[bug #1929] -Inf not detected

[avrdudes]

[bug #1929] -Inf not detected

[avrs-admin]

Anatoly Sokolov
Sun 05 Jun 2005 12:29:49 PM CEST

No this bug in gcc-4.0.0/avrlibc-1.2.3.
Test case work fine with all compare operation ( <, >, =>, <=).

volatile float foo;
volatile int res;

int
main (void)
{
foo = 3.402823466E+38;
foo = (foo * foo); /* foo = +Inf */

if(foo < 0.0)
{
res = -1;
} else {
res = 1;
}

foo = - foo; /* foo = -Inf */

if(foo < 0.0)
{
res = -1;
} else {
res = 1;
}

return (0);
}

[avrs-admin]

Dmitry Xmelkov
Mon 20 Feb 2006 11:47:29 PM CET

'fplib' is changed a little to power on IEEE 754 features.
Last acceleration by Peeter Vois (mulsf3) and numerous
improvements by Anatoly Sokolov are included also.
Try it by running a 'tst-all.sh' in tst/table directory.

[avrs-admin]

Dmitry Xmelkov
Fri 24 Feb 2006 01:27:53 AM CET

fplib-20060224: one error fixed, a bit of accuracy improvements.
Test list is expanded.

[avrs-admin]

Dmitry Xmelkov
Fri 24 Feb 2006 01:34:56 AM CET

To try the patch to avr-libc-1.4.3:

• untar a 'avr-libc-1.4.3'  tarball
• apply the patch
• make exutable: chmod a+x devtools/gen-avr-lib-tree.sh
• ./bootstrap
  Below all is standart:
• ./configure ...
• make && make install

[avrs-admin]

Dmitry Xmelkov
Sat 25 Feb 2006 11:35:27 PM CET

fplib-20060225: small inprovements

[avrs-admin]

Dmitry Xmelkov
Sun 05 Mar 2006 11:25:51 PM CET

20060305: a few of new functions (C99)

[avrs-admin]

Dmitry Xmelkov
Mon 01 May 2006 11:18:31 PM CEST

20060501: convertion 'long_long <--> float' is added.
Patch to avr-libc-1.4.4: operations with 'devtools' are not needed more.

[avrs-admin]

Dmitry Xmelkov
Wed 21 Mar 2007 02:31:08 PM CET

Fixed in CVS MAIN branch. The patch is implemented.