Re: Bounds checking - historical aside

From: Brett Glass (brettat_private)
Date: Tue Jul 21 1998 - 12:15:49 PDT

  • Next message: Craig Spannring: "Re: EMERGENCY: new remote root exploit in UW imapd"

    At 02:18 PM 7/21/98 +1200, Russell Fulton wrote:
    
    >I will add another:
    >
    >   4) Use hardware that supports bounds checking.
    >
    >OK This isn't an option for most of us since most HW architectures that
    >we are currently stuck with don't implement bounds checking.
    
    Perhaps you're not aware of it, but the x86 architecture does
    implement bounds checking. It even has a built-in BOUND instruction to
    do it! And... guess what? Most programmers don't know it exists.
    
    The x86 also has segmentation, which keeps pointers from going out
    of range. Unfortunately, OS implementors routinely turn segmentation OFF,
    perpetuating all of the many problems. We must insist that they stop doing
    this. Current CPUs are more than fast enough to handle the overhead.
    
    --Brett
    



    This archive was generated by hypermail 2b30 : Fri Apr 13 2001 - 14:07:08 PDT