>question: if the IDE does not compile but PBC does, can i trust the results?
The answer is yes.
I had exact same problem a couple of years ago with a large (by MS-DOS standards application, and the PBC-compiled version worked just ducky for many years.
MCM
Announcement
Collapse
No announcement yet.
The Elusive error 515 once again
Collapse
X
-
The Elusive error 515 once again
hello people,
i have a program that consists of a little over 13000 lines of code
(and i'm not finished yet, although close). when i compile in the IDE
i get the 515 (fixup overflow) error and the cursor jumps to the last
line of the code. when i compile using PBC everything works fine.
when running the program i have about 90Kb free, that should be enough.
the program has 7 segments that are 60k max. i've shuffled the segment
statements around, placed more etc. but that didn' have any effect.
the one thing that helped was moving 1 tiny little sub in segment 1
to the end of the program code. after that the IDE compiled fine.
you're probably wondering why i'm posting all this, since i have already
solved my problem? the thing is, this problem has surfaced before and
sofar i've not seen any satisfying solutions. it always seems to end with
the coder messing around a bit and then hey presto, error 515 is gone.
my program is going to be commercially available soon and i need to know that
i can trust the compiler. it's little things like this that make it hard
to do so. it also makes me think if the PB designers ever created test
programs that exceeded 450 Kb, as that seems to be the a sort of threshold.
now, if i know for sure that PB can't reliably handle more than 10000
lines of code, fine. i can keep that in mind and adapt my coding.
now, just to be clear: i'm not pointing fingers here. i know that at least
one person at PB gets annoyed when the word bug is used so i'm steering
clear of that. maybe i just need to be reassured.
question: if the IDE does not compile but PBC does, can i trust the results?
o, and before you ask: i'm not willing to send my code to anyone.
i'll include my meta's for reference :
$COMPILE EXE
$CPU 80386
$DEBUG MAP OFF
$DEBUG PBDEBUG OFF
$DEBUG PATH OFF
$DIM ALL
$STACK 4096
$STRING 2
$ERROR ALL ON
$EVENT ON
$LIB ALL OFF
$OPTIMIZE SIZE
$OPTION CNTLBREAK ON
$OPTION GOSUB ON
$INCLUDE "PBTCLASS.DEC" - look familiar to anyone? funstuff
$INCLUDE "EXT13H.INC"
when using error all off the IDE compiles the code okay, but at this stage
in the development i need error to be all on.
looking forward to comments,
with regards,
Tom
------------------
[URL=http://www.DiyDataRecovery.nl]Tags: None
Leave a comment: