| Age | Commit message (Collapse) | Author | 
 | 
TSAN doesn't understand the OS specific "fast" seqlock code.  Use the
pthread mutex/condvar based path when TSAN is enabled.
Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
 | 
 | 
The files converted in this commit either had some random misspelling or
formatting weirdness that made them escape automated replacement, or
have a particularly "weird" licensing setup (e.g. dual-licensed.)
This also marks a bunch of "public domain" files as SPDX License "NONE".
Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
 | 
 | 
Signed-off-by: Ruben Kerkhof <ruben@rubenkerkhof.com>
 | 
 | 
Signed-off-by: Donatas Abraitis <donatas.abraitis@gmail.com>
 | 
 | 
Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
 | 
 | 
seqlock_timedwait() puts an (absolute, CLOCK_MONOTONIC) deadline on how
long we wait.  The RCU code uses this for its watchdog implementation.
Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
 | 
 | 
When we have no contention on the seqlock, we shouldn't incur the cost
of syscalls.
Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
 | 
 | 
Manually tested rather extensively in addition to included unit tests,
should work as intended.
NB: The OpenBSD futex() code is "future"; it's not actually in OpenBSD
(yet?) and thus untested.
Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
 |