Syntax
cvtst $s_reg, $d_reg
cvtst $d_reg/$s_reg
Description
Convert S_floating to T_floating converts the S_floating contents of $s_reg to the T_floating precision. No rounding occurs and the result is exact. No underflow, overflow, or inexact result can occur. In fact, the conversion for finite values is the identity transformation.
Qualifiers
The following table describes the qualifiers for the cvtst instructions:
| Instruction | Qualifier description |
| no qualifier | Exact result, no traps. |
| Cvtsts | IEEE Trap Mode: A trap handler can convert an S_floating denormal value int the corresponding T_floating finite value by adding 896 to the exponent and normalizing. |