[Mesa-dev] [PATCH 4/4] nv50: add PostRADualIssue Pass

[karol herbst]

2016-08-13 21:33 GMT+02:00 Ilia Mirkin <imirkin at alum.mit.edu>:
> On Sat, Aug 13, 2016 at 3:26 PM, Connor Abbott <cwabbott0 at gmail.com> wrote:
>> So, I don't know much about how nv50 ir works, but to me this just
>> seems like a pretty slow implementation of a very limited instruction
>> scheduler. In addition to the runtime complexity problems you
>> mentioned, you're going to get a lot more benefit even from a very
>> simple list scheduler compared to this, and it generally only takes a
>> few hundred lines to write one. I'd send you some references, but I
>
> I agree with basically all of the above. I really don't think it's
> worth having this super-partial approach. We need a real scheduler,
> both pre- and post-ra. Which follows a reasonable strategy.
>
> I don't think this "just try to dual issue as much as possible
> post-ra" attempt really gets at that. It's a neat hack, but ultimately
> I don't think it's worth having upstream. It's not something that
> we'll be able to build on and improve -- the underlying approach is
> too brute force.
>
>   -ilia

right, but that was the code I was testing with and it gave me good
results already. Just added it with this in mind, so others could also
try this.

I was actually toying around with implementing some scheduler stuff,
but nothing really got me anywhere, so I kind of stopped.

Anyway, the dual issue fixes are worth to include and that's the main
reason I actually sent it out