Announcement

Collapse
No announcement yet.

Using Udev Without Systemd Is Going To Become Harder

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

  • #21
    Huh?

    Originally posted by Alliancemd View Post
    The guy wants to improve the experience on Linux, you will be able to make bus calls without libudev... currently this is separated and it sucks...

    I actually understand Lennart's frustration. This only will improve the experience, you don't see why someone would be against it - and here this guy comes with false claims that he doesn't hate on systemd while being opposed to something which does good, there is no other explanation to it just that the intention is to harm the project - and here comes the frustration...
    I guess some people don't understand the Gentoo approach to Linux. Gentoo is firmly focused on modularity in it's distribution. A user can build any package as needed with user-selected options, assuming the package has such options; one way to reduce OS "bloat" is to only compile the features that you need, assuming the packages provide such choices.

    The way I read that exchange was the Gentoo developer was evaluating the impact to the Gentoo distribution of merging in Lennart's latest code changes. Lennart's reply seemed to me like, "My way or the highway". If Lennart's original post really did call out the Gentoo community, and I have not yet read that original post, then this exchange and subsequent "blow up" was to be expected.

    Now let's spin this issue another way. Perhaps Lennart is a crafty marketer, using controversy to keep a spotlight and public comment on his project. That would keep his project "in the front of people's minds", preventing any "sinking into obscurity", because his project is basically "internals" stuff normally not seen by most users. So keeping the public commenting on his project, good or bad, keeps people thinking about his project, keeps them talking about it, keeps "the buzz" from fading away. That would mean that Lennart is a "puppetmaster / drama queen" that thinks the Linux community is a bunch of "tools", and these "blow ups" are nothing more than stunts to gain attention.

    Said another way, "manufactured controversy" is just another way of obtaining publicity.

    Comment


    • #22
      Originally posted by NotMine999 View Post
      Computing system architectures at any level should be modular for many reasons, but they mostly revolve around maintenance. Ever wonder why many Linux patches on Git are small? Perhaps that problem was a real small problem? Valid point. Perhaps it is easier to resolve the problem by breaking ti down into small pieces so the impact of each little piece can be evaluated? Yes, and it helps with bisecting patches if/when they cause issues further down the line.
      I agree. Maybe "compatible" was more adequate than "modular". Basically, what I want to show is that instead of hating this guy, the energy could be transfered in a more productive way like, making Lennart's patch available, but yet compatible with non-systemd systems?

      I don't know a lot about the inside of all these systems, but does the patch from Lennart makes udev completely and irreversibly tied up to systemd?

      Comment


      • #23
        Originally posted by ssuominen View Post
        Exactly. I actually like Lennart, and we are in good terms. I'm not happy that this particular thread post was referenced in this Phoronix post in this way, because it doesn't give
        an accurate picture of what is really going on.
        I wouldn't mind systemd becoming the Gentoo default, as I see systemd becoming the norm in Linux userspace, but that hasn't happened yet, and migration will need time.
        So, with all the best in mind, for everyone, for everything, I was only expressing that I'd like udev from systemd tree be usable without *an running systemd instance*
        for longer time, so users don't get too dramatical change too fast.
        i agree with you but i also agree with lennart at the same time because both are 50% right, the problem is to establish what means "longer" because the issue is people will cry to the heavens if you try to move them forward no matter if it is today or in 5 years because they are too comfortable in the current state even if it sucks for everyone else for this reason i believe lennart logic makes sense too, if this people won't accept any try to move forward ever why delay the pain for both sides?

        On the other hand lennart is not nuking anything or leaving you without options here, this change won't happens before KDBUS reach master in 3.18 most likely and this doesn't mean udev api will vanish and get isolated it just get moved to DBUS(in theory, too soon) instead libUdev(we all agree was never any pretty to work with) that you can use from basically any language.

        So the factual burden are either adopt systemd or focus some development time in migrate to the new structure in the next 6+ months or you can freeze your udev repos for non systemd init systems , etc. So is not like in the next release distros will stop to boot if they don't have systemd but well i do understand either way it means a bit more work for non systemd maintainers

        Comment


        • #24
          Originally posted by Creak View Post
          I agree. Maybe "compatible" was more adequate than "modular". Basically, what I want to show is that instead of hating this guy, the energy could be transfered in a more productive way like, making Lennart's patch available, but yet compatible with non-systemd systems?

          I don't know a lot about the inside of all these systems, but does the patch from Lennart makes udev completely and irreversibly tied up to systemd?
          udev is part of systemd and since a long time is tied up to it, what this patch changes is the API internals and methods to access it outside systemd since the horrible libudev will be no more and instead will use DBUS and KDBUS and since most non systemd init systems are linked to libudev you will need to update the code to access udev which is more work for them but should speed up and improve things a lot on the other hand.

          notice non systemd projects download and patch upstream Udev from systemd repos without anything else but this change will require KDBUS to be present and wont provide libudev but a dbus API, so their init systems won't work without modifications for udev access and is not clear yet if KDBUS will require PID1 systemd to be present(will be logical if it does for security reason but im not sure) <-- exact issue

          Comment


          • #25
            Originally posted by tpruzina View Post
            Virtually every desktop distro uses udev and thus they use systemd (apart from gentoo who forked udev and only pull relevant patches from systemd/udev).
            Even ubuntu has systemd, they just don't use it as init system.

            I think we still should be grateful that Lennart isn't politician, because this would be Hitler all over again (yay, GODWINS LAW - check).
            More Hitler than M$ or OSX architects? Give me a break, and come again when you have an actual inperson discussion with software atchitects from these 2 companies.
            And yes, their products are much more loved by users, than ours liberal os.

            Comment


            • #26
              Originally posted by jrch2k8 View Post
              i agree with you but i also agree with lennart at the same time because both are 50% right, the problem is to establish what means "longer" because the issue is people will cry to the heavens if you try to move them forward no matter if it is today or in 5 years because they are too comfortable in the current state even if it sucks for everyone else
              dbus instead of netlink for loading firmware is not "forward"
              netlink is newer and less idiotic (and made for this kind of things)

              just stop with the "future" crap
              it's the people writing the drivers for devices and people writing graphics and input protocols that drive linux forward, not one "visionary"

              Comment


              • #27
                Originally posted by Daktyl198 View Post
                Since when does "I don't agree with this change to a VERY WIDESPREAD piece of software that ALMOST EVERYBODY (including those not using systemd) uses, that makes it so that non-systemd users can't use it. Please at least give us a way to work around it." Mean systemd hater? It's very logical to ask for Lennart to at least think about people besides himself.

                But judging from Lennart's response, I'm going to assume that's impossible.

                (Note, I use and like systemd, but Lennart is one of the biggest douches I've ever seen. Please, those who praise Lennart for anything except his good ideas and code (not that they're all good, mind you), get your head out of his ass.)
                Bigger douche than Linus Torvalds himself? Hardly... And Linux massive sucess is not inspite of, but because Linus attidude.

                Comment


                • #28
                  Originally posted by Drago View Post
                  Bigger douche than Linus Torvalds himself? Hardly... And Linux massive sucess is not inspite of, but because Linus attidude.
                  linus yells at people that push bad code and do dumb things

                  lennart is an ass to people who don't think he is god

                  Comment


                  • #29
                    Originally posted by NotMine999 View Post
                    I guess some people don't understand the Gentoo approach to Linux. Gentoo is firmly focused on modularity in it's distribution. A user can build any package as needed with user-selected options, assuming the package has such options; one way to reduce OS "bloat" is to only compile the features that you need, assuming the packages provide such choices.

                    The way I read that exchange was the Gentoo developer was evaluating the impact to the Gentoo distribution of merging in Lennart's latest code changes. Lennart's reply seemed to me like, "My way or the highway". If Lennart's original post really did call out the Gentoo community, and I have not yet read that original post, then this exchange and subsequent "blow up" was to be expected.

                    Now let's spin this issue another way. Perhaps Lennart is a crafty marketer, using controversy to keep a spotlight and public comment on his project. That would keep his project "in the front of people's minds", preventing any "sinking into obscurity", because his project is basically "internals" stuff normally not seen by most users. So keeping the public commenting on his project, good or bad, keeps people thinking about his project, keeps them talking about it, keeps "the buzz" from fading away. That would mean that Lennart is a "puppetmaster / drama queen" that thinks the Linux community is a bunch of "tools", and these "blow ups" are nothing more than stunts to gain attention.

                    Said another way, "manufactured controversy" is just another way of obtaining publicity.
                    Gentoo used the wrong kernel then. They should have used something way more modular, like some micro-kernel, which they can chop off as they want. And Instead they are using that monolitic bloat - Linux.

                    Comment


                    • #30
                      Originally posted by gens View Post
                      dbus instead of netlink for loading firmware is not "forward"
                      netlink is newer and less idiotic (and made for this kind of things)

                      just stop with the "future" crap
                      it's the people writing the drivers for devices and people writing graphics and input protocols that drive linux forward, not one "visionary"
                      is not dbus but KDBUS and netlink is not especially secure or efficient for user space information exchange in the way dbus use it but i guess it kinda was good enough, check the code it will come to you eventually

                      Comment

                      Working...
                      X