You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
We often need to talk about zero-sized types (ZSTs) with a particular alignment requirement, e.g., 1, or in general, N:
#[repr(align(N))]structZst;
It was suggested in #164 (comment) that we should have an abbreviation for this defined in the glossary, where we also define the zero-sized type (ZST) abbreviation, to avoid writing "a zero-sized one-aligned type" and similar constructs all over the place.
A proposal was to use the N-ZST schema for an "N-aligned zero-sized type", e.g., 1-ZST would stand for a "one-aligned zero-sized type".
The text was updated successfully, but these errors were encountered:
N-ZST is a great name for talking about the concept in natural language, yet it has a great deficiency. Its particular instances, such as 4-ZST or 16-ZST, cannot be identifiers. aligned crate used A1, A2...A64 as names for 1-ZST ..= 64-ZST. I'm unsure whether I should change the name of A1 to, let's say, A1ZST. The latter may seem cryptic to the outsider.
We often need to talk about zero-sized types (ZSTs) with a particular alignment requirement, e.g., 1, or in general, N:
It was suggested in #164 (comment) that we should have an abbreviation for this defined in the glossary, where we also define the zero-sized type (ZST) abbreviation, to avoid writing "a zero-sized one-aligned type" and similar constructs all over the place.
A proposal was to use the
N-ZST
schema for an "N-aligned zero-sized type", e.g.,1-ZST
would stand for a "one-aligned zero-sized type".The text was updated successfully, but these errors were encountered: