summaryrefslogtreecommitdiffstats
path: root/usr.sbin/bgpd/timer.c (follow)
Commit message (Expand)AuthorAgeFilesLines
* Make the timer code independent of struct peer this way it can be usedclaudio2020-12-111-50/+51
* Change timer_nextisdue() and timer_nextduein() to take the current timeclaudio2019-05-241-5/+5
* sync log.c from relayd et al to bgpd.benno2017-01-241-1/+2
* Replace <sys/param.h> with <limits.h> and other less dirty headers wherederaadt2015-01-161-3/+4
* accept() pacing for bgpd based on similar work done on other daemons.claudio2012-04-121-3/+1
* spacing and irrelevant glitches which disrupt lintderaadt2010-10-241-2/+2
* timer_nextduein returned crap. fix it. so obvious... how could that happen.henning2009-01-211-2/+2
* use monotime instead of wallclock.henning2008-06-111-5/+18
* timer_due is not needed any morehenning2008-05-081-11/+1
* KNFhenning2008-05-081-2/+2
* take advanatge of the timewheelin timer_nextduein - we just need to lookhenning2008-05-081-11/+6
* we can stopin timer_set if new timeout equals the oldone.henning2008-05-081-1/+3
* take advantage of the timewheel: instead of checking each an everyhenning2008-05-081-1/+13
* in timer_set, when figuring out the right position for the timer,henning2008-05-081-2/+2
* change timer internals so that we use a timewheel (tailq, sorted by expiryhenning2008-05-081-29/+60
* provide timer_nextduein, which provides the number of seconds until thehenning2007-12-231-1/+15
* send timers for bgpctlshow neighbor foo timer in seperate messages afterhenning2007-12-231-1/+3
* make all timer_get() callers (except timer_set which will needhenning2007-12-231-5/+8
* rework timers.henning2007-12-201-0/+81