Re: Why are allocatable indices in main different from those in subroutine? (1:n) to (0:(n-1))
- From: "dpb" <dpbozarth@xxxxxxxx>
- Date: 28 Jul 2006 06:54:39 -0700
David Frank wrote:
You probably have discovered a bug in your YET AGAIN unidentified compiler
which appears to be
allocating the x array with a 0: base for some reason thats not apparent..
He did (a couple of times previously and in a followup here)...it's CVF
6.6 (but whether he's applied (c) hasn't been stated.
I didn't use this test on 6.6c here, but couldn't get the previous idea
to work correctly (passing back an allocated subroutine array to upper
level) in CVF 6.6c here, either.
I've not really tried to stretch the TR addition in CVF previously so
hadn't uncovered a but previously, but Mike apparently has (and not too
difficult, either). What is/isn't actually allowed under the TR I'm
not positive enough about to know where the limits are of what
can/can't be done.
To Mike, looks like you've run into a dead-end here--unfortunately, CVF
is a dead-end product and will not see any more bug fixes even from HP.
Seems like your option at this pint is to either go to one of the
non-TR workarounds or find a supported compiler, sorry.
.
- Follow-Ups:
- References:
- Prev by Date: Re: Logicals converted to integers during reading
- Next by Date: Re: Why are allocatable indices in main different from those in subroutine? (1:n) to (0:(n-1))
- Previous by thread: Re: Why are allocatable indices in main different from those in subroutine? (1:n) to (0:(n-1))
- Next by thread: Re: Why are allocatable indices in main different from those in subroutine? (1:n) to (0:(n-1))
- Index(es):
Relevant Pages
|
