freebsd-skq/sys/kern/kern_exit.c

1323 lines
33 KiB
C
Raw Normal View History

/*-
1994-05-24 10:09:53 +00:00
* Copyright (c) 1982, 1986, 1989, 1991, 1993
* The Regents of the University of California. All rights reserved.
* (c) UNIX System Laboratories, Inc.
* All or some portions of this file are derived from material licensed
* to the University of California by American Telephone and Telegraph
* Co. or Unix System Laboratories, Inc. and are reproduced herein with
* the permission of UNIX System Laboratories, Inc.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 3. Neither the name of the University nor the names of its contributors
1994-05-24 10:09:53 +00:00
* may be used to endorse or promote products derived from this software
* without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
* @(#)kern_exit.c 8.7 (Berkeley) 2/12/94
*/
2003-06-11 00:56:59 +00:00
#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");
#include "opt_compat.h"
#include "opt_ktrace.h"
1994-05-24 10:09:53 +00:00
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/sysproto.h>
#include <sys/capsicum.h>
#include <sys/eventhandler.h>
#include <sys/kernel.h>
#include <sys/malloc.h>
#include <sys/lock.h>
#include <sys/mutex.h>
1994-05-24 10:09:53 +00:00
#include <sys/proc.h>
#include <sys/procdesc.h>
#include <sys/pioctl.h>
MFp4: Bring in updated jail support from bz_jail branch. This enhances the current jail implementation to permit multiple addresses per jail. In addtion to IPv4, IPv6 is supported as well. Due to updated checks it is even possible to have jails without an IP address at all, which basically gives one a chroot with restricted process view, no networking,.. SCTP support was updated and supports IPv6 in jails as well. Cpuset support permits jails to be bound to specific processor sets after creation. Jails can have an unrestricted (no duplicate protection, etc.) name in addition to the hostname. The jail name cannot be changed from within a jail and is considered to be used for management purposes or as audit-token in the future. DDB 'show jails' command was added to aid debugging. Proper compat support permits 32bit jail binaries to be used on 64bit systems to manage jails. Also backward compatibility was preserved where possible: for jail v1 syscalls, as well as with user space management utilities. Both jail as well as prison version were updated for the new features. A gap was intentionally left as the intermediate versions had been used by various patches floating around the last years. Bump __FreeBSD_version for the afore mentioned and in kernel changes. Special thanks to: - Pawel Jakub Dawidek (pjd) for his multi-IPv4 patches and Olivier Houchard (cognet) for initial single-IPv6 patches. - Jeff Roberson (jeff) and Randall Stewart (rrs) for their help, ideas and review on cpuset and SCTP support. - Robert Watson (rwatson) for lots and lots of help, discussions, suggestions and review of most of the patch at various stages. - John Baldwin (jhb) for his help. - Simon L. Nielsen (simon) as early adopter testing changes on cluster machines as well as all the testers and people who provided feedback the last months on freebsd-jail and other channels. - My employer, CK Software GmbH, for the support so I could work on this. Reviewed by: (see above) MFC after: 3 months (this is just so that I get the mail) X-MFC Before: 7.2-RELEASE if possible
2008-11-29 14:32:14 +00:00
#include <sys/jail.h>
1994-05-24 10:09:53 +00:00
#include <sys/tty.h>
#include <sys/wait.h>
#include <sys/vmmeter.h>
#include <sys/vnode.h>
#include <sys/racct.h>
1994-05-24 10:09:53 +00:00
#include <sys/resourcevar.h>
#include <sys/sbuf.h>
#include <sys/signalvar.h>
#include <sys/sched.h>
#include <sys/sx.h>
#include <sys/syscallsubr.h>
#include <sys/syslog.h>
1994-05-24 10:09:53 +00:00
#include <sys/ptrace.h>
#include <sys/acct.h> /* for acct_process() function prototype */
#include <sys/filedesc.h>
#include <sys/sdt.h>
#include <sys/shm.h>
#include <sys/sem.h>
#include <sys/umtx.h>
#ifdef KTRACE
#include <sys/ktrace.h>
#endif
#include <security/audit/audit.h>
#include <security/mac/mac_framework.h>
1994-05-24 10:09:53 +00:00
#include <vm/vm.h>
#include <vm/vm_extern.h>
#include <vm/vm_param.h>
#include <vm/pmap.h>
#include <vm/vm_map.h>
#include <vm/vm_page.h>
#include <vm/uma.h>
Add an initial NUMA affinity/policy configuration for threads and processes. This is based on work done by jeff@ and jhb@, as well as the numa.diff patch that has been circulating when someone asks for first-touch NUMA on -10 or -11. * Introduce a simple set of VM policy and iterator types. * tie the policy types into the vm_phys path for now, mirroring how the initial first-touch allocation work was enabled. * add syscalls to control changing thread and process defaults. * add a global NUMA VM domain policy. * implement a simple cascade policy order - if a thread policy exists, use it; if a process policy exists, use it; use the default policy. * processes inherit policies from their parent processes, threads inherit policies from their parent threads. * add a simple tool (numactl) to query and modify default thread/process policities. * add documentation for the new syscalls, for numa and for numactl. * re-enable first touch NUMA again by default, as now policies can be set in a variety of methods. This is only relevant for very specific workloads. This doesn't pretend to be a final NUMA solution. The previous defaults in -HEAD (with MAXMEMDOM set) can be achieved by 'sysctl vm.default_policy=rr'. This is only relevant if MAXMEMDOM is set to something other than 1. Ie, if you're using GENERIC or a modified kernel with non-NUMA, then this is a glorified no-op for you. Thank you to Norse Corp for giving me access to rather large (for FreeBSD!) NUMA machines in order to develop and verify this. Thank you to Dell for providing me with dual socket sandybridge and westmere v3 hardware to do NUMA development with. Thank you to Scott Long at Netflix for providing me with access to the two-socket, four-domain haswell v3 hardware. Thank you to Peter Holm for running the stress testing suite against the NUMA branch during various stages of development! Tested: * MIPS (regression testing; non-NUMA) * i386 (regression testing; non-NUMA GENERIC) * amd64 (regression testing; non-NUMA GENERIC) * westmere, 2 socket (thankyou norse!) * sandy bridge, 2 socket (thankyou dell!) * ivy bridge, 2 socket (thankyou norse!) * westmere-EX, 4 socket / 1TB RAM (thankyou norse!) * haswell, 2 socket (thankyou norse!) * haswell v3, 2 socket (thankyou dell) * haswell v3, 2x18 core (thankyou scott long / netflix!) * Peter Holm ran a stress test suite on this work and found one issue, but has not been able to verify it (it doesn't look NUMA related, and he only saw it once over many testing runs.) * I've tested bhyve instances running in fixed NUMA domains and cpusets; all seems to work correctly. Verified: * intel-pcm - pcm-numa.x and pcm-memory.x, whilst selecting different NUMA policies for processes under test. Review: This was reviewed through phabricator (https://reviews.freebsd.org/D2559) as well as privately and via emails to freebsd-arch@. The git history with specific attributes is available at https://github.com/erikarn/freebsd/ in the NUMA branch (https://github.com/erikarn/freebsd/compare/local/adrian_numa_policy). This has been reviewed by a number of people (stas, rpaulo, kib, ngie, wblock) but not achieved a clear consensus. My hope is that with further exposure and testing more functionality can be implemented and evaluated. Notes: * The VM doesn't handle unbalanced domains very well, and if you have an overly unbalanced memory setup whilst under high memory pressure, VM page allocation may fail leading to a kernel panic. This was a problem in the past, but it's much more easily triggered now with these tools. * This work only controls the path through vm_phys; it doesn't yet strongly/predictably affect contigmalloc, KVA placement, UMA, etc. So, driver placement of memory isn't really guaranteed in any way. That's next on my plate. Sponsored by: Norse Corp, Inc.; Dell
2015-07-11 15:21:37 +00:00
#include <vm/vm_domain.h>
1994-05-24 10:09:53 +00:00
#ifdef KDTRACE_HOOKS
#include <sys/dtrace_bsd.h>
dtrace_execexit_func_t dtrace_fasttrap_exit;
#endif
SDT_PROVIDER_DECLARE(proc);
SDT_PROBE_DEFINE1(proc, , , exit, "int");
/* Hook for NFS teardown procedure. */
void (*nlminfo_release_p)(struct proc *p);
struct proc *
proc_realparent(struct proc *child)
{
struct proc *p, *parent;
sx_assert(&proctree_lock, SX_LOCKED);
if ((child->p_treeflag & P_TREE_ORPHANED) == 0) {
if (child->p_oppid == 0 ||
child->p_pptr->p_pid == child->p_oppid)
parent = child->p_pptr;
else
parent = initproc;
return (parent);
}
for (p = child; (p->p_treeflag & P_TREE_FIRST_ORPHAN) == 0;) {
/* Cannot use LIST_PREV(), since the list head is not known. */
p = __containerof(p->p_orphan.le_prev, struct proc,
p_orphan.le_next);
KASSERT((p->p_treeflag & P_TREE_ORPHANED) != 0,
("missing P_ORPHAN %p", p));
}
parent = __containerof(p->p_orphan.le_prev, struct proc,
p_orphans.lh_first);
return (parent);
}
void
reaper_abandon_children(struct proc *p, bool exiting)
{
struct proc *p1, *p2, *ptmp;
sx_assert(&proctree_lock, SX_LOCKED);
KASSERT(p != initproc, ("reaper_abandon_children for initproc"));
if ((p->p_treeflag & P_TREE_REAPER) == 0)
return;
p1 = p->p_reaper;
LIST_FOREACH_SAFE(p2, &p->p_reaplist, p_reapsibling, ptmp) {
LIST_REMOVE(p2, p_reapsibling);
p2->p_reaper = p1;
p2->p_reapsubtree = p->p_reapsubtree;
LIST_INSERT_HEAD(&p1->p_reaplist, p2, p_reapsibling);
if (exiting && p2->p_pptr == p) {
PROC_LOCK(p2);
proc_reparent(p2, p1);
PROC_UNLOCK(p2);
}
}
KASSERT(LIST_EMPTY(&p->p_reaplist), ("p_reaplist not empty"));
p->p_treeflag &= ~P_TREE_REAPER;
}
static void
clear_orphan(struct proc *p)
{
struct proc *p1;
sx_assert(&proctree_lock, SA_XLOCKED);
if ((p->p_treeflag & P_TREE_ORPHANED) == 0)
return;
if ((p->p_treeflag & P_TREE_FIRST_ORPHAN) != 0) {
p1 = LIST_NEXT(p, p_orphan);
if (p1 != NULL)
p1->p_treeflag |= P_TREE_FIRST_ORPHAN;
p->p_treeflag &= ~P_TREE_FIRST_ORPHAN;
}
LIST_REMOVE(p, p_orphan);
p->p_treeflag &= ~P_TREE_ORPHANED;
}
1994-05-24 10:09:53 +00:00
/*
* exit -- death of process.
1994-05-24 10:09:53 +00:00
*/
void
sys_sys_exit(struct thread *td, struct sys_exit_args *uap)
1994-05-24 10:09:53 +00:00
{
exit1(td, uap->rval, 0);
1994-05-24 10:09:53 +00:00
/* NOTREACHED */
}
/*
* Exit: deallocate address space and other resources, change proc state to
* zombie, and unlink proc from allproc and parent's lists. Save exit status
* and rusage for wait(). Check for child processes and orphan them.
1994-05-24 10:09:53 +00:00
*/
void
exit1(struct thread *td, int rval, int signo)
1994-05-24 10:09:53 +00:00
{
struct proc *p, *nq, *q, *t;
struct thread *tdt;
1994-05-24 10:09:53 +00:00
mtx_assert(&Giant, MA_NOTOWNED);
KASSERT(rval == 0 || signo == 0, ("exit1 rv %d sig %d", rval, signo));
p = td->td_proc;
/*
* XXX in case we're rebooting we just let init die in order to
* work around an unsolved stack overflow seen very late during
* shutdown on sparc64 when the gmirror worker process exists.
*/
if (p == initproc && rebooting == 0) {
printf("init died (signal %d, exit %d)\n", signo, rval);
panic("Going nowhere without my init!");
}
/*
* Deref SU mp, since the thread does not return to userspace.
*/
if (softdep_ast_cleanup != NULL)
softdep_ast_cleanup();
/*
* MUST abort all other threads before proceeding past here.
*/
PROC_LOCK(p);
/*
* First check if some other thread or external request got
* here before us. If so, act appropriately: exit or suspend.
* We must ensure that stop requests are handled before we set
* P_WEXIT.
*/
thread_suspend_check(0);
while (p->p_flag & P_HADTHREADS) {
/*
* Kill off the other threads. This requires
* some co-operation from other parts of the kernel
* so it may not be instantaneous. With this state set
* any thread entering the kernel from userspace will
* thread_exit() in trap(). Any thread attempting to
* sleep will return immediately with EINTR or EWOULDBLOCK
* which will hopefully force them to back out to userland
* freeing resources as they go. Any thread attempting
* to return to userland will thread_exit() from userret().
* thread_exit() will unsuspend us when the last of the
* other threads exits.
* If there is already a thread singler after resumption,
* calling thread_single will fail; in that case, we just
* re-check all suspension request, the thread should
* either be suspended there or exit.
*/
if (!thread_single(p, SINGLE_EXIT))
/*
* All other activity in this process is now
* stopped. Threading support has been turned
* off.
*/
break;
/*
* Recheck for new stop or suspend requests which
* might appear while process lock was dropped in
* thread_single().
*/
thread_suspend_check(0);
}
KASSERT(p->p_numthreads == 1,
("exit1: proc %p exiting with %d threads", p, p->p_numthreads));
racct_sub(p, RACCT_NTHR, 1);
/* Let event handler change exit status */
p->p_xexit = rval;
p->p_xsig = signo;
Close some races between procfs/ptrace and exit(2): - Reorder the events in exit(2) slightly so that we trigger the S_EXIT stop event earlier. After we have signalled that, we set P_WEXIT and then wait for any processes with a hold on the vmspace via PHOLD to release it. PHOLD now KASSERT()'s that P_WEXIT is clear when it is invoked, and PRELE now does a wakeup if P_WEXIT is set and p_lock drops to zero. - Change proc_rwmem() to require that the processing read from has its vmspace held via PHOLD by the caller and get rid of all the junk to screw around with the vmspace reference count as we no longer need it. - In ptrace() and pseudofs(), treat a process with P_WEXIT set as if it doesn't exist. - Only do one PHOLD in kern_ptrace() now, and do it earlier so it covers FIX_SSTEP() (since on alpha at least this can end up calling proc_rwmem() to clear an earlier single-step simualted via a breakpoint). We only do one to avoid races. Also, by making the EINVAL error for unknown requests be part of the default: case in the switch, the various switch cases can now just break out to return which removes a _lot_ of duplicated PRELE and proc unlocks, etc. Also, it fixes at least one bug where a LWP ptrace command could return EINVAL with the proc lock still held. - Changed the locking for ptrace_single_step(), ptrace_set_pc(), and ptrace_clear_single_step() to always be called with the proc lock held (it was a mixed bag previously). Alpha and arm have to drop the lock while the mess around with breakpoints, but other archs avoid extra lock release/acquires in ptrace(). I did have to fix a couple of other consumers in kern_kse and a few other places to hold the proc lock and PHOLD. Tested by: ps (1 mostly, but some bits of 2-4 as well) MFC after: 1 week
2006-02-22 18:57:50 +00:00
/*
* Wakeup anyone in procfs' PIOCWAIT. They should have a hold
* on our vmspace, so we should block below until they have
* released their reference to us. Note that if they have
* requested S_EXIT stops we will block here until they ack
* via PIOCCONT.
*/
_STOPEVENT(p, S_EXIT, 0);
Close some races between procfs/ptrace and exit(2): - Reorder the events in exit(2) slightly so that we trigger the S_EXIT stop event earlier. After we have signalled that, we set P_WEXIT and then wait for any processes with a hold on the vmspace via PHOLD to release it. PHOLD now KASSERT()'s that P_WEXIT is clear when it is invoked, and PRELE now does a wakeup if P_WEXIT is set and p_lock drops to zero. - Change proc_rwmem() to require that the processing read from has its vmspace held via PHOLD by the caller and get rid of all the junk to screw around with the vmspace reference count as we no longer need it. - In ptrace() and pseudofs(), treat a process with P_WEXIT set as if it doesn't exist. - Only do one PHOLD in kern_ptrace() now, and do it earlier so it covers FIX_SSTEP() (since on alpha at least this can end up calling proc_rwmem() to clear an earlier single-step simualted via a breakpoint). We only do one to avoid races. Also, by making the EINVAL error for unknown requests be part of the default: case in the switch, the various switch cases can now just break out to return which removes a _lot_ of duplicated PRELE and proc unlocks, etc. Also, it fixes at least one bug where a LWP ptrace command could return EINVAL with the proc lock still held. - Changed the locking for ptrace_single_step(), ptrace_set_pc(), and ptrace_clear_single_step() to always be called with the proc lock held (it was a mixed bag previously). Alpha and arm have to drop the lock while the mess around with breakpoints, but other archs avoid extra lock release/acquires in ptrace(). I did have to fix a couple of other consumers in kern_kse and a few other places to hold the proc lock and PHOLD. Tested by: ps (1 mostly, but some bits of 2-4 as well) MFC after: 1 week
2006-02-22 18:57:50 +00:00
/*
* Ignore any pending request to stop due to a stop signal.
* Once P_WEXIT is set, future requests will be ignored as
* well.
*/
p->p_flag &= ~P_STOPPED_SIG;
KASSERT(!P_SHOULDSTOP(p), ("exiting process is stopped"));
Close some races between procfs/ptrace and exit(2): - Reorder the events in exit(2) slightly so that we trigger the S_EXIT stop event earlier. After we have signalled that, we set P_WEXIT and then wait for any processes with a hold on the vmspace via PHOLD to release it. PHOLD now KASSERT()'s that P_WEXIT is clear when it is invoked, and PRELE now does a wakeup if P_WEXIT is set and p_lock drops to zero. - Change proc_rwmem() to require that the processing read from has its vmspace held via PHOLD by the caller and get rid of all the junk to screw around with the vmspace reference count as we no longer need it. - In ptrace() and pseudofs(), treat a process with P_WEXIT set as if it doesn't exist. - Only do one PHOLD in kern_ptrace() now, and do it earlier so it covers FIX_SSTEP() (since on alpha at least this can end up calling proc_rwmem() to clear an earlier single-step simualted via a breakpoint). We only do one to avoid races. Also, by making the EINVAL error for unknown requests be part of the default: case in the switch, the various switch cases can now just break out to return which removes a _lot_ of duplicated PRELE and proc unlocks, etc. Also, it fixes at least one bug where a LWP ptrace command could return EINVAL with the proc lock still held. - Changed the locking for ptrace_single_step(), ptrace_set_pc(), and ptrace_clear_single_step() to always be called with the proc lock held (it was a mixed bag previously). Alpha and arm have to drop the lock while the mess around with breakpoints, but other archs avoid extra lock release/acquires in ptrace(). I did have to fix a couple of other consumers in kern_kse and a few other places to hold the proc lock and PHOLD. Tested by: ps (1 mostly, but some bits of 2-4 as well) MFC after: 1 week
2006-02-22 18:57:50 +00:00
/*
* Note that we are exiting and do another wakeup of anyone in
* PIOCWAIT in case they aren't listening for S_EXIT stops or
* decided to wait again after we told them we are exiting.
*/
p->p_flag |= P_WEXIT;
Close some races between procfs/ptrace and exit(2): - Reorder the events in exit(2) slightly so that we trigger the S_EXIT stop event earlier. After we have signalled that, we set P_WEXIT and then wait for any processes with a hold on the vmspace via PHOLD to release it. PHOLD now KASSERT()'s that P_WEXIT is clear when it is invoked, and PRELE now does a wakeup if P_WEXIT is set and p_lock drops to zero. - Change proc_rwmem() to require that the processing read from has its vmspace held via PHOLD by the caller and get rid of all the junk to screw around with the vmspace reference count as we no longer need it. - In ptrace() and pseudofs(), treat a process with P_WEXIT set as if it doesn't exist. - Only do one PHOLD in kern_ptrace() now, and do it earlier so it covers FIX_SSTEP() (since on alpha at least this can end up calling proc_rwmem() to clear an earlier single-step simualted via a breakpoint). We only do one to avoid races. Also, by making the EINVAL error for unknown requests be part of the default: case in the switch, the various switch cases can now just break out to return which removes a _lot_ of duplicated PRELE and proc unlocks, etc. Also, it fixes at least one bug where a LWP ptrace command could return EINVAL with the proc lock still held. - Changed the locking for ptrace_single_step(), ptrace_set_pc(), and ptrace_clear_single_step() to always be called with the proc lock held (it was a mixed bag previously). Alpha and arm have to drop the lock while the mess around with breakpoints, but other archs avoid extra lock release/acquires in ptrace(). I did have to fix a couple of other consumers in kern_kse and a few other places to hold the proc lock and PHOLD. Tested by: ps (1 mostly, but some bits of 2-4 as well) MFC after: 1 week
2006-02-22 18:57:50 +00:00
wakeup(&p->p_stype);
/*
* Wait for any processes that have a hold on our vmspace to
* release their reference.
*/
while (p->p_lock > 0)
msleep(&p->p_lock, &p->p_mtx, PWAIT, "exithold", 0);
PROC_UNLOCK(p);
/* Drain the limit callout while we don't have the proc locked */
callout_drain(&p->p_limco);
#ifdef AUDIT
/*
* The Sun BSM exit token contains two components: an exit status as
* passed to exit(), and a return value to indicate what sort of exit
* it was. The exit status is WEXITSTATUS(rv), but it's not clear
* what the return value is.
*/
AUDIT_ARG_EXIT(rval, 0);
AUDIT_SYSCALL_EXIT(0, td);
#endif
/* Are we a task leader with peers? */
if (p->p_peers != NULL && p == p->p_leader) {
mtx_lock(&ppeers_lock);
q = p->p_peers;
while (q != NULL) {
PROC_LOCK(q);
kern_psignal(q, SIGKILL);
PROC_UNLOCK(q);
q = q->p_peers;
}
2003-03-19 00:33:38 +00:00
while (p->p_peers != NULL)
msleep(p, &ppeers_lock, PWAIT, "exit1", 0);
mtx_unlock(&ppeers_lock);
}
2003-03-19 00:33:38 +00:00
/*
* Check if any loadable modules need anything done at process exit.
* E.g. SYSV IPC stuff.
* Event handler could change exit status.
* XXX what if one of these generates an error?
*/
EVENTHANDLER_INVOKE(process_exit, p);
1994-05-24 10:09:53 +00:00
/*
* If parent is waiting for us to exit or exec,
* P_PPWAIT is set; we will wakeup the parent below.
*/
PROC_LOCK(p);
stopprofclock(p);
p->p_flag &= ~(P_TRACED | P_PPWAIT | P_PPTRACE);
p->p_ptevents = 0;
/*
* Stop the real interval timer. If the handler is currently
* executing, prevent it from rearming itself and let it finish.
*/
if (timevalisset(&p->p_realtimer.it_value) &&
_callout_stop_safe(&p->p_itcallout, CS_EXECUTING, NULL) == 0) {
timevalclear(&p->p_realtimer.it_interval);
msleep(&p->p_itcallout, &p->p_mtx, PWAIT, "ritwait", 0);
KASSERT(!timevalisset(&p->p_realtimer.it_value),
("realtime timer is still armed"));
}
Add implementation of robust mutexes, hopefully close enough to the intention of the POSIX IEEE Std 1003.1TM-2008/Cor 1-2013. A robust mutex is guaranteed to be cleared by the system upon either thread or process owner termination while the mutex is held. The next mutex locker is then notified about inconsistent mutex state and can execute (or abandon) corrective actions. The patch mostly consists of small changes here and there, adding neccessary checks for the inconsistent and abandoned conditions into existing paths. Additionally, the thread exit handler was extended to iterate over the userspace-maintained list of owned robust mutexes, unlocking and marking as terminated each of them. The list of owned robust mutexes cannot be maintained atomically synchronous with the mutex lock state (it is possible in kernel, but is too expensive). Instead, for the duration of lock or unlock operation, the current mutex is remembered in a special slot that is also checked by the kernel at thread termination. Kernel must be aware about the per-thread location of the heads of robust mutex lists and the current active mutex slot. When a thread touches a robust mutex for the first time, a new umtx op syscall is issued which informs about location of lists heads. The umtx sleep queues for PP and PI mutexes are split between non-robust and robust. Somewhat unrelated changes in the patch: 1. Style. 2. The fix for proper tdfind() call use in umtxq_sleep_pi() for shared pi mutexes. 3. Removal of the userspace struct pthread_mutex m_owner field. 4. The sysctl kern.ipc.umtx_vnode_persistent is added, which controls the lifetime of the shared mutex associated with a vnode' page. Reviewed by: jilles (previous version, supposedly the objection was fixed) Discussed with: brooks, Martin Simmons <martin@lispworks.com> (some aspects) Tested by: pho Sponsored by: The FreeBSD Foundation
2016-05-17 09:56:22 +00:00
PROC_UNLOCK(p);
1994-05-24 10:09:53 +00:00
Add implementation of robust mutexes, hopefully close enough to the intention of the POSIX IEEE Std 1003.1TM-2008/Cor 1-2013. A robust mutex is guaranteed to be cleared by the system upon either thread or process owner termination while the mutex is held. The next mutex locker is then notified about inconsistent mutex state and can execute (or abandon) corrective actions. The patch mostly consists of small changes here and there, adding neccessary checks for the inconsistent and abandoned conditions into existing paths. Additionally, the thread exit handler was extended to iterate over the userspace-maintained list of owned robust mutexes, unlocking and marking as terminated each of them. The list of owned robust mutexes cannot be maintained atomically synchronous with the mutex lock state (it is possible in kernel, but is too expensive). Instead, for the duration of lock or unlock operation, the current mutex is remembered in a special slot that is also checked by the kernel at thread termination. Kernel must be aware about the per-thread location of the heads of robust mutex lists and the current active mutex slot. When a thread touches a robust mutex for the first time, a new umtx op syscall is issued which informs about location of lists heads. The umtx sleep queues for PP and PI mutexes are split between non-robust and robust. Somewhat unrelated changes in the patch: 1. Style. 2. The fix for proper tdfind() call use in umtxq_sleep_pi() for shared pi mutexes. 3. Removal of the userspace struct pthread_mutex m_owner field. 4. The sysctl kern.ipc.umtx_vnode_persistent is added, which controls the lifetime of the shared mutex associated with a vnode' page. Reviewed by: jilles (previous version, supposedly the objection was fixed) Discussed with: brooks, Martin Simmons <martin@lispworks.com> (some aspects) Tested by: pho Sponsored by: The FreeBSD Foundation
2016-05-17 09:56:22 +00:00
umtx_thread_exit(td);
/*
* Reset any sigio structures pointing to us as a result of
* F_SETOWN with our pid.
*/
funsetownlst(&p->p_sigiolst);
/*
* If this process has an nlminfo data area (for lockd), release it
*/
if (nlminfo_release_p != NULL && p->p_nlminfo != NULL)
(*nlminfo_release_p)(p);
1994-05-24 10:09:53 +00:00
/*
* Close open files and release open-file table.
* This may block!
*/
Merge Capsicum overhaul: - Capability is no longer separate descriptor type. Now every descriptor has set of its own capability rights. - The cap_new(2) system call is left, but it is no longer documented and should not be used in new code. - The new syscall cap_rights_limit(2) should be used instead of cap_new(2), which limits capability rights of the given descriptor without creating a new one. - The cap_getrights(2) syscall is renamed to cap_rights_get(2). - If CAP_IOCTL capability right is present we can further reduce allowed ioctls list with the new cap_ioctls_limit(2) syscall. List of allowed ioctls can be retrived with cap_ioctls_get(2) syscall. - If CAP_FCNTL capability right is present we can further reduce fcntls that can be used with the new cap_fcntls_limit(2) syscall and retrive them with cap_fcntls_get(2). - To support ioctl and fcntl white-listing the filedesc structure was heavly modified. - The audit subsystem, kdump and procstat tools were updated to recognize new syscalls. - Capability rights were revised and eventhough I tried hard to provide backward API and ABI compatibility there are some incompatible changes that are described in detail below: CAP_CREATE old behaviour: - Allow for openat(2)+O_CREAT. - Allow for linkat(2). - Allow for symlinkat(2). CAP_CREATE new behaviour: - Allow for openat(2)+O_CREAT. Added CAP_LINKAT: - Allow for linkat(2). ABI: Reuses CAP_RMDIR bit. - Allow to be target for renameat(2). Added CAP_SYMLINKAT: - Allow for symlinkat(2). Removed CAP_DELETE. Old behaviour: - Allow for unlinkat(2) when removing non-directory object. - Allow to be source for renameat(2). Removed CAP_RMDIR. Old behaviour: - Allow for unlinkat(2) when removing directory. Added CAP_RENAMEAT: - Required for source directory for the renameat(2) syscall. Added CAP_UNLINKAT (effectively it replaces CAP_DELETE and CAP_RMDIR): - Allow for unlinkat(2) on any object. - Required if target of renameat(2) exists and will be removed by this call. Removed CAP_MAPEXEC. CAP_MMAP old behaviour: - Allow for mmap(2) with any combination of PROT_NONE, PROT_READ and PROT_WRITE. CAP_MMAP new behaviour: - Allow for mmap(2)+PROT_NONE. Added CAP_MMAP_R: - Allow for mmap(PROT_READ). Added CAP_MMAP_W: - Allow for mmap(PROT_WRITE). Added CAP_MMAP_X: - Allow for mmap(PROT_EXEC). Added CAP_MMAP_RW: - Allow for mmap(PROT_READ | PROT_WRITE). Added CAP_MMAP_RX: - Allow for mmap(PROT_READ | PROT_EXEC). Added CAP_MMAP_WX: - Allow for mmap(PROT_WRITE | PROT_EXEC). Added CAP_MMAP_RWX: - Allow for mmap(PROT_READ | PROT_WRITE | PROT_EXEC). Renamed CAP_MKDIR to CAP_MKDIRAT. Renamed CAP_MKFIFO to CAP_MKFIFOAT. Renamed CAP_MKNODE to CAP_MKNODEAT. CAP_READ old behaviour: - Allow pread(2). - Disallow read(2), readv(2) (if there is no CAP_SEEK). CAP_READ new behaviour: - Allow read(2), readv(2). - Disallow pread(2) (CAP_SEEK was also required). CAP_WRITE old behaviour: - Allow pwrite(2). - Disallow write(2), writev(2) (if there is no CAP_SEEK). CAP_WRITE new behaviour: - Allow write(2), writev(2). - Disallow pwrite(2) (CAP_SEEK was also required). Added convinient defines: #define CAP_PREAD (CAP_SEEK | CAP_READ) #define CAP_PWRITE (CAP_SEEK | CAP_WRITE) #define CAP_MMAP_R (CAP_MMAP | CAP_SEEK | CAP_READ) #define CAP_MMAP_W (CAP_MMAP | CAP_SEEK | CAP_WRITE) #define CAP_MMAP_X (CAP_MMAP | CAP_SEEK | 0x0000000000000008ULL) #define CAP_MMAP_RW (CAP_MMAP_R | CAP_MMAP_W) #define CAP_MMAP_RX (CAP_MMAP_R | CAP_MMAP_X) #define CAP_MMAP_WX (CAP_MMAP_W | CAP_MMAP_X) #define CAP_MMAP_RWX (CAP_MMAP_R | CAP_MMAP_W | CAP_MMAP_X) #define CAP_RECV CAP_READ #define CAP_SEND CAP_WRITE #define CAP_SOCK_CLIENT \ (CAP_CONNECT | CAP_GETPEERNAME | CAP_GETSOCKNAME | CAP_GETSOCKOPT | \ CAP_PEELOFF | CAP_RECV | CAP_SEND | CAP_SETSOCKOPT | CAP_SHUTDOWN) #define CAP_SOCK_SERVER \ (CAP_ACCEPT | CAP_BIND | CAP_GETPEERNAME | CAP_GETSOCKNAME | \ CAP_GETSOCKOPT | CAP_LISTEN | CAP_PEELOFF | CAP_RECV | CAP_SEND | \ CAP_SETSOCKOPT | CAP_SHUTDOWN) Added defines for backward API compatibility: #define CAP_MAPEXEC CAP_MMAP_X #define CAP_DELETE CAP_UNLINKAT #define CAP_MKDIR CAP_MKDIRAT #define CAP_RMDIR CAP_UNLINKAT #define CAP_MKFIFO CAP_MKFIFOAT #define CAP_MKNOD CAP_MKNODAT #define CAP_SOCK_ALL (CAP_SOCK_CLIENT | CAP_SOCK_SERVER) Sponsored by: The FreeBSD Foundation Reviewed by: Christoph Mallon <christoph.mallon@gmx.de> Many aspects discussed with: rwatson, benl, jonathan ABI compatibility discussed with: kib
2013-03-02 00:53:12 +00:00
fdescfree(td);
1994-05-24 10:09:53 +00:00
/*
* If this thread tickled GEOM, we need to wait for the giggling to
* stop before we return to userland
*/
if (td->td_pflags & TDP_GEOM)
g_waitidle();
/*
* Remove ourself from our leader's peer list and wake our leader.
*/
if (p->p_leader->p_peers != NULL) {
mtx_lock(&ppeers_lock);
if (p->p_leader->p_peers != NULL) {
q = p->p_leader;
while (q->p_peers != p)
q = q->p_peers;
q->p_peers = p->p_peers;
wakeup(p->p_leader);
}
mtx_unlock(&ppeers_lock);
}
vmspace_exit(td);
killjobc();
(void)acct_process(td);
Integrate the new MPSAFE TTY layer to the FreeBSD operating system. The last half year I've been working on a replacement TTY layer for the FreeBSD kernel. The new TTY layer was designed to improve the following: - Improved driver model: The old TTY layer has a driver model that is not abstract enough to make it friendly to use. A good example is the output path, where the device drivers directly access the output buffers. This means that an in-kernel PPP implementation must always convert network buffers into TTY buffers. If a PPP implementation would be built on top of the new TTY layer (still needs a hooks layer, though), it would allow the PPP implementation to directly hand the data to the TTY driver. - Improved hotplugging: With the old TTY layer, it isn't entirely safe to destroy TTY's from the system. This implementation has a two-step destructing design, where the driver first abandons the TTY. After all threads have left the TTY, the TTY layer calls a routine in the driver, which can be used to free resources (unit numbers, etc). The pts(4) driver also implements this feature, which means posix_openpt() will now return PTY's that are created on the fly. - Improved performance: One of the major improvements is the per-TTY mutex, which is expected to improve scalability when compared to the old Giant locking. Another change is the unbuffered copying to userspace, which is both used on TTY device nodes and PTY masters. Upgrading should be quite straightforward. Unlike previous versions, existing kernel configuration files do not need to be changed, except when they reference device drivers that are listed in UPDATING. Obtained from: //depot/projects/mpsafetty/... Approved by: philip (ex-mentor) Discussed: on the lists, at BSDCan, at the DevSummit Sponsored by: Snow B.V., the Netherlands dcons(4) fixed by: kan
2008-08-20 08:31:58 +00:00
1994-05-24 10:09:53 +00:00
#ifdef KTRACE
ktrprocexit(td);
1994-05-24 10:09:53 +00:00
#endif
/*
* Release reference to text vnode
*/
if (p->p_textvp != NULL) {
vrele(p->p_textvp);
p->p_textvp = NULL;
}
/*
* Release our limits structure.
*/
lim_free(p->p_limit);
Locking for the per-process resource limits structure. - struct plimit includes a mutex to protect a reference count. The plimit structure is treated similarly to struct ucred in that is is always copy on write, so having a reference to a structure is sufficient to read from it without needing a further lock. - The proc lock protects the p_limit pointer and must be held while reading limits from a process to keep the limit structure from changing out from under you while reading from it. - Various global limits that are ints are not protected by a lock since int writes are atomic on all the archs we support and thus a lock wouldn't buy us anything. - All accesses to individual resource limits from a process are abstracted behind a simple lim_rlimit(), lim_max(), and lim_cur() API that return either an rlimit, or the current or max individual limit of the specified resource from a process. - dosetrlimit() was renamed to kern_setrlimit() to match existing style of other similar syscall helper functions. - The alpha OSF/1 compat layer no longer calls getrlimit() and setrlimit() (it didn't used the stackgap when it should have) but uses lim_rlimit() and kern_setrlimit() instead. - The svr4 compat no longer uses the stackgap for resource limits calls, but uses lim_rlimit() and kern_setrlimit() instead. - The ibcs2 compat no longer uses the stackgap for resource limits. It also no longer uses the stackgap for accessing sysctl's for the ibcs2_sysconf() syscall but uses kernel_sysctl() instead. As a result, ibcs2_sysconf() no longer needs Giant. - The p_rlimit macro no longer exists. Submitted by: mtm (mostly, I only did a few cleanups and catchups) Tested on: i386 Compiled on: alpha, amd64
2004-02-04 21:52:57 +00:00
p->p_limit = NULL;
tidhash_remove(td);
1994-05-24 10:09:53 +00:00
/*
* Remove proc from allproc queue and pidhash chain.
* Place onto zombproc. Unlink from parent's child list.
*/
sx_xlock(&allproc_lock);
LIST_REMOVE(p, p_list);
LIST_INSERT_HEAD(&zombproc, p, p_list);
LIST_REMOVE(p, p_hash);
sx_xunlock(&allproc_lock);
1994-05-24 10:09:53 +00:00
/*
* Call machine-dependent code to release any
* machine-dependent resources other than the address space.
* The address space is released by "vmspace_exitfree(p)" in
* vm_waitproc().
*/
cpu_exit(td);
WITNESS_WARN(WARN_PANIC, NULL, "process (pid %d) exiting", p->p_pid);
2006-02-03 21:09:40 +00:00
/*
* Reparent all children processes:
* - traced ones to the original parent (or init if we are that parent)
* - the rest to init
2006-02-03 21:09:40 +00:00
*/
sx_xlock(&proctree_lock);
q = LIST_FIRST(&p->p_children);
if (q != NULL) /* only need this if any child is S_ZOMB */
wakeup(q->p_reaper);
for (; q != NULL; q = nq) {
nq = LIST_NEXT(q, p_sibling);
PROC_LOCK(q);
q->p_sigparent = SIGCHLD;
if (!(q->p_flag & P_TRACED)) {
proc_reparent(q, q->p_reaper);
if (q->p_state == PRS_ZOMBIE) {
PROC_LOCK(q->p_reaper);
pksignal(q->p_reaper, SIGCHLD, q->p_ksi);
PROC_UNLOCK(q->p_reaper);
}
} else {
/*
* Traced processes are killed since their existence
* means someone is screwing up.
*/
t = proc_realparent(q);
if (t == p) {
proc_reparent(q, q->p_reaper);
} else {
PROC_LOCK(t);
proc_reparent(q, t);
PROC_UNLOCK(t);
}
/*
* Since q was found on our children list, the
* proc_reparent() call moved q to the orphan
* list due to present P_TRACED flag. Clear
* orphan link for q now while q is locked.
*/
clear_orphan(q);
q->p_flag &= ~(P_TRACED | P_STOPPED_TRACE);
When a debugger attaches to the process, SIGSTOP is sent to the target. Due to a way issignal() selects the next signal to deliver and report, if the simultaneous or already pending another signal exists, that signal might be reported by the next waitpid(2) call. This causes minor annoyance for debuggers, which must be prepared to take any signal as the first event, then filter SIGSTOP later. More importantly, for tools like gcore(1), which attach and then detach without processing events, SIGSTOP might leak to be delivered after PT_DETACH. This results in the process being unintentionally stopped after detach, which is fatal for automatic tools. The solution is to force SIGSTOP to be the first signal reported after the attach. Attach code is modified to set P2_PTRACE_FSTP to indicate that the attaching ritual was not yet finished, and issignal() prefers SIGSTOP in that condition. Also, the thread which handles P2_PTRACE_FSTP is made to guarantee to own p_xthread during the first waitpid(2). All that ensures that SIGSTOP is consumed first. Additionally, if P2_PTRACE_FSTP is still set on detach, which means that waitpid(2) was not called at all, SIGSTOP is removed from the queue, ensuring that the process is resumed on detach. In issignal(), when acting on STOPing signals, remove the signal from queue before suspending. Otherwise parallel attach could result in ptracestop() acting on that STOP as if it was the STOP signal from the attach. Then SIGSTOP from attach leaks again. As a minor refactoring, some bits of the common attach code is moved to new helper proc_set_traced(). Reported by: markj Reviewed by: jhb, markj Tested by: pho Sponsored by: The FreeBSD Foundation MFC after: 2 weeks Differential revision: https://reviews.freebsd.org/D7256
2016-07-28 08:41:13 +00:00
q->p_flag2 &= ~P2_PTRACE_FSTP;
q->p_ptevents = 0;
When a debugger attaches to the process, SIGSTOP is sent to the target. Due to a way issignal() selects the next signal to deliver and report, if the simultaneous or already pending another signal exists, that signal might be reported by the next waitpid(2) call. This causes minor annoyance for debuggers, which must be prepared to take any signal as the first event, then filter SIGSTOP later. More importantly, for tools like gcore(1), which attach and then detach without processing events, SIGSTOP might leak to be delivered after PT_DETACH. This results in the process being unintentionally stopped after detach, which is fatal for automatic tools. The solution is to force SIGSTOP to be the first signal reported after the attach. Attach code is modified to set P2_PTRACE_FSTP to indicate that the attaching ritual was not yet finished, and issignal() prefers SIGSTOP in that condition. Also, the thread which handles P2_PTRACE_FSTP is made to guarantee to own p_xthread during the first waitpid(2). All that ensures that SIGSTOP is consumed first. Additionally, if P2_PTRACE_FSTP is still set on detach, which means that waitpid(2) was not called at all, SIGSTOP is removed from the queue, ensuring that the process is resumed on detach. In issignal(), when acting on STOPing signals, remove the signal from queue before suspending. Otherwise parallel attach could result in ptracestop() acting on that STOP as if it was the STOP signal from the attach. Then SIGSTOP from attach leaks again. As a minor refactoring, some bits of the common attach code is moved to new helper proc_set_traced(). Reported by: markj Reviewed by: jhb, markj Tested by: pho Sponsored by: The FreeBSD Foundation MFC after: 2 weeks Differential revision: https://reviews.freebsd.org/D7256
2016-07-28 08:41:13 +00:00
FOREACH_THREAD_IN_PROC(q, tdt) {
tdt->td_dbgflags &= ~(TDB_SUSPEND | TDB_XSIG |
TDB_FSTP);
}
kern_psignal(q, SIGKILL);
}
PROC_UNLOCK(q);
1994-05-24 10:09:53 +00:00
}
/*
* Also get rid of our orphans.
*/
while ((q = LIST_FIRST(&p->p_orphans)) != NULL) {
PROC_LOCK(q);
CTR2(KTR_PTRACE, "exit: pid %d, clearing orphan %d", p->p_pid,
q->p_pid);
clear_orphan(q);
PROC_UNLOCK(q);
}
/* Save exit status. */
PROC_LOCK(p);
Add code to support debugging threaded process. 1. Add tm_lwpid into kse_thr_mailbox to indicate which kernel thread current user thread is running on. Add tm_dflags into kse_thr_mailbox, the flags is written by debugger, it tells UTS and kernel what should be done when the process is being debugged, current, there two flags TMDF_SSTEP and TMDF_DONOTRUNUSER. TMDF_SSTEP is used to tell kernel to turn on single stepping, or turn off if it is not set. TMDF_DONOTRUNUSER is used to tell kernel to schedule upcall whenever possible, to UTS, it means do not run the user thread until debugger clears it, this behaviour is necessary because gdb wants to resume only one thread when the thread's pc is at a breakpoint, and thread needs to go forward, in order to avoid other threads sneak pass the breakpoints, it needs to remove breakpoint, only wants one thread to go. Also, add km_lwp to kse_mailbox, the lwp id is copied to kse_thr_mailbox at context switch time when process is not being debugged, so when process is attached, debugger can map kernel thread to user thread. 2. Add p_xthread to proc strcuture and td_xsig to thread structure. p_xthread is used by a thread when it wants to report event to debugger, every thread can set the pointer, especially, when it is used in ptracestop, it is the last thread reporting event will win the race. Every thread has a td_xsig to exchange signal with debugger, thread uses TDF_XSIG flag to indicate it is reporting signal to debugger, if the flag is not cleared, thread will keep retrying until it is cleared by debugger, p_xthread may be used by debugger to indicate CURRENT thread. The p_xstat is still in proc structure to keep wait() to work, in future, we may just use td_xsig. 3. Add TDF_DBSUSPEND flag, the flag is used by debugger to suspend a thread. When process stops, debugger can set the flag for thread, thread will check the flag in thread_suspend_check, enters a loop, unless it is cleared by debugger, process is detached or process is existing. The flag is also checked in ptracestop, so debugger can temporarily suspend a thread even if the thread wants to exchange signal. 4. Current, in ptrace, we always resume all threads, but if a thread has already a TDF_DBSUSPEND flag set by debugger, it won't run. Encouraged by: marcel, julian, deischen
2004-07-13 07:20:10 +00:00
p->p_xthread = td;
/* Tell the prison that we are gone. */
prison_proc_free(p->p_ucred->cr_prison);
MFp4: Bring in updated jail support from bz_jail branch. This enhances the current jail implementation to permit multiple addresses per jail. In addtion to IPv4, IPv6 is supported as well. Due to updated checks it is even possible to have jails without an IP address at all, which basically gives one a chroot with restricted process view, no networking,.. SCTP support was updated and supports IPv6 in jails as well. Cpuset support permits jails to be bound to specific processor sets after creation. Jails can have an unrestricted (no duplicate protection, etc.) name in addition to the hostname. The jail name cannot be changed from within a jail and is considered to be used for management purposes or as audit-token in the future. DDB 'show jails' command was added to aid debugging. Proper compat support permits 32bit jail binaries to be used on 64bit systems to manage jails. Also backward compatibility was preserved where possible: for jail v1 syscalls, as well as with user space management utilities. Both jail as well as prison version were updated for the new features. A gap was intentionally left as the intermediate versions had been used by various patches floating around the last years. Bump __FreeBSD_version for the afore mentioned and in kernel changes. Special thanks to: - Pawel Jakub Dawidek (pjd) for his multi-IPv4 patches and Olivier Houchard (cognet) for initial single-IPv6 patches. - Jeff Roberson (jeff) and Randall Stewart (rrs) for their help, ideas and review on cpuset and SCTP support. - Robert Watson (rwatson) for lots and lots of help, discussions, suggestions and review of most of the patch at various stages. - John Baldwin (jhb) for his help. - Simon L. Nielsen (simon) as early adopter testing changes on cluster machines as well as all the testers and people who provided feedback the last months on freebsd-jail and other channels. - My employer, CK Software GmbH, for the support so I could work on this. Reviewed by: (see above) MFC after: 3 months (this is just so that I get the mail) X-MFC Before: 7.2-RELEASE if possible
2008-11-29 14:32:14 +00:00
#ifdef KDTRACE_HOOKS
/*
* Tell the DTrace fasttrap provider about the exit if it
* has declared an interest.
*/
if (dtrace_fasttrap_exit)
dtrace_fasttrap_exit(p);
#endif
/*
* Notify interested parties of our demise.
*/
When filt_proc() removes event from the knlist due to the process exiting (NOTE_EXIT->knlist_remove_inevent()), two things happen: - knote kn_knlist pointer is reset - INFLUX knote is removed from the process knlist. And, there are two consequences: - KN_LIST_UNLOCK() on such knote is nop - there is nothing which would block exit1() from processing past the knlist_destroy() (and knlist_destroy() resets knlist lock pointers). Both consequences result either in leaked process lock, or dereferencing NULL function pointers for locking. Handle this by stopping embedding the process knlist into struct proc. Instead, the knlist is allocated together with struct proc, but marked as autodestroy on the zombie reap, by knlist_detach() function. The knlist is freed when last kevent is removed from the list, in particular, at the zombie reap time if the list is empty. As result, the knlist_remove_inevent() is no longer needed and removed. Other changes: In filt_procattach(), clear NOTE_EXEC and NOTE_FORK desired events from kn_sfflags for knote registered by kernel to only get NOTE_CHILD notifications. The flags leak resulted in excessive NOTE_EXEC/NOTE_FORK reports. Fix immediate note activation in filt_procattach(). Condition should be either the immediate CHILD_NOTE activation, or immediate NOTE_EXIT report for the exiting process. In knote_fork(), do not perform racy check for KN_INFLUX before kq lock is taken. Besides being racy, it did not accounted for notes just added by scan (KN_SCAN). Some minor and incomplete style fixes. Analyzed and tested by: Eric Badger <eric@badgerio.us> Reviewed by: jhb Sponsored by: The FreeBSD Foundation MFC after: 2 weeks Approved by: re (gjb) Differential revision: https://reviews.freebsd.org/D6859
2016-06-27 21:52:17 +00:00
KNOTE_LOCKED(p->p_klist, NOTE_EXIT);
2004-09-22 15:24:33 +00:00
#ifdef KDTRACE_HOOKS
int reason = CLD_EXITED;
if (WCOREDUMP(signo))
reason = CLD_DUMPED;
else if (WIFSIGNALED(signo))
reason = CLD_KILLED;
SDT_PROBE1(proc, , , exit, reason);
#endif
1994-05-24 10:09:53 +00:00
/*
* If this is a process with a descriptor, we may not need to deliver
* a signal to the parent. proctree_lock is held over
* procdesc_exit() to serialize concurrent calls to close() and
* exit().
1994-05-24 10:09:53 +00:00
*/
if (p->p_procdesc == NULL || procdesc_exit(p)) {
/*
* Notify parent that we're gone. If parent has the
* PS_NOCLDWAIT flag set, or if the handler is set to SIG_IGN,
* notify process 1 instead (and hope it will handle this
* situation).
*/
PROC_LOCK(p->p_pptr);
mtx_lock(&p->p_pptr->p_sigacts->ps_mtx);
if (p->p_pptr->p_sigacts->ps_flag &
(PS_NOCLDWAIT | PS_CLDSIGIGN)) {
struct proc *pp;
mtx_unlock(&p->p_pptr->p_sigacts->ps_mtx);
pp = p->p_pptr;
PROC_UNLOCK(pp);
proc_reparent(p, p->p_reaper);
p->p_sigparent = SIGCHLD;
PROC_LOCK(p->p_pptr);
/*
* Notify parent, so in case he was wait(2)ing or
* executing waitpid(2) with our pid, he will
* continue.
*/
wakeup(pp);
} else
mtx_unlock(&p->p_pptr->p_sigacts->ps_mtx);
if (p->p_pptr == p->p_reaper || p->p_pptr == initproc)
childproc_exited(p);
else if (p->p_sigparent != 0) {
if (p->p_sigparent == SIGCHLD)
childproc_exited(p);
else /* LINUX thread */
kern_psignal(p->p_pptr, p->p_sigparent);
}
} else
PROC_LOCK(p->p_pptr);
sx_xunlock(&proctree_lock);
/*
* The state PRS_ZOMBIE prevents other proesses from sending
* signal to the process, to avoid memory leak, we free memory
* for signal queue at the time when the state is set.
*/
sigqueue_flush(&p->p_sigqueue);
sigqueue_flush(&td->td_sigqueue);
/*
2003-04-17 22:22:47 +00:00
* We have to wait until after acquiring all locks before
* changing p_state. We need to avoid all possible context
* switches (including ones from blocking on a mutex) while
* marked as a zombie. We also have to set the zombie state
* before we release the parent process' proc lock to avoid
* a lost wakeup. So, we first call wakeup, then we grab the
* sched lock, update the state, and release the parent process'
* proc lock.
*/
wakeup(p->p_pptr);
cv_broadcast(&p->p_pwait);
sched_exit(p->p_pptr, td);
PROC_SLOCK(p);
p->p_state = PRS_ZOMBIE;
PROC_UNLOCK(p->p_pptr);
/*
* Save our children's rusage information in our exit rusage.
*/
PROC_STATLOCK(p);
ruadd(&p->p_ru, &p->p_rux, &p->p_stats->p_cru, &p->p_crux);
PROC_STATUNLOCK(p);
/*
* Make sure the scheduler takes this thread out of its tables etc.
* This will also release this thread's reference to the ucred.
2003-03-19 00:33:38 +00:00
* Other thread parts to release include pcb bits and such.
*/
thread_exit();
1994-05-24 10:09:53 +00:00
}
#ifndef _SYS_SYSPROTO_H_
struct abort2_args {
char *why;
int nargs;
void **args;
};
#endif
int
sys_abort2(struct thread *td, struct abort2_args *uap)
{
struct proc *p = td->td_proc;
struct sbuf *sb;
void *uargs[16];
int error, i, sig;
/*
* Do it right now so we can log either proper call of abort2(), or
* note, that invalid argument was passed. 512 is big enough to
* handle 16 arguments' descriptions with additional comments.
*/
sb = sbuf_new(NULL, NULL, 512, SBUF_FIXEDLEN);
sbuf_clear(sb);
sbuf_printf(sb, "%s(pid %d uid %d) aborted: ",
p->p_comm, p->p_pid, td->td_ucred->cr_uid);
/*
* Since we can't return from abort2(), send SIGKILL in cases, where
* abort2() was called improperly
*/
sig = SIGKILL;
/* Prevent from DoSes from user-space. */
if (uap->nargs < 0 || uap->nargs > 16)
goto out;
if (uap->nargs > 0) {
if (uap->args == NULL)
goto out;
error = copyin(uap->args, uargs, uap->nargs * sizeof(void *));
if (error != 0)
goto out;
}
/*
* Limit size of 'reason' string to 128. Will fit even when
* maximal number of arguments was chosen to be logged.
*/
if (uap->why != NULL) {
error = sbuf_copyin(sb, uap->why, 128);
if (error < 0)
goto out;
} else {
sbuf_printf(sb, "(null)");
}
if (uap->nargs > 0) {
sbuf_printf(sb, "(");
for (i = 0;i < uap->nargs; i++)
sbuf_printf(sb, "%s%p", i == 0 ? "" : ", ", uargs[i]);
sbuf_printf(sb, ")");
}
/*
* Final stage: arguments were proper, string has been
* successfully copied from userspace, and copying pointers
* from user-space succeed.
*/
sig = SIGABRT;
out:
if (sig == SIGKILL) {
sbuf_trim(sb);
sbuf_printf(sb, " (Reason text inaccessible)");
}
sbuf_cat(sb, "\n");
sbuf_finish(sb);
log(LOG_INFO, "%s", sbuf_data(sb));
sbuf_delete(sb);
exit1(td, 0, sig);
return (0);
}
#ifdef COMPAT_43
/*
* The dirty work is handled by kern_wait().
*/
int
2003-03-19 00:49:40 +00:00
owait(struct thread *td, struct owait_args *uap __unused)
1994-05-24 10:09:53 +00:00
{
int error, status;
1994-05-24 10:09:53 +00:00
error = kern_wait(td, WAIT_ANY, &status, 0, NULL);
if (error == 0)
td->td_retval[1] = status;
return (error);
1994-05-24 10:09:53 +00:00
}
#endif /* COMPAT_43 */
1994-05-24 10:09:53 +00:00
/*
* The dirty work is handled by kern_wait().
*/
int
sys_wait4(struct thread *td, struct wait4_args *uap)
1994-05-24 10:09:53 +00:00
{
Rework how we store process times in the kernel such that we always store the raw values including for child process statistics and only compute the system and user timevals on demand. - Fix the various kern_wait() syscall wrappers to only pass in a rusage pointer if they are going to use the result. - Add a kern_getrusage() function for the ABI syscalls to use so that they don't have to play stackgap games to call getrusage(). - Fix the svr4_sys_times() syscall to just call calcru() to calculate the times it needs rather than calling getrusage() twice with associated stackgap, etc. - Add a new rusage_ext structure to store raw time stats such as tick counts for user, system, and interrupt time as well as a bintime of the total runtime. A new p_rux field in struct proc replaces the same inline fields from struct proc (i.e. p_[isu]ticks, p_[isu]u, and p_runtime). A new p_crux field in struct proc contains the "raw" child time usage statistics. ruadd() has been changed to handle adding the associated rusage_ext structures as well as the values in rusage. Effectively, the values in rusage_ext replace the ru_utime and ru_stime values in struct rusage. These two fields in struct rusage are no longer used in the kernel. - calcru() has been split into a static worker function calcru1() that calculates appropriate timevals for user and system time as well as updating the rux_[isu]u fields of a passed in rusage_ext structure. calcru() uses a copy of the process' p_rux structure to compute the timevals after updating the runtime appropriately if any of the threads in that process are currently executing. It also now only locks sched_lock internally while doing the rux_runtime fixup. calcru() now only requires the caller to hold the proc lock and calcru1() only requires the proc lock internally. calcru() also no longer allows callers to ask for an interrupt timeval since none of them actually did. - calcru() now correctly handles threads executing on other CPUs. - A new calccru() function computes the child system and user timevals by calling calcru1() on p_crux. Note that this means that any code that wants child times must now call this function rather than reading from p_cru directly. This function also requires the proc lock. - This finishes the locking for rusage and friends so some of the Giant locks in exit1() and kern_wait() are now gone. - The locking in ttyinfo() has been tweaked so that a shared lock of the proctree lock is used to protect the process group rather than the process group lock. By holding this lock until the end of the function we now ensure that the process/thread that we pick to dump info about will no longer vanish while we are trying to output its info to the console. Submitted by: bde (mostly) MFC after: 1 month
2004-10-05 18:51:11 +00:00
struct rusage ru, *rup;
int error, status;
Rework how we store process times in the kernel such that we always store the raw values including for child process statistics and only compute the system and user timevals on demand. - Fix the various kern_wait() syscall wrappers to only pass in a rusage pointer if they are going to use the result. - Add a kern_getrusage() function for the ABI syscalls to use so that they don't have to play stackgap games to call getrusage(). - Fix the svr4_sys_times() syscall to just call calcru() to calculate the times it needs rather than calling getrusage() twice with associated stackgap, etc. - Add a new rusage_ext structure to store raw time stats such as tick counts for user, system, and interrupt time as well as a bintime of the total runtime. A new p_rux field in struct proc replaces the same inline fields from struct proc (i.e. p_[isu]ticks, p_[isu]u, and p_runtime). A new p_crux field in struct proc contains the "raw" child time usage statistics. ruadd() has been changed to handle adding the associated rusage_ext structures as well as the values in rusage. Effectively, the values in rusage_ext replace the ru_utime and ru_stime values in struct rusage. These two fields in struct rusage are no longer used in the kernel. - calcru() has been split into a static worker function calcru1() that calculates appropriate timevals for user and system time as well as updating the rux_[isu]u fields of a passed in rusage_ext structure. calcru() uses a copy of the process' p_rux structure to compute the timevals after updating the runtime appropriately if any of the threads in that process are currently executing. It also now only locks sched_lock internally while doing the rux_runtime fixup. calcru() now only requires the caller to hold the proc lock and calcru1() only requires the proc lock internally. calcru() also no longer allows callers to ask for an interrupt timeval since none of them actually did. - calcru() now correctly handles threads executing on other CPUs. - A new calccru() function computes the child system and user timevals by calling calcru1() on p_crux. Note that this means that any code that wants child times must now call this function rather than reading from p_cru directly. This function also requires the proc lock. - This finishes the locking for rusage and friends so some of the Giant locks in exit1() and kern_wait() are now gone. - The locking in ttyinfo() has been tweaked so that a shared lock of the proctree lock is used to protect the process group rather than the process group lock. By holding this lock until the end of the function we now ensure that the process/thread that we pick to dump info about will no longer vanish while we are trying to output its info to the console. Submitted by: bde (mostly) MFC after: 1 month
2004-10-05 18:51:11 +00:00
if (uap->rusage != NULL)
rup = &ru;
else
rup = NULL;
error = kern_wait(td, uap->pid, &status, uap->options, rup);
if (uap->status != NULL && error == 0 && td->td_retval[0] != 0)
error = copyout(&status, uap->status, sizeof(status));
if (uap->rusage != NULL && error == 0 && td->td_retval[0] != 0)
error = copyout(&ru, uap->rusage, sizeof(struct rusage));
return (error);
1994-05-24 10:09:53 +00:00
}
int
sys_wait6(struct thread *td, struct wait6_args *uap)
{
struct __wrusage wru, *wrup;
siginfo_t si, *sip;
idtype_t idtype;
id_t id;
int error, status;
idtype = uap->idtype;
id = uap->id;
if (uap->wrusage != NULL)
wrup = &wru;
else
wrup = NULL;
if (uap->info != NULL) {
sip = &si;
bzero(sip, sizeof(*sip));
} else
sip = NULL;
/*
* We expect all callers of wait6() to know about WEXITED and
* WTRAPPED.
*/
error = kern_wait6(td, idtype, id, &status, uap->options, wrup, sip);
if (uap->status != NULL && error == 0 && td->td_retval[0] != 0)
error = copyout(&status, uap->status, sizeof(status));
if (uap->wrusage != NULL && error == 0 && td->td_retval[0] != 0)
error = copyout(&wru, uap->wrusage, sizeof(wru));
if (uap->info != NULL && error == 0)
error = copyout(&si, uap->info, sizeof(si));
return (error);
}
/*
* Reap the remains of a zombie process and optionally return status and
* rusage. Asserts and will release both the proctree_lock and the process
* lock as part of its work.
*/
void
proc_reap(struct thread *td, struct proc *p, int *status, int options)
{
struct proc *q, *t;
sx_assert(&proctree_lock, SA_XLOCKED);
PROC_LOCK_ASSERT(p, MA_OWNED);
PROC_SLOCK_ASSERT(p, MA_OWNED);
KASSERT(p->p_state == PRS_ZOMBIE, ("proc_reap: !PRS_ZOMBIE"));
q = td->td_proc;
PROC_SUNLOCK(p);
if (status)
*status = KW_EXITCODE(p->p_xexit, p->p_xsig);
if (options & WNOWAIT) {
/*
* Only poll, returning the status. Caller does not wish to
* release the proc struct just yet.
*/
PROC_UNLOCK(p);
sx_xunlock(&proctree_lock);
return;
}
PROC_LOCK(q);
sigqueue_take(p->p_ksi);
PROC_UNLOCK(q);
/*
* If we got the child via a ptrace 'attach', we need to give it back
* to the old parent.
*/
if (p->p_oppid != 0 && p->p_oppid != p->p_pptr->p_pid) {
PROC_UNLOCK(p);
t = proc_realparent(p);
PROC_LOCK(t);
PROC_LOCK(p);
CTR2(KTR_PTRACE,
"wait: traced child %d moved back to parent %d", p->p_pid,
t->p_pid);
proc_reparent(p, t);
p->p_oppid = 0;
PROC_UNLOCK(p);
pksignal(t, SIGCHLD, p->p_ksi);
wakeup(t);
cv_broadcast(&p->p_pwait);
PROC_UNLOCK(t);
sx_xunlock(&proctree_lock);
return;
}
p->p_oppid = 0;
PROC_UNLOCK(p);
/*
* Remove other references to this process to ensure we have an
* exclusive reference.
*/
sx_xlock(&allproc_lock);
LIST_REMOVE(p, p_list); /* off zombproc */
sx_xunlock(&allproc_lock);
LIST_REMOVE(p, p_sibling);
reaper_abandon_children(p, true);
LIST_REMOVE(p, p_reapsibling);
PROC_LOCK(p);
clear_orphan(p);
PROC_UNLOCK(p);
leavepgrp(p);
if (p->p_procdesc != NULL)
procdesc_reap(p);
sx_xunlock(&proctree_lock);
When filt_proc() removes event from the knlist due to the process exiting (NOTE_EXIT->knlist_remove_inevent()), two things happen: - knote kn_knlist pointer is reset - INFLUX knote is removed from the process knlist. And, there are two consequences: - KN_LIST_UNLOCK() on such knote is nop - there is nothing which would block exit1() from processing past the knlist_destroy() (and knlist_destroy() resets knlist lock pointers). Both consequences result either in leaked process lock, or dereferencing NULL function pointers for locking. Handle this by stopping embedding the process knlist into struct proc. Instead, the knlist is allocated together with struct proc, but marked as autodestroy on the zombie reap, by knlist_detach() function. The knlist is freed when last kevent is removed from the list, in particular, at the zombie reap time if the list is empty. As result, the knlist_remove_inevent() is no longer needed and removed. Other changes: In filt_procattach(), clear NOTE_EXEC and NOTE_FORK desired events from kn_sfflags for knote registered by kernel to only get NOTE_CHILD notifications. The flags leak resulted in excessive NOTE_EXEC/NOTE_FORK reports. Fix immediate note activation in filt_procattach(). Condition should be either the immediate CHILD_NOTE activation, or immediate NOTE_EXIT report for the exiting process. In knote_fork(), do not perform racy check for KN_INFLUX before kq lock is taken. Besides being racy, it did not accounted for notes just added by scan (KN_SCAN). Some minor and incomplete style fixes. Analyzed and tested by: Eric Badger <eric@badgerio.us> Reviewed by: jhb Sponsored by: The FreeBSD Foundation MFC after: 2 weeks Approved by: re (gjb) Differential revision: https://reviews.freebsd.org/D6859
2016-06-27 21:52:17 +00:00
PROC_LOCK(p);
knlist_detach(p->p_klist);
p->p_klist = NULL;
PROC_UNLOCK(p);
/*
* Removal from allproc list and process group list paired with
* PROC_LOCK which was executed during that time should guarantee
* nothing can reach this process anymore. As such further locking
* is unnecessary.
*/
p->p_xexit = p->p_xsig = 0; /* XXX: why? */
PROC_LOCK(q);
ruadd(&q->p_stats->p_cru, &q->p_crux, &p->p_ru, &p->p_rux);
PROC_UNLOCK(q);
/*
* Decrement the count of procs running with this uid.
*/
(void)chgproccnt(p->p_ucred->cr_ruidinfo, -1, 0);
/*
* Destroy resource accounting information associated with the process.
*/
#ifdef RACCT
if (racct_enable) {
PROC_LOCK(p);
racct_sub(p, RACCT_NPROC, 1);
PROC_UNLOCK(p);
}
#endif
racct_proc_exit(p);
/*
* Free credentials, arguments, and sigacts.
*/
crfree(p->p_ucred);
proc_set_cred(p, NULL);
pargs_drop(p->p_args);
p->p_args = NULL;
sigacts_free(p->p_sigacts);
p->p_sigacts = NULL;
/*
* Do any thread-system specific cleanups.
*/
thread_wait(p);
/*
* Give vm and machine-dependent layer a chance to free anything that
* cpu_exit couldn't release while still running in process context.
*/
vm_waitproc(p);
#ifdef MAC
mac_proc_destroy(p);
#endif
Add an initial NUMA affinity/policy configuration for threads and processes. This is based on work done by jeff@ and jhb@, as well as the numa.diff patch that has been circulating when someone asks for first-touch NUMA on -10 or -11. * Introduce a simple set of VM policy and iterator types. * tie the policy types into the vm_phys path for now, mirroring how the initial first-touch allocation work was enabled. * add syscalls to control changing thread and process defaults. * add a global NUMA VM domain policy. * implement a simple cascade policy order - if a thread policy exists, use it; if a process policy exists, use it; use the default policy. * processes inherit policies from their parent processes, threads inherit policies from their parent threads. * add a simple tool (numactl) to query and modify default thread/process policities. * add documentation for the new syscalls, for numa and for numactl. * re-enable first touch NUMA again by default, as now policies can be set in a variety of methods. This is only relevant for very specific workloads. This doesn't pretend to be a final NUMA solution. The previous defaults in -HEAD (with MAXMEMDOM set) can be achieved by 'sysctl vm.default_policy=rr'. This is only relevant if MAXMEMDOM is set to something other than 1. Ie, if you're using GENERIC or a modified kernel with non-NUMA, then this is a glorified no-op for you. Thank you to Norse Corp for giving me access to rather large (for FreeBSD!) NUMA machines in order to develop and verify this. Thank you to Dell for providing me with dual socket sandybridge and westmere v3 hardware to do NUMA development with. Thank you to Scott Long at Netflix for providing me with access to the two-socket, four-domain haswell v3 hardware. Thank you to Peter Holm for running the stress testing suite against the NUMA branch during various stages of development! Tested: * MIPS (regression testing; non-NUMA) * i386 (regression testing; non-NUMA GENERIC) * amd64 (regression testing; non-NUMA GENERIC) * westmere, 2 socket (thankyou norse!) * sandy bridge, 2 socket (thankyou dell!) * ivy bridge, 2 socket (thankyou norse!) * westmere-EX, 4 socket / 1TB RAM (thankyou norse!) * haswell, 2 socket (thankyou norse!) * haswell v3, 2 socket (thankyou dell) * haswell v3, 2x18 core (thankyou scott long / netflix!) * Peter Holm ran a stress test suite on this work and found one issue, but has not been able to verify it (it doesn't look NUMA related, and he only saw it once over many testing runs.) * I've tested bhyve instances running in fixed NUMA domains and cpusets; all seems to work correctly. Verified: * intel-pcm - pcm-numa.x and pcm-memory.x, whilst selecting different NUMA policies for processes under test. Review: This was reviewed through phabricator (https://reviews.freebsd.org/D2559) as well as privately and via emails to freebsd-arch@. The git history with specific attributes is available at https://github.com/erikarn/freebsd/ in the NUMA branch (https://github.com/erikarn/freebsd/compare/local/adrian_numa_policy). This has been reviewed by a number of people (stas, rpaulo, kib, ngie, wblock) but not achieved a clear consensus. My hope is that with further exposure and testing more functionality can be implemented and evaluated. Notes: * The VM doesn't handle unbalanced domains very well, and if you have an overly unbalanced memory setup whilst under high memory pressure, VM page allocation may fail leading to a kernel panic. This was a problem in the past, but it's much more easily triggered now with these tools. * This work only controls the path through vm_phys; it doesn't yet strongly/predictably affect contigmalloc, KVA placement, UMA, etc. So, driver placement of memory isn't really guaranteed in any way. That's next on my plate. Sponsored by: Norse Corp, Inc.; Dell
2015-07-11 15:21:37 +00:00
/*
* Free any domain policy that's still hiding around.
*/
vm_domain_policy_cleanup(&p->p_vm_dom_policy);
KASSERT(FIRST_THREAD_IN_PROC(p),
("proc_reap: no residual thread!"));
uma_zfree(proc_zone, p);
atomic_add_int(&nprocs, -1);
}
static int
proc_to_reap(struct thread *td, struct proc *p, idtype_t idtype, id_t id,
int *status, int options, struct __wrusage *wrusage, siginfo_t *siginfo,
int check_only)
{
struct rusage *rup;
sx_assert(&proctree_lock, SA_XLOCKED);
PROC_LOCK(p);
switch (idtype) {
case P_ALL:
if (p->p_procdesc != NULL) {
PROC_UNLOCK(p);
return (0);
}
break;
case P_PID:
if (p->p_pid != (pid_t)id) {
PROC_UNLOCK(p);
return (0);
}
break;
case P_PGID:
if (p->p_pgid != (pid_t)id) {
PROC_UNLOCK(p);
return (0);
}
break;
case P_SID:
if (p->p_session->s_sid != (pid_t)id) {
PROC_UNLOCK(p);
return (0);
}
break;
case P_UID:
if (p->p_ucred->cr_uid != (uid_t)id) {
PROC_UNLOCK(p);
return (0);
}
break;
case P_GID:
if (p->p_ucred->cr_gid != (gid_t)id) {
PROC_UNLOCK(p);
return (0);
}
break;
case P_JAILID:
if (p->p_ucred->cr_prison->pr_id != (int)id) {
PROC_UNLOCK(p);
return (0);
}
break;
/*
* It seems that the thread structures get zeroed out
* at process exit. This makes it impossible to
* support P_SETID, P_CID or P_CPUID.
*/
default:
PROC_UNLOCK(p);
return (0);
}
if (p_canwait(td, p)) {
PROC_UNLOCK(p);
return (0);
}
if (((options & WEXITED) == 0) && (p->p_state == PRS_ZOMBIE)) {
PROC_UNLOCK(p);
return (0);
}
/*
* This special case handles a kthread spawned by linux_clone
* (see linux_misc.c). The linux_wait4 and linux_waitpid
* functions need to be able to distinguish between waiting
* on a process and waiting on a thread. It is a thread if
* p_sigparent is not SIGCHLD, and the WLINUXCLONE option
* signifies we want to wait for threads and not processes.
*/
if ((p->p_sigparent != SIGCHLD) ^
((options & WLINUXCLONE) != 0)) {
PROC_UNLOCK(p);
return (0);
}
if (siginfo != NULL) {
bzero(siginfo, sizeof(*siginfo));
siginfo->si_errno = 0;
/*
* SUSv4 requires that the si_signo value is always
* SIGCHLD. Obey it despite the rfork(2) interface
* allows to request other signal for child exit
* notification.
*/
siginfo->si_signo = SIGCHLD;
/*
* This is still a rough estimate. We will fix the
* cases TRAPPED, STOPPED, and CONTINUED later.
*/
if (WCOREDUMP(p->p_xsig)) {
siginfo->si_code = CLD_DUMPED;
siginfo->si_status = WTERMSIG(p->p_xsig);
} else if (WIFSIGNALED(p->p_xsig)) {
siginfo->si_code = CLD_KILLED;
siginfo->si_status = WTERMSIG(p->p_xsig);
} else {
siginfo->si_code = CLD_EXITED;
siginfo->si_status = p->p_xexit;
}
siginfo->si_pid = p->p_pid;
siginfo->si_uid = p->p_ucred->cr_uid;
/*
* The si_addr field would be useful additional
* detail, but apparently the PC value may be lost
* when we reach this point. bzero() above sets
* siginfo->si_addr to NULL.
*/
}
/*
* There should be no reason to limit resources usage info to
* exited processes only. A snapshot about any resources used
* by a stopped process may be exactly what is needed.
*/
if (wrusage != NULL) {
rup = &wrusage->wru_self;
*rup = p->p_ru;
PROC_STATLOCK(p);
calcru(p, &rup->ru_utime, &rup->ru_stime);
PROC_STATUNLOCK(p);
rup = &wrusage->wru_children;
*rup = p->p_stats->p_cru;
calccru(p, &rup->ru_utime, &rup->ru_stime);
}
if (p->p_state == PRS_ZOMBIE && !check_only) {
PROC_SLOCK(p);
proc_reap(td, p, status, options);
return (-1);
}
return (1);
}
int
2004-09-22 15:24:33 +00:00
kern_wait(struct thread *td, pid_t pid, int *status, int options,
struct rusage *rusage)
{
struct __wrusage wru, *wrup;
idtype_t idtype;
id_t id;
int ret;
/*
* Translate the special pid values into the (idtype, pid)
* pair for kern_wait6. The WAIT_MYPGRP case is handled by
* kern_wait6() on its own.
*/
if (pid == WAIT_ANY) {
idtype = P_ALL;
id = 0;
} else if (pid < 0) {
idtype = P_PGID;
id = (id_t)-pid;
} else {
idtype = P_PID;
id = (id_t)pid;
}
if (rusage != NULL)
wrup = &wru;
else
wrup = NULL;
/*
* For backward compatibility we implicitly add flags WEXITED
* and WTRAPPED here.
*/
options |= WEXITED | WTRAPPED;
ret = kern_wait6(td, idtype, id, status, options, wrup, NULL);
if (rusage != NULL)
*rusage = wru.wru_self;
return (ret);
}
static void
report_alive_proc(struct thread *td, struct proc *p, siginfo_t *siginfo,
int *status, int options, int si_code)
{
bool cont;
PROC_LOCK_ASSERT(p, MA_OWNED);
sx_assert(&proctree_lock, SA_XLOCKED);
MPASS(si_code == CLD_TRAPPED || si_code == CLD_STOPPED ||
si_code == CLD_CONTINUED);
cont = si_code == CLD_CONTINUED;
if ((options & WNOWAIT) == 0) {
if (cont)
p->p_flag &= ~P_CONTINUED;
else
p->p_flag |= P_WAITED;
PROC_LOCK(td->td_proc);
sigqueue_take(p->p_ksi);
PROC_UNLOCK(td->td_proc);
}
sx_xunlock(&proctree_lock);
if (siginfo != NULL) {
siginfo->si_code = si_code;
siginfo->si_status = cont ? SIGCONT : p->p_xsig;
}
if (status != NULL)
*status = cont ? SIGCONT : W_STOPCODE(p->p_xsig);
PROC_UNLOCK(p);
td->td_retval[0] = p->p_pid;
}
int
kern_wait6(struct thread *td, idtype_t idtype, id_t id, int *status,
int options, struct __wrusage *wrusage, siginfo_t *siginfo)
1994-05-24 10:09:53 +00:00
{
struct proc *p, *q;
pid_t pid;
int error, nfound, ret;
1994-05-24 10:09:53 +00:00
AUDIT_ARG_VALUE((int)idtype); /* XXX - This is likely wrong! */
AUDIT_ARG_PID((pid_t)id); /* XXX - This may be wrong! */
AUDIT_ARG_VALUE(options);
q = td->td_proc;
if ((pid_t)id == WAIT_MYPGRP && (idtype == P_PID || idtype == P_PGID)) {
PROC_LOCK(q);
id = (id_t)q->p_pgid;
PROC_UNLOCK(q);
idtype = P_PGID;
}
/* If we don't know the option, just return. */
if ((options & ~(WUNTRACED | WNOHANG | WCONTINUED | WNOWAIT |
WEXITED | WTRAPPED | WLINUXCLONE)) != 0)
return (EINVAL);
if ((options & (WEXITED | WUNTRACED | WCONTINUED | WTRAPPED)) == 0) {
/*
* We will be unable to find any matching processes,
* because there are no known events to look for.
* Prefer to return error instead of blocking
* indefinitely.
*/
return (EINVAL);
}
1994-05-24 10:09:53 +00:00
loop:
if (q->p_flag & P_STATCHILD) {
PROC_LOCK(q);
q->p_flag &= ~P_STATCHILD;
PROC_UNLOCK(q);
}
1994-05-24 10:09:53 +00:00
nfound = 0;
sx_xlock(&proctree_lock);
LIST_FOREACH(p, &q->p_children, p_sibling) {
pid = p->p_pid;
ret = proc_to_reap(td, p, idtype, id, status, options,
wrusage, siginfo, 0);
if (ret == 0)
continue;
else if (ret == 1)
nfound++;
else {
td->td_retval[0] = pid;
return (0);
}
PROC_LOCK_ASSERT(p, MA_OWNED);
if ((options & (WTRAPPED | WUNTRACED)) != 0)
PROC_SLOCK(p);
if ((options & WTRAPPED) != 0 &&
(p->p_flag & P_TRACED) != 0 &&
(p->p_flag & (P_STOPPED_TRACE | P_STOPPED_SIG)) != 0 &&
p->p_suspcount == p->p_numthreads &&
(p->p_flag & P_WAITED) == 0) {
PROC_SUNLOCK(p);
CTR4(KTR_PTRACE,
"wait: returning trapped pid %d status %#x "
"(xstat %d) xthread %d",
p->p_pid, W_STOPCODE(p->p_xsig), p->p_xsig,
p->p_xthread != NULL ?
p->p_xthread->td_tid : -1);
report_alive_proc(td, p, siginfo, status, options,
CLD_TRAPPED);
return (0);
}
if ((options & WUNTRACED) != 0 &&
(p->p_flag & P_STOPPED_SIG) != 0 &&
p->p_suspcount == p->p_numthreads &&
(p->p_flag & P_WAITED) == 0) {
PROC_SUNLOCK(p);
report_alive_proc(td, p, siginfo, status, options,
CLD_STOPPED);
return (0);
1994-05-24 10:09:53 +00:00
}
if ((options & (WTRAPPED | WUNTRACED)) != 0)
PROC_SUNLOCK(p);
if ((options & WCONTINUED) != 0 &&
(p->p_flag & P_CONTINUED) != 0) {
report_alive_proc(td, p, siginfo, status, options,
CLD_CONTINUED);
return (0);
}
PROC_UNLOCK(p);
1994-05-24 10:09:53 +00:00
}
/*
* Look in the orphans list too, to allow the parent to
* collect it's child exit status even if child is being
* debugged.
*
* Debugger detaches from the parent upon successful
* switch-over from parent to child. At this point due to
* re-parenting the parent loses the child to debugger and a
* wait4(2) call would report that it has no children to wait
* for. By maintaining a list of orphans we allow the parent
* to successfully wait until the child becomes a zombie.
*/
if (nfound == 0) {
LIST_FOREACH(p, &q->p_orphans, p_orphan) {
ret = proc_to_reap(td, p, idtype, id, NULL, options,
NULL, NULL, 1);
if (ret != 0) {
KASSERT(ret != -1, ("reaped an orphan (pid %d)",
(int)td->td_retval[0]));
PROC_UNLOCK(p);
nfound++;
break;
}
}
}
if (nfound == 0) {
sx_xunlock(&proctree_lock);
return (ECHILD);
}
if (options & WNOHANG) {
sx_xunlock(&proctree_lock);
td->td_retval[0] = 0;
return (0);
}
PROC_LOCK(q);
sx_xunlock(&proctree_lock);
if (q->p_flag & P_STATCHILD) {
q->p_flag &= ~P_STATCHILD;
error = 0;
} else
error = msleep(q, &q->p_mtx, PWAIT | PCATCH, "wait", 0);
PROC_UNLOCK(q);
if (error)
return (error);
1994-05-24 10:09:53 +00:00
goto loop;
}
/*
* Make process 'parent' the new parent of process 'child'.
* Must be called with an exclusive hold of proctree lock.
1994-05-24 10:09:53 +00:00
*/
void
2003-03-19 00:49:40 +00:00
proc_reparent(struct proc *child, struct proc *parent)
1994-05-24 10:09:53 +00:00
{
sx_assert(&proctree_lock, SX_XLOCKED);
PROC_LOCK_ASSERT(child, MA_OWNED);
1994-05-24 10:09:53 +00:00
if (child->p_pptr == parent)
return;
PROC_LOCK(child->p_pptr);
sigqueue_take(child->p_ksi);
PROC_UNLOCK(child->p_pptr);
LIST_REMOVE(child, p_sibling);
LIST_INSERT_HEAD(&parent->p_children, child, p_sibling);
clear_orphan(child);
if (child->p_flag & P_TRACED) {
if (LIST_EMPTY(&child->p_pptr->p_orphans)) {
child->p_treeflag |= P_TREE_FIRST_ORPHAN;
LIST_INSERT_HEAD(&child->p_pptr->p_orphans, child,
p_orphan);
} else {
LIST_INSERT_AFTER(LIST_FIRST(&child->p_pptr->p_orphans),
child, p_orphan);
}
child->p_treeflag |= P_TREE_ORPHANED;
}
1994-05-24 10:09:53 +00:00
child->p_pptr = parent;
}