Andrey Semashev 34a2bc134c Fixed compilation on 32-bit x86, reduced code duplication. ...

On 32-bit x86, 64-bit integers have 4-byte alignment, which resulted in
a non-integral type being selected for storage type in case of lock-based
atomic_ref. This broke arithmetic and bitwise operations on atomic_ref.

We now select an integral type based on its native alignment, not the storage
alignment we require for atomic operations. This is fine in case of lock-based
backend.

Also, extended buffer_storage with support for specifying alignment and removed
aligned_buffer_storage and storage128_t.

2020-03-01 03:11:33 +03:00

3.1 KiB

Raw Blame History

View Raw