5.x test request (FWD: Repairing ext2fs stat(2), fts(3) in 6.0-current, please test on 5.x)

From: Martin Cracauer (cracauer_at_cons.org)
Date: 07/09/05

  • Next message: Chris Hodgins: "Re: gmirror, sparc and SCSI problems" 
    Date: Sat, 9 Jul 2005 11:09:51 -0400
To: freebsd-stable@freebsd.org

    Can somebody running FreeBSD 5-stable please test whether this patch
    is needed in 5.x?

    I can provide a small file with an ext2fs in it so that you can try in
    a vnode mount.

    Martin

    attached mail follows:

    Date: Fri, 8 Jul 2005 18:15:27 -0400
To: freebsd-current@freebsd.org

    ext2fs fails to set the device in the stat(2) system call.

    Subsequently, that makes fts(3) fail, which goes as far as make ls(1)
    fail (which uses fts).

    The appended diff fixes it for me and looks correct to me. Unless
    somebody objects I will submit it to re@ for commit approval.

    %%

    I don't have a 5.x system anywhere, can somebody please test whether
    the problem exists in 5.x?

    Here is how:
    - compile appended test program
    - mount an ext2fs to -say- /mnt/tmp
    - `./stattests /mnt/tmp`
    ==> must return identical device ids

    Quicker test:
    - mount ext2fs to /mnt/tmp
    - `ls -F /mnt/tmp`
    ==> will fail with file not found errors.

    Martin 

    -- 
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
Martin Cracauer <cracauer@cons.org>   http://www.cons.org/cracauer/
 No warranty.    This email is probably produced by one of my cats 
 stepping on the keys. No, I don't have an infinite number of cats.

    _______________________________________________
    freebsd-stable@freebsd.org mailing list
    http://lists.freebsd.org/mailman/listinfo/freebsd-stable
    To unsubscribe, send any mail to "freebsd-stable-unsubscribe@freebsd.org"

  • Next message: Chris Hodgins: "Re: gmirror, sparc and SCSI problems"

    Relevant Pages

    • Re: the Halting Problem has a lot to it
      ... >> solve for (instead of looping). ... >these parameter-doubling loops. ... It's very easy to show that Halts will fail ... Martin ...
      (sci.logic)
    • Re: Home early
      ... Martin Bowers has been pushing me around ... and doing all the lock wheeling too. ... Dreamcatcher is now back at Lower Heyford awaiting a BSS examination ... It will fail. ...
      (uk.rec.waterways)
    • Re: dynamic type checking - a pauline conversion?
      ... Robert C. Martin wrote: ... >>example in a common superclass like Object) or do we expect the method ... >>send to fail at runtime when the specific class doesn't implement ...
      (comp.object)
    • Re: BOKO death
      ... Martin wrote: ... decided to fail today :-( ... Questions - what's the easiest method to get the data off the HD? ... Put someone else's battery into it, and get off all the data while you ...
      (uk.comp.sys.mac)
    • Re: Very strange transistor problem.... Sys 11A
      ... Most manufacturers don't specify gain with the collector and emitter ... Thanks, Martin. ... without fail, until you sell it and then it flakes out the very next ...
      (rec.games.pinball)