Announcement

Collapse
No announcement yet.

EXT4 Prepared To Switch To Linux's New Mount API

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

  • EXT4 Prepared To Switch To Linux's New Mount API

    Phoronix: EXT4 Prepared To Switch To Linux's New Mount API

    It looks like EXT4 come Linux 5.17 could be making use of the kernel's new mount API. Queued up into EXT4's "dev" branch is transitioning the EXT4 file-system driver to using the kernel's modern mount API...

    Phoronix, Linux Hardware Reviews, Linux hardware benchmarks, Linux server benchmarks, Linux benchmarking, Desktop Linux, Linux performance, Open Source graphics, Linux How To, Ubuntu benchmarks, Ubuntu hardware, Phoronix Test Suite

  • #2
    Thanks to David Howells for developing New Mount API and to Brauner for explaining it so thoroughly. That sounds like a lot of hard and thoughtful work, and much more flexible than Old Mount!

    Several questions. First, at about 14:00, Brauner comments that New Mount API isn't completely finished yet, that at least one more syscall will be added in the future. Does New Mount API replace Old Mount in the kernel before New Mount is finished?

    Second, how soon are distros likely to adopt New Mount API, or perhaps more relevantly, how soon is code likely to be rewritten to make use of New Mount?

    EDIT: Finally, what benefit might a stupid end user (e.g., me) notice?

    Thanks,
    Eric
    Last edited by Eumaios; 14 December 2021, 07:04 AM.

    Comment


    • #3
      This smells like overengineering.

      Comment


      • #4
        Yay, ext4 bugs!

        Comment


        • #5
          Originally posted by Eumaios View Post
          First, congratulations and thanks to Brauner and team for developing New Mount API. That sounds like a lot of hard and thoughtful work, and much more flexible than Old Mount!
          The mount API itself was developed by David Howells (Red Hat) a while back and Ext4 patches were also proposed after




          (I haven't watched the presentation, so I am not sure what it covers)

          Comment


          • #6
            Originally posted by Weasel View Post
            This smells like overengineering.
            I agree with you, it is very hard to believe that an extra complexity in the kernel interface will actually make things better.

            But o the same note, having more flexibility is not an issue in itself. If they manage to get an "one-shot effort", who uses this new syscalls and do the mounting quickly, for the people who don't need the extra flexibility this can actually be a good thing.

            Comment


            • #7
              Originally posted by Weasel View Post
              This smells like overengineering.
              Maybe, or maybe mounting needs to be re-thought from the rotating platter hard disk drives on cylinder-head-sector (CHS) technology in a world that SSD, asynchronous, cached, remote, distributed, and new technology like software-defined storage, etc.

              Comment


              • #8
                Originally posted by RahulSundaram View Post

                The mount API itself was developed by David Howells (Red Hat) a while back and Ext4 patches were also proposed after
                So default in the next Fedora drop.

                Comment


                • #9
                  Originally posted by uid313 View Post
                  maybe mounting needs to be re-thought from the rotating platter hard disk drives on cylinder-head-sector (CHS) technology
                  i'm not sure how chs or rotating platter affects mount. we had network/fuse/tmpfs mounts for decades

                  Comment


                  • #10
                    Non programmers are the major users, and the main readers of these Phoronix articles. So we needed to better understand the worthiness of this press announcement.
                    This development automates and speeds the infrastructure that the devices need to communicate with its inside units. Similar to the synchronized lighting sequences used in traffic transport systems for drivers of cars, trucks, pedestrians, and others.

                    Comment

                    Working...
                    X