wireguard-openbsd/sys/arch/amd64/include/mutex.h, branch jd/queueboosts WireGuard implementation for the OpenBSD kernel https://git.zx2c4.com/wireguard-openbsd/atom/sys/arch/amd64/include/mutex.h?h=jd%2Fqueueboosts 2018-01-25T15:06:29Z Move common mutex implementations to a MI place. 2018-01-25T15:06:29Z mpi mpi@openbsd.org 2018-01-25T15:06:29Z urn:sha1:710c2d98ef6b48aed52cf1712826535b248055bf Archs not yet converted can to the jump by defining __USE_MI_MUTEX. ok visa@ Define and use IPL_MPFLOOR in our common mutex implementation. 2018-01-13T15:18:11Z mpi mpi@openbsd.org 2018-01-13T15:18:11Z urn:sha1:4965d1a4536ae31f44fece9aedf52ba3da2c14fb ok kettenis@, visa@ Unify <machine/mutex.h> a bit further. 2018-01-12T09:19:32Z mpi mpi@openbsd.org 2018-01-12T09:19:32Z urn:sha1:ee04a0b4c7b4d80bfec00ccaa22436570edc25dd `mtx_owner' becomes the first field of 'struct mutex' on i386/amd64/arm64. ok visa@ Hook up mutex(9) to witness(4). 2017-04-20T13:57:29Z visa visa@openbsd.org 2017-04-20T13:57:29Z urn:sha1:c4925651028d655edda3c8b85b8d1b4a0cda6fb9 It's been a quarter century: we can assume volatile is present with that name. 2014-03-29T18:09:28Z guenther guenther@openbsd.org 2014-03-29T18:09:28Z urn:sha1:2df76cc2a0aee2cb2166e3799b7bb5c3f883f83d ok dlg@ mpi@ deraadt@ Since interrupts at IPL_AUDIO no longer grab the kernel lock, and nobody 2013-06-26T19:53:50Z kettenis kettenis@openbsd.org 2013-06-26T19:53:50Z urn:sha1:a12998ad0e7dde4da2a3473a5320dff115becdf0 is supposed to establish interrupts at IPL_VM, we only need to "flatten" mutex ipls up until IPL_TTY. Add an assertion to enforce that IPL_AUDIO interrupts are establish with the IPL_MPSAFE flag set. Should improve latency of audio interrupts a tiny bit. ok matthew@ To prevent lock ordering problems with the kernel lock, we need to make sure 2013-06-02T01:55:52Z kettenis kettenis@openbsd.org 2013-06-02T01:55:52Z urn:sha1:ccca563d8700526dd92edef719f08e1cf6a44c85 we block all interrupts that can grab the kernel lock. The simplest way to achieve this is to make sure mutexes always raise the ipl to the highest level that has interrupts that grab the kernel lock. This will allow us to have "mpsafe" interrupt handlers at lower priority levels. No change for non-MULTIPROCESSOR kernels. discussed with many ok mpi@, tedu@ Normalize sentinel. Use _MACHINE_*_H_ and _<ARCH>_*_H_ properly and consitently. 2011-03-23T16:54:33Z pirofti pirofti@openbsd.org 2011-03-23T16:54:33Z urn:sha1:2fa72412907ae78c4787a4a8f888c5f0486e225c Discussed and okay drahn@. Okay deraadt@. remove trailing newline in panic(9); ok millert@ and deraadt@ 2005-07-18T02:43:24Z fgsch fgsch@openbsd.org 2005-07-18T02:43:24Z urn:sha1:62a36cb007c2ec846c21ec2042386e6231228966 add rcs ids 2005-04-08T07:07:06Z jolan jolan@openbsd.org 2005-04-08T07:07:06Z urn:sha1:966b911426dc4c3605a457155522434b7b8af762