Re: Format string bugs & race conditions

[Ole-Hjalmar Kristensen <ole-hjalmar.kristensen@substitute_employer_here.com>]

>>>>> "BK" == Benjamin Ketcham <bketcham@drizzle.com> writes:

<snip>

    BK> I consider tasking difficult to separate from issues of process
    BK> management; and that in turn is difficult to separate from dependence
    BK> on the OS.  I.e., it's not always a great idea to build this into
    BK> the language, any more than it'd be to build in graphical I/O.
    BK> In both cases, many people would be happy to have the things built-in,
    BK> and many would find uses for the built-in constructs, but others will
    BK> find deficiencies and will have to go outside the language.

    BK> --Benjamin

Actually, having tasking built-in is an excellent idea. It gives you a
much higher probability that programs which use tasking actually will
be portable. And if you have access to the source of the tasking run
time, or it has a published interface, you can always reimplement the
tasking run-time to suit your needs. If you don't like the language's
tasking *model*, that's a bit tougher. In this case some languages
(including Ada, see annex D.10) let you have access to more primitive
operations which can be used to roll your own.

-- 
   C++: The power, elegance and simplicity of a hand grenade.