I hadn't realised that on big-endian targets, VEC_UNPACK*HI_EXPR unpacks
the low-numbered lanes and VEC_UNPACK*LO_EXPR unpacks the high-numbered
lanes. This meant that both the SVE patterns and the handling of
fully-masked loops were wrong.
The patch deals with that by making sure that all vec_un
On Fri, Jan 26, 2018 at 3:21 PM, Richard Sandiford
wrote:
> I hadn't realised that on big-endian targets, VEC_UNPACK*HI_EXPR unpacks
> the low-numbered lanes and VEC_UNPACK*LO_EXPR unpacks the high-numbered
> lanes. This meant that both the SVE patterns and the handling of
> fully-masked loops we
Richard Biener writes:
> On Fri, Jan 26, 2018 at 3:21 PM, Richard Sandiford
> wrote:
>> I hadn't realised that on big-endian targets, VEC_UNPACK*HI_EXPR unpacks
>> the low-numbered lanes and VEC_UNPACK*LO_EXPR unpacks the high-numbered
>> lanes. This meant that both the SVE patterns and the hand
On Fri, Jan 26, 2018 at 4:04 PM, Richard Sandiford
wrote:
> Richard Biener writes:
>> On Fri, Jan 26, 2018 at 3:21 PM, Richard Sandiford
>> wrote:
>>> I hadn't realised that on big-endian targets, VEC_UNPACK*HI_EXPR unpacks
>>> the low-numbered lanes and VEC_UNPACK*LO_EXPR unpacks the high-numbe
On Fri, Jan 26, 2018 at 02:21:11PM +, Richard Sandiford wrote:
> I hadn't realised that on big-endian targets, VEC_UNPACK*HI_EXPR unpacks
> the low-numbered lanes and VEC_UNPACK*LO_EXPR unpacks the high-numbered
> lanes. This meant that both the SVE patterns and the handling of
> fully-masked