2.2.72 NumericValue
NumericValue is any numerical value, whether integer or real, that is valid within the CIM type system.
-
NumericValue = BYTE / SINT16 / UINT16 / SINT32 / UINT32 / SINT64 / UINT64 / REAL32 / REAL64
For each of these types, the binary encoding rules are specified in the following table.
The CIM model defines standard numerical data types, as specified in [DMTF-DSP0004] section 2.2.
|
CIM type as specified in [DMTF-DSP0004] |
ABNF representation |
Binary representation |
|---|---|---|
|
uint8 |
OCTET, BYTE |
An 8-bit unsigned integer. |
|
sint8 |
OCTET, BYTE |
An 8-bit signed integer. |
|
uint16 |
UINT16 |
A 16-bit unsigned integer. |
|
sint16 |
SINT16 |
A 16-bit signed integer. |
|
uint32 |
UINT32 |
A 32-bit unsigned integer. |
|
sint32 |
SINT32 |
A 32-bit signed integer. |
|
real32 |
REAL32 |
A 4-byte floating-point format, as specified in [IEEE754]. |
|
real64 |
REAL64 |
An 8-byte floating-point format, as specified in [IEEE754]. |
|
sint64 |
SINT64 |
A 64-bit signed integer. |
|
uint64 |
UINT64 |
A 64-bit unsigned integer. |
The binary representations MUST be used to encode the specified CIM data types.
All signed and unsigned integer types that consist of more than one octet MUST be encoded as little-endian.
The CIM Boolean type has its own encoding that is specified in the BOOL (section 2.2.75) encoding rule.
Floating point values, as specified in [IEEE754], MUST be encoded as little-endian.