sperber(a)informatik.uni-tuebingen.de (Michael Sperber [Mr. Preprocessor]) writes:
Well, it's just an example for the really serious problem:
*Everything* has gotten noticeably slower. What I'm saying is: it
didn't use to be this way. Something has changed. The GC algorithm
doesn't explain the problem. It's not the GC which has gotten slower,
it's the computation between GCs.
Could you please state more exactly what you mean and how you
conclude that from the test you posted. I'm guessing that you mean
that
"In an old XEmacs 21.2 everything is slower than in an fresh XEmacs
21.2. This slowdown was not present in previous versions."
My profiling data looked sane and there the slow down looked mostly
due to spending more time in GC.
Jan
P.S. We currently use 50% more memory per cons cell.