View previous topic :: View next topic |
Author |
Message |
Zrblm
Joined: 21 Jun 2005 Posts: 2
|
Posted: Tue Jun 21, 2005 12:27 pm Post subject: ftn95 handling of IO units? |
|
|
Does FTN95 handle IO units in some unusual way? I'm testing the compiler by attempting to port over some code written using Sun's f95, and I'm getting a result that doesn't make sense to me. Specifically, I have an IO unit open for writing to an output file:
open(unit=iounit2,file=tracename,status='unknown')
write (*,*) 'Output file opened'
and then when I write to standard output further down in the code, it ends up in my output file instead. Am I doing something subtly wrong? |
|
Back to top |
|
|
PaulLaidler Site Admin
Joined: 21 Feb 2005 Posts: 7924 Location: Salford, UK
|
Posted: Wed Jun 22, 2005 12:30 am Post subject: ftn95 handling of IO units? |
|
|
The value of iounit2 must not be 2 or 6.
For simplicity user values for IO units should be greater than 6. |
|
Back to top |
|
|
Zrblm
Joined: 21 Jun 2005 Posts: 2
|
Posted: Wed Jun 22, 2005 10:12 am Post subject: ftn95 handling of IO units? |
|
|
Aha, that'd be why. Reserved unit numbers seem to be compiler-dependent. Thanks! |
|
Back to top |
|
|
|