| 1 | ?RCS: $Id$ |
| 2 | ?RCS: |
| 3 | ?RCS: Copyright (c) 2000 Jarkko Hietaniemi |
| 4 | ?RCS: |
| 5 | ?RCS: You may distribute under the terms of either the GNU General Public |
| 6 | ?RCS: License or the Artistic License, as specified in the README file. |
| 7 | ?RCS: |
| 8 | ?MAKE:d_fpclass: Inlibc |
| 9 | ?MAKE: -pick add $@ %< |
| 10 | ?S:d_fpclass: |
| 11 | ?S: This variable conditionally defines the HAS_FPCLASS symbol, which |
| 12 | ?S: indicates to the C program that the fpclass() routine is available. |
| 13 | ?S:. |
| 14 | ?C:HAS_FPCLASS: |
| 15 | ?C: This symbol, if defined, indicates that the fpclass routine is |
| 16 | ?C: available to classify doubles. Available for example in Solaris/SVR4. |
| 17 | ?C: The returned values are defined in <ieeefp.h> and are: |
| 18 | ?C: |
| 19 | ?C: FP_SNAN signaling NaN |
| 20 | ?C: FP_QNAN quiet NaN |
| 21 | ?C: FP_NINF negative infinity |
| 22 | ?C: FP_PINF positive infinity |
| 23 | ?C: FP_NDENORM negative denormalized non-zero |
| 24 | ?C: FP_PDENORM positive denormalized non-zero |
| 25 | ?C: FP_NZERO negative zero |
| 26 | ?C: FP_PZERO positive zero |
| 27 | ?C: FP_NNORM negative normalized non-zero |
| 28 | ?C: FP_PNORM positive normalized non-zero |
| 29 | ?C:. |
| 30 | ?H:#$d_fpclass HAS_FPCLASS /**/ |
| 31 | ?H:. |
| 32 | ?LINT:set d_fpclass |
| 33 | : see if fpclass exists |
| 34 | set fpclass d_fpclass |
| 35 | eval $inlibc |
| 36 | |