1. LCC port
-------Phoenix-Boundary-07081998-
You wrote on 2/14/02 12:37:43 PM:
>While we're on the subject, didnt you write 2X threaded code for
>LCC? Mind sharing what/how you did this?
>
>Euman
>euman at bellsouth.net
I did and my submitted interpreter is threaded (unless I
mis-typed). This makes it much faster, but still 15-20%
slower than Watcom (Watcom can use register-passing
conventions).
Implementation involves a bit of assembly code to do the
threading ('jmp esi'), a subroutine that returns the code
address of its call, and a bunch of debugging.
In a number of places, a dummy instruction must be placed
!after! a 'jmp esi' instruction. Rob's Watcom code needed
them in some places, my LCC version needed them in
different places. There seems to be no way to know where
other than by tracing the crashes.
I'm willing to share my port with other LCC users subject
to verification of source licencing. Anybody know how
we do that?
Karl Bochert
-------Phoenix-Boundary-07081998---
2. Re: LCC port
- Posted by euman at bellsouth.net
Feb 14, 2002
----- Original Message -----
From: <kbochert at ix.netcom.com>
You wrote on 2/14/02 12:37:43 PM:
>While we're on the subject, didnt you write 2X threaded code for
>LCC? Mind sharing what/how you did this?
>
>Euman
>euman at bellsouth.net
I did and my submitted interpreter is threaded (unless I
mis-typed). This makes it much faster, but still 15-20%
slower than Watcom (Watcom can use register-passing
conventions).
Implementation involves a bit of assembly code to do the
threading ('jmp esi'), a subroutine that returns the code
address of its call, and a bunch of debugging.
In a number of places, a dummy instruction must be placed
!after! a 'jmp esi' instruction. Rob's Watcom code needed
them in some places, my LCC version needed them in
different places. There seems to be no way to know where
other than by tracing the crashes.
I'm willing to share my port with other LCC users subject
to verification of source licencing. Anybody know how
we do that?
Karl Bochert
Maybe a line# in the Source as confirmation unless
Rob wants to post a list of Reg'd users.
Euman
3. Re: LCC port
Karl Bochert writes:
> I'm willing to share my port with other LCC users subject
> to verification of source licencing. Anybody know how
> we do that?
I suggest that if someone asks you for source code,
you should ask him to tell you some random detail
from the source, e.g. what's on line #xxx of source file yyy.
I can confirm that you Karl and Euman both have the source,
along with many others. For various reasons I don't want to
publish and keep updating a full list of names.
Regards,
Rob Craig
Rapid Deployment Software
http://www.RapidEuphoria.com
