On 4/7/08, Simon Marlow <[EMAIL PROTECTED]> wrote: > Samuel Bronson wrote: > > Sun Apr 6 09:47:55 EDT 2008 Samuel Bronson <[EMAIL PROTECTED]> > > * Implement #2191 (traceCcs# -- prints CCS of a value when available) > > Apparantly, tagged objects are passed, contrary to the comments for > > UNTAG and GETTAG in Cmm.h. So I untag them. *sigh*. > > Sorry about that comment, I think it was just out of date (actually I don't > remember writing it, so maybe it wasn't me). FWIW, the following wiki page > is a correct description of how pointer tagging works, I believe: > http://hackage.haskell.org/trac/ghc/wiki/Commentary/Rts/HaskellExecution/PointerTagging > > Regarding your patch, it looks like the primitive should be entering R2, > rather than returning it, since it might not be evaluated.
Yeah, I eventually figured that out. Oh, trac is back up now? I guess that means I can attach the actually-working version of my patch to the ticket... _______________________________________________ Cvs-ghc mailing list [email protected] http://www.haskell.org/mailman/listinfo/cvs-ghc
