http://dev.starlink.ac.uk/bugzilla/show_bug.cgi?id=42
Summary: Converting usercode to 64bit Starlink
Product: DOCS
Version: unspecified
Platform: PC
OS/Version: Linux
Status: NEW
Severity: normal
Priority: P2
Component: DOCS
AssignedTo: [log in to unmask]
ReportedBy: [log in to unmask]
QAContact: [log in to unmask]
Not sure what category to file this under. We have some self-written code in the
SuperWASP which uses the IMG and PAR libraries to do various things. I have just
tried compiling and running it a 64bit Starlink program and it segfaults. I
racked it down to a subroutine call and it segfaulted as soon the subroutine
touched one of the passed arrays:
call acc_vignet( xaxis,yaxis,nx,ny,%val(imgptr),qrad,dfac,
: %val(rdiptr),%val(rdpptr),%val(pptr),%val(igdptr),
: %val(wdgptr),%val(dimptr),%val(outptr) )
Presumably it's the %val's that are causing the problem for the same reasons as
the FIGARO tasks have been segfaulting.
So the question is are there any docs in SC12 or elsewhere that say what users
need to do to convert the codes so they work properly with Starlink on 64bit
machines ? I found Mark's docs
(http://www.star.bris.ac.uk/~mbt/docs/ssn73.htx/ssn73.html) on EXTREME which
seems to dicuss this sort of thing. Is this document still applicable and useful
for dumb endusers or just smart programmers ?
While I'm at, I have a general question about the 64bit conversions. How come
lots of the 64bit conversions are being done now ? Hasn't Starlink been running
successfully on 64bit Alphas for years ?
Many thanks,
Tim
------- You are receiving this mail because: -------
You are the QA contact for the bug, or are watching the QA contact.
|