Re: panic: kern/52718

From: Terry Lambert (tlambert2_at_mindspring.com)
Date: 05/30/03

  • Next message: Matt: "4BSD vs ULE scheduler question" 
    Date: Fri, 30 May 2003 02:27:27 -0700
To: David Xu <davidxu@freebsd.org>

    David Xu wrote:
    > > This was caused by rev. 1.3 of a commit by Jeff Robertson to
    > > kern_utmx.c. The problem is that the proc struct is not locked
    > > for:
    > >
    > > FOREACH_THREAD_IN_PROC(td->td_proc, td0)
    > >
    > > in the lock and unlock.
    > >
    > > Either lock the proc before and unlock it after this, in both
    > > _utmx_lock() and _utmx_unlock(), or revert the code to 1.2.
    >
    > kern_sig.c has same issue in several places.

    Just looked... YUCK! The Process group code and the code in
    the filt_sigdetach() have got to be what you are talking about,
    right?

    I'm constantly surprised at some of the race windows I find in
    production code (not just FreeBSD), that are just waiting there
    to chew someone's leg off the first chance they get... 8-(.

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

  • Next message: Matt: "4BSD vs ULE scheduler question"

    Relevant Pages

    • Re: Continuous Forms Formatting
      ... David wrote: ... Check a check box and lock that check box after it is checked ... Unlock the check box, allowing the user the option of checking it ... On a continuous or datasheet form, you need to use CF for ...
      (microsoft.public.access.formscoding)
    • Re: Refresh Problem
      ... If I wait 3 seconds between mouse clicks on the lock / unlock button, ... workstation is looking at a current copy of the data. ... MESSAGEBOX statement after all of the database lookups have been done ...
      (microsoft.public.fox.programmer.exchange)
    • Re: [PATCH] Remove softlockup from invalidate_mapping_pages.
      ... These block devices are very likely to have just one ... Being locked for read is very unlikely because mdadm would have already ... memory reclaim could lock the page, ... core is free to unlock the page when the IO completes, ...
      (Linux-Kernel)
    • Re: Behavior of flock() oriented procedures
      ... happens when a file descriptor is locked. ... descriptore without explicitly calling flockwith the unlock message, ... does this mean that the file will still have the lock on it, ... without any explicit unlocking. ...
      (comp.unix.programmer)
    • Re: CSingleLock - known behaviour?
      ... While a CRITICAL_SECTION has recursive acquisition semantics, the MFC layers introduce so ... many bugs that it would not at all surprise me if CCriticalSection was as bug-laden as the ... Now look at the unlock code: ... only do an unlock if it believes the lock is set, and the result of a successful unlock is ...
      (microsoft.public.vc.mfc)