Re: [colorforth] Strage ColorForth results
- Subject: Re: [colorforth] Strage ColorForth results
- From: "Roman Pavlyuk \(personal\)" <john@xxxxxxxxxxxxx>
- Date: Tue, 24 Feb 2004 17:55:27 +0200
(Best viewed in monospace font)
You're right, I'm wrong.
Anyway, the original wrong one
: 0? ? drop if 0 ; then 1 ;
00900811 A9 EB 07 90 00 test eax,9007EBh
00900816 AD lods dword ptr [esi]
00900817 74 00 je 00900819
00900819 8D 76 FC lea esi,[esi-4]
0090081C 89 06 mov dword ptr [esi],eax
0090081E B8 01 00 00 00 mov eax,1
00900823 C3 ret
00900824 8D 76 FC lea esi,[esi-4]
00900827 89 06 mov dword ptr [esi],eax
00900829 B8 00 00 00 00 mov eax,0
0090082E C3 ret
TEST uses some junk off stack.
Now, my preferred one:
: 0? [ 0 ] + drop if 0 ; then 1 ;
where [ 0 ] is a yellow zero
0090079E 05 00 00 00 00 add eax,0
009007A3 AD lods dword ptr [esi]
009007A4 74 0B je 009007B1
009007A6 8D 76 FC lea esi,[esi-4]
009007A9 89 06 mov dword ptr [esi],eax
009007AB B8 00 00 00 00 mov eax,0
009007B0 C3 ret
009007B1 8D 76 FC lea esi,[esi-4]
009007B4 89 06 mov dword ptr [esi],eax
009007B6 B8 01 00 00 00 mov eax,1
009007BB C3 ret
You can see, btw, that this is not optimized properly
So,
: 0? [ 0 ] + if drop 0 ; then 1 ;
009007CC 05 00 00 00 00 add eax,0
009007D1 74 06 je 009007D9
009007D3 B8 00 00 00 00 mov eax,0
009007D8 C3 ret
009007D9 B8 01 00 00 00 mov eax,1
009007DE C3 ret
looks much better
This time tested (a bit)
BR,
Roman
---------------------------------------------------------------------
To unsubscribe, e-mail: colorforth-unsubscribe@xxxxxxxxxxxxxxxxxx
For additional commands, e-mail: colorforth-help@xxxxxxxxxxxxxxxxxx
Main web page - http://www.colorforth.com
