Bruce Evans wrote:
On Tue, 15 Aug 2006, John Baldwin wrote:
On Tuesday 15 August 2006 09:55, Kostik Belousov wrote:
So, it seems that xorq %rax, %rax and xorl %eax, %eax will make the
same results, but in the different ways. And xorq requires REX prefix,
that shall make the decoding longer.
Ok, thanks! David, can you revert your change? I had almost gone
through
earlier and fixed all the places that did this a while back until I
realized
that it must have been intentional.
It is logged as intentional in the commit that went through and optimized
all places that used xorq:
%%%
RCS file: /home/ncvs/src/sys/amd64/amd64/support.S,v
Working file: support.S
head: 1.118
...
----------------------------
revision 1.116
date: 2005/09/27 18:32:46; author: peter; state: Exp; lines: +11 -11
Fix a minor nit that has been bugging me for a while. Fix the obvious
cases of using a 64 bit operation to zero a register. 32 bit opcodes are
smaller and supposedly faster, and clear the upper 32 bits for free.
----------------------------
%%%
Bruce
I really, truly, am constantly surprised at the number of people who
'use' CVS, or any version control system for that matter, but don't look
at annotations or revision history when they are making changes. I have
often had to remind co-workers that version control software is more
than just a fancy/expensive FTP server.
Scott
_______________________________________________
cvs-all@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/cvs-all
To unsubscribe, send any mail to "[EMAIL PROTECTED]"
