"We Never Use Assembly Language"
in [ASM]
|
From: Frank Kotler on 14 Mar 2006 22:40 o//annabee wrote: > Talk is cheap Master PDF. And machine code don't lie. (and it's pretty cheap, too!) I don't doubt that the code does what you say, on your machine (that does a CR on "shift-M"), but I was so "flabbergasted" that I had to try it myself. Maybe I didn't get it right - the RosAsm macro is expanded from memory of what you posted (too lazy to go back and look). Is that correct? I put the "align"s where I thought they should be, too. I'm thinking that maybe the right thing to do is to "cpuid" (or some other serializing instruction?) before the ending rdtsc, too? Anyway, here's what I came up with. Results differ from yours: 7751h vs 4E89h (pretty consistant) Best, Frank global _start %assign COUNT 10000 section .text _start: nop ; "Wannabee's loop" cpuid rdtsc push eax mov ecx, COUNT xor eax, eax align 16 top: cmp ecx, 0 jna endw add eax, ecx dec ecx jmp top endw: rdtsc pop ebx sub eax, ebx call showeaxh mov al, 10 call putc ; "Randy's loop" cpuid rdtsc push eax mov ecx, COUNT xor eax, eax jecxz endw2 align 16 top2: add eax, ecx dec ecx jnz top2 endw2: rdtsc pop ebx sub eax, ebx call showeaxh mov al, 10 call putc exit: mov eax, 1 xor ebx, ebx int 80h ;---------------------- ; Linux cruft from here ;--------------------------- putc: push edx push ecx push ebx push eax mov eax, 4 mov ebx, 1 mov ecx, esp mov edx, 1 int 80h pop eax pop ebx pop ecx pop edx ret ;----------------------------- ;------------------------------ showeaxh: push eax push ebx push ecx push edx sub esp, 10h mov ecx, esp xor edx, edx mov ebx, eax ..top: rol ebx, 4 mov al, bl and al, 0Fh cmp al, 0Ah sbb al, 69h das mov [ecx + edx], al inc edx cmp edx, 8 jnz .top mov ebx, 1 mov eax, 4 int 80h add esp, 10h pop edx pop ecx pop ebx pop eax ret ;------------------------------
From: sevagK on 15 Mar 2006 00:46 o//annabee wrote: > På Wed, 15 Mar 2006 01:49:42 +0100, skrev sevagK <kahlinor(a)yahoo.com>: > > >> < http://rosasm.org > > > > You and your > > cronies can blurt out this or that without a shred of evidence, at > > least what I say can be found on this newsgroup by anyone who looks. > > When did I blurt out something without a shred of evidence? Well, let's see, just a few from this very thread: "Fastest assebler" - benchmarks prove otherwise. .. "Best source organization" - more like the worst. "Best macros" - when this illusion was shattered, you changed your tune and claimed you're not interested in sophisticated macros and Rosasm macros are good enough for you. Good enough does not equal "best macros" The sad fact is, you've become so entrenched in this warfare that you only see what you want to see. Anything that contradicts with your limited viewpoint is thrown out and conveniently forgotten. That's why you endlessly bring out the same tired arguments as soon as the counter-arguments against them fall off the bottom of the thread list. -sevag.k www.geocities.com/kahlinor
From: o//annabee on 15 Mar 2006 02:56 På Wed, 15 Mar 2006 04:40:36 +0100, skrev Frank Kotler <fbkotler(a)comcast.net>: > o//annabee wrote: > >> Talk is cheap Master PDF. > > And machine code don't lie. (and it's pretty cheap, too!) I don't doubt > that the code does what you say, on your machine (that does a CR on > "shift-M"), but I was so "flabbergasted" that I had to try it myself. > Maybe I didn't get it right - the RosAsm macro is expanded from memory > of what you posted (too lazy to go back and look). Is that correct? I > put the "align"s where I thought they should be, too. I'm thinking that > maybe the right thing to do is to "cpuid" (or some other serializing > instruction?) before the ending rdtsc, too? Anyway, here's what I came > up with. Results differ from yours: 7751h vs 4E89h (pretty consistant) Ok. I test it on the athlon as well later on. I have an old intel P3 in parts as well, maybe I can test there as well. In any case, a "master of assembly", should know that there are diffrences between processors, and not makes claims like this !#"¤!"#¤"!# does. I take your words for the test as gospel for now. Mine is 100% autentic. Its tested on an AMD64 - 3700, running win32. Now I got a more interessting question for you Frank. I need some win32 skeleton file for NASM. Just want to have a minimum skeleton. Single file. Is that possible? I look at at 3 forums yet no luck. I didnt find anything of use in the docs or the download. I am a bit impatient, sorry. It must show how to import an API function. Could you make me a file equivalent to this one, for Win32 ? This is a RosAsm win32 code, fuly running skeleton. I dont care of the NASM is longer, I just need to see it. In the mean time I try to find something on my own. _______________________________________ [OneEquate 100 SizeOf_Array 256] [ByteVariable: b$ 0 WordVariable: w$ 0 Dword: D$ 0 Float: F$ 1.0 Real: r$ 2.2 ArrayOfByte: b$ 0 #SizeOf_Array ] main: mov al b$ByteVariable mov bx w$WordVariable mov ecx D$DWord mov edx F$Float fild D$Float fistp D$Float fld r$Real ftp r$Real call 'kerneø32.ExitProcess' 0 _______________________________________ > > Best, > Frank Hei! Is this sort of "before I have asked, you have answered trick? Will this run compile as win32 file? I check it out. Thanks. > global _start > > %assign COUNT 10000 > > section .text > _start: > > nop > > > ; "Wannabee's loop" > cpuid > rdtsc > push eax > > mov ecx, COUNT > xor eax, eax > align 16 > top: > cmp ecx, 0 > jna endw > add eax, ecx > dec ecx > jmp top > endw: > > rdtsc > pop ebx > sub eax, ebx > call showeaxh > > mov al, 10 > call putc > > > ; "Randy's loop" > > cpuid > rdtsc > push eax > > > mov ecx, COUNT > xor eax, eax > jecxz endw2 > align 16 > top2: > add eax, ecx > dec ecx > jnz top2 > endw2: > > rdtsc > pop ebx > sub eax, ebx > call showeaxh > > mov al, 10 > call putc > > exit: > mov eax, 1 > xor ebx, ebx > int 80h > ;---------------------- > > ; Linux cruft from here > > ;--------------------------- > putc: > push edx > push ecx > push ebx > push eax > > mov eax, 4 > mov ebx, 1 > mov ecx, esp > mov edx, 1 > int 80h > > pop eax > pop ebx > pop ecx > pop edx > ret > ;----------------------------- > > > ;------------------------------ > showeaxh: > push eax > push ebx > push ecx > push edx > > sub esp, 10h > > mov ecx, esp > xor edx, edx > mov ebx, eax > .top: > rol ebx, 4 > mov al, bl > and al, 0Fh > cmp al, 0Ah > sbb al, 69h > das > mov [ecx + edx], al > inc edx > cmp edx, 8 > jnz .top > mov ebx, 1 > mov eax, 4 > int 80h > > add esp, 10h > > pop edx > pop ecx > pop ebx > pop eax > ret > ;------------------------------
From: o//annabee on 15 Mar 2006 03:04 P? Wed, 15 Mar 2006 04:55:14 +0100, skrev Isaac Bosompem <x86asm(a)gmail.com>: > I do think it is a very good idea to use a scripting language to code > game logic. It makes it easy to debug and maintain. The only scripting language for games should be assembly language. I think assembly language should be the scripting language for Visual Basic as well. I'd like to add a smiley, but I am serious.
From: o//annabee on 15 Mar 2006 03:20
P? Wed, 15 Mar 2006 06:46:14 +0100, skrev sevagK <kahlinor(a)yahoo.com>: > > o//annabee wrote: >> P? Wed, 15 Mar 2006 01:49:42 +0100, skrev sevagK <kahlinor(a)yahoo.com>: >> >> >> < http://rosasm.org > >> >> > You and your >> > cronies can blurt out this or that without a shred of evidence, at >> > least what I say can be found on this newsgroup by anyone who looks. >> >> When did I blurt out something without a shred of evidence? > > > Well, let's see, just a few from this very thread: > > "Fastest assebler" - benchmarks prove otherwise. Which benchmark is that? > "Best source organization" - more like the worst. No it IS the best. What on earth makes you say its the worst? > "Best macros" - when this illusion was shattered, you changed your tune > and claimed you're not interested in sophisticated macros and Rosasm > macros are good enough for you. Good enough does not equal "best > macros" The macros I like are the best macros ! > The sad fact is, you've become so entrenched in this warfare that you > only see what you want to see. Not at all, I see you as well. > Anything that contradicts with your > limited viewpoint is thrown out and conveniently forgotten. If so, why you are still here? > That's why > you endlessly bring out the same tired arguments as soon as the > counter-arguments against them fall off the bottom of the thread list. :))) Hehehe. What a nerd you are SevagK! > > > -sevag.k > www.geocities.com/kahlinor > |