- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Using IVF 8.1, and was converting a CVF program that used USE DFLIB, but USE IFLIB produces an error. When I search the Include files I find DFLIB but not IFLIB. I noticed in past forum posts thatreferences are made to IFLIB so I assume it exists.
Thanks for any help
Link Copied
9 Replies
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
try:
use ifport
hans
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Thank you
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
IFPORT may or may not be an appropriate replacement. You can continue to use DFLIB or look to see if the routines you're using are in IFCORE or IFQWIN instead.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Thanks for info. Tried ifport and it did not work in my particular case, so I will look at IFCORE or IFQWIN.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
If you get errors, look up the routine name in the Intel Fortran Libraries Reference and make note of the module name listed. Symbols that were in DFLIB in CVF have been split among IFCORE, IFQWIN and IFPORT.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
I am trying to compile a code using `dn4qbj' and `dneqbf' in the dflib by Intel compiler (ifort) on Fedora, but I can not find how to get it work. Could any one advise me any module equivalent to dflib or any way to get arround this problem. Thanks.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Those are IMSL routines, not in DFLIB or IFLIB. For Linux, you will need to purchase IMSL from Visual Numerics. I have not looked to see what those routines do, but you might find something similar in the Intel Math Kernel Library, which is included with Intel Fortran. Ask in the MKL section of the forum if you want advice on that.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
To add to what Steve Lionel wrote in #7:
The IMSL routines that you cited are for (i) computing the Jacobian using finite difference approximations, which is itself a subproblem of (ii) solving a set of nonlinear equations in the least-squares sense.
Recent versions of MKL provide jacobi_solve() and ?trnlsp_solve(). These routines correspond in functionality to the IMSL routines. The argument lists, of course are different as are the routines for set-up, optional parameters setting and printing intermediate results.
MINPACK in Netlib provides open-source alternatives.
The IMSL routines that you cited are for (i) computing the Jacobian using finite difference approximations, which is itself a subproblem of (ii) solving a set of nonlinear equations in the least-squares sense.
Recent versions of MKL provide jacobi_solve() and ?trnlsp_solve(). These routines correspond in functionality to the IMSL routines. The argument lists, of course are different as are the routines for set-up, optional parameters setting and printing intermediate results.
MINPACK in Netlib provides open-source alternatives.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Thank all for the information.
Reply
Topic Options
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page