https://gcc.gnu.org/bugzilla/show_bug.cgi?id=87548
Bug ID: 87548
Summary: Optimize fetch atomics with unused results
Product: gcc
Version: 8.2.1
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: other
Assignee: unassigned at gcc dot gnu.org
Reporter: david.feuer at gmail dot com
Target Milestone: ---
Atomic fetch and and/or/xor/etc., are not natively supported on amd64/x86_64,
so they're simulated with CAS loops. There's a special case when this is
unnecessary: when the fetched result is never used. In this case, it might be
better to use a non-fetching operation instead.
