[PATCH] D106292: [flang] Implement the runtime portion of the CSHIFT intrinsic

[Pete Steinfeld via Phabricator via llvm-commits]

PeteSteinfeld created this revision.
PeteSteinfeld requested review of this revision.
Herald added a project: LLVM.
Herald added a subscriber: llvm-commits.

This change implements the runtime portion of the CSHIFT intrinsic.  There are
two aspects to it.

In the process of implementing lowering for CSHIFT, I realized that always
treating the SHIFT argument as an array simplified the lowering implementation.

Also, I fixed a bug in the vector version of CSHIFT that happens when the value
of the SHIFT argument is negative.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D106292

Files:
  flang/runtime/transformational.cpp
  flang/runtime/transformational.h
  flang/unittests/RuntimeGTest/Transformational.cpp

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D106292.359853.patch
Type: text/x-patch
Size: 4434 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20210719/3a7183f1/attachment.bin>