Christophe Rhodes <[EMAIL PROTECTED]> writes:

> It looks like somewhere in the compiler there is an O(n^2) algorithm
> which is the bottleneck in the compilation for your program.  This
> isn't entirely surprising to me: many of the phases in a compiler have
> problems with solutions easily expressed in O(n^2) time or worse.
> Have you tried profiling bits of the compiler to see where most of the
> time is spent?

I did it now.  The culprit is the function IR1-OPTIMIZE in
"compiler/ir1opt.lisp".

In this routine there is a nested loop over blocks at the beginning which
looks like O(n^2).  I would also be interested to profile
IR1-OPTIMIZE-BLOCK, but this does not work:

  "Warning:  Ignoring undefined function C::IR1-OPTIMIZE-BLOCK"

Yours, Nicolas.


Reply via email to