David Xu 3f76af0f4a 1. use per-chain mutex instead of global mutex to reduce ... lock collision. 2. Fix two race conditions. One is between _umtx_unlock and signal, also a thread was marked TDF_UMTXWAKEUP by _umtx_unlock, it is possible a signal delivered to the thread will cause msleep returns EINTR, and the thread breaks out of loop, this causes umtx ownership is not transfered to the thread. Another is in _umtx_unlock itself, when the function sets the umtx to UMTX_UNOWNED state, a new thread can come in and lock the umtx, also the function tries to set contested bit flag, but it will fail. Although the function will wake a blocked thread, if that thread breaks out of loop by signal, no contested bit will be set.		2004-11-30 12:02:53 +00:00
..
bus_if.m
clock_if.m
device_if.m
genassym.sh
imgact_aout.c	Axe a.out core dump support. Neither older gdb binaries nor current	2004-11-27 06:46:59 +00:00
imgact_elf32.c
imgact_elf64.c
imgact_elf.c	On arm, set the default elf brand to FreeBSD, until the binutils do it for us.	2004-09-23 23:29:24 +00:00
imgact_gzip.c
imgact_shell.c	Don't treat # as a comment in interpreter specification line.	2004-10-31 11:12:59 +00:00
inflate.c
init_main.c	Don't include sys/user.h merely for its side-effect of recursively	2004-11-27 06:51:39 +00:00
init_sysent.c	Regen.	2004-11-25 12:08:16 +00:00
kern_acct.c	Rework how we store process times in the kernel such that we always store	2004-10-05 18:51:11 +00:00
kern_acl.c
kern_alq.c
kern_clock.c	Rework how we store process times in the kernel such that we always store	2004-10-05 18:51:11 +00:00
kern_condvar.c	Refine the turnstile and sleep queue interfaces just a bit:	2004-10-12 18:36:20 +00:00
kern_conf.c	Get rid of the magic "stash" of cdev structures, we no longer call	2004-10-25 13:12:06 +00:00
kern_context.c
kern_descrip.c	Don't acquire Giant before calling closef() in close() (and elsewhere);	2004-11-28 14:37:17 +00:00
kern_environment.c	Retire TUNABLE_QUAD_*.	2004-11-08 18:22:22 +00:00
kern_event.c	Move a FILEDESC_UNLOCK upwards to silence witness.	2004-11-16 14:41:31 +00:00
kern_exec.c	Don't include sys/user.h merely for its side-effect of recursively	2004-11-27 06:51:39 +00:00
kern_exit.c	Don't include sys/user.h merely for its side-effect of recursively	2004-11-27 06:51:39 +00:00
kern_fork.c	Don't include sys/user.h merely for its side-effect of recursively	2004-11-27 06:51:39 +00:00
kern_idle.c
kern_intr.c	Don't bother exiting storming mode once a second to see if it has gone	2004-11-17 14:39:41 +00:00
kern_jail.c
kern_kse.c	Remove local definitions of RANGEOF() and use __rangeof() instead.	2004-11-20 23:00:59 +00:00
kern_kthread.c
kern_ktr.c	- Change the ddb paging "support" to use a variable (db_lines_per_page) to	2004-11-01 22:15:15 +00:00
kern_ktrace.c
kern_linker.c
kern_lock.c	When upgrading the shared lock to an exclusive lock, if we discover	2004-11-29 22:58:32 +00:00
kern_lockf.c
kern_mac.c	Bump MAC Framework version to 2 in preparation for the upcoming API/ABI	2004-11-09 11:28:40 +00:00
kern_malloc.c	Turn VM_KMEM_SIZE_MAX and VM_KMEM_SIZE_SCALE into tunables.	2004-09-29 14:21:40 +00:00
kern_mbuf.c
kern_mib.c
kern_module.c
kern_mtxpool.c
kern_mutex.c	Refine the turnstile and sleep queue interfaces just a bit:	2004-10-12 18:36:20 +00:00
kern_ntptime.c	Place function comment above the right function.	2004-11-19 00:58:30 +00:00
kern_physio.c	Remove buf->b_dev field.	2004-11-04 07:59:57 +00:00
kern_poll.c
kern_proc.c	Axe a.out core dump support. Neither older gdb binaries nor current	2004-11-27 06:46:59 +00:00
kern_prot.c
kern_resource.c	Rework how we store process times in the kernel such that we always store	2004-10-05 18:51:11 +00:00
kern_sema.c
kern_shutdown.c	Replace a printf with a KASSERT that we are indeed running on the BSP.	2004-11-30 06:21:38 +00:00
kern_sig.c	Fix an off by one error. MAXPATHLEN already has +1.	2004-11-15 20:51:32 +00:00
kern_subr.c	Eliminate an unused argument to vm_pgmoveco().	2004-11-08 06:57:31 +00:00
kern_switch.c	Remove local definitions of RANGEOF() and use __rangeof() instead.	2004-11-20 23:00:59 +00:00
kern_sx.c
kern_synch.c	Unlock mutex if PDROP was set by caller.	2004-11-27 11:43:31 +00:00
kern_syscalls.c
kern_sysctl.c	Move the 'debug' sysctl tree under options SYSCTL_DEBUG. It generates	2004-10-27 19:26:01 +00:00
kern_tc.c	Put on my peril sensitive sunglasses and add a flags field to the internal	2004-10-11 22:04:16 +00:00
kern_thr.c	Remove local definitions of RANGEOF() and use __rangeof() instead.	2004-11-20 23:00:59 +00:00
kern_thread.c	Remove local definitions of RANGEOF() and use __rangeof() instead.	2004-11-20 23:00:59 +00:00
kern_time.c	Rework how we store process times in the kernel such that we always store	2004-10-05 18:51:11 +00:00
kern_timeout.c
kern_umtx.c	1. use per-chain mutex instead of global mutex to reduce	2004-11-30 12:02:53 +00:00
kern_uuid.c
kern_xxx.c	In osethostname(), don't need to call suser() directly as	2004-10-22 12:10:50 +00:00
ksched.c
link_elf_obj.c
link_elf.c
linker_if.m
Make.tags.inc
Makefile
makesyscalls.sh
md4c.c
md5c.c
p1003_1b.c
posix4_mib.c
sched_4bsd.c	When preempting a thread, put it back on the HEAD of its run queue.	2004-10-05 22:03:10 +00:00
sched_ule.c	- Temporarily disable the nice -20 throttling code. It has some interaction	2004-11-11 19:48:57 +00:00
subr_acl_posix1e.c
subr_autoconf.c
subr_blist.c
subr_bus.c	Don't print the singularly unhelpful message:	2004-11-03 09:06:45 +00:00
subr_clist.c
subr_clock.c
subr_devstat.c
subr_disk.c
subr_eventhandler.c
subr_hints.c
subr_kdb.c	Zero terminate empty sting in kdb_sysctl_available.	2004-10-21 01:11:25 +00:00
subr_kobj.c
subr_log.c
subr_mbpool.c
subr_mchain.c
subr_module.c
subr_msgbuf.c
subr_param.c	Fix the build.	2004-11-30 03:23:35 +00:00
subr_pcpu.c
subr_power.c
subr_prf.c
subr_prof.c
subr_rman.c
subr_rtc.c
subr_sbuf.c
subr_scanf.c
subr_sleepqueue.c	- Store threads on sleep queues in FIFO order rather than sorted by	2004-11-05 20:19:58 +00:00
subr_smp.c
subr_taskqueue.c	Add taskqueue_drain. This waits for the specified task to finish, if	2004-10-05 04:16:01 +00:00
subr_trap.c	Add a new per-thread private flag: TDP_GEOM.	2004-10-23 20:49:17 +00:00
subr_turnstile.c	Refine the turnstile and sleep queue interfaces just a bit:	2004-10-12 18:36:20 +00:00
subr_unit.c	Add delete_unrhdr() function.	2004-10-25 12:27:03 +00:00
subr_witness.c	clean up some tunables that should of been removed a while ago...	2004-11-09 06:46:14 +00:00
sys_generic.c	Push Giant down through ioctl.	2004-11-17 09:09:55 +00:00
sys_pipe.c	Correct a bug introduced in sys_pipe.c:1.179: in pipe_ioctl(),	2004-11-23 22:15:08 +00:00
sys_process.c	Don't include sys/user.h merely for its side-effect of recursively	2004-11-27 06:51:39 +00:00
sys_socket.c	Push Giant down through ioctl.	2004-11-17 09:09:55 +00:00
syscalls.c	Regen.	2004-11-25 12:08:16 +00:00
syscalls.master	Mark mount, unmount and nmount MPSAFE	2004-11-25 12:07:28 +00:00
sysv_ipc.c	Add an annotation to the comment for sysv_ipc.c to indicate that the	2004-10-22 12:12:40 +00:00
sysv_msg.c	Second of several commits to allow kernel System V IPC data structures	2004-11-12 13:23:47 +00:00
sysv_sem.c	Correct two incorrectly merged changes introduced in sysv_sem.c:1.71:	2004-11-13 08:06:40 +00:00
sysv_shm.c	Second of several commits to allow kernel System V IPC data structures	2004-11-12 13:23:47 +00:00
tty_compat.c
tty_conf.c
tty_cons.c	Hold threadcount reference when we call into the underlying console	2004-09-24 07:16:56 +00:00
tty_pty.c	Add new function ttyinitmode() which sets our systemwide default	2004-10-18 21:51:27 +00:00
tty_subr.c
tty_tty.c
tty.c	Improvements and fixes in the 1.241 commit:	2004-11-16 17:41:16 +00:00
uipc_accf.c
uipc_cow.c	Introduce two new options, "CPU private" and "no wait", to sf_buf_alloc().	2004-11-08 00:43:46 +00:00
uipc_domain.c	- Introduce protosw_init().	2004-11-11 19:19:54 +00:00
uipc_jumbo.c	The synchronization provided by vm object locking has eliminated the	2004-11-03 20:17:31 +00:00
uipc_mbuf2.c	Rename _m_tag_free() to m_tag_free_default() and make it non-static.	2004-10-11 18:40:19 +00:00
uipc_mbuf.c	improve the mbuf m_print function.. Only pull length from pkthdr if there	2004-09-28 18:40:18 +00:00
uipc_proto.c
uipc_sem.c
uipc_sockbuf.c	Fix a typo in a comparison appeared in rev. 1.125.	2004-10-27 05:37:58 +00:00
uipc_socket2.c	Fix a typo in a comparison appeared in rev. 1.125.	2004-10-27 05:37:58 +00:00
uipc_socket.c	If soreceive() is called from a socket callback, there's no reason	2004-11-29 23:10:59 +00:00
uipc_syscalls.c	Introduce an alias for FILEDESC_{UN}LOCK() with the suffix _FAST.	2004-11-13 11:53:02 +00:00
uipc_usrreq.c	Initialize struct pr_userreqs in new/sparse style and fill in common	2004-11-08 14:44:54 +00:00
vfs_acl.c
vfs_aio.c	Remove buf->b_dev field.	2004-11-04 07:59:57 +00:00
vfs_bio.c	- Eliminate the acquisition and release of the bqlock in bremfree() by	2004-11-18 08:44:09 +00:00
vfs_cache.c
vfs_cluster.c	Don't manually set b_bufobj, pbgetvp() does this for us.	2004-11-15 08:38:50 +00:00
vfs_default.c	Make more sense out of vop_stdcreatevobject()	2004-11-18 07:55:05 +00:00
vfs_export.c	Sigh. I really need to get an internet connection which is less than	2004-11-29 14:00:08 +00:00
vfs_extattr.c	Eliminate pointless goto.	2004-11-16 08:22:06 +00:00
vfs_init.c
vfs_lookup.c
vfs_mount.c	Use FILEDESC_LOCK_FAST in checkdirs()	2004-11-28 11:26:43 +00:00
vfs_subr.c	Move pbgetvp() and pbrelvp() to vm_pager.c with the rest of the pbuf stuff.	2004-11-15 08:12:50 +00:00
vfs_syscalls.c	Eliminate pointless goto.	2004-11-16 08:22:06 +00:00
vfs_vnops.c	Ok, first blunder: ioctls are not entirely unused on vnodes anymore :-)	2004-11-18 17:15:04 +00:00
vnode_if.src	Make VOP_BMAP return a struct bufobj for the underlying storage device	2004-11-15 09:18:27 +00:00