[Nasm-devel] Re: nasm - debugging

[Frank K. <fbk...@at...>]

Ralf Eigl wrote:
> Cc: <ei...@us...>
> 
> Message-Id: <E18...@sc...>
> Date: Sun, 30 Mar 2003 14:13:26 -0800
> 
> 
> hello,
> 
> 
> 
> my name is ralf eigl and i'm studying computer science
> 
> in germany-esslingen. at the moment i'm writing a student
> 
> research project about assembly-programming under linux.
> 
> there i have to choose a assembly for the students which 
> 
> the have to use. i would like to suggest nasm. but my
> 
> lecturer needs debugging, especially symbolic debugging.
> 
> i tried to debug with the ddd (gnu-dbg) but i didn't work.
> 
> is there any other debug-tool, which works under linux ?
> 
> have i done something wrong ? will symbolic debugging
> 
> implemented in the next versions ? what can you suggest 
> 
> me ?
> 
> 
> 
> it would be great if you could send me a small feedback.
> 
> thanks in advance .

Hello Ralf,

Besides gdb, there's Patrick Alken's ALD - unfortunately the link I've 
got for it seems to be down, at the moment.

http://ellipse.mcs.drexel.edu/ald.html

Martin Wawro has written a patch to add debugging info to elf output:

http://ls7-www.cs.uni-dortmund.de/~wawro/nasmstabs.html

Attached is my attempt to bring his patch up to 0.98.36, and eliminate 
the "#IFDEF"s and stuff - an amateur work, beware! It does seem to make 
gdb a little friendlier. I'm not familiar enough with gdb's "normal" 
behavior to judge if it's "right" or not.

I don't know what version of Nasm it might appear in. If you can 
evaluate it, and find it's worthwhile, or improve it so it *is*, I 
suppose it could be cleaned up and go in "soon".

There may be problems with it. Something I just noticed that worries me 
a little (not part of Martin's patch):

    elf_sects = nasm_malloc(sizeof(*elf_sects) * (2 * nsects + 10));

What's the 10 for? Does that assume there will be 5 sections? If we 
increase by the debugging sections, is that still 10? If something like 
that doesn't bite us, I guess it's okay.

Best,
Frank