CHANGES BETWEEN VERSIONS 0.00 AND 1.00:

-1. Added support for printing NaN-s and infinities.


CHANGES BETWEEN VERSIONS 1.00 AND 1.01:

0.  Added this file.
1.  Make-target "naninf" changed to "check-naninf". Comments about "make check-naninf" added at top of Makefile.
2.  Comments added at the top of test-all.sh explaining its purpose and use.
3.  Corrected small bug in test-all.sh involving a wrongly placed comment.
4.  Changed version number to 1.01 in dispmodule.f90.
5.  To facilitate changes, the real kinds and matrix dimensions in dispdemo.f90 are now parameters.
6.  Changed Makefile, test-all.sh and test_dispmodule_fpp.f90 to test quad precision (use target check-quadprec).
7.  make-package now includes changelog.txt in tgz file, and makes a copy in the file dispmodule.vX.XX.tgz.
8.  The file portability-evidence.txt now consists of output from test-all.sh on four platforms.
9.  User manual updated to reflect addition of nan and inf displaying and testing, and quad precision testing.
10. README file updated to reflect these changes.
11. Added expected-demo-output-txt to package (it was missing by mistake).

CHANGES BETWEEN VERSIONS 1.01 AND 1.02:

12. Corrected error in replace_zero_naninf that displayed nans/infs wrongly when using zeroas = '0.0'
13. Added test to test_naninf_mod to check change number 12.
14. Changed tar file to have a directory with all the files of the package.
15. Created a version compatible with the Absoft af95 compiler version 10.1 (see file README)
