Re: [PATCH] lock troubles in pre6-2

From: Andy Chou (acc@CS.Stanford.EDU)
Date: Sat Aug 05 2000 - 18:06:07 EDT

  • Next message: James Stevenson: "Re: 2.2.16 - Crash"

    Here is a more complete listing of the issues found by a compiler-based
    checker after being filtered by me. The checker essentially checks that
    spin locks, interrupts, and bh's are paired up in functions, warning when
    the function can exit holding a lock, with interrupts disabled, or while
    bh's are disabled. It also tries to check for double locks/unlocks within
    the same function, and also restoring of flags not saved within the same
    function.

    I had to filter out ~250 false positives, the ones here are ones I've
    looked at at least briefly so there should be at least a few bugs :)

    Issues to be aware of:

    1. The context given for each potential bug is incomplete; it's necessary
    to consult the source.

    2. The version we checked is 2.3.99-? -- I'm not sure exactly which one
    but I can verify if needed.

    3. Some of the "bugs" are at most poor style, and there are probably false
    positives. The checker isn't global so it knows nothing about what goes
    on in function calls. It may be made global at some point (suggestions
    on how to do this effectively are welcome).

    4. If you can verify any of these, please cc mc@cs.stanford.edu.

    -Andy



    -
    To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
    the body of a message to majordomo@vger.rutgers.edu
    Please read the FAQ at http://www.tux.org/lkml/



    This archive was generated by hypermail 2b29 : Sat Aug 05 2000 - 18:09:56 EDT