[PATCH] D43970: [InstCombine] Allow fptrunc (fpext X)) to be reduced to a single fpext/ftrunc

[Craig Topper via Phabricator via llvm-commits]

craig.topper updated this revision to Diff 136617.
craig.topper retitled this revision from "[InstCombine] Allow fptrunc (fpext X)) to be reduced to a single fpext if the trunc result type is larger than the original size" to "[InstCombine] Allow fptrunc (fpext X)) to be reduced to a single fpext/ftrunc".
craig.topper added a comment.

Allow it to be turned into fptrunc as well. Which I believe makes it equivalent to the handler for integers.


https://reviews.llvm.org/D43970

Files:
  lib/IR/Instructions.cpp
  test/Transforms/InstCombine/double-float-shrink-2.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D43970.136617.patch
Type: text/x-patch
Size: 4705 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20180301/6e2c2b91/attachment.bin>