Infuriating vasm error

[DanielAllsopp]

Evening each!

I've got a strange error using vasm, which is happening more and and more often at the minute.

The following simple instruction:

add.b #1,d0

Outputs the following error:

error 9 in line 122 of "main.s": instruction not supported on selected architecture
>  add.b #1,d0

Build flags are: -m68020 -Fhunk
Version is:

vasm 1.8h (c) in 2002-2020 Volker Barthelmann
vasm M68k/CPU32/ColdFire cpu backend 2.3k (c) 2002-2020 Frank Wille
vasm motorola syntax module 3.14 (c) 2002-2020 Frank Wille
vasm hunk format output module 2.12 (c) 2002-2020 Frank Wille

Any got any ideas as to why this would be? I'm about to go

[Cyprian]

for me works ok

vasm 1.8k (c) in 2002-2021 Volker Barthelmann
vasm M68k/CPU32/ColdFire cpu backend 2.3o (c) 2002-2021 Frank Wille
vasm motorola syntax module 3.14c (c) 2002-2020 Frank Wille
vasm tos output module 1.1 (c) 2009-2016,2020 Frank Wille

text(acrx2):               4 bytes

vasm 1.8k (c) in 2002-2021 Volker Barthelmann
vasm M68k/CPU32/ColdFire cpu backend 2.3o (c) 2002-2021 Frank Wille
vasm motorola syntax module 3.14c (c) 2002-2020 Frank Wille
vasm hunk format output module 2.13 (c) 2002-2020 Frank Wille

CODE(acrx2):               4 bytes

I would suggest to you update VASM to the latest version

[gigabates]

It could be due to a non-ascii character, especially if you've copy/pasted the code from somewhere. I get the same error if I put a non-breaking space in place of a regular space between the instruction and the operands. Try deleting and re-typing all whitespace.

[DanielAllsopp]

vasm 1.8m (c) in 2002-2021 Volker Barthelmann
vasm M68k/CPU32/ColdFire cpu backend 2.4d (c) 2002-2021 Frank Wille
vasm motorola syntax module 3.15b (c) 2002-2021 Frank Wille
vasm hunk format output module 2.13a (c) 2002-2021 Frank Wille

This is the latest nightly source code and it's producing the same error!

[DanielAllsopp]

Originally Posted by gigabates

It could be due to a non-ascii character, especially if you've copy/pasted the code from somewhere. I get the same error if I put a non-breaking space in place of a regular space between the instruction and the operands. Try deleting and re-typing all whitespace.

Thank you so much! That seems to have been the issue! Jesus wept, I nearly through my iMac through the window earlier.

Life saver

[gigabates]

No problem. I've encountered this kind of issue before and it's absolutely maddening! What appears to be the exact same code typed by someone else compiles/assembles just fine!

On a related note, there was a security exploit reported recently that uses hidden characters to insert malicious code that's effectively invisible to human code reviewers.

[Tigerskunk]

Damn, that's a horrible bug to have.

[roondar]

Yeah, I've seen similar things happen with several languages/compilers/assemblers. Really annoying.

For those who use Notepad++ to code, you can find such characters using Search ==> Find characters in range ==> Non-ASCII. Quite useful
Other editors might also have similar options.

[DanielAllsopp]

Originally Posted by roondar

Yeah, I've seen similar things happen with several languages/compilers/assemblers. Really annoying.

For those who use Notepad++ to code, you can find such characters using Search ==> Find characters in range ==> Non-ASCII. Quite useful
Other editors might also have similar options.

That's a very good point roondar... I've just found this for the Atom editor if anyone else uses it:

https://atom.io/packages/highlight-bad-chars

Seems to be a more common problem than I thought!

[coldacid]

If I'm not mistaken, support to identify BiDi directional characters and other such Unicode miscreants has been added to VSCode as of late and enabled by default.