I wrote:
 > probably better than anyone just what it is. Some months ago, this
 > very issue of why dynamically-linked binaries ran so slowly on sparc
 > came up. After quite a bit of intricate work, he discovered and
 > removed a feature of the way mmap and shared libs were interacting
 > with the cpu cache.

Which is missing quite a bit of the story. The part about "quite a bit
of intricate work" refers to work (especially) by chuck and theo, and
I had intended to state that. This work was to identify, understand
and find the first experimental fixes for the problem.

My apologies for the error of omission, it was damn fine work by all,
that led to impressively successful results.

--
Dan.