[PATCH] D84877: Support for soft fp16 to fp64 IEEE conversions

[Luke Drummond via Phabricator via llvm-commits]

ldrumm created this revision.
ldrumm added reviewers: ab, joerg, howard.hinnant.
Herald added subscribers: Sanitizers, mgorny.
Herald added projects: Sanitizers, LLVM.
ldrumm requested review of this revision.

Support for soft fp16 to fp64 IEEE conversions

      

This patch adds initial support for lowering conversions from `half` to
`double` with single rounding.

      

No backends have been updated to use this builtin yet.

I'm particularly interested whether the exhaustive testcases make sense for non-identical bittpatterns for NaN values


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D84877

Files:
  compiler-rt/lib/builtins/CMakeLists.txt
  compiler-rt/lib/builtins/extendhfdf2.c
  compiler-rt/test/builtins/Unit/extendhfdf2_test.c
  compiler-rt/test/builtins/Unit/extendhfdf2_testcases.inc
  llvm/include/llvm/IR/RuntimeLibcalls.def