wireguard-openbsd/sys/arch/hppa/include/atomic.h, branch jd/queueboostsWireGuard implementation for the OpenBSD kernelhttps://git.zx2c4.com/wireguard-openbsd/atom/sys/arch/hppa/include/atomic.h?h=jd%2Fqueueboosts2018-05-14T09:33:20ZExpose memory barriers to userland.2018-05-14T09:33:20Zketteniskettenis@openbsd.org2018-05-14T09:33:20Zurn:sha1:ca950ec43995bb7d9640f1f39d8be4c78a3e7360
ok visa@, mpi@
make __cpu_simple_lock provide serialisation of the critical section.2016-02-09T06:46:02Zdlgdlg@openbsd.org2016-02-09T06:46:02Zurn:sha1:219fa1133a6cab8f5c345bfbe2016ac4da803913
that in turn makes atomic sequences actually atomic, which in turn
means the refcnt api asserts wont fire erronously when if_get and
if_put are actually used correctly.
such embarrassment.
reported by landry@ who also let me debug on the affected machines
ok jmatthew@
jmatthew@ points out atomic_swap_foo takes two arguments, not three.2014-09-24T06:34:37Zdlgdlg@openbsd.org2014-09-24T06:34:37Zurn:sha1:3480ef410d836ac34acc833ddbf19d74957a15e7implement atomic_{cas,swap}_{uint,ulong,ptr} and2014-09-22T12:12:23Zdlgdlg@openbsd.org2014-09-22T12:12:23Zurn:sha1:8bf031b6c9d14bce74cd445f935fd079a0ae10a6
atomic_{add,sub}_{int,long}_nv. sys/atomic.h turns these into the
rest of the atomic api.
on uniprocessor hppa systems "atomic" operations are implemented
as a non-interruptable sequence by disabling all interrupts on the
cpu, doing the operation, and then restoring the interrupt mask.
this isnt enough on MP systems, so we added a global atomic memory
mutex that is taken inside the interrupt disabling above to coordinate
operations between cpus.
this is a lot of overhead though cos mutexes dance around with ipls,
which is unecessary in our case because of the interrupt disabling
that is already done. also, hppa spinlocks are implemented with
ldcw which requires the word it operates on to be 16 byte aligned.
mutexes arent guaranteed to have this alignment so they compensate
by having lots of words inside themselves so they can hit the
appropriate one to use for the ldcw op.
with this in mind, this change pulls __cpu_simple_locks, which are
simply ldcw spinlocks with a 16 byte aligned word, out of
src/sys/arch/hppa/include/lock.h into src/sys/arch/hppa/include/atomic.h
so atomic.h can use them. lock.h includes atomic.h, so it still
gets and provides the same functionality as before.
finally, this also pulls the rwlock cas implementation apart. cas
ops now share the same serialising lock on MP systems as the other
memory operations, and rw_cas is defined as a wrapper around
atomic_cas_uint.
ok kettenis@
Implement the membar(9) API for hppa.2014-06-17T19:49:53Zketteniskettenis@openbsd.org2014-06-17T19:49:53Zurn:sha1:095af81feeb0931df8b31371d073c848b431d697
ok miod@, dlg@, guenther@
It's been a quarter century: we can assume volatile is present with that name.2014-03-29T18:09:28Zguentherguenther@openbsd.org2014-03-29T18:09:28Zurn:sha1:2df76cc2a0aee2cb2166e3799b7bb5c3f883f83d
ok dlg@ mpi@ deraadt@
Normalize sentinel. Use _MACHINE_*_H_ and _<ARCH>_*_H_ properly and consitently.2011-03-23T16:54:33Zpiroftipirofti@openbsd.org2011-03-23T16:54:33Zurn:sha1:2fa72412907ae78c4787a4a8f888c5f0486e225c
Discussed and okay drahn@. Okay deraadt@.
Use a mutex to make atomic operations atomic on multiprocessor kernels.2010-06-27T00:04:44Zjsingjsing@openbsd.org2010-06-27T00:04:44Zurn:sha1:7dda6ca755a19f3180569d751f06136030e33332
ok kettenis@
Proper atomic_{set,clear}bits_{int,long} implementation, and use it for2007-04-26T20:52:48Zmiodmiod@openbsd.org2007-04-26T20:52:48Zurn:sha1:f8b1aca581460905e05ce5918c46bc607fcd64d5
softintr(); ok mickey@
only make this interface available to the kernel for now, discussed witha2007-02-19T17:18:40Zderaadtderaadt@openbsd.org2007-02-19T17:18:40Zurn:sha1:8aa3ef09f5622dd8eb8f6d30eb7512c48ec87f8b
rt and such; tested and ok miod drahn