[PATCH] D75817: [NVPTX] Fix instruction selection for addresses in case of addrspacecasts

[Thomas Faingnaert via Phabricator via llvm-commits]

thomasfaingnaert abandoned this revision.
thomasfaingnaert added a comment.

I thought some more on this, and I don't think this is the best solution.
Besides, in most cases (such as the simple example posted above), the CUDA driver seems to do a pretty good job of folding away these extra additions anyway.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D75817/new/

https://reviews.llvm.org/D75817