7.13 Comparison of condition code meanings in integer and floating-point code
The meaning of the condition code mnemonic suffixes depends on whether the condition flags were set by a floating-point instruction or by an A32 or T32 data processing instruction.
This is because:
- Floating-point values are never unsigned, so the unsigned conditions are not required.
- Not-a-Number (NaN) values have no ordering relationship with numbers or with each other, so additional conditions are required to account for unordered results.
The meaning of the condition code mnemonic suffixes is shown in the following table:
Table 7-3 Condition codes
|Suffix||Meaning after integer data processing instruction||Meaning after floating-point instruction|
|Not equal||Not equal, or unordered|
|Carry set||Greater than or equal, or unordered|
|Unsigned higher or same||Greater than or equal, or unordered|
|Carry clear||Less than|
|Unsigned lower||Less than|
|Positive or zero||Greater than or equal, or unordered|
|Overflow||Unordered (at least one NaN operand)|
|No overflow||Not unordered|
|Unsigned higher||Greater than, or unordered|
|Unsigned lower or same||Less than or equal|
|Signed greater than or equal||Greater than or equal|
|Signed less than||Less than, or unordered|
|Signed greater than||Greater than|
|Signed less than or equal||Less than or equal, or unordered|
|Always (normally omitted)||Always (normally omitted)|