Hi, On Tue, Oct 8, 2019 at 11:48 AM TK Chia <[email protected]> wrote: > > I did a quick disassembly of the precompiled nasm.exe and ndisasm.exe in > nsm09839.zip, and indeed --- for some reason --- they were built to use > 186+ instructions such as `leave' and immediate pushes. > > So perhaps it will be good to try to rebuild this version of nasm, for > the 8086.
(TINYASM) First, let me mention a few things I've noticed about Tinyasm: BTW, what is the proper way to build Tinyasm with IA16-GCC? Is there a specific set of switches you use? Or have you not tested (much) with it yet? Tinyasm seems to work okay (although line number / error reporting is a bit backwards). It still lacks a few things, but I know it's still very early in development. I'm no expert, so please forgive my misuse of terminology. But it's missing "[var+BX]" Effective Address support. Also, "MOV AX,..." (etc.) instructions use the bigger, non-optimal encodings. I haven't quite identified some other quirks yet (beyond simple stuff like hex number syntax [0Ah not supported], quirky "mylabel: dw 5", seg overrides only at beginning of line, etc). I'm not opposed to uploading Tinyasm to iBiblio, but since I still want to get PSR Invaders working with it, I haven't quite done it yet. (I may or may not be able to mentally wrangle around the quirks with Sed scripting. I'm only half successful so far. Yes, I know Oscar has his own smaller invaders clone, but the difficulty is much harder. I still regret that the old one is VGA-only, but apparently he doesn't!) I've done still more tweaking (indirectly) to PSR Invaders in recent months, so I'll probably update the FD Package "yet again" for us soon. (NASM) Now, let me mention NASM (old 0.98.39 from 2005, from which we have a pre-existing [broken?] 16-bit build): I did rebuild it with both Turbo C++ 1.01 and OpenWatcom 1.9. I found that TC++ (mkfiles\makefile.bor) was harder to get working (and it didn't fit into Large model, so I switched to Huge). For both, I only included outputs "bin" and "obj", which I think is reasonable. They do work, but I haven't tested under 8086tinyplus (again) yet. I'll try doing that tomorrow. The makefiles are messy, and I'm no make expert, but they should both be 100% reproducible (if anyone is honestly interested). If I don't find any obvious problems, I'll probably upload my weak efforts to iBiblio for us. Hopefully someone else can then further verify and improve upon my attempt. Obligatory link: https://www.stevemorse.org/8086/index.html _______________________________________________ Freedos-user mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/freedos-user
