On 10/11/2024 21.45, Mykola Garkusha via Std-Discussion wrote:
> Hi all,
>
> Currently(unlike C), the standard does not guarantee that padding bits may retain specific values even between sequential reads. Let's exclude tail padding re-use for potentially-overlapping sub-objects. as it is a slightly different matter related to OOP in C++ .
>
> In C, padding bits are guaranteed to say the same unless you modify the value!
Yes.
> Also looks like Clang and GCC also follow it, particularly to facilitate atomic operations on objects with padding bits.
> Clang and GCC even go further by providing intrinsics to clear padding bits (for example *__builtin_clear_padding* in GCC).
>
> Is there any chance C++ standard would allow this officially to come closer to the behaviour in C?
How would you want to inspect those padding bits portably? And why?
What's the use-case?
I don't see anything radically different compared to C, hence I hope portability would not be an issue. The main benefit of this change would be the ability to compare objects using object representation which is much faster. Say one could ensure padding bits set to 0 on creation or during the lifetime of an object and later compare the object with memcmp knowing ahead that the object value does not change and even it does one could again reset padding bits to 0 to use memcmp
Unless I've misunderstood, the C23 draft appears to say that as soon as you write to any member of a struct, all padding bits are reset to unspecified values.
6.2.6.1/6:
When a value is stored in an object of structure or union type, including in a member object, the bytes of the object representation that correspond to any padding bytes take unspecified values [...]
That seems to prevent the use of `memcmp` for comparing the values of struct objects in most cases.
Or is the C standard only talking about writes of struct type? (Which, as you probably know, don't exist in C++: an assignment operator, possibly a trivial one, is always called.)
--
Std-Discussion mailing list
Std-Discussion@lists.isocpp.org
https://lists.isocpp.org/mailman/listinfo.cgi/std-discussion