Bug#173074: problems with uint8 on powerpc

[David Paul BELANGER <dbelan2@CS.McGill.CA>]

Package: libffi2
Version: 1:3.2.1-0pre3
Severity: normal

ffi_type_uint8 and other arguments shorter than one word are
not passed in correctly to the function called by ffi_call.

Also, returns values are not passed correctly.  For example, instead of
finding the returned byte value where the pointer points to, it is found
at an offset of 3 from the pointer.

This is related to the fact that the powerpc is a big endian machine and
somewhere, bytes need to be moved.

I have included code and output I get on my powerpc machine.

Thanks,
David

-----

#include <stdio.h>
#include <ffi.h>


unsigned char identity(unsigned char c) {
  printf("identity: %d\n", c);
  return c;
}


int main(void) {
  ffi_cif cif;
  ffi_type *args[1];
  void *values[1];
  int b;
  unsigned char result;
  int i;

  args[0] = &ffi_type_uint;
  /*
   * args[0] = &ffi_type_uint8;  - if this type is specified,
   * function always receives 0.
   */


  b = 24;
  values[0] = &b;


  if (ffi_prep_cif(&cif, FFI_DEFAULT_ABI, 1,
		   &ffi_type_uint8, args) != FFI_OK) {
    printf("error\n");
    return 1;
  }


  ffi_call(&cif, identity, &result, values);

  printf("at result: %d\n", result);
  printf("at result + 3: %d\n", *(&result + 3));

  /*
  for (i = 0; i < sizeof(result); i++) {
    printf("byte %d = %d\n", i, *((unsigned char *) &result));
  }
  */

  return 0;
}

-----
output:

identity: 24
at result: 0
at result + 3: 24

-----


David


-- System Information
Debian Release: testing/unstable
Architecture: powerpc
Kernel: Linux ibou 2.4.20-rc1-ben0 #3 Thu Nov 14 22:24:15 EST 2002 ppc
Locale: LANG=C, LC_CTYPE=C

Versions of packages libffi2 depends on:
ii  gcc-3.2-base               1:3.2.1-0pre3 The GNU Compiler Collection (base 
ii  libc6                      2.2.5-14.3    GNU C Library: Shared libraries an