On Fri, 30 Aug 2019, Akshat Garg wrote: > > The first question for any new thing that is syntactically a qualifier is: > > is it intended generally to be counted as a qualifier where the standard > > refers to qualified type, the unqualified version of a type, etc.? Or is > > it, like _Atomic, a qualifier only syntactically and generally excluded > > from references to qualifiers? > > > Can you help me in understanding why the _Atomic is excluded from the > standard references. I referred to the C standard draft N2310 ( > http://www.open-std.org/jtc1/sc22/wg14/www/docs/n2310.pdf) but I couldn't > understand how it is excluded? I want to know what properties should a > qualifier have to be in standard qualifiers list?
In the case of _Atomic, it can affect the size and alignment of the type to which it is applied, which means it can't be considered a qualifier in the nornal semantic sense. In general you need to consider questions such as: is it safe to convert a pointer to unqualified type to a pointer to the corresponding _Dependent_ptr-qualified type? Are conditional expressions between pointers whose target types differ in presence or absence of _Dependent_ptr safe? If in general, in such places where the standard refers to qualifiers, the existing logic there is also appropriate for _Dependent_ptr, that suggests it should be a qualifier semantically. If the standard logic often seems inappropriate for _Dependent_ptr, that indicates it's not a qualifier in standard terms. -- Joseph S. Myers jos...@codesourcery.com
