Thanks! This was committed to trunk last week as r247671. As we discussed
offline, I've also backported to GCC 7 (r248010) and GCC 6 (r248011).
Bill
> On May 5, 2017, at 10:30 AM, Segher Boessenkool
> wrote:
>
> Hi Bill,
>
> On Wed, May 03, 2017 at 02:43:09PM -0500, Bill Schmidt wrote:
>> W
Hi Bill,
On Wed, May 03, 2017 at 02:43:09PM -0500, Bill Schmidt wrote:
> We recently became aware of some poor code generation as a result of
> unprofitable (for POWER) loop vectorization. When a loop is simply copying
> data with 64-bit loads and stores, vectorizing with 128-bit loads and stores
...only without the typo in the ChangeLog below...
> On May 3, 2017, at 2:43 PM, Bill Schmidt wrote:
>
> Hi,
>
> We recently became aware of some poor code generation as a result of
> unprofitable (for POWER) loop vectorization. When a loop is simply copying
> data with 64-bit loads and stores
Hi,
We recently became aware of some poor code generation as a result of
unprofitable (for POWER) loop vectorization. When a loop is simply copying
data with 64-bit loads and stores, vectorizing with 128-bit loads and stores
generally does not provide any benefit on modern POWER processors.
Furth
