OpenE2K
/
gcc
2
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

* config/m32c/m32c.h (PCC_BITFIELD_TYPE_MATTERS): Define to zero. 

From-SVN: r143981
This commit is contained in:
Nick Clifton 2009-02-06 09:02:37 +00:00 committed by Nick Clifton
parent 851691140a
commit 8506528ca2
2 changed files with 21 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
gcc/ChangeLog
View File

@ -1,3 +1,8 @@
2009-02-06 Nick Clifton <nickc@redhat.com>
* config/m32c/m32c.h (PCC_BITFIELD_TYPE_MATTERS): Define to
zero.
2009-02-06 Paolo Bonzini <bonzini@gnu.org>
PR tree-optimization/35659

16
gcc/config/m32c/m32c.h
View File

@ -143,6 +143,17 @@ machine_function;
#define UNITS_PER_WORD 2
#define POINTER_SIZE (TARGET_A16 ? 16 : 32)
#define POINTERS_EXTEND_UNSIGNED 1
/* We have a problem with libgcc2. It only defines two versions of
each function, one for "int" and one for "long long". Ie it assumes
that "sizeof (int) == sizeof (long)". For the M32C this is not true
and we need a third set of functions. We explicitly define
LIBGCC2_UNITS_PER_WORD here so that it is clear that we are expecting
to get the SI and DI versions from the libgcc2.c sources, and we
provide our own set of HI functions in m32c-lib2.c, which is why this
definition is surrounded by #ifndef..#endif. */
#ifndef LIBGCC2_UNITS_PER_WORD
#define LIBGCC2_UNITS_PER_WORD 4
#endif
/* These match the alignment enforced by the two types of stack operations. */
#define PARM_BOUNDARY (TARGET_A16 ? 8 : 16)
@ -154,6 +165,11 @@ machine_function;
#define FUNCTION_BOUNDARY 8
#define BIGGEST_ALIGNMENT 8
/* Since we have a maximum structure alignment of 8 there
is no need to enforce any alignment of bitfield types. */
#undef PCC_BITFIELD_TYPE_MATTERS
#define PCC_BITFIELD_TYPE_MATTERS 0
#define STRICT_ALIGNMENT 0
#define SLOW_BYTE_ACCESS 1