[all-commits] [llvm/llvm-project] 2b6b8c: [APFloat] Prevent construction of APFloat with Sem...

[Ehud Katz via All-commits]

  Branch: refs/heads/master
  Home:   https://github.com/llvm/llvm-project
  Commit: 2b6b8cb10c870d64f7cc29d21fc27cef3c7e0056
      https://github.com/llvm/llvm-project/commit/2b6b8cb10c870d64f7cc29d21fc27cef3c7e0056
  Author: Ehud Katz <ehudkatz at gmail.com>
  Date:   2019-12-04 (Wed, 04 Dec 2019)

  Changed paths:
    M llvm/include/llvm/ADT/APFloat.h
    M llvm/lib/Transforms/InstCombine/InstCombineCalls.cpp
    M llvm/lib/Transforms/Utils/SimplifyLibCalls.cpp
    M llvm/unittests/ADT/APFloatTest.cpp

  Log Message:
  -----------
  [APFloat] Prevent construction of APFloat with Semantics and FP value

Constructor invocations such as `APFloat(APFloat::IEEEdouble(), 0.0)`
may seem like they accept a FP (floating point) value, but the overload
they reach is actually the `integerPart` one, not a `float` or `double`
overload (which only exists when `fltSemantics` isn't passed).

This may lead to possible loss of data, by the conversion from `float`
or `double` to `integerPart`.

To prevent future mistakes, a new constructor overload, which accepts
any FP value and marked with `delete`, to prevent its usage.

Fixes PR34095.

Differential Revision: https://reviews.llvm.org/D70425