Hi Mike,
on 2023/11/28 12:34, Michael Meissner wrote:
> On Fri, Nov 24, 2023 at 05:31:20PM +0800, Kewen.Lin wrote:
>> Hi Ajit,
>>
>> Don't forget to CC David (CC-ed) :), some comments are inlined below.
>>
>> on 2023/10/8 03:04, Ajit Agarwal wrote:
>>> Hello All:
>>>
>>> This patch add new pass to replace contiguous addresses vector load lxv
>>> with mma instruction
>>> lxvp.
>>
>> IMHO the current binding lxvp (and lxvpx, stxvp{x,}) to MMA looks wrong,
>> it's only
>> Power10 and VSX required, these instructions should perform well without MMA
>> support.
>> So one patch to separate their support from MMA seems to go first.
>
> I tend to agree with you, but I recall the decision being made because at the
> time, vector pairs and vector quads were only used with MMA. We now have
> various attempts to improve things for using vector pairs for non-MMA code. In
Thanks for the comments! Yeah, so this time seems a good timing to make it
separated
from MMA support.
> my patches, I keeped the MMA requirement, but if we decide to make it ISA 3.1
> only if is fairly straight forward to look at all of the TARGET_MMA tests.
>
> Now in the GCC 13 days, it was useful that -mmma controlled vector pair.
> There
> was an issue if we enabled memcpy to use store vector pair, it would lead to
> one slow down. When I was doing the tests, it was easy to use -mno-mma and it
> would stop memcpy from using load/store vector pair since GCC doesn't generate
> code to use MMA without using the built-ins.
OK, maybe a vector store pair specific option can be added for the disablement
need.
BR,
Kewen
