elf_xlate(3E)

elf_xlate: elf32_xlatetof, elf32_xlatetom,elf64_xlatetof, elf64_xlatetom - class-dependent data translation

As shipped in IRIX 6.5. First release of IRIX 6.5.

NAME
     elf_xlate: elf32_xlatetof, elf32_xlatetom,elf64_xlatetof, elf64_xlatetom
     - class-dependent data translation

SYNOPSIS
     cc [flag ...] file ...  -lelf [library ...]

     #include <libelf.h>
     Elf_Data *elf32_xlatetof(Elf_Data *dst, const Elf_Data *src,
          unsigned encode);
     Elf_Data *elf32_xlatetom(Elf_Data *dst, const Elf_Data *src,
          unsigned encode);
     Elf_Data *elf64_xlatetof(Elf_Data *dst, const Elf_Data *src,
          unsigned encode);
     Elf_Data *elf64_xlatetom(Elf_Data *dst, const Elf_Data *src,
          unsigned encode);

DESCRIPTION
     elf32_xlatetom translates various data structures from their 32-bit class
     file representations to their memory representations; elf32_xlatetof
     provides the inverse.  This conversion is particularly important for
     cross development environments.  src is a pointer to the source buffer
     that holds the original data; dst is a pointer to a destination buffer
     that will hold the translated copy.  encode gives the byte encoding in
     which the file objects are (to be) represented and must have one of the
     encoding values defined for the ELF header's e_ident[EI_DATA] entry [see
     elf_getident(3E)].  If the data can be translated, the functions return
     dst.  Otherwise, they return null because an error occurred, such as
     incompatible types, destination buffer overflow, etc.

     elf_getdata(3E) describes the Elf_Data descriptor, which the translation
     routines use as follows.

     d_buf         Both the source and destination must have valid buffer
                   pointers.

     d_type        This member's value specifies the type of the data to which
                   d_buf points and the type of data to be created in the
                   destination.  The program supplies a d_type value in the
                   source; the library sets the destination's d_type to the
                   same value.  These values are summarized below.

     d_size        This member holds the total size, in bytes, of the memory
                   occupied by the source data and the size allocated for the
                   destination data.  If the destination buffer is not large
                   enough, the routines do not change its original contents.
                   The translation routines reset the destination's d_size
                   member to the actual size required, after the translation
                   occurs.  The source and destination sizes may differ.
     d_version     This member holds version number of the objects (desired)
                   in the buffer.  The source and destination versions are
                   independent.

     Translation routines allow the source and destination buffers to
     coincide.  That is, dst->d_buf may equal src->d_buf.  Other cases where
     the source and destination buffers overlap give undefined behavior.

                          Elf_Type     32-Bit Memory Type
                         ________________________________
                         ELF_T_ADDR    Elf32_Addr
                         ELF_T_BYTE    unsigned char
                         ELF_T_DYN     Elf32_Dyn
                         ELF_T_EHDR    Elf32_Ehdr
                         ELF_T_HALF    Elf32_Half
                         ELT_T_OFF     Elf32_Off
                         ELF_T_PHDR    Elf32_Phdr
                         ELF_T_REL     Elf32_Rel
                         ELF_T_RELA    Elf32_Rela
                         ELF_T_SHDR    Elf32_Shdr
                         ELF_T_SWORD   Elf32_Sword
                         ELF_T_SYM     Elf32_Sym
                         ELF_T_WORD    Elf32_Word
                         ________________________________
                                     |















     ``Translating'' buffers of type ELF_T_BYTE does not change the byte
     order.

     The 64-bit class works identically, simply replacing all instances of 32
     in the description and table with 64.

SEE ALSO
     elf(3E), elf_fsize(3E), elf_getdata(3E), elf_getident(3E).

CAVEAT
     By default, the Elf_Data used for the 64-bit class in a 32-bit
     application restricts one to building 64-bit class applications with only
     32-bit fields unless the entire application is compiled with
     _LIBELF_XTND_64 defined (see elf.3e).