[PATCH] D95139: [SVE][LoopVectorize] Add support for extracting the last lane of a scalable vector

[Florian Hahn via Phabricator via llvm-commits]

fhahn added inline comments.


================
Comment at: llvm/test/Transforms/LoopVectorize/AArch64/sve-extract-last-veclane.ll:25
+
+for.cond.for.cond.cleanup_crit_edge:              ; preds = %for.body
+  %mul.lcssa = phi i32 [ %mul, %for.body ]
----------------
david-arm wrote:
> fhahn wrote:
> > nit: the names of the blocks could be improved.
> OK, to be honest I don't really know what they should be called. :) This is the name that LLVM generates. How about `for.cond.pre-cleanup`?
how about something like just `exit`. It can also directly return; you don't need the `%mul.lcssa` phi I think, LV will insert them if needed.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D95139/new/

https://reviews.llvm.org/D95139