P0660 Stop Tokens and a Joining Thread about papers HOT 8 CLOSED

P0660R7 Interrupt Tokens and a Joining Thread, Rev 7 (Nicolai Josuttis, Lewis Baker, Billy O’Neal, Herb Sutter, Anthony Williams)

from papers.

P0660R8 Stop Tokens and a Joining Thread (Nicolai Josuttis, Lewis Baker, Billy O’Neal, Herb Sutter, Anthony Williams)

from papers.

LEWG in Kona:
stop_source(nullptr_t) should be nostopstate_t
comparison operators should be nodiscard
jthread::get_stop_source() returns stop_source;
jthread’s exposition-only member should be stop_source.
jthread::request_stop() should not be nodiscard
Unanimous consent

naming for wait_until()

11 wait_until/wait_until/wait_for
5 wait/wait_until/wait_for
0 wait_with_stop/wait_until_with_stop/wait_for_with_stop
2 wait_interruptible/wait_until_interruptible/wait_for_interruptible
4 wait_with_stop/wait_until/wait_for
6 wait_interruptible/wait_until/wait_for

jthread::request_stop() should be non-const, so should jthread::get_stop_source(), add a jthread::get_stop_token() (const)
SF F N A SA
9 5 1 0 0

We want this for C++20 (we’ll need to resolve the above questions).
SF F N A SA
10 6 3 1 0

Make the above changes and forward to LWG for C++20.
Unanimous consent

from papers.

Discussed in SG1 in Kona:

• revise paragraph 5 of 30.4 [thread.stop_token] to only forbid data races and dropping the phrasing about atomic operations
• add "atomically" to the Effects clauses of request_stop(), stop_requested(), and stop_prossible()
• wording for Effects of ~stop_callback() (30.4.2.1 p3) 2nd and last sentence replaced by new sentence written
• reword "atmoic state" to "shared state"

SF F N A SA
5 7 2 0 0

from papers.

P0660R9 Stop Token and Joining Thread (Nicolai Josuttis, Lewis Baker, Billy O’Neal, Herb Sutter, Anthony Williams)

from papers.

D0660R10 discussed in small group on Monday in Cologne

from papers.

P0660R10 Stop Token and Joining Thread (Nicolai Josuttis, Lewis Baker, Billy O'Neal, Herb Sutter, Anthony Williams)

from papers.

Adopted 2019-07.

from papers.