C/C++ help

[Zero102]

Stupid question, but I'm in the midst of translating some C code to assembler, and I'm stuck on one line.
return bufp > 0 ? buf[--bufp] : getchar();

Does this roughly translate to
if (bufp > 0)
buf[--bufp] = getchar();
return buf[--bufp]
??
I just don't want to translate it wrong, but I always get messed up with these stupid question marks in the code.

If it's of any consequence, bufp is an int, and bufp is an array of char's.

[rc2002]

Basically that just says:

if (bufp> 0)
return buf[--bufp];
else
return getchar();

What assembly language are you using?

[Zero102]

Ahhh, good thing I asked, I did mis-read it.

I'm programming in SPARC assembly.

It's for a CPSC 355 assignment. That's one of the last bits holding me up. I seem to have a typo in another function that's giving me a bit of grief tracking it down.

[Zero102]

Well, I thought I had a typo, but perhaps I'm just mis-reading something again.

return val[sp++] = f;
is that the same as:
sp++;
val[sp] = f;
return val[sp];

??
Or is that more like
return val[sp];
val[sp++] = f;
?? (I don't think so...)

Or, the third choice:
val[sp] = f;
sp++;
return f;

While I'm asking, return val[--sp] does?....
return val[sp]
sp--
?
sp--;
return val[sp];
?

I did it the third way... perhaps that's my problem?

[D'z Nutz]

Originally posted by Zero102

return val[sp++] = f;

It's essentially:

val[sp] = f;, which is what is returned
sp += 1;

So your third choice would be the closest.

While I'm asking, return val[--sp] does?....
return val[sp]
sp--
?
sp--;
return val[sp];
?

sp -= 1;
return val[sp];

So your second one is the closest. Don't get your pre/post-increment/decrement's mixed up. It can cause you some grief if you dont' know the difference

[Zero102]

Would explain the stack over/underflows I've been getting.
I was doing both the first way.
I'll fix that up now.

I know assembler, but not C, lol.
Although I do know java pretty well, but I know it well enough to stay the hell away from pre/post increment/decrement's. I just actually add the number, lol.
It all reduces to roughly the same machine code anyways, but my way is 10x easier to read to the untrained reader.

Well, I'll fix this up, and if my program still is buggered up, I'll be back here with the next thing I may have mis-interpreted.

[Zero102]

Ugh, bloody seg faults now..... if you guys know anything about SPARC assembly, you might be as confused as I am....

mov 1, %o1
set sp, %o0
st [%o0], %o1

and I get a seg fault....
sp is an integer.

yet, this works just fine
set sp, %o0
ld [%o0], %o1

I can touch the same section of memory with no seg faults....

I can even take it as far as...


str: .asciz "%d\n"

....
....

set sp, %o0
ld [%o0], %o1
set str, %o0
call printf
nop

And I'll get the correct value for sp, but for some reason, I can't set the value of it....