[LLVMdev] X86 LowerVECTOR_SHUFFLE Question
David A. Greene
greened at obbligato.org
Mon Feb 28 09:59:38 PST 2011
Bruno Cardoso Lopes <bruno.cardoso at gmail.com> writes:
>> There's certainly been improvement on the TableGen side of things. I
>> really liked the unpck*, shufp, etc. nodes and the ShuffleVectorSDNode.
>> That's a huge help. It's too bad we're getting rid of them. But
>> legalization still looks about the same to me.
>
> The idea is to use tablegen again once we have a clean implementation.
Ah, good. Having a clearer idea of the roadmap helps.
> It would be good to have all tables and per-processor logic in a
> tablegen file, but I think a clean implementation needs to be done
> first.
Do you mean a clean implementation of lowering?
> Anyway, I don't have time to work in any of this at the moment, and
> I'm not also the one with final word on this. I'm happy to help with
> the bugs that may appear and with some design ideas though.
Understandable. Since I'm working in this area now I'll do what I can
to clean things up. Do let me know if I start heading off in the wrong
direction. :)
-Dave
More information about the llvm-dev
mailing list