On Mon, Jan 2, 2012 at 9:31 PM, Richard Henderson <r...@redhat.com> wrote:
> On 01/03/2012 06:47 AM, Uros Bizjak wrote:
>> if (d->testing_p)
>> return true;
>>
>> + hi = shift < nelt ? d->op1 : d->op0;
>> + lo = shift < nelt ? d->op0 : d->op1;
>> +
>> + shift %= nelt;
>
> This bit only works for little-endian. It's why I had that assert
> for shift range 1-63, for one thing.
>
> I guess an extra check for big-endian before the loop could fix that.
> I.e.
>
> shift = d->perm[0];
> + if (BYTES_BIG_ENDIAN && shift > nelt)
> + return false;
I tried to investigate -mbig-endian a bit, but unfortunately almost
all gcc.dg/torture/vshuf-*.c tests FAIL with -O2 on unpatched gcc.
Tests pass with -O0, though.
Due to this, I committed only the obvious part of the patch, that is:
2012-01-02 Uros Bizjak <ubiz...@gmail.com>
* config/ia64/ia64.c (expand_vec_perm_broadcast): Use correct
operands for extzv pattern.
Uros.
