On 09/26/14 09:31, Ilya Tocar wrote:
It's not a question of performance, but of design.
Obviously, but I still fail to see why honoring HARD_REGNO_MODE_OK is
bad design. I suspect that even without avx512 changes not honoring it will
bite us sooner or later.
If we look at maybe_mode_change:
On 25 Sep 13:14, Jeff Law wrote:
> On 09/01/14 04:29, Ilya Tocar wrote:
> >>>
> >>>AVX512 added new 16 xmm registers (xmm16-xmm31).
> >>>Those registers require evex encoding.
> >>>Only 512-bit wide versions of instructions have evex encoding with
> >>>avx512f, but all versions have it with avx512v
On 09/01/14 04:29, Ilya Tocar wrote:
AVX512 added new 16 xmm registers (xmm16-xmm31).
Those registers require evex encoding.
Only 512-bit wide versions of instructions have evex encoding with
avx512f, but all versions have it with avx512vl.
Most instructions have same macroized pattern for 128/2
> >
> >AVX512 added new 16 xmm registers (xmm16-xmm31).
> >Those registers require evex encoding.
> >Only 512-bit wide versions of instructions have evex encoding with
> >avx512f, but all versions have it with avx512vl.
> >Most instructions have same macroized pattern for 128/256/512 vector
> >leng
On 08/14/14 07:16, Ilya Tocar wrote:
I've observed SPEC2006 failure on avx512-vlbwdq branch.
It was caused by hardreg_cprop. In maybe_mode_change it was
assumed, that all values of the same register class and same mode.
are ok. This is not the case for i386/avx512. We need to honor
HARD_REGNO_MO
> >I've observed SPEC2006 failure on avx512-vlbwdq branch.
> >It was caused by hardreg_cprop. In maybe_mode_change it was
> >assumed, that all values of the same register class and same mode.
> >are ok. This is not the case for i386/avx512. We need to honor
> >HARD_REGNO_MODE_OK.
> One could argu
On 08/11/14 07:49, Ilya Tocar wrote:
Hi,
I've observed SPEC2006 failure on avx512-vlbwdq branch.
It was caused by hardreg_cprop. In maybe_mode_change it was
assumed, that all values of the same register class and same mode.
are ok. This is not the case for i386/avx512. We need to honor
HARD_REG
Hi,
I've observed SPEC2006 failure on avx512-vlbwdq branch.
It was caused by hardreg_cprop. In maybe_mode_change it was
assumed, that all values of the same register class and same mode.
are ok. This is not the case for i386/avx512. We need to honor
HARD_REGNO_MODE_OK.
Patch bellow does it.
Ok fo
