Why was 1 << 31 changed to be implementation-defined in C++14?

Solution 1:

The relevant issue is CWG 1457, where the justification is that the change allows 1 << 31 to be used in constant expressions:

The current wording of 5.8 [expr.shift] paragraph 2 makes it undefined behavior to create the most-negative integer of a given type by left-shifting a (signed) 1 into the sign bit, even though this is not uncommonly done and works correctly on the majority of (twos-complement) architectures:

...if E1 has a signed type and non-negative value, and E1 * 2^E2 is representable in the result type, then that is the resulting value; otherwise, the behavior is undefined.

As a result, this technique cannot be used in a constant expression, which will break a significant amount of code.

Constant expressions can't contain undefined behavior, which means that using an expression containing UB in a context requiring a constant expression makes the program ill-formed. libstdc++'s numeric_limits::min, for example, once failed to compile in clang due to this.

Ampersand (&) operator in a SQL Server WHERE Clause

Double Pipe Symbols in Ruby Variable Assignment? [duplicate]

Using JQuery and Prototype in the same page

Domain strictly contained in the intersection of localizations at the primes of height one

Cantor - No set is equinumerous with its power set.

Burnside's formula on a hexagon

$A=\{(x_1,x_2,x_3): x_1\le x_2 \le x_3 \}, B=\{(x_1,x_2,x_3): x_2\le x_1 \le x_3 \}$. Show $P(Z+\mu \in A) \ge P(Z+ \mu \in B)$ for $\mu \in A$.

If $f$ is a non-constant analytic function on $B$ such that $|f|$ is a constant on $\partial B$, then $f$ must have a zero in $B$ [duplicate]

A simple algebra that is not semisimple [duplicate]

Determinants of products of binary matrices and binomial coefficients

Infinite series involving $\sqrt{n}$

Proof of $\sum_{m=0}^{n}\binom{m}{j}\binom{n-m}{k-j}=\binom{n+1}{k+1}$ (Another form of the Chu–Vandermonde identity)