In many (all?) file systems that Linux uses, including ext2/3/4, the maximum block size is the page size of the architecture Linux is running on, ie. 4KiB on x86. Why is there this limit? Might this limit ever be removed?

  • I don't think they hanged together, but maybe somebody knows more. Until then, here is a bookmarking and an upvote. Do you know any filesystem whose block size were bigger as 4k? – peterh Aug 13 '14 at 8:27
  • @PeterHorvath: XFS (not on Linux), BSD's FFS, ZFS (although that one's a bit weird), VxFS (not sure on Linux), UFS on Solaris/Sparc (but that one's "cheating" - 8k block size but 8k page size too) – Mat Aug 13 '14 at 8:59
  • @Mat Uhm, thanks. This makes the question really interesting. – peterh Aug 13 '14 at 9:00
  • 5
    lwn.net/Articles/469805 - "Raising the filesystem block size in the kernel is a dauntingly difficult task involving major changes to memory management, the page cache, and more." - so the answer to your question is probably a bit complicated... – Mat Aug 13 '14 at 9:30
  • 1
    I wonder what happens if you make a large block file-system on a system with large pages. Then move it to, and mount it on, a system with smaller pages. – ctrl-alt-delor Aug 13 '14 at 10:22

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.