| 摘要 |
A compiler capable of facilitating calculation description in a source program and simplifying the source program description so as to reduce generation of bugs in calculation according to the block floating method by software. When a source program is given with description of calculation formula using a block floating variable, a calculation formula performing addition is extracted from the source program. A portion related to an addition in the extracted calculation formula is expanded as an instruction code constituting the processing performing addition using a data block corresponding to a plurality of block floating variables to be added. The expanded instruction code is embedded in an object program, thus creating and outputting an object program.
|
