C++ programming (was Newbie Alert: Beginning parallel programming with Scyld)

Dean Johnson dtj at uberh4x0r.org
Wed Oct 16 07:31:38 PDT 2002

On Wed, 2002-10-16 at 05:50, graham.mullier at syngenta.com wrote:
> I do find these religious wars amusing. A small point following the F******
> is better/oh no it isn't stuff:

It kinda scares me that it hasn't degenerated into that already. There
are usually language zealots lurking waiting to become a full fledged
language nazi's when such stuff comes up. In my local user group I often
take a whack at the hornets nest by casually mentioning vi's superiority
over emacs. ;-) If we ever get enough organization together to go play
paintball, that is the odds-on favorite for how to divide into teams.

> compiler writers can no doubt correct me but I'm reasonably convinced that
> in many cases Fortran (77 rather than 90 or later) is sufficiently limited
> in the language constructs possible that the compiler is better able to
> optimise the code. That's better than a random sample of code written in,
> say, C, not code written really carefully knowing exactly how the compiler
> will map code onto the underlying hardware architecture.

You are indeed correct. Aliasing is one of the biggest problems that
inhibits optimizations, whether its there or not. It also depends on the
compiler folks as well. For instance, Cray compiler folks typically were
of the safety first mentality. If there was a possible problem, opt for
the most conservative route. SGI compiler folks were more of the risk
management style. If there was a safety issue, figure out how many
people it might hit and if the number is reasonably low, do the unsafe
thing and give them a command line option to disable it.

> (hitting send during one of these debates feels a lot like lighting the blue
> touchpaper and retiring...)

Its often during those times of energetic arguing that the most
knowledge is exchanged. Sometimes, sadly, it is learning that your
friend is a complete idiot. ;-)


More information about the Beowulf mailing list