NASM - The Netwide Assembler
NASM Forum => Using NASM => Topic started by: STF92 on July 12, 2011, 09:59:59 AM
-
nasm 2.03.01
Hi:
I have noticed that when nasm finds errors in the source file, it does not output the listing file, regardless of the presence of the -l option in the command line. Is this normal behavior?
-
Normal for 2.03.01, yes. Try a newer version:
http://www.nasm.us/pub/nasm/releasebuilds/?C=M;O=D
Best,
Frank
-
Thank you very much.
EDIT:
I am running nasm 2.09.09 but still no listing in the presence of errors.
-
Just tried explicitly 2.09.09 - works for me. Lower case 'l', of course! What error are you encountering? Possible it doesn't work for all of 'em?
"parser: instruction expected" does generate the list file, but does not write the error to the listing file as expected, so it may make a difference...
Best,
Frank
-
Ah! "unrecognised special symbol '..start'" does not generate a .lst file! Solution is to make a different error! :)
Best,
Frank
-
Thank you for your post. Of course I used -l <filename>. The data you asked for:
Linux kernel 2.6.21.5, Slackware 12.0 (mostly or all GNU). Spin I guess is the version.
However, I found this, assembling with 2.09.09 (may be 2.03.01, the previous version I had, has identical behavior): After having fixed the last two errors displayed, a new compilation threw a lot more errors. Quite understandable for a one pass assembler. But, alas, a listing file was now output. This is quite another thing. But the listing does not show any errors.
As I was writing a new post arrived: Yes, I had a lot of "parser: instruction expected" but no listing. And I used lower 'L' (-l <filename).
-
Sorry you're having this problem! Seems like you've discovered a "bug", or more properly an "incompletely implemented feature". We'll look into it (that is, the active developers will look into it... I hope. :) ). Might be a while... Thanks for the feedback!
Best,
Frank
-
Thanks to you for your kind posts.
-
guys, if you found a bug or unimplemented feature :D please file a bug, this would allow us to track it and not loose
-
Thanks, Mahmoud! Sorry, Cyrill. I was "gonna" do it (honest), but thought I'd try a few more errors first. Nasm64developer seems to have some insight into it...
Best,
Frank