HAY Herbert, wanna write a legal PE executable ?
in [ASM]
|
From: hutch-- on 7 Sep 2007 07:32 Herbert, > > I don't particularly care what you requested, the masm32 project has > > been available for years and it contains code that I have written. > > So you refuse to enter the ring. I have long ago been in the ring with the MASM32 project, you are just hedging because you know you are wrong. > No, I posted pure binary code. Sometime it happens that > a binary byte is displayed as a 'a' or 'Z'. Your talking nonsense, ascii does not display binary, it displays ascii. > The other way: show me a single 32 bit Windows version where it > doesn't run. I don't have to, Microsoft LINK and DUMPPE both show your EXE is broken junk. ========================================================================= Microsoft (R) COFF Binary File Dumper Version 5.12.8078 Copyright (C) Microsoft Corp 1992-1998. All rights reserved. Dump of file H:\herbert\demo.exe H:\herbert\demo.exe : warning LNK4048: Invalid format file; ignored ========================================================================= DumpPE v2.20 (c) Copyright Tenth Planet Software Intl., C Turvey 1995-2005. All rights reserved. Non-Commercial use only. 'demo.exe' not an .EXE or .DBG file. ======================================================================== > > In a HEX editor it looks like encrypted data. Run the file and is > > startup is laggy and is a lot slower than the console startup time. > > Most binary file looks like "encrypted data" in a hex editor. Try telling the truth, every 32 bit PE has a DOS header starting with MZ. It appears you are unfamiliar with this. > > Who said it need the sheer brutal power of MASM to write a 1024 byte > > window. It just happens to be the legal minimum size for a 32 bit PE > > file. > > Because you wanted to give us some examples of the "sheer brutal > power of MASM" and then posted this example. No, I gave you more than one 1k legal PE example built in masm. Something you not yet managed to do with your hand coded junk. > > > Ok Hutch, let's see one of your MASM examples so we can decide if it's > > > possible to improve it using a better assembler. But don't give us a link, > > > post it right here, the complete source (with all necessary include files) > > > for a short demo which can be assembled with MASM. > > > Done years ago, the MASM32 Project at www.masm32.com. > > Then extract the one example which best shows us the "sheer brutal > power of MASM" and post it here. You are in no position to tell me what to do. There are more than enough examples in the masm32 project. > I always hand code the binaries when doing assembly programming. You are talking nonsense here, if you code binary you are not writing assembler. > Ok Hutch, let's see one of your MASM examples so we can decide > if it's possible to improve it using a better assembler. You were also being dishonest about doing a comparison with an assembler. > Only MASM and HLA user let the "assembler" generate code which > they didn't write explicitly themselves. You are also ignorant of the technical details here, MASM assembles object modules, the LINKER writes the MZ and PE header and sections. If you know how to write masm you get what you write. > > What is the big deal, its done with a DOS stub for a standard PE file. > > Sure, but I want to see the "sheer brutal power of MASM" to do it > in a simple and readable way. In 16 bit linker use the undocumented /KNOWEAS option from MSDN. In 32 bit linker use /STUB:yourfile.exe Don't you know how to use a linker ?
From: CodeMonk on 7 Sep 2007 09:05 Frank Kotler wrote: > Herbert Kleebauer wrote: > > .... >> gver.com>gver.exe >> del gever.com > > Funny thing is, with this typo, you've given 'em a clue. They're gonna > figure it out! Mmmmm, maybe not... (I thought I was disassembling data, > too :) > > Best, > Frank Who needed a clue. We we're all DOS programmer's once. But the original message seemed to be the opposite of the clue you're referring to. :) - Scott
From: //o//annabee on 7 Sep 2007 11:05 P� Fri, 07 Sep 2007 10:24:05 +0100, skrev hutch-- <hutch(a)movsd.com>: > You are simply mistaken here. When you can verify that your 922 byte > application runs on EVERY Win32 version from the earliest 32 bit > version of WinNT to current Vista then you may have something to say > but the PE specification actually DOES run on all versions of 32 bit > Windows. This is pretty weak logic. Any app with a significant enough bug, and a fully "legal" PE, will not run in ANY windows version. You really are scary stupid, Hutch-- > > Regards, > > hutch at movsd dot com >
From: CodeMonk on 7 Sep 2007 10:47 With respect to the 922 byte executable in the other thread, for which hints aren't/weren't necessary, I just want to give you the Kode.Fu Black Belt. I don't think I've seen anyone play with the *vdm* that way before. Very impressive and skillful Kode-Fu indeed !!! But, with respect to that 922 byte executable's executable executable, the resulting code seems contrary to what I experienced. It wouldn't be the first time I've seen programs get off on the wrong thread. Hopefully, there's an exception handler somewhere to catch such unfortunate events and route them to their appropriate service routine. My only hope is that said routine has the ability to recognize and respond to such errant events. If not, Frank might be able to provide the necessary code. However, since he and I use different assemblers, sometimes things get lost in translation. I've had the same issue with Keith too, twice, but whoa, he uses NASM also. Maybe one day the *elusive dream* will be a reality, and all code will be logical, if not useful. Then again, maybe no-one really cares for such a reality, hence the apparent lack of interest in Menuet32, which is the closest I've seen so far. - Scott
From: hutch-- on 7 Sep 2007 10:48
Wannabee, > You really are scary stupid, Hutch-- Just shuddup and keep chewin. |