Much of this is done already. Simon Marlow's done the register mapping (ghc/includes/MachRegs.h) and done other registerisation work. There was a bug in gcc-3.3x that halted things for a while, but this has been fixed in 6.4
All the files mentioned except for the Adjustor appear to have x86_64 bits.
If you felt up to it, it would be a good way to learn about ghc internals. Feel free to ask questions on the mailing list, we'll help as much as possible.
Sounds interesting, at the moment I just want to get an unregistered build going. I've followed the instructions to the letter. After adding an x86_64-* entry to config.sub in gmp I get the following on the target: copying ./mpn/generic/gcdext.c to mpn/gcdext.c copying ./mpn/generic/tdiv_qr.c to mpn/tdiv_qr.c copying ./mpn/generic/bz_divrem_n.c to mpn/bz_divrem_n.c copying ./mpn/generic/sb_divrem_mn.c to mpn/sb_divrem_mn.c copying ./mpn/generic/jacbase.c to mpn/jacbase.c copying ./mpn/generic/gmp-mparam.h to gmp-mparam.h gmake: *** No rule to make target `AutoApply.hc', needed by `depend'. Stop. gmake: Leaving directory `/home/kmacy/ghc-6.2.2/ghc/rts' The makefile indicates that genapply is called for, but it isn't there, oddly enough the one on the host failed initially: kmacy@shemp ./ghc/utils/genapply/genapply genapply: internal error: stg_ap_v_ret Please report this as a bug to glasgow-haskell-bugs@haskell.org, or http://www.sourceforge.net/projects/ghc/ but when I rebuilt it it worked - I then copied that over - Then I got complaints about all the *_hsc.c files missing in the libraries - hoping that the compiler itself didn't need them I just touched those files so the build could continue - the same occurred for .o files under base/control ... Am I way off track or is building an unregistered target just really unfriendly?