Changes between Version 17 and Version 18 of GcnOperands


Ignore:
Timestamp:
Jun 6, 2017, 12:00:29 PM (21 months ago)
Author:
trac
Comment:

--

Legend:

Unmodified
Added
Removed
Modified
  • GcnOperands

    v17 v18  
    283283(both  integer and floating points) and second is 32-bit immediate. Some type encoding
    284284allow to supply immediate with various size (16-bit or 12-bit).</p>
    285 <p>The literals are differently treated for scalar instructions and for vector instructions.
    286 In scalar instructions if operand is 64-bit, the literal value is exact value 64-bit value
    287 (sign or zero extended). By contrast, in vector instructions, for 64-bit operand, the
    288 literal is higher 32-bits of value (lower 32-bit are zero). Unhapilly, the CLRX assembler
    289 always encodes and decodes literal immediate as 32-bit value (except floating values).
     285<p>The literals are differently treated for scalar/vector instructions and for
     286double floating point operands in vector instructions.
     287In scalar or vector instructions if operand is 64-bit, the literal value is exact value
     28864-bit value (sign or zero extended). By contrast, in FP64 operands in vector instructions,
     289for 64-bit operand, the literal is higher 32-bits of value (lower 32-bit are zero). Unhapilly, the CLRX assembler always encodes and decodes literal immediate as 32-bit
     290value (except floating values).
    290291The immediate constants are always exact value, either for 32-bit and 64-bit operands.
    291292For example, instructions <code>v_frexp_exp_i32_f64 v3, lit(45)</code> and