source: rtems/c/src/lib/libbsp/m68k/idp/README @ df25c998

4.104.114.84.95
Last change on this file since df25c998 was eb5a7e07, checked in by Joel Sherrill <joel.sherrill@…>, on Oct 6, 1995 at 8:48:38 PM

fixed missing CVS IDs

  • Property mode set to 100644
File size: 1.5 KB
Line 
1#
2#  $Id$
3#
4
5This board support package has not been tested with multiprocessor
6or the timing support.  The uniprocessor IDP package was tested though
7with a fairly large application (although bugs may exist).  The
8$RTEMS_ROOT/src/tests/Makefile should therefore have the samples and
9the tests directories compiled first with the tmtests (timing) and
10multiprocessor directories optional.
11
12Let me know if you have any problems or bug fixes.  Bug fixes are greatly
13appreciated.  I do not work for RTEMS or am a member of the RTEMS support
14group in any way, however.  I am just one of many that appreciate
15"free" software and enjoy contributing when possible :).  You can do it too!
16 
17                -- doug mcbride
18                   mcbride@rodin.colorado.edu
19
20==============================================================
21Notes: Make sure that -msoft-float is defined when compiling in the
22newlib-beta-rtems/newlib/msoft-float directory subtree.  You also probably
23want to add the following line to line 413 of
24newlib-beta-rtems/newlib/msoft-float/libc/stdio/vfprintf.c:
25
26#define INTEGER_ONLY
27
28That allows you to pass most of the paranoia test in the samples
29directory of RTEMS although you can't see the floating point values actually
30printed (software floating point tends to make that difficult anyway).
31In order to pass the whole paranoia test (with one flaw), however, I had to
32comment out the following line in milestone 140 (why?):
33
34/*  printf ("Testing X^((X + 1) / (X - 1)) vs. exp(2) = %.17e as X -> 1.\n",
35    Exp2); */
Note: See TracBrowser for help on using the repository browser.