Re: clang memory sanitizer report unsigned int overflow
[
Date Prev][
Date Next][
Thread Prev][
Thread Next]
[
Date Index]
[
Thread Index]
- Subject: Re: clang memory sanitizer report unsigned int overflow
- From: Lorenzo Donati <lorenzodonatibz@...>
- Date: Fri, 6 Jan 2023 11:33:45 +0100
I said "in some context", not "in general". However
On 06/01/2023 08:51, bil til wrote:
If you would count this as "overflow/invalid", then presumably every
Lua (or C) Code like this also would show some overflow:
i= 1 << 65;
this is most probably UB [1] unless, as Francisco Olarte said in his
post, you have ints with 65 bits or more (presumably 128?).
Quite unusual even today on 64 bit system I guess, because otherwise
"long" and "long long" would be probably redundant (OK, they could have
still more bits, but a rare beast for sure, I guess).
... or e. g.
i= 1
while i do i= i<<1 ... some code... end
... but this clearly should be valid.
[1] C prohibits shifting for more than an integer object width. Citing
from C99 draft standard (N1256; pag.84):
6.5.7 Bitwise shift operators
[...]
Semantics
The integer promotions are performed on each of the operands. The
type of the result is that of the promoted left operand. If the value
of the right operand is negative or is greater than or equal to the
width of the promoted left operand, the behavior is undefined.