Hi Guys, I am checking in the patch below (to both the mainline and the 4.9 branch) in order to fix some unexpected failures in the gcc testsuite with the msp430-elf toolchain. With this patch in place the following tests now pass:
gcc.c-torture/execute/builtins/memcpy-chk.c
gcc.c-torture/execute/builtins/memmove.c
gcc.c-torture/execute/builtins/mempcpy-chk.c
gcc.c-torture/execute/builtins/mempcpy.c
gcc.c-torture/execute/builtins/strcpy-chk.c
gcc.c-torture/execute/builtins/strcpy.c
gcc.c-torture/execute/20020412-1.c
gcc.c-torture/execute/20050826-2.c
gcc.c-torture/execute/20050826-2.c
gcc.c-torture/execute/920908-1.c
gcc.c-torture/execute/pr56205.c
gcc.c-torture/execute/strct-varg-1.c
gcc.c-torture/execute/va-arg-22.c
No regressions are introduced by the patch.
The patch fixes two bugs - the first was that the zero_extendpsisi2
pattern was confusing gcc about the number of hard registers needed to
hold an SImode value.
The second fixes a problem with expanding va_arg when pointers are not
a multiple of the word size.
Cheers
Nick
gcc/ChangeLog
2014-05-20 Nick Clifton <ni...@redhat.com>
* config/msp430/msp430.c (TARGET_GIMPLIFY_VA_ARG_EXPR): Define.
(msp430_gimplify_va_arg_expr): New function.
(msp430_print_operand): Handle (CONST (ZERO_EXTRACT)).
* config/msp430/msp430.md (zero_extendpsisi2): Use + constraint on
operand 0 in order to prevent confusion about the number of
registers involved.
msp430.patch.xz
Description: application/xz
