projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: ddb8e06) | patch
[APInt] Use MathExtras.h BitsToFloat/Double and Float/DoubleToBits instead of type...
authorCraig Topper <craig.topper@gmail.com>
Fri, 12 May 2017 16:28:21 +0000 (16:28 +0000)
committerCraig Topper <craig.topper@gmail.com>
Fri, 12 May 2017 16:28:21 +0000 (16:28 +0000)
commit51416caf343a56754c06f64843eda12698f1a91a
treee03317b73830f3387d3bec17cbc43b76b5599281tree | snapshot
parentddb8e06a8e5cf4964d1f87e51ba5541235dd6a48commit | diff
[APInt] Use MathExtras.h BitsToFloat/Double and Float/DoubleToBits instead of type punning through a union

The functions in MathExtras.h uses a safer memcpy instead of going through a union.

Differential Revision: https://reviews.llvm.org/D33116

llvm-svn: 302916
llvm/include/llvm/ADT/APInt.h diff | blob | history
Domain: System / Toolchain;