aboutsummaryrefslogtreecommitdiff
path: root/sys/kern
Commit message (Expand)AuthorAgeFilesLines
* This patch adds an M_NOFREE flag which allows one to mark an mbuf asKip Macy2007-10-062-0/+12
* Allow drivers to free an mbuf without having the mbuf be touched ifKip Macy2007-10-061-2/+5
* Fix sx_try_slock(), so it only fails when there is an exclusive owner.Pawel Jakub Dawidek2007-10-021-9/+12
* - Reassign the thread queue lock to newtd prior to switching. AssigningJeff Roberson2007-10-021-4/+6
* - Move the rebalancer back into hardclock to prevent potential softclockJeff Roberson2007-10-021-55/+86
* - Honor the PREEMPTION and FULL_PREEMPTION flags by setting the defaultJeff Roberson2007-09-271-2/+10
* Fix the description of the formula used to autosize the number ofRuslan Ermilov2007-09-261-1/+1
* Change the management of cached pages (PQ_CACHE) in two fundamentalAlan Cox2007-09-252-11/+5
* - Bound the interactivity score so that it cannot become negative.Jeff Roberson2007-09-241-1/+1
* - Improve grammar. s/it's/its/.Jeff Roberson2007-09-221-5/+13
* Fix some locking cases where we ask for exclusively locked vnode, but we getPawel Jakub Dawidek2007-09-212-4/+25
* - Redefine p_swtime and td_slptime as p_swtick and td_slptick. ThisJeff Roberson2007-09-214-26/+29
* - Call sched_sleep() before we suspend threads. sched_wakeup() is alreadyJeff Roberson2007-09-211-0/+2
* Fix some entries in the locks static table of witness.Attilio Rao2007-09-203-11/+9
* - Move all of the PS_ flags into either p_flag or td_flags.Jeff Roberson2007-09-1713-69/+47
* Remove the definition and implementation of 'CALLOUT_NETGIANT', a now- (andRobert Watson2007-09-151-11/+2
* Currently the LO_NOPROFILE flag (which is masked on upper level code byAttilio Rao2007-09-141-1/+1
* subr_sleepqueue.c presents a thread lock missing which leads to dangerousAttilio Rao2007-09-131-0/+2
* When restoring the mount after umount failed, the MNTK_UNMOUNT flagKonstantin Belousov2007-09-122-5/+10
* This is a follow-up, cleaning-up commit about recent changes involvingAttilio Rao2007-09-111-1/+1
* Rename mac_check_vnode_delete() MAC Framework and MAC Policy entryRobert Watson2007-09-101-2/+2
* In userland_sysctl(), call useracc() with the actual newlen value to beRobert Watson2007-09-021-1/+1
* Close a race that snuck in with the recent changes to fix a LOR betweenJohn Baldwin2007-08-311-13/+27
* Add missing newline in the log message of the previous commit.Diomidis Spinellis2007-08-311-1/+1
* Don't panic. When encountering a negative value call log(LOG_NOTICE, ...)Diomidis Spinellis2007-08-311-1/+7
* Partially revert the previous change. I failed to notice that whereJohn Baldwin2007-08-291-2/+0
* Rework the routines to convert a 5.x+ statfs structure (with fixed-sizeJohn Baldwin2007-08-281-4/+46
* - During shutdown pending, when the last sack came in andRandall Stewart2007-08-271-8/+5
* Destroy the kaio_mtx on the freeing the struct kaioinfo in theKonstantin Belousov2007-08-201-1/+5
* - Improve runq_findbit_from() which is used by ULE's circular queue. MaskJeff Roberson2007-08-201-32/+22
* - Set steal_thresh to log2(ncpus). This improves idle-time load balancingJeff Roberson2007-08-201-0/+6
* Always call sched_bind(), even if on the CPU in question. It is wrong toNate Lawson2007-08-201-25/+15
* Use a different loop variable for the inner loop. This previous reuse couldNate Lawson2007-08-191-4/+4
* Regenerate.David Xu2007-08-163-0/+11
* Add thr_kill2 syscall which sends a signal to a thread in another process.David Xu2007-08-162-0/+56
* On 6.x this works:John Baldwin2007-08-151-11/+56
* Improve vn_printf() by:Pawel Jakub Dawidek2007-08-131-7/+45
* Do not call free() while holding vnode interlock.Konstantin Belousov2007-08-071-27/+44
* Remove the now-unused NET_{LOCK,UNLOCK,ASSERT}_GIANT() macros, whichRobert Watson2007-08-065-107/+25
* - Fix one line that erroneously crept in my last commit.Jeff Roberson2007-08-041-1/+0
* - Share scheduler locks between hyper-threaded cores to protect theJeff Roberson2007-08-031-114/+200
* - Set SW_PREEMPT when we preempt in critical_exit().Jeff Roberson2007-08-031-1/+1
* First in a series of changes to remove the now-unused Giant compatibilityRobert Watson2007-07-272-10/+2
* Actually, upcalls cannot be freed while destroying the thread because weAttilio Rao2007-07-272-0/+20
* When we do open, we should lock the vnode exclusively. This fixes few races:Pawel Jakub Dawidek2007-07-262-3/+3
* The v_mountedhere field is protected by the vnode lock, not vnode's internalPawel Jakub Dawidek2007-07-261-1/+1
* upcall_free() was only used in kse_GC() which has been removed so it nowAttilio Rao2007-07-231-8/+0
* Actually, KSE kernel bits locking is broken and can lead likely toAttilio Rao2007-07-232-82/+71
* If clock_ct_to_ts fails to convert time time from the real time clock,David Malone2007-07-231-1/+1
* ttyfree() frees the cdev(). But if there are pending kevents,Konstantin Belousov2007-07-201-7/+17