[PATCH] D60545: [DAGCombiner] narrow shuffle of concatenated vectors

[Sanjay Patel via Phabricator via llvm-commits]

spatel added a comment.

In D60545#1464207 <https://reviews.llvm.org/D60545#1464207>, @sdesmalen wrote:

> LGTM. The compiler will have more opportunity to use cheaper shuffles if part of the (sub)vector is undef, so this looks like a general improvement.


Thanks!

Re: code comment organization - I see your point, and I'll take that as something to be more aware of in general...I've been hacking away in here long enough that it's hard to remember the shock of seeing this mess for the 1st time. :)


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D60545/new/

https://reviews.llvm.org/D60545