"John S. Dyson" writes:
> Ted Lemon said:
> > The only way I can see that they can make this claim is if their
> > buffer cache code performs a lot better than ours, or if they've got
> > some hack going in their stdio code that makes things work better.
> > These are both easy suppositions to test.

> Our buffer cache does work much better, as I stated in a very recent
> email.

If this is actually true, and this actually speeds up compiles of
reasonable size, and you actually have benchmark data to prove this,
then surely it would be trivial for you to shut me up by posting it.

Perry