Re: compat work.

[christos%zoulas.com@localhost (Christos Zoulas)]

On Mar 15,  6:41pm, paul%whooppee.com@localhost (Paul Goyette) wrote:
-- Subject: Re: compat work.

| Yeah, it was a bit trickier than just resolving the few conflicts that 
| CVS found, but not as difficult as I'd feared.

Yes, there is only non compat code in uvm_swap.c (except the lines that
declare the compat function pointers and the lines that call them).

| BTW, if you're looking for a challenging candidate to untangle, have a
| look at the rtsock.c code!  The entire rtsock.c file is #include'd in
| the rtsock_50.c compat code, and bits and pieces of rtsock.c are
| basically disabled by #ifdef COMPAT_RTSOCK which in turn is defined in
| by rtsock_50.c

Yes, this is the poster boy of a mess and the most difficult to deal with.
Please don't refactor that. I need to really think what's the best way to
untagle this one...

christos