[intro.races] p5 Terminological imprecision regarding "modification order" in the definition of release sequence #854
Description
Full name of submitter (unless configured in github; will be published with the issue): Jim X
In the definition of release sequence, the standard currently states:
A release sequence headed by a release operation A on an atomic object M is a maximal contiguous sub-sequence of side effects in the modification order of M
Mathematically, the "modification order of M" is defined as a total order (a binary relation over the set of modifications), not a sequence or a set itself. Therefore, technically speaking, a relation cannot have a "contiguous sub-sequence".
While the English idiom "in the modification order" might convey the intent (i.e., the sequence formed by sorting all side effects according to the modification order), this wording loosely conflates a mathematical relation with the sequence generated by it. Tightening this wording would make the concurrency model mathematically stricter.
Proposed Suggestion:
A release sequence headed by a release operation A on an atomic object M is a maximal contiguous sub-sequence of the sequence of side effects ordered by the modification order of M ...
This may be editorial.