Re: [Freedos-user] Desmet C 3 archived and put on Github

[Rugxulo]

Hi again,

On Thu, Nov 9, 2017 at 9:34 PM, Random Liegh via Freedos-user
 wrote:
>
> I think I have figured out what goes where, though I'm not sure.

I've only scratched the surface of reading the manual (which is 400+ pages).

> I've only done a quick "hello world" test, but it looks like it works **if
> you stay in the same directory.**

The manual implies that it can also find auxiliary binaries (ASM88,
GEN) in your %PATH%.

> Doing "c88 -I " didn't seem to work.

You can set %DSINC% or %INCLUDE% as well. BTW, it seems options come
after the filename, e.g. "C88 myfile.c -Ic:\whatever\" (with mandatory
backslash).

So you have to use "-b" (and BBIND) for Large model.

(Not to mention a bunch of other utils and separate libs [*.S] for
software and hardware floating point.)

> Anyway, here's desmet c with a tiny install (just the programs), a complete
> install (with sources etc) and the 10 meg pdf manual:
>
> May work, may not ...but at least it should be easier to test!

The main question is how much is bootstrappable. Do we also need MASM
for anything? Can the libs and utils be reliably rebuilt, byte-exact,
to their existing versions? And we should probably focus on "latest"
3.1N with as many ANSI headers as are available (11?).

So, for the main binaries: C88, ASM88, GEN, BIND and BBIND, LIB88 (but
ignore D88, SEE, TOOBJ, PCMAKE, LATER, MERGE, etc. for now).

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Freedos-user mailing list
Freedos-user@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/freedos-user

Re: [Freedos-user] confirmation and compiler

[Rugxulo]

Hi,

On Mon, Nov 13, 2017 at 6:51 PM, Random Liegh via Freedos-user
 wrote:
>
> I appreciate that! I wonder if that also happens with the exes/zipfiles that
> come from desmet-c.com too?

Oops, you misunderstood me. I was only talking about SmallerC (since
DeSmet doesn't include PE/Win32 .EXEs) from his Github. I'm thinking
about making a package for that compiler as well (one of these days
...) since it too has DOS support.

> I appreciate knowing what's going on. Github complained when I tried to
> upload the source and other files, but I can try using a traditional git
> client instead.

It's none of that, just that anything atypical in PE files is usually
(falsely) flagged by bad heuristics (guesses).

> I haven't really used this compiler myself (I believe I found it either from
> osdev.org or a thread on hacker news; but I don't remember) so I'm not sure
> if it's self-hosting or not. I assume it is.

DeSmet? Presumably yes. (But SmallerC definitely is ... although I
haven't tried lately.)

> Any way it goes, I'll take a shot at figuring out how to compile the source
> but it won't be tomorrow. After I have that worked out, I'll upload a Proper
> git tree vi linux.

I didn't mean you personally had to do it, nor that it had to be soon.
No pressure! FreeDOS is very hands off.

Just in general, it's hard to keep things organized, and it's hard to
be 100% airtight with licenses. And of course, practically speaking,
rebuilding everything is better overall.

Not counting the .PDF manual, your archive is 377 files (but some of
those are nested .ZIPs, so it's actually more). Looks like roughly 500
files (with some duplication of LIB8831H.ZIP). Not sure about
NEWBIND.EXE and Large model or whatever.

So yeah, this needs further cleanup and clarification. I haven't read
the manual yet, though, so who knows what is contains.

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Freedos-user mailing list
Freedos-user@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/freedos-user