Stories
Slash Boxes
Comments

SoylentNews is people

posted by janrinok on Friday April 04 2014, @07:48PM   Printer-friendly
from the never-one-to-stay-silent dept.

From an email from Linus Torvalds to GKH

"Greg just for your information, I will *not* be merging any code from Kay [Sievers] into the kernel until this constant pattern is fixed.

This has been going on for *years*, and doesn't seem to be getting any better. This is relevant to you because I have seen you talk about the kdbus patches, and this is a heads-up that you need to keep them separate from other work. Let distributions merge it as they need to and maybe we can merge it once it has been proven to be stable by whatever distro that was willing to play games with the developers.

But I'm not willing to merge something where the maintainer is known to not care about bugs and regressions and then forces people in other projects to fix their project. Because I am *not* willing to take patches from people who don't clean up after their problems, and don't admit that it's their problem to fix."

 
This discussion has been archived. No new comments can be posted.
Display Options Threshold/Breakthrough Mark All as Read Mark All as Unread
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
  • (Score: 2) by efitton on Saturday April 05 2014, @01:34AM

    by efitton (1077) on Saturday April 05 2014, @01:34AM (#26487) Homepage

    I like the title a bit better than the summary. kdbus is blocked, not just patches and pull requests from Kay. Kay has not been a kernel developer for a year plus. However, between udev and systemd he has managed to make life difficult enough for Linus that now Kay's association with a project is enough to taint the entire project.

  • (Score: 1) by kevinl on Saturday April 05 2014, @01:26AM

    by kevinl (3951) on Saturday April 05 2014, @01:26AM (#26484)

    Yes, the systemd dude needs to look for systemd.debug rather than debug, the primary problem is his code.

    BUT, why is it OK for the kernel to not boot a system if a process is writing too much to /dev/kmsg? Even the ratelimiting stuff they added could be circumvented. Granted, by default only root can write to kmsg and it's stupid to protect the system from its own administrator, yet I always thought it was a circular buffer anyway so how was systemd able to fuck things up anyway?

    • (Score: 2) by neagix on Saturday April 05 2014, @09:15AM

      by neagix (25) on Saturday April 05 2014, @09:15AM (#26594)

      You've got a good point, I would like to know as well why it can be killed this way

  • (Score: 0) by Anonymous Coward on Saturday April 05 2014, @12:05AM

    by Anonymous Coward on Saturday April 05 2014, @12:05AM (#26464)

    Guess this is nothing new, but the excess dependence on Red Hat is showing its ugly side. On the kernel, Linus still holds down the fort, but on the rest of the OS, Red Hat with their asinine developers are bulldozing they way imperiously.

  • (Score: 5, Informative) by Snotnose on Friday April 04 2014, @10:26PM

    by Snotnose (1623) on Friday April 04 2014, @10:26PM (#26432)

    From reading the email threads, I can summarize the problem as follows:

    1) Kernel devs use debug to debug the kernel.
    2) Module devs use foo.debug to debug their module
    3) Systemd is a module
    4) Systemd used debug to debug their module
    5) This crashes the kernel
    6) The systemd folks never use debug themselves, else they'd have noticed it breaks the kernel
    7) Kay's response is "debug is a generic term, anybody should be able to use it. Not my problem."
    8) Linus said "sod off, with that attitude I'm not taking *any* code from you"

    Imho, Linux is right on this one.

    --
    Why shouldn't we judge a book by it's cover? It's got the author, title, and a summary of what the book's about.
    • (Score: 1) by jamesbond on Saturday April 05 2014, @05:04AM

      by jamesbond (2383) on Saturday April 05 2014, @05:04AM (#26551)

      The attitude seems to be pervasive in the systemd dev team. Look at another bug report linked from that LKML thread: https://bugs.freedesktop.org/show_bug.cgi?id=74589 [freedesktop.org]. Somebody reported that systemd won't work in under LXC (Linux OS-level virtualisation similar to OpenVZ/Virtuozzo); and the response is essentially "We don't use it that way, go fix it yourself." But because of this bug - you can't run *any* systemd-based OS under LXC virtualisation.

      This is not the only case, I found another one from systemd-udev (that I had to personally deal with) where compatibility-breaking change was announced for no good reason, with note that said change wasn't completed and there was no documentation around it, and 4 days later said change got committed *AND* at the same time dropping the code which worked for years and years on. It was rather easy to have both old and new code working at the same time but no, they had to drop it. In case you're wondering, I'm referring to this: http://lists.freedesktop.org/archives/systemd-deve l/2013-July/011773.html [freedesktop.org].

    • (Score: 0) by Anonymous Coward on Saturday April 05 2014, @03:33AM

      by Anonymous Coward on Saturday April 05 2014, @03:33AM (#26529)
      5) This crashes the kernel This is a fault with the kernel then. Too much logging crashes the kernel...oh boy
  • (Score: 5, Interesting) by tomtomtom on Friday April 04 2014, @10:09PM

    by tomtomtom (340) on Friday April 04 2014, @10:09PM (#26427)

    Following the threads, the ultimate origin of the bug is more illuminating than the attitude of systemd's developers here. The reason this bug happened is that apparently systemd is defaulting to stashing its early log messages in the kernel dmesg ring buffer using the /dev/kmsg interface. That is so incredibly stupid I don't even know what to say about it. The constant extension of systemd to do more and more things which have nothing to do with init was bad enough (udev, logging, dbus, etcd .... where does it end?). But it's just astounding that someone could think it was an even remotely sensible idea to put log messages there, let alone someone who is apparently in charge of a subsystem as important as this. That pretty much makes up my mind that I won't EVER be using systemd if I can help it (unless something drastic changes).

    • (Score: 0) by Anonymous Coward on Friday April 04 2014, @10:50PM

      by Anonymous Coward on Friday April 04 2014, @10:50PM (#26438)

      Did you know systemd is the brainchild of the same guy who came up with pulseaudio?

      Even George W. Bush knows "Fool me - you can't get fooled again."

  • (Score: -1, Offtopic) by Khyber on Friday April 04 2014, @08:38PM

    by Khyber (54) on Friday April 04 2014, @08:38PM (#26372) Journal

    "I will *not* be merging any code from Kay [Sievers] into the kernel until this constant pattern is fixed."

    You can't even fix a kernel memory map bug.

    When I can make an encrypted swap file and make it crash the kernel via an out of memory issue, you're fucking up, Linus.

    Wake the fuck up to reality. Your own ego is the true threat to anyone running slashcode.

    As NCommander thinks I'm an actual threat (despite me having no real experience besides logical thinking) you should probably tuck tail and run your ass off, as this entire site has no clue what a real threat is.

    --
    Destroying Semiconductors With Style Since 2008, and scaring you ill-educated fools since 2013.
    • (Score: 0) by Anonymous Coward on Friday April 04 2014, @11:25PM

      by Anonymous Coward on Friday April 04 2014, @11:25PM (#26449)

      Hehehe, this feels like the old anti-drug TV commercial: "This is your brain on drugs..."

    • (Score: 5, Interesting) by Ethanol-fueled on Friday April 04 2014, @08:45PM

      by Ethanol-fueled (2792) on Friday April 04 2014, @08:45PM (#26379) Homepage

      If you were in charge of Linux, you'd bet your ass you'd be every bit as curt and abrasive as Linus is just as you are being right now.

      You need a tough sonofabitch to keep the project on track, whatever that track may be, and Linus is doing an awesome job for just one person. You can't have a bed-wetting crybaby or a soft-talking appeaser as a leader for such a project. If that makes him wrong once or twice, then tough shit -- you're free to run Windows or BeOS if you don't agree with Linus the Messiah.

      p.s. Welcome back. I knew you'd come back, they always do, bwahahahah.

      • (Score: 2) by Geotti on Friday April 04 2014, @10:27PM

        by Geotti (1146) on Friday April 04 2014, @10:27PM (#26433) Journal

        you're free to run Windows or BeOS if you don't agree with Linus the Messiah

        You forgot BSD, which actually works quite well as an alternative as opposed to the other two weird, out of date OSs that you proposed.

        Now, if you would have mentioned Haiku instead of BeOS, then it would be just one weird, out of date OS, but no, you just wanted to play fanboi...

      • (Score: 0, Flamebait) by Khyber on Friday April 04 2014, @08:56PM

        by Khyber (54) on Friday April 04 2014, @08:56PM (#26385) Journal

        "If you were in charge of Linux, you'd bet your ass you'd be every bit as curt and abrasive as Linus is just as you are being right now."

        I see the shades of your eyes getting more brown by the minute.

        You don't even know me, yet you assume that I'd make the same mistakes Linus is currently making?

        I'm not that socially inept, sir.

        --
        Destroying Semiconductors With Style Since 2008, and scaring you ill-educated fools since 2013.
        • (Score: 1) by EQ on Saturday April 05 2014, @04:18AM

          by EQ (1716) on Saturday April 05 2014, @04:18AM (#26540)

          Thanks. My first filtered asshole on a Soy. Buhbye

        • (Score: 4, Insightful) by hybristic on Friday April 04 2014, @09:52PM

          by hybristic (10) on Friday April 04 2014, @09:52PM (#26420) Journal

          I don't believe anyone, including Linus is saying that they don't mess up. If you read the mailing list like I do, you will see that he proposes a fix, and someone else comes up with a better idea. Linus then admits that the other idea is better. The issue here is Kay has consistently produced code that causes bugs and then refuses to own up to them or fix them. He never said no one can ever fuck up, just that he noticed a trend and he refuses to perpetuate it.

          outside of that, I am curious where you think Linus if messing up? While he is the authority on the kernel, plenty of other people develop code, create their own solutions and hope it gets merged. These pushes get seen by a few people before making it to Linus in a lot of cases. He just has final say as to whether something is stable enough, or makes enough sense to make it into the code. While I agree he can be a dick, hes consistent. I have been reading LKML for a few years, and I have always seen a consistent response from him. If you break user space, expect to get chewed out. This is his project, and he has made his idea of success very clear. People can either get on board with that, or they can go work on something else. I don't find that to be unreasonable.

          • (Score: 1, Redundant) by Geotti on Friday April 04 2014, @10:23PM

            by Geotti (1146) on Friday April 04 2014, @10:23PM (#26430) Journal

            Mod parent up!
            Thanks for the context, hybristic!

        • (Score: 4, Funny) by Anonymous Coward on Friday April 04 2014, @09:11PM

          by Anonymous Coward on Friday April 04 2014, @09:11PM (#26393)

          I'm not that socially inept, sir.

          I don't know about that. I just read both of your posts in this thread and I think they prove that you are.

          • (Score: 1, Informative) by Anonymous Coward on Friday April 04 2014, @09:21PM

            by Anonymous Coward on Friday April 04 2014, @09:21PM (#26400)

            Not to mention his posts elsewhere...

    • (Score: 2, Informative) by Lukehasnoname on Friday April 04 2014, @08:43PM

      by Lukehasnoname (3303) on Friday April 04 2014, @08:43PM (#26378) Homepage

      What are you going on about? Your comment has less context than the summary!

    • (Score: 1, Insightful) by MechaStreisand on Friday April 04 2014, @08:40PM

      by MechaStreisand (1550) on Friday April 04 2014, @08:40PM (#26374)

      If you have a karma-bonus modifier, there's something seriously wrong with how they're giving those out.

      • (Score: 2) by zocalo on Friday April 04 2014, @11:22PM

        by zocalo (302) on Friday April 04 2014, @11:22PM (#26448)
        Seems to have started off OK with a +5 insightful post, then gone off the rails when the spat over site management kicked off. NCommander posted the default scores for rankings a few days back and there's a fair way to got before he gets the "Bad Karma" rating (-25).
        --
        UNIX? They're not even circumcised! Savages!
        • (Score: 2) by zocalo on Saturday April 05 2014, @12:05AM

          by zocalo (302) on Saturday April 05 2014, @12:05AM (#26465)
          Found the original post [soylentnews.org] and I had it backwards; 25 is for "Good", -10 for "Terrible", so not that far to go after all.
          --
          UNIX? They're not even circumcised! Savages!
  • (Score: 1, Troll) by isostatic on Friday April 04 2014, @08:11PM

    by isostatic (365) on Friday April 04 2014, @08:11PM (#26357) Journal

    Who is GKH, who is Kay Sievers and why should I care about Kernel politics?

    • (Score: 0) by Anonymous Coward on Friday May 02 2014, @06:23AM

      by Anonymous Coward on Friday May 02 2014, @06:23AM (#38770)

      Because the surgeon who first diagnosed me, listen to prime minister with a thunderous roar.And it capped nearly a thousand years., free online casino bonus [onlinecasi...iazone.com], [url="http://onlinecasinoaustraliazone.com/ "]free online casino bonus[/url], =]]], casino online [bestcasinoclubcom.com], [url="http://bestcasinoclubcom.com/"]casino online[/url], :]],

    • (Score: 3, Informative) by Anonymous Coward on Friday April 04 2014, @08:52PM

      by Anonymous Coward on Friday April 04 2014, @08:52PM (#26382)

      -Grek Koah Hartman , more or less 'the second in command after Linux Torvalds' in the kernel community.
        -Kay Sievers, the creator of solutions in search of problems. Created udev, which is ok(ish) though somewhat arcane. Created systemd which is an abomination. Lets put loads and loads of complicated functionality in the single process which may *never* crash because that would hang your computer. Sounds like a swell idea! And now creator of kernel-dbus.

      RANT:
        I kid you not, a dbus server *inside* the kernel. Because it will be slightly faster than outside maybe? I'm not a microkernel purist but come on what's next: put the compiler inside the kernel, or maybe a wordprocessor? Dbus might have it's place (though the dependency hell it generates because all sorts of apps start to rely on other apps being running makes me sometimes yearn for simpler times when each app would be more self contained), but *not in the kernel*. Why anybody would think that the speed benefits would outweight the stablity costs I really don't understand (same for systemd). EOR

      -You should care, because Having a stubborn friendly asshole like Linus at the helm is the only thing from linux degenerating info a mess of featuritis.

      This is very much news for nerds. Don't like it, don't read it. Some other people do.

      • (Score: 1) by GeminiDomino on Saturday April 05 2014, @07:07PM

        by GeminiDomino (661) on Saturday April 05 2014, @07:07PM (#26785)

        I thought systemD was created by whatshisface... the one who shat pulseaudio at us years ago. Lennart Pottering.

        --
        "We've been attacked by the intelligent, educated segment of our culture"
      • (Score: 0) by Anonymous Coward on Saturday April 05 2014, @12:23AM

        by Anonymous Coward on Saturday April 05 2014, @12:23AM (#26470)

        Greg Koah-Hartman

        Outside of your typo and missing hyphen, excellent post.

        Linux has the best hardware support of *any* OS;
        it has better support for new hardware than old EULAware OSes (notably, EoL'd OSes)
        and Linux has better support for old hardware than new EULAware OSes.
        This is largely due to gregkh. [googleusercontent.com] (orig) [lwn.net]

        When your oddball device Just Works(tm) under Linux, [googleusercontent.com] (orig) [archlinux.org] it's probably Greg whom you owe a beer.

        -- gewg_

      • (Score: 0) by Anonymous Coward on Friday April 04 2014, @09:28PM

        by Anonymous Coward on Friday April 04 2014, @09:28PM (#26406)

        This is very much news for nerds.

        Our motto is SoylentNews is people. So, yeh...

    • (Score: 2) by fliptop on Friday April 04 2014, @08:41PM

      by fliptop (1666) on Friday April 04 2014, @08:41PM (#26377) Journal

      who is Kay Sievers

      Works for Red Hat and is the developer of udev and systemd [wikipedia.org].

      why should I care about Kernel politics?

      Because you're a nerd?

      --
      Our Constitution was made only for a moral and religious people. It is wholly inadequate to the government of any other.
      • (Score: 2) by zocalo on Friday April 04 2014, @11:39PM

        by zocalo (302) on Friday April 04 2014, @11:39PM (#26454)
        You left of the merging of "/lib", "/bin" and "/sbin" into "/usr", thus essentially preventing Linux from mounting putting "/usr" on its own partition and mounting it read-only as a security hardening measure. One of the people I genuinely wonder whether they are actually working for Microsoft since they've done so much to compromise the OS.

        On the plus side, they've driven so many people to BSD that Netcraft is *still* getting it wrong. :)
        --
        UNIX? They're not even circumcised! Savages!
        • (Score: 2) by TheRaven on Saturday April 05 2014, @11:38AM

          by TheRaven (270) on Saturday April 05 2014, @11:38AM (#26628) Journal
          Speaking as a FreeBSD developer, I welcome all of these things that people are doing to make FreeBSD a more attractive operating system to people with *NIX experience...
          --
          sudo mod me up
          • (Score: 1) by GeminiDomino on Saturday April 05 2014, @07:02PM

            by GeminiDomino (661) on Saturday April 05 2014, @07:02PM (#26784)

            As someone who's been out of the FreeBSD arena for a (long) while:

            Do you still have to compile from scratch to do a package/system update?

            I used to love FreeBSD, but having a couple of dozen heterogenous hardware servers to keep current, that's a lot of downtime.

            --
            "We've been attacked by the intelligent, educated segment of our culture"
            • (Score: 1) by David_W on Sunday April 06 2014, @01:35AM

              by David_W (3469) on Sunday April 06 2014, @01:35AM (#26901)

              Do you still have to compile from scratch to do a package/system update?

              Not really, no. Binary system updates can be done with freebsd-update [freebsd.org], and the new pkg (pkgng) [freebsd.org] stuff is more focused around binary packages. I still build my stuff from source, so I can't vouch for how good these tools are at their jobs, but they definitely exist. As always, the Handbook [freebsd.org] has much more info.

        • (Score: 2) by frojack on Saturday April 05 2014, @03:05AM

          by frojack (1554) on Saturday April 05 2014, @03:05AM (#26523) Journal

          Not to mention tmp not being tmp anymore, and you aren't safe in cleaning tmp between boots.

          Systemd might be needed, because initd has become somewhat unworkable in larger machines, but it goes way too far and messes up way too much of the existing system.

          --
          No, you are mistaken. I've always had this sig.
          • (Score: 2) by neagix on Saturday April 05 2014, @09:07AM

            by neagix (25) on Saturday April 05 2014, @09:07AM (#26593)

            Wait aaaa minute since when this is not allowed anymore? and what do you mean specifically?

            • (Score: 2) by zocalo on Saturday April 05 2014, @09:49AM

              by zocalo (302) on Saturday April 05 2014, @09:49AM (#26603)
              There's more in this in this writeup [freedesktop.org]. Basically, "/usr" is required to be pre-mounted because the Udev developers thought it would be a good idea to put force all the tools in "/sbin" onto the same physical partition as "/usr". The problem with that idea is that you *need* those tools to do things in the start up sequence upto and including the point where you can mount partitions, so "/usr" now has to go on the root partition. Since that contains stuff that must be writable, you can no longer mount the "/usr" partition read-only and use that as a hardening technique.

              And yes, there's a lot more of the "not our fault!" attitude, and although that write up is by the SystemD team and Udev caused the breakage, there's such a heavy overlap between the members of the two teams it doesn't really make much difference.
              --
              UNIX? They're not even circumcised! Savages!
              • (Score: 2) by neagix on Monday April 07 2014, @06:32AM

                by neagix (25) on Monday April 07 2014, @06:32AM (#27309)

                Yeah, I heard about this. I was wondering about the possibility to delete /tmp across reboots.

                In general, it's like hearing the loud screech of the chains of legacy...not saying that everything should have been left untouched, but that not all the problems have been addressed with the new solutions.

                Is this the chance for a new init project (Epoch?) or maturity will be reached afterwards?

                Disclaimer: I would have forbidden the author to access any PC just because of PulseAudio already

      • (Score: 0) by Anonymous Coward on Friday April 04 2014, @08:58PM

        by Anonymous Coward on Friday April 04 2014, @08:58PM (#26386)

        Don't let Kay get all the credit for systemd, the other primary author is the same genius who gave us pulseaudio.

        Yeah, this is going to go well.

        Looks like I'll be switching to Slackware once a conventional init mechanism is no longer feasible in Debian.

        • (Score: 1) by Subsentient on Saturday April 05 2014, @02:51AM

          by Subsentient (1111) on Saturday April 05 2014, @02:51AM (#26516) Homepage Journal

          Uhh

          http://universe2.us/epoch.html [universe2.us]

          (Boy, I'm having a field day!)

          --
          "It is no measure of health to be well adjusted to a profoundly sick society." -Jiddu Krishnamurti
        • (Score: 2) by forsythe on Friday April 04 2014, @09:45PM

          by forsythe (831) on Friday April 04 2014, @09:45PM (#26415)

          I actually switched to Gentoo a while ago because the ports-like package management allows me to use Wine without consolekit. Since the systemd thing hit, I've been very pleased with its sysvinit + openrc init system. To me, systemd is a rather poor solution in frantic search of a problem.

    • (Score: 3, Insightful) by Grishnakh on Friday April 04 2014, @08:40PM

      by Grishnakh (2831) on Friday April 04 2014, @08:40PM (#26375)

      Who is GKH, who is Kay Sievers and why should I care about Kernel politics?

      If you don't know who Greg Kroah-Hartman is, then you're obviously completely clueless about the Linux kernel and its development. That's fine; you don't have to be an expert in all things tech or software to hang out on this site. For people who are interested in kernel politics, this is a very important and interesting story. If that doesn't describe you, then just move along. Not every story is going to be of interest to every single reader.

      • (Score: 2) by isostatic on Monday April 07 2014, @07:05PM

        by isostatic (365) on Monday April 07 2014, @07:05PM (#27669) Journal

        If you don't know who Greg Kroah-Hartman is, then you're obviously completely clueless about the Linux kernel and its development.

        Quite, I haven't paid much attention to development since about 2000, and it looks like he only started writing kernel code in 1999. In fact I stopped compiling kernels around 2001, as the stock debian ones tended to meet my needs. I think that was the time Alan Cox was finishing his maintenance duties on 2.2 and some whippersnapper was maintaining 2.4?

        A far better summary would have included something about the fact Kay Sievers wrote systemd, and the interaction between systemd and the kernel is causing a lot political problems, so Linus has "discussed" the matter with his chief lieutenant, that there's no technical issue here, it seems that the systemd guy is kicking up a fuss because he doesn't like to be corrected, or shown to be wrong. Something like context for those of us that don't follow LKML.

      • (Score: 3, Insightful) by jimshatt on Friday April 04 2014, @09:22PM

        by jimshatt (978) on Friday April 04 2014, @09:22PM (#26402) Journal

        you're obviously completely clueless about the Linux kernel and its development

        Well, I am clueless about Linux kernel development, but I might still be interested. Explaining things to clueless but interested people is a good thing, methinks. But now that you have at least provided his full name, so I can look it up. So, thanks, I guess.

        • (Score: 3, Informative) by edIII on Friday April 04 2014, @09:47PM

          by edIII (791) on Friday April 04 2014, @09:47PM (#26417)

          Don't feel bad at all.

          The original poster just asked for context and was specific in what they wanted to know.

          It was a question, not a statement.

          We should treat it like a question, and there really are no stupid questions. Just stupid answers.

          --
          Technically, lunchtime is at any moment. It's just a wave function.
        • (Score: 2) by Grishnakh on Friday April 04 2014, @09:47PM

          by Grishnakh (2831) on Friday April 04 2014, @09:47PM (#26416)

          Sorry, I didn't mean to come off as snarky, but generally I just assume that if someone isn't already familiar with the leading names in kernel development by now, they're probably just not that interested in it in general. By all means, if you want to learn more, go right ahead; the LKML at tux.org/lkml is the site of the official mailing list where you can read all about it, and lwn.net is even better if you're not directly involved and just want to read about it, as they have great in-depth articles about the goings-on in the community (specifically at lwn.net/kernel) and in development.

          • (Score: 1, Flamebait) by frojack on Saturday April 05 2014, @02:58AM

            by frojack (1554) on Saturday April 05 2014, @02:58AM (#26521) Journal

            Saying you didn't mean to be snarky (when in fact your first replay was a virtuoso study in snark) doesn't let you off the hook, nor does handing out a reading assignment.

            Take your own advice.

            If you don't want to help out new users and explain a few things, just move along. Don't sling thinly veiled insults. It would have taken you less verbiage to explain who these people are than to tell the OP to piss-off.

            --
            No, you are mistaken. I've always had this sig.
      • (Score: 3, Insightful) by TK on Friday April 04 2014, @08:55PM

        by TK (2760) on Friday April 04 2014, @08:55PM (#26384)

        Remember that pet peeve from the other site about not explaining acronyms? I think it applies to initials too.

        --
        The fleas have smaller fleas, upon their backs to bite them, and those fleas have lesser fleas, and so ad infinitum
    • (Score: 5, Informative) by Bob The Cowboy on Friday April 04 2014, @08:19PM

      by Bob The Cowboy (2019) on Friday April 04 2014, @08:19PM (#26364)

      1) Looks like a Kernel Dev
      2) Looks like a systemd Dev
      3) Don't know, but no one's making you ;)

      Basically the issue boiled down to something like:

      If you pass 'debug' to the linux kernel boot command (presumably/historically to debug the kernel), systemd parses that and floods dmesg with its own debugging info. Apparently it floods dmesg with so much of its own debugging info it locks up the machine.

      The systemd dev (Kay) basically said "you guys don't own the word debug, if someone passes debug to the kernel, we're going to assume they want to debug systemd as well."

      This didn't go over well...

      • (Score: 1) by Subsentient on Saturday April 05 2014, @02:48AM

        by Subsentient (1111) on Saturday April 05 2014, @02:48AM (#26515) Homepage Journal

        Do I dare advertise the glory that is Epoch with this opportunity?

        Ahh, screw it. http://universe2.us/epoch.html [universe2.us]

        --
        "It is no measure of health to be well adjusted to a profoundly sick society." -Jiddu Krishnamurti
      • (Score: 1) by mverwijs on Friday April 04 2014, @09:56PM

        by mverwijs (2457) on Friday April 04 2014, @09:56PM (#26421) Homepage

        Not only does systemd parses that, it also renders the system unbootable at that point.

        Another gem of systemd: disable CGROUPS and it segfaults. Again: rendering the system unusable.

        The problem here is not the bugs though. It's the attitude of the developers: "Works for me. Your problem. WONTFIX."

      • (Score: 2) by TheGratefulNet on Friday April 04 2014, @09:06PM

        by TheGratefulNet (659) on Friday April 04 2014, @09:06PM (#26390)

        if he knew (or later, found out) that the shared word 'debug' caused so much trouble at boot time, why on earth would anyone want to argue to keep it that way instead of just making it debug_systemd or something not in collision with the standard debug word?

        sheesh. there are things to argue for; but this is not one of them. linus was right. thou shalt NOT break userland via any kernel change!

        --
        "It is now safe to switch off your computer."
        • (Score: 3, Insightful) by emg on Friday April 04 2014, @09:23PM

          by emg (3464) on Friday April 04 2014, @09:23PM (#26403)

          Because the goal of systemd appears to be to take over the entire operating system until the kernel is just a minor part of systemd?

          • (Score: 2, Funny) by GeminiDomino on Saturday April 05 2014, @06:57PM

            by GeminiDomino (661) on Saturday April 05 2014, @06:57PM (#26783)

            What?!

            EMACS won't stand for that sort of thing for very long, let me tell you.

            --
            "We've been attacked by the intelligent, educated segment of our culture"
          • (Score: 3, Insightful) by frojack on Saturday April 05 2014, @02:52AM

            by frojack (1554) on Saturday April 05 2014, @02:52AM (#26518) Journal

            Pretty much this.
            Poettering is about as arrogant an ass as you will ever see.

            Systemd's hooks go way too deep and very little of it has been reviewed from a security standpoint. There is no telling what it can do or might be doing.

            It wasn't written for you.

            --
            No, you are mistaken. I've always had this sig.
      • (Score: 3, Insightful) by Open4D on Friday April 04 2014, @09:01PM

        by Open4D (371) on Friday April 04 2014, @09:01PM (#26387) Journal

        The systemd bug report is: https://bugs.freedesktop.org/show_bug.cgi?id=76935 [freedesktop.org]

        But that was closed.

        So there was an LKML thread here [iu.edu], about possible Linux patches to deal with the problem instead. Torvalds's announcement about Sievers is the first reply.

        But now the systemd bug has a status of "REOPENED". So maybe Torvalds's announcement had the desired effect?

        • (Score: 2) by zocalo on Friday April 04 2014, @11:47PM

          by zocalo (302) on Friday April 04 2014, @11:47PM (#26456)
          The activity log [freedesktop.org] of the SystemD bug report is worth a look. Yep, that sure looks like the right kind of people to trust with the replacement for init and arguably the most important part of the OS after the kernel...
          --
          UNIX? They're not even circumcised! Savages!
          • (Score: 1) by blackpaw on Saturday April 05 2014, @08:24AM

            by blackpaw (2554) on Saturday April 05 2014, @08:24AM (#26584) Journal

            Wow, that's just plain childish. How embarrassing. Its like a five year sticking their fingers in their ears going "NYAH NYAH NAYH! CAN'T HEAR YOU!"

        • (Score: 3, Funny) by Anonymous Coward on Friday April 04 2014, @09:48PM

          by Anonymous Coward on Friday April 04 2014, @09:48PM (#26418)

          And here's a fix:

          + if (!strcmp(p->p_name, "systemd")) killproc(p, "fuck you");

          • (Score: 1) by Subsentient on Saturday April 05 2014, @02:46AM

            by Subsentient (1111) on Saturday April 05 2014, @02:46AM (#26512) Homepage Journal

            Oh man, that's a funny one.

            --
            "It is no measure of health to be well adjusted to a profoundly sick society." -Jiddu Krishnamurti
          • (Score: 0) by Anonymous Coward on Saturday April 05 2014, @01:39AM

            by Anonymous Coward on Saturday April 05 2014, @01:39AM (#26489)

            fuck me?

            not syncing: attempt to kill init!

            fuck you!

      • (Score: 2) by isostatic on Friday April 04 2014, @08:39PM

        by isostatic (365) on Friday April 04 2014, @08:39PM (#26373) Journal

        2) Looks like a systemd Dev

        Ahh, now it makes sense. I've heard of systemd, and the "controversial" use of it.

        Seems like the obvious thing would be to pass "debugsysd" to allow systemd to be debugged.

        • (Score: 1, Funny) by Anonymous Coward on Saturday April 05 2014, @04:50AM

          by Anonymous Coward on Saturday April 05 2014, @04:50AM (#26548)

          > Seems like the obvious thing would be to

          Actually, what the kernel needs is:

          if ( strcmp( pid[1], "systemd" ) {
              kernel_panic();
          }

          and just stop the systemd craziness at its start.

          • (Score: 2) by nukkel on Friday April 18 2014, @04:39PM

            by nukkel (168) on Friday April 18 2014, @04:39PM (#33127)

            Either you are Kay, or you forgot a "!" operator.

          • (Score: 1) by GeminiDomino on Saturday April 05 2014, @06:55PM

            by GeminiDomino (661) on Saturday April 05 2014, @06:55PM (#26781)

            No no no! That's exactly why Linus tore this jerk a new asshole to begin with! He's apparently had the attitude of "change your code to suit mine" for years, and that's just what you're playing into! ;)

            --
            "We've been attacked by the intelligent, educated segment of our culture"
        • (Score: 2) by forsythe on Friday April 04 2014, @09:19PM

          by forsythe (831) on Friday April 04 2014, @09:19PM (#26399)

          That solution (actually the variant 'systemd.debug', to match a few other 'systemd.xyz' parameters) was actually presented, and a patch was even submitted for it. The rejection of that suggestion may very well have been a significant factor in Linus' ongoing irritation.

    • (Score: 5, Insightful) by naubol on Friday April 04 2014, @08:17PM

      by naubol (1918) on Friday April 04 2014, @08:17PM (#26362)

      I think it was really better said to be an email to the LKML (Linux Kernel Mailing List).

      Aside from being our version of page 6, I think kernel politics is interesting because it exposes the issues that come up in open source and how they're handled in one of the most (THE most?) important open source project in the world where the stakes are really high.

      Reading through them discussing how to handle one misbehaving stakeholder without punishing the others and to do so in a clean way is a fascinating bit of software management problem solving. That last bit strikes me as important to the SN community.

    • (Score: 2) by tynin on Friday April 04 2014, @08:15PM

      by tynin (2013) on Friday April 04 2014, @08:15PM (#26360) Journal

      Agreed. While I do love hearing about Linus ripping on people, some context is needed.

      • (Score: 3, Informative) by Thexalon on Friday April 04 2014, @08:47PM

        by Thexalon (636) on Friday April 04 2014, @08:47PM (#26380)

        My impression is that periodically, Linus makes a big decision about what will and will not be included. Back on the other site, there were a bunch of these stories that would show up whenever it happened from miffed kernel devs and/or their supporters. In some cases, it's been so frequent as to appear to be a concerted effort to remove Linus as BDFL of Linux.

        And Linus has been pretty clear about his basic mantra: Don't Break Userspace. And this post is presenting a pretty good corollary: If you break userspace, apologize and do your very best to fix it. Both of these seem like very good rules to have for kernel development.

        --
        The only thing that stops a bad guy with a compiler is a good guy with a compiler.
        • (Score: 2) by edIII on Friday April 04 2014, @09:43PM

          by edIII (791) on Friday April 04 2014, @09:43PM (#26414)

          I've heard he can be a real jerk, but from everything in TFS, it sounds very reasonable and he is just calling out a piss-poor programmer.

          There should be more people like him, not less, involved in Linux.

          --
          Technically, lunchtime is at any moment. It's just a wave function.