Changes between Version 22 and Version 23 of GcnOperands
- Timestamp:
- 12/29/17 09:00:33 (6 years ago)
Legend:
- Unmodified
- Added
- Removed
- Modified
-
GcnOperands
v22 v23 296 296 </table> 297 297 <h3>Operand syntax</h3> 298 <p> THeSingle operands can be given by their name: <code>s0</code>, <code>v54</code>.298 <p>Single operands can be given by their name: <code>s0</code>, <code>v54</code>. 299 299 CLRX assembler accepts the syntax with 300 300 brackets: <code>s[0]</code>, <code>s[z]</code>, <code>v[66]</code>. In many instructions operands are … … 303 303 last register's number.</p> 304 304 <p>The names of the registers are case-insensitive.</p> 305 <p> The constant values are automatically resolved if an expression have already value.305 <p>Constant values are automatically resolved if an expression have already value. 306 306 The 1/(2*PI), 1.0, -2.0 and other floating point constant values will be 307 307 resolved if that accurate floating point value will be given.</p> 308 308 <p>In instruction syntax, operands are listed by name of the encoding field. Optionally, in 309 parentheses is given number of the registers. The ranges of number of aregisters are in309 parentheses is given number of the registers. The ranges of number of registers are in 310 310 form 'START:LAST'. Example:</p> 311 311 <p>Syntax: S_SUB_I32 SDST, SSRC0, SSRC1<br />