Announcement

Collapse
No announcement yet.

Linus Ends Up Accepting The DRM Changes For Linux 4.11

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

  • Linus Ends Up Accepting The DRM Changes For Linux 4.11

    Phoronix: Linus Ends Up Accepting The DRM Changes For Linux 4.11

    While Linus Torvalds yesterday was criticizing the DRM code quality using colorful language and threatening not to accept the DRM changes for Linux 4.11, he ended up merging the code to mainline...

    http://www.phoronix.com/scan.php?pag...1-DRM-Accepted

  • #2
    Linus is a feature of the merge process, not a bug.

    But I am a bit curious. Other than the compilation issues, what about code quality? Was the code finally deemed good enough?

    Comment


    • #3
      if the contributors are incompetents they should criticize themselves. What to improve if something has been made poor or bad!?

      Comment


      • #4
        Originally posted by [email protected] View Post
        But I am a bit curious. Other than the compilation issues, what about code quality? Was the code finally deemed good enough?
        The main problem was that the code was in such a state that it wouldn't even compile, meaning that it hadn't even been build tested before it was pushed to Linus for being merged into the mainline kernel. If you know anything about the development of major software projects you'll know that a complete lack of testing like this is an absolute no-no. Apparently Linus himself has gone ahead and fixed the bugs preventing it from compiling, but it's still throwing up some rather nasty warnings during compilation.

        When Linus loses his temper like this it's usually because a senior maintainer has pushed code to him in such a bad state that it doesn't even compile, forcing him to undo and re-do loads of merge-related work. He expects senior maintainers to catch problems like this and fix them or have them fixed rather than kicking the problem up the ladder to him. Additionally he makes a point of chewing out the maintainer, not the developer who actually made the mistake, as he considers the maintainers to ultimately be responsible for any and all the code they push to him for merging into the kernel.

        Still, I was him I'd have left the offending TinyDRM functionality out of this version of the kernel and told David Airlie that the code needs to mature for at least another kernel cycle before it can be considered ready to be merged into the kernel proper. Hell, I'd have told David to apply the same level of scrutiny as with AMD's DAL/DC code before trying to push it again.
        "Why should I want to make anything up? Life's bad enough as it is without wanting to invent any more of it."

        Comment


        • #5
          David Airlie is the DRM subsystem maintainer and he did not do this job properly. It his responsibility as a subsystem maintainer to verify that the code quality is adequate and that it builds. He should not just blindly accept any commits then push it to Linus Torvalds without even testing that it builds. It is irresponsible behavior.

          Comment


          • #6
            If people are turned off by his criticisms, they should just stay away. Anyone who wants to be a contributor should know by now that Linus is very "expressive", and that he takes his work very seriously. Though I personally wouldn't do the same thing, I do not blame him at all for his reaction to the DRM code submitted. Honestly, he should not have had to clean up someone else's code warnings. If your code has warnings, it is incomplete.

            Comment


            • #7
              Or at the very least do a basic build test of your code before sending to Linus, I'm sure he can wait a hour for you to do it!

              Comment


              • #8
                Originally posted by [email protected] View Post
                Other than the compilation issues, what about code quality?
                who cares if it is not used for anything?

                Comment


                • #9
                  Why everybody talk about Linus, maybe Dave was just drunk at the moment

                  Comment


                  • #10
                    Originally posted by Daniel Vetter
                    Don't shit on (new) contributors like this. Noralf has done good work (go look), Kconfig is an impossible combinatorial maze, and I'd be rather annoyed if you managed to drive away Noralf (and other contributors) with your mail.
                    Sure, Daniel has a very good point about driving people away - that would suck. However, the world would be terribly annoyed if buggy code gets into the Linux kernel and causes whatever damage it can do. Feelings are important, but (go look) doesn't cover for newly written crappy code, no matter how polite (or not) we act about it. And the rant was about both the bad code, AND the fact that it got past review, which is not to be desired. What if that introduced yet another heartbleed-like bug or something else critical? With great commit power also comes great responsibility, so people pushing code to the official Linux kernel should take responsibility for it and code like their life depends on it. Because there are also systems running Linux, that actually support the life and/or security of people. We don't want them to fail. As much as I dislike strong language, I'm rather annoyed by the fact that bad code went under the radar.

                    Comment

                    Working...
                    X