Latest Win32 VASM Build?

[guy lateur]

Originally Posted by Leffmann

EDIT: attached GNU Make 4.1, vbcc 0.9f, vasm 1.8, and vlink 0.16

Thanks again for the binaries, Leffmann, vasm an vlink work perfectly. However, I'm having problems running vbccm68k.exe. I've made a separate thread about this here: http://eab.abime.net/showthread.php?t=87965

Anybody else experiencing this issue?

[clenched]

phx - If you get a minute could you test this out on the 8-bit oldstyle modules? Check the tail end of a.out. PC is where the data would be loaded. Whenever PC (not offset into file) would surpass $8001 the pad bytes are truncated.

Change PC to 0 and the whole file is correct. Change PC to $8000 and only the first byte is correct. Z80 could assemble this as long as $ffff is never passed.

pc  set    $4000
    rept   128
    org    pc
pc  set    pc+$100
    byte   $d6
    endr
    end

[phx]

Thanks for reporting. Another sign-problem in the binary-output. Fixed it. Updated snapshot is available tomorrow.

BTW, this is off topic and I also have an email address.

[dalton]

Originally Posted by Leffmann

The makefile works fine with VS 2017 for both Alkis and myself. Otherwise, the solution needs the Windows 8.1 SDK and the Windows XP platform toolset, maybe you need to install those first, or you can edit the solution to ditch the resource scripts and use the regular Windows 10 toolset.

I also have an installer in this thread: http://eab.abime.net/showthread.php?t=83113 and if you don't need the whole thing then you can open the installer in 7-zip and just extract the executables you need.

EDIT: attached GNU Make 4.1, vbcc 0.9f, vasm 1.8, and vlink 0.16

Thanks for uploading binaries again! Really saves me a lot of trouble.

[jotd]

BTW found a minor issue. Do this and you get a "branch out of range":

      bra.b	.open
.open:

not that it is very useful but $6000 is still in range

[Toni Wilen]

It is not out of range, it does not exist

6000 xxxx -> BRA.W XXXX
60xx where xx != 00 -> BRA.S xx (and if 68020+ xx=FF -> BRA.L)

[jotd]

yeah ok trying to play smartass (no pun intended ). I mean $6002.

[phx]

You disabled optimizations. Otherwise vasm would have told you this:

warning 2058 in line 1 of "tst.s": short-branch to following instruction turned into a nop
>       bra.b   .open

Without the possibility to change your code it can only respond with an error. "out of range" may sound strange, but the following address is also kind of "out of range" for a short-branch.

[Toni Wilen]

Originally Posted by jotd

yeah ok trying to play smartass (no pun intended ). I mean $6002.

$6002 would have been:

bra.s .label
dc.b something,anything
.label

[NorthWay]

I never really looked at the encoding of the instructions, but this is kinda weird... I would have expected the offset to be shifted right in the encoding so as to point to something word aligned.
So the 68000 has 128+32768 BRA encodings that result in an exception? And 1 was redefined for the 68020 for long branches?
That sounds very wasteful. Unless they were reserved for future expansions?