NAME
docs/dev/byteorder.pod - Byteorder Conversion Functions
Overview
The platform byteorder is stored for C code in include/parrot/config.h
#define PARROT_BYTEORDER 0x1234
for parrot code in
$P0 = _config() $P0["byteorder"]
or for perl code via
use Parrot::Config; $PConfig{byteorder};
The byteorder values are analog to perl, see "pack" in perlfunc:
1234 little-endian 32-bit, 12345678 little-endian 64-bit 4321 big-endian 32-bit, 87654321 big-endian 64-bit
When reading a pbc stored in a different architecture, the pbc header defines the pbc byteorder for the architecture which stored the pbc, and the src/packfile/pf_items.c functions are used to convert the values to the native endianness, wordsize and ptrsize.
The byteorder code will check the endianness of an INTVAL
or an opcode_t
value and swap from little to big, or big to little when appropriate. Functions also exist to convert a 4, 8, 12, or 16 byte character buffer to big or little endian. The functions are be placed in the PackFile vtable and are be called when necessary. The Parrot interpreter is be smart enough to avoid calling these functions when converting from and to the same byteorder.
Data Structures and Algorithms
The algorithm to change from one endianness to another is identical and simple to understand. Basically, the size of an INTVAL
or opcode_t
is used to determine at compile time how many bits should be shifted around. Then the correct bits are shifted by the correct amounts (please look at source code for exact amounts). The buffer change functions are implemented by a straight forward algorithm that assigns swaps all of the bytes.
Important Functions
fetch_iv_le
This function will convert an fetch_iv_be
This function will convert an fetch_op_le
This function will convert an fetch_op_be
This function will convert an fetch_buf_le_
(4,8,12,16)
This set of functions will convert an unsigned character buffer into little endian format. Only a fetch_buf_be_
(4,8,12,16)
This set of functions will convert an unsigned character buffer into big endian format. Only a
INTVAL
into little endian format. It is a no-op if the native format is already little endian.
INTVAL
into big endian format. It is a no-op if the native format is already big endian.
opcode_t
into little endian format. It is a no-op if the native format is already little endian.
opcode_t
into big endian format. It is a no-op if the native format is already big endian.
memcpy
is performed if the native format is already little endian.
memcpy
is performed if the native format is already big endian.Low level FLOATVAL fetch and convert functions
We support two different floattypes, stored in the pbc header as 0 or 1.
Floattype 0 = IEEE-754 8 byte double Floattype 1 = x86 little endian 12 byte long double
cvt_num12_num8
Converts i386 LE 12-byte long double to IEEE 754 8 byte double.
cvt_num12_num8_be
Converts a 12-byte i386 long double into a big-endian IEEE 754 8-byte double. Converting to BE is not yet implemented (throws internal_exception).
cvt_num12_num8_le
Converts a 12-byte i386 long double into a little-endian IEEE 754 8-byte double.
Unimplemented Functions
endianize_fetch_int
Fetch an endianize_put_int
Put an
INTVAL
directly from a bytestream
INTVAL
directly on a bytestreamHistory
Initial version by Melvin on 2002-01-05, more byteorder explanations by Reini Urban 2009-02-02
Notes
This assumes big or little endianness...other, more esoteric forms (such as middle endian) are not supported. Also, an assumption of 4 or 8 byte INTVAL
's and opcode_t
's is made.
References
The fetch and transformer functions are implemented in src/packfile/pf_items.c