Re: HAVE_FAST_UNALIGNED_ACCESS

[Mattias Engdegård]

30 mars 2023 kl. 11.34 skrev Robert Pluim <rpluim@gmail.com>:

> but even if unaligned access is normally permitted by a machine, it is
> still undefined behavior to dereference an unaligned pointer.

That's not necessarily a problem; we have plenty of formally undefined 
behaviour. What's important is whether we can be confident that it works.

> Instead, HAVE_FAST_UNALIGNED_ACCESS and UNALIGNED_LOAD_SIZE should be
> removed and memcpy used instead:

No, this isn't something we should be doing at all unless the platform allows 
fast unaligned access.

Ideally we shouldn't do any of this nonsense but I haven't worked out how to 
get the autovectorisers to kick in.

> Emacs currently
> crashes when built with various compilers performing pointer alignment
> checks.

Which compilers, exactly?