replied to this.

                              JuvenalUrbino It's not for drives like you would find on a desktop or server system, it's for raw flash - like on some embedded boards.

                              Word on the street is that it's pretty slow and inefficient, but it might be worth extracting its raw flash capabilities and adapting them for FFS.

                                  nia

                                  I can only hope after my own struggles with it lol

                                    2 months later