Date: Fri, 23 Jan 2026 15:03:38 -0800
On Friday, 23 January 2026 14:01:35 Pacific Standard Time Andrey Semashev via
Std-Proposals wrote:
> > [1] https://gcc.godbolt.org/z/Wxdxbjao5
>
> Interesting, thanks for the code sample.
>
> However, I do not find where the standard permits padding bits (or, more
> generally, the object representation) to change out of thin air, without
> a modifying operation on the object. I tend to think this might be a
> compiler bug.
But is there language saying they must be retained?
It's possible that under *current* rules saying they are indeterminate, any
attempt to compare them is itself UB/EB, and therefore it doesn't matter if
they change or not.
This may be worth a Defect Report.
Std-Proposals wrote:
> > [1] https://gcc.godbolt.org/z/Wxdxbjao5
>
> Interesting, thanks for the code sample.
>
> However, I do not find where the standard permits padding bits (or, more
> generally, the object representation) to change out of thin air, without
> a modifying operation on the object. I tend to think this might be a
> compiler bug.
But is there language saying they must be retained?
It's possible that under *current* rules saying they are indeterminate, any
attempt to compare them is itself UB/EB, and therefore it doesn't matter if
they change or not.
This may be worth a Defect Report.
-- Thiago Macieira - thiago (AT) macieira.info - thiago (AT) kde.org Principal Engineer - Intel Data Center - Platform & Sys. Eng.
Received on 2026-01-23 23:03:49
