Path: utzoo!attcan!uunet!ncrlnk!ncr-sd!hp-sdd!hplabs!ucbvax!bloom-beacon!CITI.UMICH.EDU!martin
From: martin@CITI.UMICH.EDU (Martin Friedman)
Newsgroups: comp.windows.x
Subject: (none)
Message-ID: <8811171309.AA25465@EXPO.LCS.MIT.EDU>
Date: 17 Nov 88 13:08:00 GMT
Sender: daemon@bloom-beacon.MIT.EDU
Organization: The Internet
Lines: 17

Newsgroups: citi.xpert
Subject: Re: To that guy earle at sun doing speedups
References: <26195@citi.umich.edu>
Reply-To: martin@citi.umich.edu (Martin Friedman)
Organization: U of M Center for Information Technology Integration, Ann Arbor, MI

I have been spending much time in the past days merging the speedups contained
on the R3 tape into the mfb code for R3.  I have had great luck so far.  There
were bugs some of the code, and I suppose they survived because they were 
Sun-byte-order dependent bugs.  #ifdef vax  compiled good code.

Not being finished, nor done, I can't post diffs, but I will later...
I am in addition going to include some asm statements for getbits and putbits
(the inner-inner-loop macros) on the 68020.  Should cause the affected servers
to fly.

.signature