src/sys/mips/include/atomic.h, branch release/13.2.0FreeBSD source treehttps://cgit-dev.freebsd.org/src/atom?h=release%2F13.2.02020-01-02T23:20:37Z[PowerPC] [MIPS] Implement 32-bit kernel emulation of atomic64 operations2020-01-02T23:20:37ZBrandon Bergrenbdragon@FreeBSD.org2020-01-02T23:20:37Zurn:sha1:9aafc7c0523456c8b5ee8919c97f75277cf4d70b
This is a lock-based emulation of 64-bit atomics for kernel use, split off
from an earlier patch by jhibbits.
This is needed to unblock future improvements that reduce the need for
locking on 64-bit platforms by using atomic updates.
The implementation allows for future integration with userland atomic64,
but as that implies going through sysarch for every use, the current
status quo of userland doing its own locking may be for the best.
Submitted by: jhibbits (original patch), kevans (mips bits)
Reviewed by: jhibbits, jeff, kevans
Differential Revision: https://reviews.freebsd.org/D22976
Two minor issues:2019-12-17T03:20:37ZWarner Loshimp@FreeBSD.org2019-12-17T03:20:37Zurn:sha1:fa9b4635f0874eedc3d908f2cd21746680c8c2ed
(1) Don't define load/store 64 atomics for o32. They aren't atomic
there.
(2) Add comment about why we need 64 atomic define on n32 only.
add atomic_load_64 for mipsn322019-10-07T07:42:26ZAndriy Gaponavg@FreeBSD.org2019-10-07T07:42:26Zurn:sha1:eab7984cfe35a346becfe203414be2437878b7ef
It's just an alias for atomic_load_acq_64 (same as on i386).
MFC after: 1 week
mips: use generic sub-word atomic *cmpset2019-10-02T17:07:59ZKyle Evanskevans@FreeBSD.org2019-10-02T17:07:59Zurn:sha1:281ec62c977aba21b4aa60ed0f532d6a20b669e0
Most of this diff is refactoring to reduce duplication between the different
acq_ and rel_ variants.
Differential Revision: https://reviews.freebsd.org/D21822
mips: fcmpset: do not spin on sc failure2019-10-02T15:13:40ZKyle Evanskevans@FreeBSD.org2019-10-02T15:13:40Zurn:sha1:22c2c971a614a72875787e0fd0001906451ee245
For ll/sc architectures, atomic(9) allows failure modes where *old == val
due to write failure and callers should compensate for this. Do not retry on
failure, just leave 0 in ret and fail the operation if we couldn't sc it.
This lets the caller determine if it should retry or not.
Reviewed by: kib
Looks ok: imp
Differential Revision: https://reviews.freebsd.org/D21836
atomic_cmpset return value is also an int.2018-12-14T19:48:42ZWarner Loshimp@FreeBSD.org2018-12-14T19:48:42Zurn:sha1:3d060215a5013c72b968aa5163833fa31fc92913atomic_fcmpset* return int, not the type of *.2018-12-14T19:14:51ZWarner Loshimp@FreeBSD.org2018-12-14T19:14:51Zurn:sha1:2fb9d3808a3a0a8939502600cde69b04550d1e7d
fcmpset returns true/false as a int, so make the return types and
variables match the int to be consistent with other arch.
Reviewed by: cognet@
Differential Revision: https://reviews.freebsd.org/D18557
Correctly implemenet atomic_swap_long for mips64.2018-12-13T00:42:26ZWarner Loshimp@FreeBSD.org2018-12-13T00:42:26Zurn:sha1:a1128e850e3692714057c0a3632ef5b0436106ad
MIPS64 has 64-bit longs, so use uint64_t for it, otherwise uint32_t.
sizeof(long) == sizeof(ptr) for all platforms, so define
atomic_swap_ptr in terms of atomic_swap_long.
Submitted by: hps@
Implement atomic_swap_xxx() for all platforms.2018-12-10T13:38:13ZHans Petter Selaskyhselasky@FreeBSD.org2018-12-10T13:38:13Zurn:sha1:d7a9bfee8f10e6a787ea24581ac0db1040f409aa
Differential Revision: https://reviews.freebsd.org/D18450
Reviewed by: kib@
MFC after: 3 days
Sponsored by: Mellanox Technologies
Remove mips MD atomic_load_64 and atomic_store_64.2017-12-22T23:27:03ZKonstantin Belousovkib@FreeBSD.org2017-12-22T23:27:03Zurn:sha1:7aea69e54a9df49786f5cf56fcd86dfb4c513b6f
The only users of the functions were db_read_bytes() and
db_write_bytes() ddb(4) interfaces. Replace the calls with direct
reads and writes, which are automatically atomic on 64bits and n32.
Note that removed assembler implementation for mips32 is not atomic
anyway.
Reviewed by: jhb
Discussed with: imp
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Differential revision: https://reviews.freebsd.org/D13586