lkml.org 
[lkml]   [2006]   [Nov]   [4]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
    /
    Date
    From
    SubjectRe: New filesystem for Linux
    On Sat, 2006-11-04 14:59:53 -0500, Albert Cahalan <acahalan@gmail.com> wrote:
    > Grzegorz Kulewski writes:
    > > On Thu, 2 Nov 2006, Mikulas Patocka wrote:
    > > > As my PhD thesis, I am designing and writing a filesystem,
    > > > and it's now in a state that it can be released. You can
    > > > download it from http://artax.karlin.mff.cuni.cz/~mikulas/spadfs/
    > >
    > > "Disk that can atomically write one sector (512 bytes) so that
    > > the sector contains either old or new content in case of crash."
    >
    > New drives will soon use 4096-byte sectors. This is a better
    > match for the normal (non-VAX!) page size and reduces overhead.

    Well... VAXen use physical PTEs for 512 byte pages, Linux uses 8
    consecutive pages to simulate 4K pages.

    On top of that, some of today's machines have configurable page sizes.
    Besides that, 512 byte sectors are quite a clever thing: Drives
    probably can write a number of consecutive sectors, so if you want to
    send a page, just send eight sectors.

    > BTW, a person with disk recovery experience told me that drives
    > will sometimes reorder the sectors. Sector 42 becomes sector 7732,
    > sector 880880 becomes sector 12345, etc. The very best filesystems
    > can handle with without data loss. (for example, ZFS) Merely great
    > filesystems will at least recognize that the data has been trashed.

    Uh? This should be transparent to the host computer, so logical sector
    numbers won't change.

    MfG, JBG

    --
    Jan-Benedict Glaw jbglaw@lug-owl.de +49-172-7608481
    Signature of: http://catb.org/~esr/faqs/smart-questions.html
    the second :
    [unhandled content-type:application/pgp-signature]
    \
     
     \ /
      Last update: 2006-11-04 22:03    [W:2.516 / U:0.032 seconds]
    ©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site