cvs commit: fptools/ghc/compiler/coreSyn CoreSat.lhs fptools/
Thu, 21 Dec 2000 03:31:56 -0800
> qrczak 2000/12/20 10:32:00 PST
> Modified files:
> ghc/compiler/coreSyn CoreSat.lhs
> ghc/compiler/stgSyn CoreToStg.lhs
> import PprCore
Did you need to do this because GHC was complaining about missing
instances during bootstrapping? If so, it could be a bug in GHC. Simon
fixed at least two related bugs in the renamer earlier this week, but
there could be more.
FWIW, the HEAD is reasonably stable now. Yesterday I did a full build
in my working tree (including hslibs), a nofib run, a bootstrap, and
finally a test run with the bootstrapped compiler. There are one or two
small issues shown up by the tests, but nothing major.
One thing that might catch people out: You can't make boot in hslibs
until you've made the compiler. I *still* don't have a good solution to
Have a great Christmas folks!