This is the mail archive of the gcc-bugs@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]

Re: Failure of gcc.c-torture/execute/991014-1.c on irix6.5 with -mabi =64


"Billinghurst, David (CRTS)" <David.Billinghurst@riotinto.com> writes:

> Compilation of gcc.c-torture/execute/991014-1.c fails on irix6.5 with
> -mabi=64.   I think this test should be an XFAIL for any 32 bit -> 64 bit
> cross compile.
> 
> This test case can be reduced to
> 
> typedef typeof(sizeof(char)) Size_t;
> #define bufsize ((1L << (8 * sizeof(Size_t) - 2))-256)
> struct huge_struct
> {
>   short buf[bufsize];
> };
> 
> which fails with "size of array `buf' is too large"
> 
> This seems reasonable as bufsize is:
>   * 1073741568 for default -n32 ABI
>   * 4611686018427387648 for -64 ABI
> and we are doing a 32 bit -> 64 bit cross compile
> 
> Thoughts?

I think this is just about to be fixed, with the HOST_WIDE_INT patch,
so xfailing it would be premature.

-- 
- Geoffrey Keating <geoffk@geoffk.org>


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]