summaryrefslogtreecommitdiffstats
path: root/sys/kern
Commit message (Expand)AuthorAgeFilesLines
* The "free-lance" timer in the i8254 is only used for the speakerphk2008-03-261-0/+51
* Regen.dfr2008-03-263-4/+14
* Add the new kernel-mode NFS Lock Manager. To use it instead of thedfr2008-03-264-473/+1969
* Implement taskqueue_block() and taskqueue_unblock(). These functions allowscottl2008-03-251-1/+28
* Replaced the misleading uses of a historical artefact M_TRYWAIT with M_WAIT.ru2008-03-254-52/+28
* Regen after changing prototypes of cpuset_{get,set}affinity().ru2008-03-253-7/+6
* Fixed type of the fourth argument of cpuset_{get,set}affinity(2) to be size_t.ru2008-03-253-16/+15
* - Greatly simplify vget() by removing the guarantee that any newjeff2008-03-241-32/+18
* - Don't acquire the vnode interlock in _vn_lock() unless no lock typejeff2008-03-241-19/+13
* Yield the cpu in the kernel while iterating the list of thekib2008-03-231-0/+6
* Remove commented out code, thread suspension is done in thread library.davidxu2008-03-231-2/+1
* - Only return 1 from sync_vnode() in cases where the vnode is stilljeff2008-03-231-1/+1
* - Pass BO_MTX(bo) to lockmgr in vtruncbuf, we don't own the vnodejeff2008-03-231-1/+1
* In abort2(2): Accept a NULL arg pointer if nargs == 0phk2008-03-221-6/+8
* - Complete part of the unfinished bufobj work by consistently usingjeff2008-03-224-57/+64
* Fix a race where timeout/untimeout could cause crashes for Giant lockedalfred2008-03-221-4/+19
* Reduce contention on the vnode interlock by not acquiring the BO_LOCKkib2008-03-211-12/+10
* - Reduce contention on the global bdonelock and bpinlock by usingjeff2008-03-211-30/+34
* - Add a new td flag TDF_NEEDSUSPCHK that is set whenever a thread needsjeff2008-03-215-19/+18
* Implement a BUS_BIND_INTR() method in the bus interface to bind an IRQjhb2008-03-202-0/+48
* Fix the leak of the vmspace on the fork when the process limitskib2008-03-201-2/+4
* - Restore runq to manipulating threads directly by putting runq links andjeff2008-03-203-241/+196
* - Remove the unused and redundant sched_newproc() function.jeff2008-03-201-34/+0
* - There is no sense in calling sched_newthread() at thread_init() andjeff2008-03-201-2/+0
* - ULE and 4BSD share only one line of code from sched_newthread() so implementjeff2008-03-202-7/+10
* - Don't call the empty sched_newproc() function. sched_newproc() alreadyjeff2008-03-201-1/+0
* - Move maybe_preempt() from kern_switch.c to sched_4bsd.c. This is functionjeff2008-03-202-114/+115
* - Directly include opt_sched.h in sched_4bsd.jeff2008-03-201-0/+1
* Revert previous change - it appears that the limit I was hitting was asobomax2008-03-192-37/+3
* Remove extra uihold() call that accidentally sneak in during perforcepjd2008-03-191-1/+0
* - Remove some dead code and comments related to KSE.jeff2008-03-191-56/+16
* - At the top of sleepq_catch_signals() lock the thread and check TDF_NEEDSIGCHKjeff2008-03-191-4/+12
* - Remove stale comment.jeff2008-03-191-8/+0
* - Add a facility similar to LOCK_PROFILING under SLEEPQUEUE_PROFILING. Keepjeff2008-03-191-1/+159
* - Fix the last of the threading bugs that were introduced as far back asjeff2008-03-191-6/+12
* - Restore the NULL check for td_cpuset. This can happen if a partiallyjeff2008-03-191-1/+2
* - Relax requirements for p_numthreads, p_threads, p_swtick, and p_nice fromjeff2008-03-1914-113/+42
* Simplify the interrupt code a bit:jhb2008-03-171-46/+6
* Fix two races in the handling of the d_gianttrick for the D_NEEDGIANTkib2008-03-171-35/+121
* - There is no more "uidinfo struct" mutex.pjd2008-03-171-2/+1
* Whitespace cleanups.pjd2008-03-161-7/+7
* - Use wait-free method to manage ui_sbsize and ui_proccnt fields in thepjd2008-03-161-58/+48
* Consistently use ANSI C declarationsfor all functions in kern_synch.c.rwatson2008-03-161-19/+7
* Style fixes.pjd2008-03-161-11/+7
* Fix information leak. We can find PIDs of running processes from withinpjd2008-03-161-1/+2
* In keeping with style(9)'s recommendations on macros, use a ';'rwatson2008-03-1638-64/+74
* Properly set size of the file_zone to match kern.maxfiles parameter.sobomax2008-03-162-3/+37
* Fix panic on e.g. "kldload /dev/null".ru2008-03-152-0/+9
* Add preliminary support for binding interrupts to CPUs:jhb2008-03-141-4/+92
* Make the function prototype for cpu_search() match the declaration so thatjhb2008-03-141-2/+2
OpenPOWER on IntegriCloud