| |
| |
| |
|
Page: 1 2 3 4 5 6 7 8 9
Comments:
<0> pword
<1> ah man this is screwed up.
<1> http://www.rafb.net/paste/results/0hgpgb96.html its messy
<1> well masm doesn't seem to recognise a pword
<0> Uhm. Alright. I used it in tasm
<0> But I thought they were more compatible
<0> Wait.. you're using masm32?
<1> i'm using link16 so no
<1> actually it is ***embling using ml (32bit) ***embler
<2> guys .. using gcc under linux .. for some reason when i do ..
<2> xchg (%esp),%eax
<2> it's not exchanging an entire dword
<0> Sounds like a bug in gcc
<2> damnit
<0> Do you have to use xchg?
<0> That's a very slow way of doing things
<3> because it uses 4 instructions?
<0> No, because it locks the bus
<3> it also uses 4 instructions
<4> hey TCA : )
<4> no, no, I'm not going to har*** you for your master turbo ***embler book
<1> ahhh
<1> TCA its fword
<4> grr i'm talking to this girl who frustrates the **** out of me, and not in the ***ual sense
<4> i have a perfect analogy for her, she's like a windshield wiper, she just won't make a complete 360
<0> How does xchg use 4 instructions?
<3> xchg reg1, reg2 will be:
<0> Stop there
<3> push reg1; push reg2; pop reg1; pop reg2
<1> 3 clocks heh heh
<0> That's not how it works
<5> crazy
<3> i dis***embled in 16bit, and thats how it works
<0> The internal microcode that is executed might be many instruction (and this information is printed in the intel manual, I suggest you pick upa copy)
<0> xchg is an instruction. You can't dis***emble it further
<1> lol
<5> it's hardwired that way
<3> i saw the push and pop in the dis***emble
<0> You were mistaken
<0> Or it was a macro
<0> Or, it was a high level construct and that's how it turned out
<3> im using a86 16bit ***embler
<0> But xchg is an ***embler instruction.
<0> It doesn't get smaller than that from your perspective
<3> ok so the problem is with the ***embler
<0> Sure. Blame that if you'd like.
<5> heh
<3> download it and try it
<0> .asm xchg eax, ebx
<6> TCA: 93
<0> Okay, how is a one byte opcode 4 instructions?
<0> .dasm 93
<6> TCA:
<6> -------------------------------------------------
<5> xchgl %eax,%eax == nop :)
<3> download it and try it
<6> 00000000 93 xchg eax,ebx
<6> -------------------------------------------------
<5> .asm xchg eax,eax
<6> nijz: 90
<5> i like u wisebot.
<0> You need to "download it and try it", becuase you're the one who is mistaken
<3> dasm AAAAAAAA
<3> i did and im telling you what i saw when i dis***embled the binary file
<3> .dasm AAAAAAAA
<6> Alexande-:
<6> -------------------------------------------------
<6> 00000000 AA stosb
<6> 00000001 AA stosb
<0> Did you write the original code?
<6> 00000002 AA stosb
<6> 00000003 AA stosb
<6> -------------------------------------------------
<3> yes
<0> And it was in ***embly?
<3> yes
<1> its faster using push and pops
<0> Yeah, blame the ***embler
<0> It's faster using a register
<7> i need help
<7> in i/o
<0> Go on, abbs. You have a captive audience
<7> reading a string needs 41h interrupt right?
<0> In DOS?
<7> yes
<0> no
<7> 16 bit programming
<0> 21h
<0> http://www.ctyme.com/intr/int-21.htm
<7> tnx
<0> http://www.ctyme.com/intr/rb-2563.htm
<0> specifically that one
<0> Alexande-.. you want to throw together a short program which demonstrates your push/pop for xchg?
<3> just write a xchg instruction, ***emble it with a86 and then dis***emble it
<0> fine
<1> push ebx
<1> mov eax, ebx
<1> pop eax
<1> mines only 3 intructions hehe
<1> er wait i got it wrong there
<1> mov ebx, eax
<0> http://www.monstersoft.com/xchg.jpg
<0> Proof enough for you?
<3> ok, but this is not a86 ***embler
<0> I see. You're saying that a86 will -- instead of using the actual xchg instruction -- use push/pop to exchange registers?
<0> Is that your ***ertion?
<0> stool: using xchg on registers is fine. It's when you xchg things in memory that things get yucky.
<3> yes
<0> There are two possibilities here, Alexande-. One is that you're an idiot. The other is that the maker of a86 is an idiot.
<0> Which do you think is more likely?
<3> <3> just write a xchg instruction, ***emble it with a86 and then dis***emble it
<0> I don't have a86
<0> You want to prove you're not an idiot?
<0> I'm giving you plenty of chances to do so
<3> <3> download it and try it
<1> you confused the poor guy
<0> I gave him plenty of chances to prove or disavow his claim
<1> but now we don't have something to laugh at :|
<0> We have you.
<0> Don't worry
<0> He'll be back
<0> They always do the ban evasion thing
<0> What they don't realize is that I get off on kickbanning people
<5> lol
<0> He told me to **** off
<0> I might cry
<1> hehe
<8> good morning
<8> utro =)
<1> .weather toronto, ontario
<1> morning
<6> Weather for Toronto, Ontario(last update: 2:00 AM EST on January 25, 2006) | Temperature: 37F | Windchill: 28F | Humidity: 81% | Dew Point: 32F | Sunrise: 7:41 AM EST | Sunset: 5:19 PM EST | Moon Rise: 4:14 AM EST | Moon Set: 12:44 PM EST
<0> http://www.monstersoft.com/xchg2.jpg
<0> There's proof
<0> that he's an idiot
<5> haha
<0> Some people are forever n00b
<0> He stopped responding
<0> I think he shot himself
<1> well the fact that he said he dis***embled his binary and saw push/pops and still thought the instruction expanded to that tells us much.
<0> It could have been a macro. I would have accepted that as an explanation
<0> It could have been a high-level language construct. I would have accepted that.
<0> But, no
<0> He insisted
<0> Now he's gone
<0> So sad, too bad.
<1> Or maybe its just built in that stinking a86 ***embler
<0> He was doing xchg ds, es
Return to
#asm
or
Go to some related
logs:
Warning hedear
mutilated dicks
#moviestogo logs
#linux
#skype
victorionox wallet
usjeans
#AllNiteCafe
you my secret pasion and I have no
#london
|
|
