1
0
mirror of https://sourceware.org/git/glibc.git synced 2026-10-24 01:33:52 +03:00
Files
glibc/include
Adhemerval Zanella 304b22d7f9 atomic: Consolidate atomic_read_barrier implementation
All ABIs, except alpha, powerpc, and x86_64, define it to
atomic_full_barrier/__sync_synchronize, which can be mapped to
__atomic_thread_fence (__ATOMIC_SEQ_CST) in most cases, with the
exception of aarch64 (where the acquire fence is generated as
'dmb ishld' instead of 'dmb ish').

For s390x, it defaults to a memory barrier where __sync_synchronize
emits a 'bcr 15,0' (which the manual describes as pipeline
synchronization).

For PowerPC, it allows the use of lwsync for additional chips
(since _ARCH_PWR4 does not cover all chips that support it).

Tested on aarch64-linux-gnu, where the acquire produces a different
instruction that the current code.

Co-authored-by: Wilco Dijkstra <Wilco.Dijkstra@arm.com>
Reviewed-by: Wilco Dijkstra  <Wilco.Dijkstra@arm.com>
2025-11-04 04:14:01 -03:00
..
…
2021-06-22 12:09:52 -03:00
…
2023-05-01 19:33:21 +02:00
2025-09-11 14:46:07 -03:00
…
2021-06-15 10:42:11 -03:00
2021-06-15 10:42:11 -03:00
…
2017-04-07 07:45:53 -04:00
…
2025-10-29 12:53:50 -03:00
…
…
2021-07-22 18:38:04 +05:30
…
…
2021-06-30 08:43:37 +02:00
2023-10-24 12:30:59 +02:00
2023-05-01 09:08:59 -04:00
2019-05-02 11:42:51 +02:00
…
2023-10-24 12:30:59 +02:00
2019-10-09 17:48:41 -03:00
…
2024-01-03 12:07:14 +00:00
…
…
…
…
…
…
…
2021-06-30 08:43:37 +02:00
2019-11-06 09:20:52 -03:00
2024-02-01 11:02:01 +00:00
…