SHASM Floating Point Data Blocks
.FDATAB is the assembler directive that reserves the specified number of floating-point data consecutively in memory. The block count specification must be an absolute value, and forward reference symbols must not appear in the specification. Arbitrary values, including relative values and forward reference symbols, can be used to specify the floating-point data.
The range of values that can be specified as the block size and as the floating-point data varies with the operation size. Block size limits are theoretical. The maximum values that can be successfully used are smaller and depend on the amount of memory available.
The following table shows the operation size specified by block size range.
| Operation Size | Block Size Range |
|---|---|
| S | H'00000001 to H'3FFFFFFF |
| D | H'00000001 to H'1FFFFFFF |
The following table shows the operation size specified by floating point range.
| Type | Single-Precision | Double Precision |
|---|---|---|
| sNaN | 7FFFFFFF to 7FC00000 | 7FFFFFFF FFFFFFFF to 7FF80000 00000000 |
| qNaN | 7FBFFFFF to 7F800001 | 7FF7FFFF FFFFFFFF to 7FF00000 00000001 |
| +Infinity | 7F800000 | 7FF00000 00000000 |
| +Normalized | 7F7FFFFF to 00800000 | 7FEFFFFF FFFFFFFF to 00100000 00000000 |
| +Denormalized | 007FFFFF to 00000001 | 000FFFFF FFFFFFFF to 00000000 00000001 |
| +Zero | 00000000 | 00000000 00000000 |
| -Zero | 80000000 | 80000000 00000000 |
| -Denormalized | 80000001 to 807FFFFF | 80000000 00000001 to 800FFFFF FFFFFFFF |
| -Infinity | FF800000 | FFF00000 00000000 |
| qNaN | FF800001 to FFBFFFFF | FFF00000 00000001 to FFF7FFFF FFFFFFFF |
| sNaN | FFC00000 to FFFFFFFF | FFF90000 00000000 to FFFFFFFF FFFFFFFF |
See Also
SHASM Data and Data Area Directives | SHASM Floating Point Data
Last updated on Thursday, April 08, 2004
© 1992-2003 Microsoft Corporation. All rights reserved.