Announcement

Collapse
No announcement yet.

ZFS On Linux Landing Workaround For Linux 5.0 Kernel Support

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

  • #21
    Originally posted by starshipeleven View Post
    The "valuable fs" is not part of the Linux project, and is using a license that was specifically made to not allow said "valuable fs" to be merged in Linux.

    Asking to make concessions "just because" is arrogant. Rules are rules.
    Tell that to ubuntu legal team.

    Comment


    • #22
      Originally posted by bill_mcgonigle View Post
      That's really ugly - disabling symbols just to hurt one open source project that an individual developer doesn't like. What good is this new Code of Conduct if on the other side of its implementation devs are targeting specific projects for trouble? I liked it better when Linus said 'fuck' and projects competed on merits, not thumbs on scales.

      I get it, the in-kernel filesystems aren't getting many resources because ZFS is just better in most respects and most high-end users have wound up there, but not only does it hurt users, this has an ecosystem-wide energy penalty by preventing the use of SIMD instructions. I better not hear any of those jokers pretending to be "Green".
      Thing is, these functions were very useful in theory for any module to use, have a small footprint, and removing them is retarded and a pathetic reason like "don't keep symbols for non-in-kernel users" or "code purity" or whatever is obvious.

      In short there's no real technical reason to remove them, just arbitrary or political or crap like "code cleanup" (seriously who the fuck cares about unused code with a small footprint?).

      It's like saying "oh, you have a need for X part of your hardware? Well I don't, so I'm removing access to that part of the hardware."

      Comment


      • #23
        Originally posted by hreindl View Post
        since there are no benchmarks until now i am curious how you shithead will argue when it turns out that it works now even faster and nomerous people on several forums with more brain as you have pointed out that this is likely possible
        Gee, maybe because they could have used the normal (non-SIMD) path without having the functions removed, and yet didn't do it for obvious reasons?

        I didn't do benchmarks, but I'm pretty sure the ZFS guys who wrote the SIMD checksumming (or ported it, w/e) did when implementing it. It's there for a reason.

        Comment


        • #24
          Originally posted by hreindl View Post
          listen: until someone has benchmarks it's all a useless discussion
          "did when implementing it. It's there for a reason" - surely, but how long dates that reason back and is it still true?
          Google the date when AVX2 was introduced in CPUs. There's your answer, at the minimum.

          Comment


          • #25
            Originally posted by some_canuck View Post

            because nobody is using freebsd, linux has been the future for at least a decade now.
            Long live your bubble world.

            Comment


            • #26
              Originally posted by cen1 View Post
              Tell that to ubuntu legal team.
              The fact that you can get away with something does not make it a good idea.

              Comment


              • #27
                Originally posted by Weasel View Post
                Thing is, these functions were very useful in theory for any module to use, have a small footprint, and removing them is retarded and a pathetic reason like "don't keep symbols for non-in-kernel users" or "code purity" or whatever is obvious.
                If this is true then please explain why NO OTHER module inside the linux kernel was using that. The only user for a long while was some UEFI-related infrastructure thing.

                Are you saying Torvalds used his thought-police to "convince" them not to? Is everyone but ZOL developers (and you) a moron?

                Comment


                • #28
                  Originally posted by ArthurBorsboom View Post
                  Do I understand that if Oracle would change the license to GPL, ZFS would be main-lined?

                  If so, how big are the chances that Oracle would do this?
                  None, zero. it's a license cold war. Why? A few reason but Oracle only has so much to do with this.

                  We have to understand what we are talking about. "ZFS" is different things now the fork happened 10 years ago so there are some big differences between them.
                  Sun ZFS ~ Solaris 10 (pool version 28)
                  Oracle ZFS = Solaris 10+ (pool version 37~)
                  OpenZFS = FreeBSD, Illumos, MacOS X, ZoL (Pool version 5000, forked at 28)

                  ZFS's code base was actually released as OpenSolaris as a whole. My understanding is it's integrated into the OS not separate. Oracle also has to abide by the CDDL meaning that if they want to open up their branch of the code they must release Solaris 10+ changes. The CDDL bites both ways. Yes, they can re-license it but I don't even think we want the Oracle version of it as OpenZFS is better, faster etc and Oracle does not own copyright on the changes made in OpenZFS. They only own Solaris (ie Oracle ZFS). This is the result of closing the source on an open source project. And the CDDL isn't really a bad license on it's own, it's MPL like and I tend to prefer that style over GPL.

                  tl;dr No, they can't just "open source it" even *if* Oracle was a good player in the industry, and they are not. If Linux wants ZFS, they have to live with this situation.

                  Comment


                  • #29
                    There is disagreement over whether the CDDL-GPL incompatibility was intentional. The evidence seems to be elusive, just a claim that became a "fact". There's a bit of discussion on Wikipedia. The relevance either way to kernel dev policy on ZoL is questionable.

                    Comment


                    • #30
                      Originally posted by feydun View Post
                      There is disagreement over whether the CDDL-GPL incompatibility was intentional. The evidence seems to be elusive, just a claim that became a "fact". There's a bit of discussion on Wikipedia. The relevance either way to kernel dev policy on ZoL is questionable.
                      https://en.wikipedia.org/wiki/Common..._compatibility
                      This is true too. It is rumor by the Linux base that it was intentional but Sun developers empathically claim the opposite. The "Oh they did this on purpose!" claim because they *knew*.. :: roll eyes :: Unix people tend not to like the restrictions of the GPL so the went with something else. Who knows the truth really.. The CDDL also is probably the best open source license to ever come out of a huge monolith like Sun so.. they had their heart in the right place.. and they did give us a lot of open source software (OpenOffice, ZFS, Dtrace, VirtualBox etc)

                      There is legal dispute over this and quite a few lawyers agree they actually are compatible licenses. It's complicated but the GPL applies to works, and the CDDL applies to files. There is also a case of "derivative works" and a "license intent" case. Their is a lot of threat with sue happy Oracle though.. When Ian Murdock (Debian Founder) worked at Sun supposedly they were mixing GPL and CDDL internally.. And currently Ubuntu is standing in violation of this "incompatibility" if there is any.
                      Last edited by k1e0x; 17 January 2019, 02:52 PM.

                      Comment

                      Working...
                      X