[sdcc-devel] [ sdcc-Bugs-827117 ] problem with passing pointer as parameter

[SourceForge.net <no...@so...>]

Bugs item #827117, was opened at 2003-10-20 15:31
Message generated for change (Settings changed) made by epetrich
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100599&aid=827117&group_id=599

Category: z80 port
>Group: unreproducable
>Status: Closed
>Resolution: Rejected
Priority: 5
Submitted By: Philipp Krause (spth)
Assigned to: Erik Petrich (epetrich)
Summary: problem with passing pointer as parameter

Initial Comment:
The function call in line 162, calling the function at
line 132 doesn't work as expected. The value of t
remains unchanged. It seems like the first function
parameter is not passed correctly.


----------------------------------------------------------------------

Comment By: Philipp Krause (spth)
Date: 2003-10-25 06:59

Message:
Logged In: YES 
user_id=564030

It seems it was a problem with the emulator I use, not sdcc.
Sorry.


----------------------------------------------------------------------

Comment By: brian witt (wittb)
Date: 2003-10-21 15:25

Message:
Logged In: YES 
user_id=592116

I'm a bozo....
please ignore my comment : 2003-10-21 20:19
(in fact delete it if you can! back to lurking...)

lines 1382-83 do indeed store the value.
Hmmm... stack corruption, then maybe?


----------------------------------------------------------------------

Comment By: brian witt (wittb)
Date: 2003-10-21 15:19

Message:
Logged In: YES 
user_id=592116

WRT Eriks' comment, it appears to be pushing the integer
value "t" and not the address of "t" ...  The code should:
	LD  HL,#-10
	ADD HL, IX
	PUSH HL
to compute the address.  I don't see where -10(IX) is being
loaded with the address of "t".  Lines 1891-92 of assem code
read the value from -10(IX) ; not a store.  (Feel free to
rebuff me ; it's my first comment.)

Phillip: before line 162 print "t" and the addr of "t", then
in your function print "dest" and see how they relate.

*brian


----------------------------------------------------------------------

Comment By: Erik Petrich (epetrich)
Date: 2003-10-20 23:17

Message:
Logged In: YES 
user_id=635249

This is a rather complex function, so I could be overlooking
something, but I can't seem to find a problem. Local
variable t is at offset -8(IX) or equivilently 26(SP), since
SP=IX-34 by the function entry code. In line 151, &t is
evaluated correctly and a copy of this value is kept in
-9(IX) and -10(IX). The value in -9(IX) and -10(IX) is
reused when &t is referenced in line 162.

Please help point me in the right direction.


----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100599&aid=827117&group_id=599