MSC100ABIRM Motorola / Freescale Semiconductor, MSC100ABIRM Datasheet - Page 38

• Manufacturer Part Number: MSC100ABIRM
• Description: SC100 Application Binary Interface Reference Manual
• Manufacturer: Motorola / Freescale Semiconductor

SC100 ELF Object File Format
The final field in the sequence, described in Example 4-6, is the value field. The value field, after it is
evaluated, determines the contents of the memory location. The other fields of the relocation may set
attributes of the memory location and may do general or architecture-based checking as shown in
Example 4-7.
Example 4-7. Relocation Attribute Setting
The relocation in Example 4-7 performs a check of symbol number 5 to make sure that its signed value
will fit into a bit field 8 bits wide. The memory location value is calculated by bit-wise ORing the section
data (represented by the pre-defined identifier “op”) and the value generated by the architecture-dependent
function, “F5.”
4.3.4.2 Functions
Functions are categorized as generic or architecture-specific. Within the architecture-specific category,
functions may be further categorized as encoding or other.
Following are descriptions of the functions used for relocations.
4.3.4.2.1 Check Function
Category:
Arguments:
Description:
Returns:
4.3.4.2.2 Pack Function
Category:
Arguments:
Description:
Returns:
4-6
“check(sym(5),s,8), op|F5(sym(5))”
generic
value, (s | u), <len> [,<align>=0]
where:
s=signed range check
u=unsigned range check
This function checks that the value passed will fit into the number of bits specified
by <len>. The second argument specifies the type of value, signed or unsigned
integer. The optional parameter, <align>, specifies the number of constant least
significant bits in value. These constant bits are disregarded when calculating
whether the value will fit into <len> bits. The <align> argument, if omitted,
defaults to 0 (no alignment).
0, generates error is value is out of range.
generic
value, (s | u | n), <len> [,<align>=0]
where:
s=signed range check
u=unsigned range check
n=truncate to range
This function is similar to the check function. It returns the significant bits which
can be placed in a <len> size bit field. If the second argument is ‘n’, the value is
truncated to the number of <len> bits and no error is generated.
<len> significant bits of value, generates error is value is out of range (s or u) or
truncated value if second argument is ‘n’.
Preliminary (April 2000)
SC100 Application Binary Interface