TASM revisited
in [ASM]
|
Prev: opcode of mov instruction
Next: Free Disassembler
From: k_jh77 on 24 Oct 2006 20:02 Hello all! Just one quick survey. Suppose that I have a modified source code for Borland's Turbo Assembler 32bit version. Do you think that it would be beneficial to port it to for example to Linux or FreeBSD? Because that would be relatively simple. Just replace those ca. 20 operating system interface primitives in the assembler, and reassembling the whole package, maybe using FASM or NASM, or what have you. Textual conversion of sources would take some time (currently MASM syntax), but it's nothing special. Of course it still would produce an OMF modules for 80x86 architectures. But at least it would run on Linux and variants. If we don't consider TASM's current copyright situation (which belongs to Paradigm), what do you think in general? I think it's abandonware, YMMV. I have spent my last two years disassembling and reverse engineering this wonderful assembler, and now it is in a point where I have a pretty much full specs for it's inner workings, commented all it's ca. 2000 functions, it's data structures, the whole 75000 lines of assembly, and the byproduct is what some might call a source code. I'm just interested in general, what do you think about TASM? Would you use it if it was updated with latest instructions and if it works under Linux or something? br, KJH
From: randyhyde@earthlink.net on 24 Oct 2006 23:42 k_jh77(a)yahoo.com wrote: > Hello all! > > Just one quick survey. Suppose that I have a modified source code for > Borland's Turbo Assembler 32bit version. Do you think that it would be > beneficial to port it to for example to Linux or FreeBSD? > > Because that would be relatively simple. Just replace those ca. 20 > operating system interface primitives in the assembler, and > reassembling the whole package, maybe using FASM or NASM, or what have > you. Textual conversion of sources would take some time (currently MASM > syntax), but it's nothing special. Of course it still would produce an > OMF modules for 80x86 architectures. But at least it would run on Linux > and variants. > > If we don't consider TASM's current copyright situation (which belongs > to Paradigm), what do you think in general? I think it's abandonware, > YMMV. I have spent my last two years disassembling and reverse > engineering this wonderful assembler, and now it is in a point where I > have a pretty much full specs for it's inner workings, commented all > it's ca. 2000 functions, it's data structures, the whole 75000 lines of > assembly, and the byproduct is what some might call a source code. > > I'm just interested in general, what do you think about TASM? Would you > use it if it was updated with latest instructions and if it works under > Linux or something? Well, abandonware or otherwise, I wouldn't want to be the one posting that code. That's a lawsuit waiting to happen. :-( Yes, it would be cool to port it to *nix. But simply modifying a disassembled source listing is a sure copyright violation in every sense of the word. What would be *better* is for you to create specifications for a TASM-work-alike based on your reverse engineering and get a different team to implement those specs (i.e., clean-room reverse engineering). I don't know of there are enough people willing to put in that kind of effort around here (75KLOC is a *lot* of code to write, test, debug, and document), but you might try. The NASM project started up with less. Cheers, Randy Hyde
From: sevag.krikorian@gmail.com on 25 Oct 2006 02:17 k_jh77(a)yahoo.com wrote: > Hello all! > >. I have spent my last two years disassembling and reverse > engineering this wonderful assembler, and now it is in a point where I > have a pretty much full specs for it's inner workings, commented all > it's ca. 2000 functions, it's data structures, the whole 75000 lines of > assembly, and the byproduct is what some might call a source code. Wouldn't it have been easier to write a TASM compatible assembler from scratch or even from FASM sources (with Tomasz's permission of course)? > I'm just interested in general, what do you think about TASM? Would you > use it if it was updated with latest instructions and if it works under > Linux or something? > > br, > KJH I once floated the idea of a MASM compatible assembler that runs on 'nix but it wasn't received well by the 'nix crowd. I don't know if TASM would fare any better in that department. Besides, you'll have legal issues on your hands unless the copywrite is owned by a defunct company. I'm partial to FASM myself for its clean syntax and easy integration with HLA. -sevag.k www.geocities.com/kahlinor
From: Betov on 25 Oct 2006 02:51 k_jh77(a)yahoo.com ?crivait news:1161734563.132061.289640 @b28g2000cwb.googlegroups.com: > I think it's abandonware No, it is not. Go to FASM or NASM, instead. Betov. < http://rosasm.org >
From: johnzulu[at]yahoo.com on 25 Oct 2006 04:16
KJH, Do not distribute the source code or software when it is under a license! The lawyer firms will breath down your neck if they found out that you are "stealing" from their clients. Sorry to be harsh but that is the fact of life. I too DON'T dare go against them. Regards, John On 24 Oct 2006 17:02:43 -0700, k_jh77(a)yahoo.com wrote: >Hello all! > >Just one quick survey. Suppose that I have a modified source code for >Borland's Turbo Assembler 32bit version. Do you think that it would be >beneficial to port it to for example to Linux or FreeBSD? > >Because that would be relatively simple. Just replace those ca. 20 >operating system interface primitives in the assembler, and >reassembling the whole package, maybe using FASM or NASM, or what have >you. Textual conversion of sources would take some time (currently MASM >syntax), but it's nothing special. Of course it still would produce an >OMF modules for 80x86 architectures. But at least it would run on Linux >and variants. > >If we don't consider TASM's current copyright situation (which belongs >to Paradigm), what do you think in general? I think it's abandonware, >YMMV. I have spent my last two years disassembling and reverse >engineering this wonderful assembler, and now it is in a point where I >have a pretty much full specs for it's inner workings, commented all >it's ca. 2000 functions, it's data structures, the whole 75000 lines of >assembly, and the byproduct is what some might call a source code. > >I'm just interested in general, what do you think about TASM? Would you >use it if it was updated with latest instructions and if it works under >Linux or something? > >br, >KJH |