Use Bohem's GC for compiler proper in 4.1?

[Robert Dewar]

Gabriel Dos Reis wrote:

> I wasn't aware that people were exclusively concentrating on small
> linear gains. 

Since no one said they were, and since they aren't, it is not surprising
that you would be unaware of this non-fact :-) I didn't say people were
exclusively concentrating on such gains by any means, just that it makes
more sense to look for good targets of opportunity. Some of the suggestions
in this thread seemed dubious to me from that point of view.