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.
|
|
|
|
|
*
|
|
|
|
|
* 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. All advertising materials mentioning features or use of this software
|
|
|
|
|
* must display the following acknowledgement:
|
|
|
|
|
* This product includes software developed by the University of
|
|
|
|
|
* California, Berkeley and its contributors.
|
|
|
|
|
* 4. Neither the name of the University nor the names of its contributors
|
|
|
|
|
* 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.
|
|
|
|
|
*
|
1996-03-11 06:05:03 +00:00
|
|
|
* @(#)kern_proc.c 8.7 (Berkeley) 2/14/95
|
1999-08-28 01:08:13 +00:00
|
|
|
* $FreeBSD$
|
1994-05-24 10:09:53 +00:00
|
|
|
*/
|
|
|
|
|
|
|
|
|
|
#include <sys/param.h>
|
|
|
|
|
#include <sys/systm.h>
|
|
|
|
|
#include <sys/kernel.h>
|
2001-05-01 08:13:21 +00:00
|
|
|
#include <sys/lock.h>
|
1994-05-24 10:09:53 +00:00
|
|
|
#include <sys/malloc.h>
|
2001-05-01 08:13:21 +00:00
|
|
|
#include <sys/mutex.h>
|
1999-11-16 20:31:58 +00:00
|
|
|
#include <sys/proc.h>
|
2002-02-07 20:58:47 +00:00
|
|
|
#include <sys/sysproto.h>
|
2001-05-01 08:13:21 +00:00
|
|
|
#include <sys/sysctl.h>
|
1998-11-11 10:56:07 +00:00
|
|
|
#include <sys/filedesc.h>
|
1994-05-24 10:09:53 +00:00
|
|
|
#include <sys/tty.h>
|
1994-09-25 19:34:02 +00:00
|
|
|
#include <sys/signalvar.h>
|
2001-03-28 11:52:56 +00:00
|
|
|
#include <sys/sx.h>
|
2001-05-01 08:13:21 +00:00
|
|
|
#include <sys/user.h>
|
|
|
|
|
#include <sys/jail.h>
|
|
|
|
|
|
1995-12-07 12:48:31 +00:00
|
|
|
#include <vm/vm.h>
|
|
|
|
|
#include <vm/pmap.h>
|
|
|
|
|
#include <vm/vm_map.h>
|
VM level code cleanups.
1) Start using TSM.
Struct procs continue to point to upages structure, after being freed.
Struct vmspace continues to point to pte object and kva space for kstack.
u_map is now superfluous.
2) vm_map's don't need to be reference counted. They always exist either
in the kernel or in a vmspace. The vmspaces are managed by reference
counts.
3) Remove the "wired" vm_map nonsense.
4) No need to keep a cache of kernel stack kva's.
5) Get rid of strange looking ++var, and change to var++.
6) Change more data structures to use our "zone" allocator. Added
struct proc, struct vmspace and struct vnode. This saves a significant
amount of kva space and physical memory. Additionally, this enables
TSM for the zone managed memory.
7) Keep ioopt disabled for now.
8) Remove the now bogus "single use" map concept.
9) Use generation counts or id's for data structures residing in TSM, where
it allows us to avoid unneeded restart overhead during traversals, where
blocking might occur.
10) Account better for memory deficits, so the pageout daemon will be able
to make enough memory available (experimental.)
11) Fix some vnode locking problems. (From Tor, I think.)
12) Add a check in ufs_lookup, to avoid lots of unneeded calls to bcmp.
(experimental.)
13) Significantly shrink, cleanup, and make slightly faster the vm_fault.c
code. Use generation counts, get rid of unneded collpase operations,
and clean up the cluster code.
14) Make vm_zone more suitable for TSM.
This commit is partially as a result of discussions and contributions from
other people, including DG, Tor Egge, PHK, and probably others that I
have forgotten to attribute (so let me know, if I forgot.)
This is not the infamous, final cleanup of the vnode stuff, but a necessary
step. Vnode mgmt should be correct, but things might still change, and
there is still some missing stuff (like ioopt, and physical backing of
non-merged cache files, debugging of layering concepts.)
1998-01-22 17:30:44 +00:00
|
|
|
#include <vm/vm_zone.h>
|
1994-05-24 10:09:53 +00:00
|
|
|
|
1997-10-12 20:26:33 +00:00
|
|
|
static MALLOC_DEFINE(M_PGRP, "pgrp", "process group header");
|
|
|
|
|
MALLOC_DEFINE(M_SESSION, "session", "session header");
|
1998-02-20 13:52:15 +00:00
|
|
|
static MALLOC_DEFINE(M_PROC, "proc", "Proc structures");
|
1997-10-12 20:26:33 +00:00
|
|
|
MALLOC_DEFINE(M_SUBPROC, "subproc", "Proc sub-structures");
|
1997-10-11 18:31:40 +00:00
|
|
|
|
1995-12-14 08:32:45 +00:00
|
|
|
static void pgdelete __P((struct pgrp *));
|
1994-05-25 09:21:21 +00:00
|
|
|
|
1995-12-02 18:58:56 +00:00
|
|
|
static void orphanpg __P((struct pgrp *pg));
|
|
|
|
|
|
1994-05-24 10:09:53 +00:00
|
|
|
/*
|
1996-03-11 06:05:03 +00:00
|
|
|
* Other process lists
|
|
|
|
|
*/
|
|
|
|
|
struct pidhashhead *pidhashtbl;
|
|
|
|
|
u_long pidhash;
|
|
|
|
|
struct pgrphashhead *pgrphashtbl;
|
|
|
|
|
u_long pgrphash;
|
|
|
|
|
struct proclist allproc;
|
|
|
|
|
struct proclist zombproc;
|
2001-03-28 11:52:56 +00:00
|
|
|
struct sx allproc_lock;
|
|
|
|
|
struct sx proctree_lock;
|
VM level code cleanups.
1) Start using TSM.
Struct procs continue to point to upages structure, after being freed.
Struct vmspace continues to point to pte object and kva space for kstack.
u_map is now superfluous.
2) vm_map's don't need to be reference counted. They always exist either
in the kernel or in a vmspace. The vmspaces are managed by reference
counts.
3) Remove the "wired" vm_map nonsense.
4) No need to keep a cache of kernel stack kva's.
5) Get rid of strange looking ++var, and change to var++.
6) Change more data structures to use our "zone" allocator. Added
struct proc, struct vmspace and struct vnode. This saves a significant
amount of kva space and physical memory. Additionally, this enables
TSM for the zone managed memory.
7) Keep ioopt disabled for now.
8) Remove the now bogus "single use" map concept.
9) Use generation counts or id's for data structures residing in TSM, where
it allows us to avoid unneeded restart overhead during traversals, where
blocking might occur.
10) Account better for memory deficits, so the pageout daemon will be able
to make enough memory available (experimental.)
11) Fix some vnode locking problems. (From Tor, I think.)
12) Add a check in ufs_lookup, to avoid lots of unneeded calls to bcmp.
(experimental.)
13) Significantly shrink, cleanup, and make slightly faster the vm_fault.c
code. Use generation counts, get rid of unneded collpase operations,
and clean up the cluster code.
14) Make vm_zone more suitable for TSM.
This commit is partially as a result of discussions and contributions from
other people, including DG, Tor Egge, PHK, and probably others that I
have forgotten to attribute (so let me know, if I forgot.)
This is not the infamous, final cleanup of the vnode stuff, but a necessary
step. Vnode mgmt should be correct, but things might still change, and
there is still some missing stuff (like ioopt, and physical backing of
non-merged cache files, debugging of layering concepts.)
1998-01-22 17:30:44 +00:00
|
|
|
vm_zone_t proc_zone;
|
2000-09-07 01:33:02 +00:00
|
|
|
vm_zone_t ithread_zone;
|
1996-03-11 06:05:03 +00:00
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Initialize global process hashing structures.
|
1994-05-24 10:09:53 +00:00
|
|
|
*/
|
1994-05-25 09:21:21 +00:00
|
|
|
void
|
1996-03-11 06:05:03 +00:00
|
|
|
procinit()
|
1994-05-24 10:09:53 +00:00
|
|
|
{
|
2001-03-07 06:52:12 +00:00
|
|
|
int i, j;
|
1994-05-24 10:09:53 +00:00
|
|
|
|
2001-03-28 11:52:56 +00:00
|
|
|
sx_init(&allproc_lock, "allproc");
|
|
|
|
|
sx_init(&proctree_lock, "proctree");
|
1996-03-11 06:05:03 +00:00
|
|
|
LIST_INIT(&allproc);
|
|
|
|
|
LIST_INIT(&zombproc);
|
|
|
|
|
pidhashtbl = hashinit(maxproc / 4, M_PROC, &pidhash);
|
|
|
|
|
pgrphashtbl = hashinit(maxproc / 4, M_PROC, &pgrphash);
|
VM level code cleanups.
1) Start using TSM.
Struct procs continue to point to upages structure, after being freed.
Struct vmspace continues to point to pte object and kva space for kstack.
u_map is now superfluous.
2) vm_map's don't need to be reference counted. They always exist either
in the kernel or in a vmspace. The vmspaces are managed by reference
counts.
3) Remove the "wired" vm_map nonsense.
4) No need to keep a cache of kernel stack kva's.
5) Get rid of strange looking ++var, and change to var++.
6) Change more data structures to use our "zone" allocator. Added
struct proc, struct vmspace and struct vnode. This saves a significant
amount of kva space and physical memory. Additionally, this enables
TSM for the zone managed memory.
7) Keep ioopt disabled for now.
8) Remove the now bogus "single use" map concept.
9) Use generation counts or id's for data structures residing in TSM, where
it allows us to avoid unneeded restart overhead during traversals, where
blocking might occur.
10) Account better for memory deficits, so the pageout daemon will be able
to make enough memory available (experimental.)
11) Fix some vnode locking problems. (From Tor, I think.)
12) Add a check in ufs_lookup, to avoid lots of unneeded calls to bcmp.
(experimental.)
13) Significantly shrink, cleanup, and make slightly faster the vm_fault.c
code. Use generation counts, get rid of unneded collpase operations,
and clean up the cluster code.
14) Make vm_zone more suitable for TSM.
This commit is partially as a result of discussions and contributions from
other people, including DG, Tor Egge, PHK, and probably others that I
have forgotten to attribute (so let me know, if I forgot.)
This is not the infamous, final cleanup of the vnode stuff, but a necessary
step. Vnode mgmt should be correct, but things might still change, and
there is still some missing stuff (like ioopt, and physical backing of
non-merged cache files, debugging of layering concepts.)
1998-01-22 17:30:44 +00:00
|
|
|
proc_zone = zinit("PROC", sizeof (struct proc), 0, 0, 5);
|
2000-09-05 22:11:13 +00:00
|
|
|
uihashinit();
|
Change the proc information returned from the kernel so that it
no longer contains kernel specific data structures, but rather
only scalar values and structures that are already part of the
kernel/user interface, specifically rusage and rtprio. It no
longer contains proc, session, pcred, ucred, procsig, vmspace,
pstats, mtx, sigiolst, klist, callout, pasleep, or mdproc. If
any of these changed in size, ps, w, fstat, gcore, systat, and
top would all stop working. The new structure has over 200 bytes
of unassigned space for future values to be added, yet is nearly
100 bytes smaller per entry than the structure that it replaced.
2000-12-12 07:25:57 +00:00
|
|
|
/*
|
|
|
|
|
* This should really be a compile time warning, but I do
|
|
|
|
|
* not know of any way to do that...
|
|
|
|
|
*/
|
2001-03-07 06:52:12 +00:00
|
|
|
if (sizeof(struct kinfo_proc) != KINFO_PROC_SIZE) {
|
|
|
|
|
printf("This message will repeat for the next 20 seconds\n");
|
|
|
|
|
for (i = 0; i < 20; i++) {
|
|
|
|
|
printf("WARNING: size of kinfo_proc (%ld) should be %d!!!\n",
|
|
|
|
|
(long)sizeof(struct kinfo_proc), KINFO_PROC_SIZE);
|
|
|
|
|
printf("The kinfo_proc structure was changed ");
|
|
|
|
|
printf("incorrectly in <sys/user.h>\n");
|
|
|
|
|
for (j = 0; j < 0x7ffffff; j++);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
}
|
1994-05-24 10:09:53 +00:00
|
|
|
}
|
|
|
|
|
|
2001-09-12 08:38:13 +00:00
|
|
|
/*
|
2002-02-07 20:58:47 +00:00
|
|
|
* Note that we do not link to the proc's ucred here
|
|
|
|
|
* The thread is linked as if running but no KSE assigned
|
2001-09-12 08:38:13 +00:00
|
|
|
*/
|
2002-02-07 20:58:47 +00:00
|
|
|
static void
|
|
|
|
|
thread_link(struct thread *td, struct ksegrp *kg)
|
2001-09-12 08:38:13 +00:00
|
|
|
{
|
2002-02-07 20:58:47 +00:00
|
|
|
struct proc *p = kg->kg_proc;
|
2001-09-12 08:38:13 +00:00
|
|
|
|
2002-02-07 20:58:47 +00:00
|
|
|
td->td_proc = p;
|
|
|
|
|
td->td_ksegrp = kg;
|
2002-02-08 02:56:10 +00:00
|
|
|
td->td_last_kse = &p->p_kse;
|
2001-09-12 08:38:13 +00:00
|
|
|
|
2002-02-07 20:58:47 +00:00
|
|
|
TAILQ_INSERT_HEAD(&p->p_threads, td, td_plist);
|
|
|
|
|
TAILQ_INSERT_HEAD(&kg->kg_threads, td, td_kglist);
|
|
|
|
|
td->td_critnest = 0;
|
|
|
|
|
td->td_savecrit = 0;
|
|
|
|
|
td->td_kse = NULL;
|
|
|
|
|
}
|
2001-09-12 08:38:13 +00:00
|
|
|
|
2002-02-07 20:58:47 +00:00
|
|
|
/*
|
|
|
|
|
* KSE is linked onto the idle queue.
|
|
|
|
|
*/
|
|
|
|
|
static void
|
|
|
|
|
kse_link(struct kse *ke, struct ksegrp *kg)
|
|
|
|
|
{
|
|
|
|
|
struct proc *p = kg->kg_proc;
|
|
|
|
|
|
|
|
|
|
TAILQ_INSERT_HEAD(&kg->kg_kseq, ke, ke_kglist);
|
|
|
|
|
kg->kg_kses++;
|
|
|
|
|
TAILQ_INSERT_HEAD(&kg->kg_iq, ke, ke_kgrlist);
|
|
|
|
|
ke->ke_proc = p;
|
|
|
|
|
ke->ke_ksegrp = kg;
|
|
|
|
|
ke->ke_thread = NULL;
|
|
|
|
|
ke->ke_oncpu = NOCPU;
|
|
|
|
|
}
|
2001-09-12 08:38:13 +00:00
|
|
|
|
2002-02-07 20:58:47 +00:00
|
|
|
static void
|
|
|
|
|
ksegrp_link(struct ksegrp *kg, struct proc *p)
|
|
|
|
|
{
|
2002-02-07 23:06:26 +00:00
|
|
|
|
2002-02-07 20:58:47 +00:00
|
|
|
TAILQ_INIT(&kg->kg_threads);
|
|
|
|
|
TAILQ_INIT(&kg->kg_runq); /* links with td_runq */
|
|
|
|
|
TAILQ_INIT(&kg->kg_slpq); /* links with td_runq */
|
|
|
|
|
TAILQ_INIT(&kg->kg_kseq); /* all kses in ksegrp */
|
|
|
|
|
TAILQ_INIT(&kg->kg_iq); /* all kses in ksegrp */
|
|
|
|
|
kg->kg_proc = p;
|
|
|
|
|
/* the following counters are in the -zero- section and may not need clearing */
|
|
|
|
|
kg->kg_runnable = 0;
|
|
|
|
|
kg->kg_kses = 0;
|
|
|
|
|
kg->kg_runq_kses = 0; /* XXXKSE change name */
|
|
|
|
|
/* link it in now that it's consitant */
|
|
|
|
|
TAILQ_INSERT_HEAD(&p->p_ksegrps, kg, kg_ksegrp);
|
|
|
|
|
}
|
2001-09-12 08:38:13 +00:00
|
|
|
|
2002-02-07 20:58:47 +00:00
|
|
|
/*
|
|
|
|
|
* for a newly created process,
|
|
|
|
|
* link up a the structure and its initial threads etc.
|
|
|
|
|
*/
|
|
|
|
|
void
|
|
|
|
|
proc_linkup(struct proc *p, struct ksegrp *kg,
|
|
|
|
|
struct kse *ke, struct thread *td)
|
|
|
|
|
{
|
2002-02-07 23:06:26 +00:00
|
|
|
|
2002-02-07 20:58:47 +00:00
|
|
|
TAILQ_INIT(&p->p_ksegrps); /* all ksegrps in proc */
|
|
|
|
|
TAILQ_INIT(&p->p_threads); /* all threads in proc */
|
2001-09-12 08:38:13 +00:00
|
|
|
|
2002-02-07 20:58:47 +00:00
|
|
|
ksegrp_link(kg, p);
|
|
|
|
|
kse_link(ke, kg);
|
|
|
|
|
thread_link(td, kg);
|
|
|
|
|
/* link them together for 1:1 */
|
|
|
|
|
td->td_kse = ke;
|
|
|
|
|
ke->ke_thread = td;
|
|
|
|
|
}
|
2001-09-12 08:38:13 +00:00
|
|
|
|
2002-02-07 20:58:47 +00:00
|
|
|
/* temporary version is ultra simple while we are in 1:1 mode */
|
|
|
|
|
struct thread *
|
|
|
|
|
thread_get(struct proc *p)
|
|
|
|
|
{
|
|
|
|
|
struct thread *td = &p->p_xxthread;
|
2001-09-12 08:38:13 +00:00
|
|
|
|
2002-02-07 20:58:47 +00:00
|
|
|
return (td);
|
2001-09-12 08:38:13 +00:00
|
|
|
}
|
|
|
|
|
|
1994-05-24 10:09:53 +00:00
|
|
|
/*
|
|
|
|
|
* Is p an inferior of the current process?
|
|
|
|
|
*/
|
2000-06-23 07:10:34 +00:00
|
|
|
int
|
1994-05-24 10:09:53 +00:00
|
|
|
inferior(p)
|
|
|
|
|
register struct proc *p;
|
|
|
|
|
{
|
2001-11-12 18:56:49 +00:00
|
|
|
|
|
|
|
|
sx_assert(&proctree_lock, SX_LOCKED);
|
|
|
|
|
for (; p != curproc; p = p->p_pptr)
|
2001-11-13 16:55:54 +00:00
|
|
|
if (p->p_pid == 0)
|
|
|
|
|
return (0);
|
|
|
|
|
return (1);
|
1994-05-24 10:09:53 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Locate a process by number
|
|
|
|
|
*/
|
|
|
|
|
struct proc *
|
|
|
|
|
pfind(pid)
|
|
|
|
|
register pid_t pid;
|
|
|
|
|
{
|
|
|
|
|
register struct proc *p;
|
|
|
|
|
|
2001-03-28 11:52:56 +00:00
|
|
|
sx_slock(&allproc_lock);
|
1999-11-16 16:28:58 +00:00
|
|
|
LIST_FOREACH(p, PIDHASH(pid), p_hash)
|
2001-04-24 00:51:53 +00:00
|
|
|
if (p->p_pid == pid) {
|
|
|
|
|
PROC_LOCK(p);
|
2000-11-22 07:42:04 +00:00
|
|
|
break;
|
2001-04-24 00:51:53 +00:00
|
|
|
}
|
2001-03-28 11:52:56 +00:00
|
|
|
sx_sunlock(&allproc_lock);
|
2000-11-22 07:42:04 +00:00
|
|
|
return (p);
|
1994-05-24 10:09:53 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Locate a process group by number
|
|
|
|
|
*/
|
|
|
|
|
struct pgrp *
|
|
|
|
|
pgfind(pgid)
|
|
|
|
|
register pid_t pgid;
|
|
|
|
|
{
|
|
|
|
|
register struct pgrp *pgrp;
|
|
|
|
|
|
1999-11-16 16:28:58 +00:00
|
|
|
LIST_FOREACH(pgrp, PGRPHASH(pgid), pg_hash)
|
1994-05-24 10:09:53 +00:00
|
|
|
if (pgrp->pg_id == pgid)
|
|
|
|
|
return (pgrp);
|
|
|
|
|
return (NULL);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Move p to a new or existing process group (and session)
|
|
|
|
|
*/
|
1994-05-25 09:21:21 +00:00
|
|
|
int
|
1994-05-24 10:09:53 +00:00
|
|
|
enterpgrp(p, pgid, mksess)
|
|
|
|
|
register struct proc *p;
|
|
|
|
|
pid_t pgid;
|
|
|
|
|
int mksess;
|
|
|
|
|
{
|
|
|
|
|
register struct pgrp *pgrp = pgfind(pgid);
|
2001-03-07 03:28:50 +00:00
|
|
|
struct pgrp *savegrp;
|
1994-05-24 10:09:53 +00:00
|
|
|
|
1999-01-08 17:31:30 +00:00
|
|
|
KASSERT(pgrp == NULL || !mksess,
|
1999-01-10 01:58:29 +00:00
|
|
|
("enterpgrp: setsid into non-empty pgrp"));
|
1999-01-08 17:31:30 +00:00
|
|
|
KASSERT(!SESS_LEADER(p),
|
1999-01-10 01:58:29 +00:00
|
|
|
("enterpgrp: session leader attempted setpgrp"));
|
|
|
|
|
|
1994-05-24 10:09:53 +00:00
|
|
|
if (pgrp == NULL) {
|
|
|
|
|
pid_t savepid = p->p_pid;
|
|
|
|
|
struct proc *np;
|
|
|
|
|
/*
|
|
|
|
|
* new process group
|
|
|
|
|
*/
|
1999-01-08 17:31:30 +00:00
|
|
|
KASSERT(p->p_pid == pgid,
|
1999-01-10 01:58:29 +00:00
|
|
|
("enterpgrp: new pgrp and pid != pgid"));
|
2001-04-24 00:51:53 +00:00
|
|
|
if ((np = pfind(savepid)) == NULL || np != p) {
|
|
|
|
|
if (np != NULL)
|
|
|
|
|
PROC_UNLOCK(np);
|
1994-05-24 10:09:53 +00:00
|
|
|
return (ESRCH);
|
2001-04-24 00:51:53 +00:00
|
|
|
}
|
|
|
|
|
PROC_UNLOCK(np);
|
2001-03-27 20:49:51 +00:00
|
|
|
MALLOC(pgrp, struct pgrp *, sizeof(struct pgrp), M_PGRP,
|
|
|
|
|
M_WAITOK);
|
1994-05-24 10:09:53 +00:00
|
|
|
if (mksess) {
|
|
|
|
|
register struct session *sess;
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* new session
|
|
|
|
|
*/
|
|
|
|
|
MALLOC(sess, struct session *, sizeof(struct session),
|
1996-03-11 06:05:03 +00:00
|
|
|
M_SESSION, M_WAITOK);
|
1994-05-24 10:09:53 +00:00
|
|
|
sess->s_leader = p;
|
1998-11-09 15:08:04 +00:00
|
|
|
sess->s_sid = p->p_pid;
|
1994-05-24 10:09:53 +00:00
|
|
|
sess->s_count = 1;
|
|
|
|
|
sess->s_ttyvp = NULL;
|
|
|
|
|
sess->s_ttyp = NULL;
|
|
|
|
|
bcopy(p->p_session->s_login, sess->s_login,
|
|
|
|
|
sizeof(sess->s_login));
|
2001-03-07 03:28:50 +00:00
|
|
|
PROC_LOCK(p);
|
1994-05-24 10:09:53 +00:00
|
|
|
p->p_flag &= ~P_CONTROLT;
|
2001-03-07 03:28:50 +00:00
|
|
|
PROC_UNLOCK(p);
|
1994-05-24 10:09:53 +00:00
|
|
|
pgrp->pg_session = sess;
|
1999-01-08 17:31:30 +00:00
|
|
|
KASSERT(p == curproc,
|
1999-01-10 01:58:29 +00:00
|
|
|
("enterpgrp: mksession and p != curproc"));
|
1994-05-24 10:09:53 +00:00
|
|
|
} else {
|
|
|
|
|
pgrp->pg_session = p->p_session;
|
|
|
|
|
pgrp->pg_session->s_count++;
|
|
|
|
|
}
|
|
|
|
|
pgrp->pg_id = pgid;
|
1996-03-11 06:05:03 +00:00
|
|
|
LIST_INIT(&pgrp->pg_members);
|
|
|
|
|
LIST_INSERT_HEAD(PGRPHASH(pgid), pgrp, pg_hash);
|
1994-05-24 10:09:53 +00:00
|
|
|
pgrp->pg_jobc = 0;
|
Installed the second patch attached to kern/7899 with some changes suggested
by bde, a few other tweaks to get the patch to apply cleanly again and
some improvements to the comments.
This change closes some fairly minor security holes associated with
F_SETOWN, fixes a few bugs, and removes some limitations that F_SETOWN
had on tty devices. For more details, see the description on the PR.
Because this patch increases the size of the proc and pgrp structures,
it is necessary to re-install the includes and recompile libkvm,
the vinum lkm, fstat, gcore, gdb, ipfilter, ps, top, and w.
PR: kern/7899
Reviewed by: bde, elvind
1998-11-11 10:04:13 +00:00
|
|
|
SLIST_INIT(&pgrp->pg_sigiolst);
|
1994-05-24 10:09:53 +00:00
|
|
|
} else if (pgrp == p->p_pgrp)
|
|
|
|
|
return (0);
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Adjust eligibility of affected pgrps to participate in job control.
|
|
|
|
|
* Increment eligibility counts before decrementing, otherwise we
|
|
|
|
|
* could reach 0 spuriously during the first call.
|
|
|
|
|
*/
|
|
|
|
|
fixjobc(p, pgrp, 1);
|
|
|
|
|
fixjobc(p, p->p_pgrp, 0);
|
|
|
|
|
|
2001-03-07 03:28:50 +00:00
|
|
|
PROC_LOCK(p);
|
1996-03-11 06:05:03 +00:00
|
|
|
LIST_REMOVE(p, p_pglist);
|
2001-03-07 03:28:50 +00:00
|
|
|
savegrp = p->p_pgrp;
|
1994-05-24 10:09:53 +00:00
|
|
|
p->p_pgrp = pgrp;
|
1996-03-11 06:05:03 +00:00
|
|
|
LIST_INSERT_HEAD(&pgrp->pg_members, p, p_pglist);
|
2001-03-07 03:28:50 +00:00
|
|
|
PROC_UNLOCK(p);
|
|
|
|
|
if (LIST_EMPTY(&savegrp->pg_members))
|
|
|
|
|
pgdelete(savegrp);
|
1994-05-24 10:09:53 +00:00
|
|
|
return (0);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* remove process from process group
|
|
|
|
|
*/
|
1994-05-25 09:21:21 +00:00
|
|
|
int
|
1994-05-24 10:09:53 +00:00
|
|
|
leavepgrp(p)
|
|
|
|
|
register struct proc *p;
|
|
|
|
|
{
|
2001-03-07 03:28:50 +00:00
|
|
|
struct pgrp *savegrp;
|
1994-05-24 10:09:53 +00:00
|
|
|
|
2001-03-07 03:28:50 +00:00
|
|
|
PROC_LOCK(p);
|
1996-03-11 06:05:03 +00:00
|
|
|
LIST_REMOVE(p, p_pglist);
|
2001-03-07 03:28:50 +00:00
|
|
|
savegrp = p->p_pgrp;
|
|
|
|
|
p->p_pgrp = NULL;
|
|
|
|
|
PROC_UNLOCK(p);
|
|
|
|
|
if (LIST_EMPTY(&savegrp->pg_members))
|
|
|
|
|
pgdelete(savegrp);
|
1994-05-24 10:09:53 +00:00
|
|
|
return (0);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* delete a process group
|
|
|
|
|
*/
|
1995-12-14 08:32:45 +00:00
|
|
|
static void
|
1994-05-24 10:09:53 +00:00
|
|
|
pgdelete(pgrp)
|
|
|
|
|
register struct pgrp *pgrp;
|
|
|
|
|
{
|
|
|
|
|
|
Installed the second patch attached to kern/7899 with some changes suggested
by bde, a few other tweaks to get the patch to apply cleanly again and
some improvements to the comments.
This change closes some fairly minor security holes associated with
F_SETOWN, fixes a few bugs, and removes some limitations that F_SETOWN
had on tty devices. For more details, see the description on the PR.
Because this patch increases the size of the proc and pgrp structures,
it is necessary to re-install the includes and recompile libkvm,
the vinum lkm, fstat, gcore, gdb, ipfilter, ps, top, and w.
PR: kern/7899
Reviewed by: bde, elvind
1998-11-11 10:04:13 +00:00
|
|
|
/*
|
|
|
|
|
* Reset any sigio structures pointing to us as a result of
|
|
|
|
|
* F_SETOWN with our pgid.
|
|
|
|
|
*/
|
|
|
|
|
funsetownlst(&pgrp->pg_sigiolst);
|
|
|
|
|
|
1995-05-30 08:16:23 +00:00
|
|
|
if (pgrp->pg_session->s_ttyp != NULL &&
|
1994-05-24 10:09:53 +00:00
|
|
|
pgrp->pg_session->s_ttyp->t_pgrp == pgrp)
|
|
|
|
|
pgrp->pg_session->s_ttyp->t_pgrp = NULL;
|
1996-03-11 06:05:03 +00:00
|
|
|
LIST_REMOVE(pgrp, pg_hash);
|
1994-05-24 10:09:53 +00:00
|
|
|
if (--pgrp->pg_session->s_count == 0)
|
|
|
|
|
FREE(pgrp->pg_session, M_SESSION);
|
|
|
|
|
FREE(pgrp, M_PGRP);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Adjust pgrp jobc counters when specified process changes process group.
|
|
|
|
|
* We count the number of processes in each process group that "qualify"
|
|
|
|
|
* the group for terminal job control (those with a parent in a different
|
|
|
|
|
* process group of the same session). If that count reaches zero, the
|
|
|
|
|
* process group becomes orphaned. Check both the specified process'
|
|
|
|
|
* process group and that of its children.
|
|
|
|
|
* entering == 0 => p is leaving specified group.
|
|
|
|
|
* entering == 1 => p is entering specified group.
|
|
|
|
|
*/
|
1994-05-25 09:21:21 +00:00
|
|
|
void
|
1994-05-24 10:09:53 +00:00
|
|
|
fixjobc(p, pgrp, entering)
|
|
|
|
|
register struct proc *p;
|
|
|
|
|
register struct pgrp *pgrp;
|
|
|
|
|
int entering;
|
|
|
|
|
{
|
|
|
|
|
register struct pgrp *hispgrp;
|
|
|
|
|
register struct session *mysession = pgrp->pg_session;
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Check p's parent to see whether p qualifies its own process
|
|
|
|
|
* group; if so, adjust count for p's process group.
|
|
|
|
|
*/
|
2001-03-28 11:52:56 +00:00
|
|
|
sx_slock(&proctree_lock);
|
1994-05-24 10:09:53 +00:00
|
|
|
if ((hispgrp = p->p_pptr->p_pgrp) != pgrp &&
|
1999-05-06 18:13:11 +00:00
|
|
|
hispgrp->pg_session == mysession) {
|
1994-05-24 10:09:53 +00:00
|
|
|
if (entering)
|
|
|
|
|
pgrp->pg_jobc++;
|
|
|
|
|
else if (--pgrp->pg_jobc == 0)
|
|
|
|
|
orphanpg(pgrp);
|
1999-05-06 18:13:11 +00:00
|
|
|
}
|
1994-05-24 10:09:53 +00:00
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Check this process' children to see whether they qualify
|
|
|
|
|
* their process groups; if so, adjust counts for children's
|
|
|
|
|
* process groups.
|
|
|
|
|
*/
|
1999-11-16 16:28:58 +00:00
|
|
|
LIST_FOREACH(p, &p->p_children, p_sibling)
|
1994-05-24 10:09:53 +00:00
|
|
|
if ((hispgrp = p->p_pgrp) != pgrp &&
|
|
|
|
|
hispgrp->pg_session == mysession &&
|
1999-05-06 18:13:11 +00:00
|
|
|
p->p_stat != SZOMB) {
|
1994-05-24 10:09:53 +00:00
|
|
|
if (entering)
|
|
|
|
|
hispgrp->pg_jobc++;
|
|
|
|
|
else if (--hispgrp->pg_jobc == 0)
|
|
|
|
|
orphanpg(hispgrp);
|
1999-05-06 18:13:11 +00:00
|
|
|
}
|
2001-03-28 11:52:56 +00:00
|
|
|
sx_sunlock(&proctree_lock);
|
1994-05-24 10:09:53 +00:00
|
|
|
}
|
|
|
|
|
|
1995-05-30 08:16:23 +00:00
|
|
|
/*
|
1994-05-24 10:09:53 +00:00
|
|
|
* A process group has become orphaned;
|
|
|
|
|
* if there are any stopped processes in the group,
|
|
|
|
|
* hang-up all process in that group.
|
|
|
|
|
*/
|
|
|
|
|
static void
|
|
|
|
|
orphanpg(pg)
|
|
|
|
|
struct pgrp *pg;
|
|
|
|
|
{
|
|
|
|
|
register struct proc *p;
|
|
|
|
|
|
2001-03-07 03:28:50 +00:00
|
|
|
mtx_lock_spin(&sched_lock);
|
1999-11-16 16:28:58 +00:00
|
|
|
LIST_FOREACH(p, &pg->pg_members, p_pglist) {
|
1994-05-24 10:09:53 +00:00
|
|
|
if (p->p_stat == SSTOP) {
|
2001-03-07 03:28:50 +00:00
|
|
|
mtx_unlock_spin(&sched_lock);
|
1999-11-16 16:28:58 +00:00
|
|
|
LIST_FOREACH(p, &pg->pg_members, p_pglist) {
|
2001-03-07 03:28:50 +00:00
|
|
|
PROC_LOCK(p);
|
1994-05-24 10:09:53 +00:00
|
|
|
psignal(p, SIGHUP);
|
|
|
|
|
psignal(p, SIGCONT);
|
2001-03-07 03:28:50 +00:00
|
|
|
PROC_UNLOCK(p);
|
1994-05-24 10:09:53 +00:00
|
|
|
}
|
|
|
|
|
return;
|
|
|
|
|
}
|
|
|
|
|
}
|
2001-03-07 03:28:50 +00:00
|
|
|
mtx_unlock_spin(&sched_lock);
|
1994-05-24 10:09:53 +00:00
|
|
|
}
|
|
|
|
|
|
1996-09-14 10:53:48 +00:00
|
|
|
#include "opt_ddb.h"
|
|
|
|
|
#ifdef DDB
|
|
|
|
|
#include <ddb/ddb.h>
|
|
|
|
|
|
|
|
|
|
DB_SHOW_COMMAND(pgrpdump, pgrpdump)
|
1994-05-24 10:09:53 +00:00
|
|
|
{
|
|
|
|
|
register struct pgrp *pgrp;
|
|
|
|
|
register struct proc *p;
|
1998-02-20 13:52:15 +00:00
|
|
|
register int i;
|
1994-05-24 10:09:53 +00:00
|
|
|
|
1996-03-11 06:05:03 +00:00
|
|
|
for (i = 0; i <= pgrphash; i++) {
|
1999-11-16 16:28:58 +00:00
|
|
|
if (!LIST_EMPTY(&pgrphashtbl[i])) {
|
1996-03-11 06:05:03 +00:00
|
|
|
printf("\tindx %d\n", i);
|
1999-11-16 16:28:58 +00:00
|
|
|
LIST_FOREACH(pgrp, &pgrphashtbl[i], pg_hash) {
|
1998-07-11 07:46:16 +00:00
|
|
|
printf(
|
|
|
|
|
"\tpgrp %p, pgid %ld, sess %p, sesscnt %d, mem %p\n",
|
|
|
|
|
(void *)pgrp, (long)pgrp->pg_id,
|
|
|
|
|
(void *)pgrp->pg_session,
|
1996-03-11 06:05:03 +00:00
|
|
|
pgrp->pg_session->s_count,
|
1999-11-16 16:28:58 +00:00
|
|
|
(void *)LIST_FIRST(&pgrp->pg_members));
|
|
|
|
|
LIST_FOREACH(p, &pgrp->pg_members, p_pglist) {
|
1998-07-11 07:46:16 +00:00
|
|
|
printf("\t\tpid %ld addr %p pgrp %p\n",
|
|
|
|
|
(long)p->p_pid, (void *)p,
|
|
|
|
|
(void *)p->p_pgrp);
|
1996-03-11 06:05:03 +00:00
|
|
|
}
|
|
|
|
|
}
|
1994-05-24 10:09:53 +00:00
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
}
|
1996-09-14 10:53:48 +00:00
|
|
|
#endif /* DDB */
|
1995-11-14 09:16:27 +00:00
|
|
|
|
|
|
|
|
/*
|
Change the proc information returned from the kernel so that it
no longer contains kernel specific data structures, but rather
only scalar values and structures that are already part of the
kernel/user interface, specifically rusage and rtprio. It no
longer contains proc, session, pcred, ucred, procsig, vmspace,
pstats, mtx, sigiolst, klist, callout, pasleep, or mdproc. If
any of these changed in size, ps, w, fstat, gcore, systat, and
top would all stop working. The new structure has over 200 bytes
of unassigned space for future values to be added, yet is nearly
100 bytes smaller per entry than the structure that it replaced.
2000-12-12 07:25:57 +00:00
|
|
|
* Fill in an kinfo_proc structure for the specified process.
|
1995-11-14 09:16:27 +00:00
|
|
|
*/
|
|
|
|
|
void
|
Change the proc information returned from the kernel so that it
no longer contains kernel specific data structures, but rather
only scalar values and structures that are already part of the
kernel/user interface, specifically rusage and rtprio. It no
longer contains proc, session, pcred, ucred, procsig, vmspace,
pstats, mtx, sigiolst, klist, callout, pasleep, or mdproc. If
any of these changed in size, ps, w, fstat, gcore, systat, and
top would all stop working. The new structure has over 200 bytes
of unassigned space for future values to be added, yet is nearly
100 bytes smaller per entry than the structure that it replaced.
2000-12-12 07:25:57 +00:00
|
|
|
fill_kinfo_proc(p, kp)
|
|
|
|
|
struct proc *p;
|
|
|
|
|
struct kinfo_proc *kp;
|
1995-11-14 09:16:27 +00:00
|
|
|
{
|
2001-09-12 08:38:13 +00:00
|
|
|
struct thread *td;
|
Change the proc information returned from the kernel so that it
no longer contains kernel specific data structures, but rather
only scalar values and structures that are already part of the
kernel/user interface, specifically rusage and rtprio. It no
longer contains proc, session, pcred, ucred, procsig, vmspace,
pstats, mtx, sigiolst, klist, callout, pasleep, or mdproc. If
any of these changed in size, ps, w, fstat, gcore, systat, and
top would all stop working. The new structure has over 200 bytes
of unassigned space for future values to be added, yet is nearly
100 bytes smaller per entry than the structure that it replaced.
2000-12-12 07:25:57 +00:00
|
|
|
struct tty *tp;
|
|
|
|
|
struct session *sp;
|
|
|
|
|
|
|
|
|
|
bzero(kp, sizeof(*kp));
|
|
|
|
|
|
2002-02-07 20:58:47 +00:00
|
|
|
td = FIRST_THREAD_IN_PROC(p);
|
Change the proc information returned from the kernel so that it
no longer contains kernel specific data structures, but rather
only scalar values and structures that are already part of the
kernel/user interface, specifically rusage and rtprio. It no
longer contains proc, session, pcred, ucred, procsig, vmspace,
pstats, mtx, sigiolst, klist, callout, pasleep, or mdproc. If
any of these changed in size, ps, w, fstat, gcore, systat, and
top would all stop working. The new structure has over 200 bytes
of unassigned space for future values to be added, yet is nearly
100 bytes smaller per entry than the structure that it replaced.
2000-12-12 07:25:57 +00:00
|
|
|
kp->ki_structsize = sizeof(*kp);
|
|
|
|
|
kp->ki_paddr = p;
|
2001-01-24 11:05:50 +00:00
|
|
|
PROC_LOCK(p);
|
2001-09-12 08:38:13 +00:00
|
|
|
kp->ki_addr =/* p->p_addr; */0; /* XXXKSE */
|
Change the proc information returned from the kernel so that it
no longer contains kernel specific data structures, but rather
only scalar values and structures that are already part of the
kernel/user interface, specifically rusage and rtprio. It no
longer contains proc, session, pcred, ucred, procsig, vmspace,
pstats, mtx, sigiolst, klist, callout, pasleep, or mdproc. If
any of these changed in size, ps, w, fstat, gcore, systat, and
top would all stop working. The new structure has over 200 bytes
of unassigned space for future values to be added, yet is nearly
100 bytes smaller per entry than the structure that it replaced.
2000-12-12 07:25:57 +00:00
|
|
|
kp->ki_args = p->p_args;
|
|
|
|
|
kp->ki_tracep = p->p_tracep;
|
|
|
|
|
kp->ki_textvp = p->p_textvp;
|
|
|
|
|
kp->ki_fd = p->p_fd;
|
|
|
|
|
kp->ki_vmspace = p->p_vmspace;
|
o Merge contents of struct pcred into struct ucred. Specifically, add the
real uid, saved uid, real gid, and saved gid to ucred, as well as the
pcred->pc_uidinfo, which was associated with the real uid, only rename
it to cr_ruidinfo so as not to conflict with cr_uidinfo, which
corresponds to the effective uid.
o Remove p_cred from struct proc; add p_ucred to struct proc, replacing
original macro that pointed.
p->p_ucred to p->p_cred->pc_ucred.
o Universally update code so that it makes use of ucred instead of pcred,
p->p_ucred instead of p->p_pcred, cr_ruidinfo instead of p_uidinfo,
cr_{r,sv}{u,g}id instead of p_*, etc.
o Remove pcred0 and its initialization from init_main.c; initialize
cr_ruidinfo there.
o Restruction many credential modification chunks to always crdup while
we figure out locking and optimizations; generally speaking, this
means moving to a structure like this:
newcred = crdup(oldcred);
...
p->p_ucred = newcred;
crfree(oldcred);
It's not race-free, but better than nothing. There are also races
in sys_process.c, all inter-process authorization, fork, exec, and
exit.
o Remove sigio->sio_ruid since sigio->sio_ucred now contains the ruid;
remove comments indicating that the old arrangement was a problem.
o Restructure exec1() a little to use newcred/oldcred arrangement, and
use improved uid management primitives.
o Clean up exit1() so as to do less work in credential cleanup due to
pcred removal.
o Clean up fork1() so as to do less work in credential cleanup and
allocation.
o Clean up ktrcanset() to take into account changes, and move to using
suser_xxx() instead of performing a direct uid==0 comparision.
o Improve commenting in various kern_prot.c credential modification
calls to better document current behavior. In a couple of places,
current behavior is a little questionable and we need to check
POSIX.1 to make sure it's "right". More commenting work still
remains to be done.
o Update credential management calls, such as crfree(), to take into
account new ruidinfo reference.
o Modify or add the following uid and gid helper routines:
change_euid()
change_egid()
change_ruid()
change_rgid()
change_svuid()
change_svgid()
In each case, the call now acts on a credential not a process, and as
such no longer requires more complicated process locking/etc. They
now assume the caller will do any necessary allocation of an
exclusive credential reference. Each is commented to document its
reference requirements.
o CANSIGIO() is simplified to require only credentials, not processes
and pcreds.
o Remove lots of (p_pcred==NULL) checks.
o Add an XXX to authorization code in nfs_lock.c, since it's
questionable, and needs to be considered carefully.
o Simplify posix4 authorization code to require only credentials, not
processes and pcreds. Note that this authorization, as well as
CANSIGIO(), needs to be updated to use the p_cansignal() and
p_cansched() centralized authorization routines, as they currently
do not take into account some desirable restrictions that are handled
by the centralized routines, as well as being inconsistent with other
similar authorization instances.
o Update libkvm to take these changes into account.
Obtained from: TrustedBSD Project
Reviewed by: green, bde, jhb, freebsd-arch, freebsd-audit
2001-05-25 16:59:11 +00:00
|
|
|
if (p->p_ucred) {
|
|
|
|
|
kp->ki_uid = p->p_ucred->cr_uid;
|
|
|
|
|
kp->ki_ruid = p->p_ucred->cr_ruid;
|
|
|
|
|
kp->ki_svuid = p->p_ucred->cr_svuid;
|
2001-08-16 23:43:24 +00:00
|
|
|
/* XXX bde doesn't like KI_NGROUPS */
|
|
|
|
|
kp->ki_ngroups = min(p->p_ucred->cr_ngroups, KI_NGROUPS);
|
o Merge contents of struct pcred into struct ucred. Specifically, add the
real uid, saved uid, real gid, and saved gid to ucred, as well as the
pcred->pc_uidinfo, which was associated with the real uid, only rename
it to cr_ruidinfo so as not to conflict with cr_uidinfo, which
corresponds to the effective uid.
o Remove p_cred from struct proc; add p_ucred to struct proc, replacing
original macro that pointed.
p->p_ucred to p->p_cred->pc_ucred.
o Universally update code so that it makes use of ucred instead of pcred,
p->p_ucred instead of p->p_pcred, cr_ruidinfo instead of p_uidinfo,
cr_{r,sv}{u,g}id instead of p_*, etc.
o Remove pcred0 and its initialization from init_main.c; initialize
cr_ruidinfo there.
o Restruction many credential modification chunks to always crdup while
we figure out locking and optimizations; generally speaking, this
means moving to a structure like this:
newcred = crdup(oldcred);
...
p->p_ucred = newcred;
crfree(oldcred);
It's not race-free, but better than nothing. There are also races
in sys_process.c, all inter-process authorization, fork, exec, and
exit.
o Remove sigio->sio_ruid since sigio->sio_ucred now contains the ruid;
remove comments indicating that the old arrangement was a problem.
o Restructure exec1() a little to use newcred/oldcred arrangement, and
use improved uid management primitives.
o Clean up exit1() so as to do less work in credential cleanup due to
pcred removal.
o Clean up fork1() so as to do less work in credential cleanup and
allocation.
o Clean up ktrcanset() to take into account changes, and move to using
suser_xxx() instead of performing a direct uid==0 comparision.
o Improve commenting in various kern_prot.c credential modification
calls to better document current behavior. In a couple of places,
current behavior is a little questionable and we need to check
POSIX.1 to make sure it's "right". More commenting work still
remains to be done.
o Update credential management calls, such as crfree(), to take into
account new ruidinfo reference.
o Modify or add the following uid and gid helper routines:
change_euid()
change_egid()
change_ruid()
change_rgid()
change_svuid()
change_svgid()
In each case, the call now acts on a credential not a process, and as
such no longer requires more complicated process locking/etc. They
now assume the caller will do any necessary allocation of an
exclusive credential reference. Each is commented to document its
reference requirements.
o CANSIGIO() is simplified to require only credentials, not processes
and pcreds.
o Remove lots of (p_pcred==NULL) checks.
o Add an XXX to authorization code in nfs_lock.c, since it's
questionable, and needs to be considered carefully.
o Simplify posix4 authorization code to require only credentials, not
processes and pcreds. Note that this authorization, as well as
CANSIGIO(), needs to be updated to use the p_cansignal() and
p_cansched() centralized authorization routines, as they currently
do not take into account some desirable restrictions that are handled
by the centralized routines, as well as being inconsistent with other
similar authorization instances.
o Update libkvm to take these changes into account.
Obtained from: TrustedBSD Project
Reviewed by: green, bde, jhb, freebsd-arch, freebsd-audit
2001-05-25 16:59:11 +00:00
|
|
|
bcopy(p->p_ucred->cr_groups, kp->ki_groups,
|
2001-08-16 23:43:24 +00:00
|
|
|
kp->ki_ngroups * sizeof(gid_t));
|
o Merge contents of struct pcred into struct ucred. Specifically, add the
real uid, saved uid, real gid, and saved gid to ucred, as well as the
pcred->pc_uidinfo, which was associated with the real uid, only rename
it to cr_ruidinfo so as not to conflict with cr_uidinfo, which
corresponds to the effective uid.
o Remove p_cred from struct proc; add p_ucred to struct proc, replacing
original macro that pointed.
p->p_ucred to p->p_cred->pc_ucred.
o Universally update code so that it makes use of ucred instead of pcred,
p->p_ucred instead of p->p_pcred, cr_ruidinfo instead of p_uidinfo,
cr_{r,sv}{u,g}id instead of p_*, etc.
o Remove pcred0 and its initialization from init_main.c; initialize
cr_ruidinfo there.
o Restruction many credential modification chunks to always crdup while
we figure out locking and optimizations; generally speaking, this
means moving to a structure like this:
newcred = crdup(oldcred);
...
p->p_ucred = newcred;
crfree(oldcred);
It's not race-free, but better than nothing. There are also races
in sys_process.c, all inter-process authorization, fork, exec, and
exit.
o Remove sigio->sio_ruid since sigio->sio_ucred now contains the ruid;
remove comments indicating that the old arrangement was a problem.
o Restructure exec1() a little to use newcred/oldcred arrangement, and
use improved uid management primitives.
o Clean up exit1() so as to do less work in credential cleanup due to
pcred removal.
o Clean up fork1() so as to do less work in credential cleanup and
allocation.
o Clean up ktrcanset() to take into account changes, and move to using
suser_xxx() instead of performing a direct uid==0 comparision.
o Improve commenting in various kern_prot.c credential modification
calls to better document current behavior. In a couple of places,
current behavior is a little questionable and we need to check
POSIX.1 to make sure it's "right". More commenting work still
remains to be done.
o Update credential management calls, such as crfree(), to take into
account new ruidinfo reference.
o Modify or add the following uid and gid helper routines:
change_euid()
change_egid()
change_ruid()
change_rgid()
change_svuid()
change_svgid()
In each case, the call now acts on a credential not a process, and as
such no longer requires more complicated process locking/etc. They
now assume the caller will do any necessary allocation of an
exclusive credential reference. Each is commented to document its
reference requirements.
o CANSIGIO() is simplified to require only credentials, not processes
and pcreds.
o Remove lots of (p_pcred==NULL) checks.
o Add an XXX to authorization code in nfs_lock.c, since it's
questionable, and needs to be considered carefully.
o Simplify posix4 authorization code to require only credentials, not
processes and pcreds. Note that this authorization, as well as
CANSIGIO(), needs to be updated to use the p_cansignal() and
p_cansched() centralized authorization routines, as they currently
do not take into account some desirable restrictions that are handled
by the centralized routines, as well as being inconsistent with other
similar authorization instances.
o Update libkvm to take these changes into account.
Obtained from: TrustedBSD Project
Reviewed by: green, bde, jhb, freebsd-arch, freebsd-audit
2001-05-25 16:59:11 +00:00
|
|
|
kp->ki_rgid = p->p_ucred->cr_rgid;
|
|
|
|
|
kp->ki_svgid = p->p_ucred->cr_svgid;
|
1995-11-14 09:16:27 +00:00
|
|
|
}
|
1999-11-17 12:49:22 +00:00
|
|
|
if (p->p_procsig) {
|
Change the proc information returned from the kernel so that it
no longer contains kernel specific data structures, but rather
only scalar values and structures that are already part of the
kernel/user interface, specifically rusage and rtprio. It no
longer contains proc, session, pcred, ucred, procsig, vmspace,
pstats, mtx, sigiolst, klist, callout, pasleep, or mdproc. If
any of these changed in size, ps, w, fstat, gcore, systat, and
top would all stop working. The new structure has over 200 bytes
of unassigned space for future values to be added, yet is nearly
100 bytes smaller per entry than the structure that it replaced.
2000-12-12 07:25:57 +00:00
|
|
|
kp->ki_sigignore = p->p_procsig->ps_sigignore;
|
|
|
|
|
kp->ki_sigcatch = p->p_procsig->ps_sigcatch;
|
1999-01-13 03:11:43 +00:00
|
|
|
}
|
Change and clean the mutex lock interface.
mtx_enter(lock, type) becomes:
mtx_lock(lock) for sleep locks (MTX_DEF-initialized locks)
mtx_lock_spin(lock) for spin locks (MTX_SPIN-initialized)
similarily, for releasing a lock, we now have:
mtx_unlock(lock) for MTX_DEF and mtx_unlock_spin(lock) for MTX_SPIN.
We change the caller interface for the two different types of locks
because the semantics are entirely different for each case, and this
makes it explicitly clear and, at the same time, it rids us of the
extra `type' argument.
The enter->lock and exit->unlock change has been made with the idea
that we're "locking data" and not "entering locked code" in mind.
Further, remove all additional "flags" previously passed to the
lock acquire/release routines with the exception of two:
MTX_QUIET and MTX_NOSWITCH
The functionality of these flags is preserved and they can be passed
to the lock/unlock routines by calling the corresponding wrappers:
mtx_{lock, unlock}_flags(lock, flag(s)) and
mtx_{lock, unlock}_spin_flags(lock, flag(s)) for MTX_DEF and MTX_SPIN
locks, respectively.
Re-inline some lock acq/rel code; in the sleep lock case, we only
inline the _obtain_lock()s in order to ensure that the inlined code
fits into a cache line. In the spin lock case, we inline recursion and
actually only perform a function call if we need to spin. This change
has been made with the idea that we generally tend to avoid spin locks
and that also the spin locks that we do have and are heavily used
(i.e. sched_lock) do recurse, and therefore in an effort to reduce
function call overhead for some architectures (such as alpha), we
inline recursion for this case.
Create a new malloc type for the witness code and retire from using
the M_DEV type. The new type is called M_WITNESS and is only declared
if WITNESS is enabled.
Begin cleaning up some machdep/mutex.h code - specifically updated the
"optimized" inlined code in alpha/mutex.h and wrote MTX_LOCK_SPIN
and MTX_UNLOCK_SPIN asm macros for the i386/mutex.h as we presently
need those.
Finally, caught up to the interface changes in all sys code.
Contributors: jake, jhb, jasone (in no particular order)
2001-02-09 06:11:45 +00:00
|
|
|
mtx_lock_spin(&sched_lock);
|
1995-11-14 09:16:27 +00:00
|
|
|
if (p->p_stat != SIDL && p->p_stat != SZOMB && p->p_vmspace != NULL) {
|
Change the proc information returned from the kernel so that it
no longer contains kernel specific data structures, but rather
only scalar values and structures that are already part of the
kernel/user interface, specifically rusage and rtprio. It no
longer contains proc, session, pcred, ucred, procsig, vmspace,
pstats, mtx, sigiolst, klist, callout, pasleep, or mdproc. If
any of these changed in size, ps, w, fstat, gcore, systat, and
top would all stop working. The new structure has over 200 bytes
of unassigned space for future values to be added, yet is nearly
100 bytes smaller per entry than the structure that it replaced.
2000-12-12 07:25:57 +00:00
|
|
|
struct vmspace *vm = p->p_vmspace;
|
|
|
|
|
|
|
|
|
|
kp->ki_size = vm->vm_map.size;
|
|
|
|
|
kp->ki_rssize = vmspace_resident_count(vm); /*XXX*/
|
2001-09-10 07:29:32 +00:00
|
|
|
if (p->p_sflag & PS_INMEM)
|
2001-09-12 08:38:13 +00:00
|
|
|
kp->ki_rssize += UAREA_PAGES;
|
|
|
|
|
FOREACH_THREAD_IN_PROC(p, td) /* XXXKSE: thread swapout check */
|
|
|
|
|
kp->ki_rssize += KSTACK_PAGES;
|
Change the proc information returned from the kernel so that it
no longer contains kernel specific data structures, but rather
only scalar values and structures that are already part of the
kernel/user interface, specifically rusage and rtprio. It no
longer contains proc, session, pcred, ucred, procsig, vmspace,
pstats, mtx, sigiolst, klist, callout, pasleep, or mdproc. If
any of these changed in size, ps, w, fstat, gcore, systat, and
top would all stop working. The new structure has over 200 bytes
of unassigned space for future values to be added, yet is nearly
100 bytes smaller per entry than the structure that it replaced.
2000-12-12 07:25:57 +00:00
|
|
|
kp->ki_swrss = vm->vm_swrss;
|
|
|
|
|
kp->ki_tsize = vm->vm_tsize;
|
|
|
|
|
kp->ki_dsize = vm->vm_dsize;
|
|
|
|
|
kp->ki_ssize = vm->vm_ssize;
|
|
|
|
|
}
|
2001-01-24 11:05:50 +00:00
|
|
|
if ((p->p_sflag & PS_INMEM) && p->p_stats) {
|
Change the proc information returned from the kernel so that it
no longer contains kernel specific data structures, but rather
only scalar values and structures that are already part of the
kernel/user interface, specifically rusage and rtprio. It no
longer contains proc, session, pcred, ucred, procsig, vmspace,
pstats, mtx, sigiolst, klist, callout, pasleep, or mdproc. If
any of these changed in size, ps, w, fstat, gcore, systat, and
top would all stop working. The new structure has over 200 bytes
of unassigned space for future values to be added, yet is nearly
100 bytes smaller per entry than the structure that it replaced.
2000-12-12 07:25:57 +00:00
|
|
|
kp->ki_start = p->p_stats->p_start;
|
|
|
|
|
kp->ki_rusage = p->p_stats->p_ru;
|
|
|
|
|
kp->ki_childtime.tv_sec = p->p_stats->p_cru.ru_utime.tv_sec +
|
|
|
|
|
p->p_stats->p_cru.ru_stime.tv_sec;
|
|
|
|
|
kp->ki_childtime.tv_usec = p->p_stats->p_cru.ru_utime.tv_usec +
|
|
|
|
|
p->p_stats->p_cru.ru_stime.tv_usec;
|
1995-11-14 09:16:27 +00:00
|
|
|
}
|
2002-02-07 20:58:47 +00:00
|
|
|
if (td->td_wmesg != NULL)
|
|
|
|
|
strncpy(kp->ki_wmesg, td->td_wmesg, sizeof(kp->ki_wmesg) - 1);
|
2001-01-24 11:05:50 +00:00
|
|
|
if (p->p_stat == SMTX) {
|
|
|
|
|
kp->ki_kiflag |= KI_MTXBLOCK;
|
2002-02-07 20:58:47 +00:00
|
|
|
strncpy(kp->ki_mtxname, td->td_mtxname,
|
2001-08-16 08:41:15 +00:00
|
|
|
sizeof(kp->ki_mtxname) - 1);
|
2001-01-24 11:05:50 +00:00
|
|
|
}
|
|
|
|
|
kp->ki_stat = p->p_stat;
|
2001-01-24 12:49:52 +00:00
|
|
|
kp->ki_sflag = p->p_sflag;
|
2001-01-24 11:05:50 +00:00
|
|
|
kp->ki_swtime = p->p_swtime;
|
|
|
|
|
kp->ki_traceflag = p->p_traceflag;
|
Change the proc information returned from the kernel so that it
no longer contains kernel specific data structures, but rather
only scalar values and structures that are already part of the
kernel/user interface, specifically rusage and rtprio. It no
longer contains proc, session, pcred, ucred, procsig, vmspace,
pstats, mtx, sigiolst, klist, callout, pasleep, or mdproc. If
any of these changed in size, ps, w, fstat, gcore, systat, and
top would all stop working. The new structure has over 200 bytes
of unassigned space for future values to be added, yet is nearly
100 bytes smaller per entry than the structure that it replaced.
2000-12-12 07:25:57 +00:00
|
|
|
kp->ki_pid = p->p_pid;
|
2001-09-12 08:38:13 +00:00
|
|
|
/* vvv XXXKSE */
|
|
|
|
|
kp->ki_runtime = p->p_runtime;
|
|
|
|
|
kp->ki_pctcpu = p->p_kse.ke_pctcpu;
|
2002-02-11 20:37:54 +00:00
|
|
|
kp->ki_estcpu = td->td_ksegrp->kg_estcpu;
|
|
|
|
|
kp->ki_slptime = td->td_ksegrp->kg_slptime;
|
2002-02-07 20:58:47 +00:00
|
|
|
kp->ki_wchan = td->td_wchan;
|
2002-02-11 20:37:54 +00:00
|
|
|
kp->ki_pri.pri_level = td->td_priority;
|
|
|
|
|
kp->ki_pri.pri_user = td->td_ksegrp->kg_user_pri;
|
|
|
|
|
kp->ki_pri.pri_class = td->td_ksegrp->kg_pri_class;
|
|
|
|
|
kp->ki_pri.pri_native = td->td_base_pri;
|
|
|
|
|
kp->ki_nice = td->td_ksegrp->kg_nice;
|
2001-09-12 08:38:13 +00:00
|
|
|
kp->ki_rqindex = p->p_kse.ke_rqindex;
|
|
|
|
|
kp->ki_oncpu = p->p_kse.ke_oncpu;
|
2002-02-07 20:58:47 +00:00
|
|
|
kp->ki_lastcpu = td->td_lastcpu;
|
|
|
|
|
kp->ki_tdflags = td->td_flags;
|
|
|
|
|
kp->ki_pcb = td->td_pcb;
|
|
|
|
|
kp->ki_kstack = (void *)td->td_kstack;
|
2001-09-12 08:38:13 +00:00
|
|
|
/* ^^^ XXXKSE */
|
Change and clean the mutex lock interface.
mtx_enter(lock, type) becomes:
mtx_lock(lock) for sleep locks (MTX_DEF-initialized locks)
mtx_lock_spin(lock) for spin locks (MTX_SPIN-initialized)
similarily, for releasing a lock, we now have:
mtx_unlock(lock) for MTX_DEF and mtx_unlock_spin(lock) for MTX_SPIN.
We change the caller interface for the two different types of locks
because the semantics are entirely different for each case, and this
makes it explicitly clear and, at the same time, it rids us of the
extra `type' argument.
The enter->lock and exit->unlock change has been made with the idea
that we're "locking data" and not "entering locked code" in mind.
Further, remove all additional "flags" previously passed to the
lock acquire/release routines with the exception of two:
MTX_QUIET and MTX_NOSWITCH
The functionality of these flags is preserved and they can be passed
to the lock/unlock routines by calling the corresponding wrappers:
mtx_{lock, unlock}_flags(lock, flag(s)) and
mtx_{lock, unlock}_spin_flags(lock, flag(s)) for MTX_DEF and MTX_SPIN
locks, respectively.
Re-inline some lock acq/rel code; in the sleep lock case, we only
inline the _obtain_lock()s in order to ensure that the inlined code
fits into a cache line. In the spin lock case, we inline recursion and
actually only perform a function call if we need to spin. This change
has been made with the idea that we generally tend to avoid spin locks
and that also the spin locks that we do have and are heavily used
(i.e. sched_lock) do recurse, and therefore in an effort to reduce
function call overhead for some architectures (such as alpha), we
inline recursion for this case.
Create a new malloc type for the witness code and retire from using
the M_DEV type. The new type is called M_WITNESS and is only declared
if WITNESS is enabled.
Begin cleaning up some machdep/mutex.h code - specifically updated the
"optimized" inlined code in alpha/mutex.h and wrote MTX_LOCK_SPIN
and MTX_UNLOCK_SPIN asm macros for the i386/mutex.h as we presently
need those.
Finally, caught up to the interface changes in all sys code.
Contributors: jake, jhb, jasone (in no particular order)
2001-02-09 06:11:45 +00:00
|
|
|
mtx_unlock_spin(&sched_lock);
|
Change the proc information returned from the kernel so that it
no longer contains kernel specific data structures, but rather
only scalar values and structures that are already part of the
kernel/user interface, specifically rusage and rtprio. It no
longer contains proc, session, pcred, ucred, procsig, vmspace,
pstats, mtx, sigiolst, klist, callout, pasleep, or mdproc. If
any of these changed in size, ps, w, fstat, gcore, systat, and
top would all stop working. The new structure has over 200 bytes
of unassigned space for future values to be added, yet is nearly
100 bytes smaller per entry than the structure that it replaced.
2000-12-12 07:25:57 +00:00
|
|
|
sp = NULL;
|
1995-11-14 09:16:27 +00:00
|
|
|
if (p->p_pgrp) {
|
Change the proc information returned from the kernel so that it
no longer contains kernel specific data structures, but rather
only scalar values and structures that are already part of the
kernel/user interface, specifically rusage and rtprio. It no
longer contains proc, session, pcred, ucred, procsig, vmspace,
pstats, mtx, sigiolst, klist, callout, pasleep, or mdproc. If
any of these changed in size, ps, w, fstat, gcore, systat, and
top would all stop working. The new structure has over 200 bytes
of unassigned space for future values to be added, yet is nearly
100 bytes smaller per entry than the structure that it replaced.
2000-12-12 07:25:57 +00:00
|
|
|
kp->ki_pgid = p->p_pgrp->pg_id;
|
|
|
|
|
kp->ki_jobc = p->p_pgrp->pg_jobc;
|
|
|
|
|
sp = p->p_pgrp->pg_session;
|
|
|
|
|
|
|
|
|
|
if (sp != NULL) {
|
|
|
|
|
kp->ki_sid = sp->s_sid;
|
2001-08-16 08:41:15 +00:00
|
|
|
strncpy(kp->ki_login, sp->s_login,
|
|
|
|
|
sizeof(kp->ki_login) - 1);
|
Change the proc information returned from the kernel so that it
no longer contains kernel specific data structures, but rather
only scalar values and structures that are already part of the
kernel/user interface, specifically rusage and rtprio. It no
longer contains proc, session, pcred, ucred, procsig, vmspace,
pstats, mtx, sigiolst, klist, callout, pasleep, or mdproc. If
any of these changed in size, ps, w, fstat, gcore, systat, and
top would all stop working. The new structure has over 200 bytes
of unassigned space for future values to be added, yet is nearly
100 bytes smaller per entry than the structure that it replaced.
2000-12-12 07:25:57 +00:00
|
|
|
if (sp->s_ttyvp)
|
2002-01-05 17:18:59 +00:00
|
|
|
kp->ki_kiflag |= KI_CTTY;
|
Change the proc information returned from the kernel so that it
no longer contains kernel specific data structures, but rather
only scalar values and structures that are already part of the
kernel/user interface, specifically rusage and rtprio. It no
longer contains proc, session, pcred, ucred, procsig, vmspace,
pstats, mtx, sigiolst, klist, callout, pasleep, or mdproc. If
any of these changed in size, ps, w, fstat, gcore, systat, and
top would all stop working. The new structure has over 200 bytes
of unassigned space for future values to be added, yet is nearly
100 bytes smaller per entry than the structure that it replaced.
2000-12-12 07:25:57 +00:00
|
|
|
if (SESS_LEADER(p))
|
|
|
|
|
kp->ki_kiflag |= KI_SLEADER;
|
1996-05-30 01:21:50 +00:00
|
|
|
}
|
1995-11-14 09:16:27 +00:00
|
|
|
}
|
Change the proc information returned from the kernel so that it
no longer contains kernel specific data structures, but rather
only scalar values and structures that are already part of the
kernel/user interface, specifically rusage and rtprio. It no
longer contains proc, session, pcred, ucred, procsig, vmspace,
pstats, mtx, sigiolst, klist, callout, pasleep, or mdproc. If
any of these changed in size, ps, w, fstat, gcore, systat, and
top would all stop working. The new structure has over 200 bytes
of unassigned space for future values to be added, yet is nearly
100 bytes smaller per entry than the structure that it replaced.
2000-12-12 07:25:57 +00:00
|
|
|
if ((p->p_flag & P_CONTROLT) && sp && ((tp = sp->s_ttyp) != NULL)) {
|
|
|
|
|
kp->ki_tdev = dev2udev(tp->t_dev);
|
|
|
|
|
kp->ki_tpgid = tp->t_pgrp ? tp->t_pgrp->pg_id : NO_PID;
|
|
|
|
|
if (tp->t_session)
|
|
|
|
|
kp->ki_tsid = tp->t_session->s_sid;
|
1995-11-14 09:16:27 +00:00
|
|
|
} else
|
Change the proc information returned from the kernel so that it
no longer contains kernel specific data structures, but rather
only scalar values and structures that are already part of the
kernel/user interface, specifically rusage and rtprio. It no
longer contains proc, session, pcred, ucred, procsig, vmspace,
pstats, mtx, sigiolst, klist, callout, pasleep, or mdproc. If
any of these changed in size, ps, w, fstat, gcore, systat, and
top would all stop working. The new structure has over 200 bytes
of unassigned space for future values to be added, yet is nearly
100 bytes smaller per entry than the structure that it replaced.
2000-12-12 07:25:57 +00:00
|
|
|
kp->ki_tdev = NOUDEV;
|
2001-08-16 08:41:15 +00:00
|
|
|
if (p->p_comm[0] != '\0') {
|
|
|
|
|
strncpy(kp->ki_comm, p->p_comm, sizeof(kp->ki_comm) - 1);
|
|
|
|
|
strncpy(kp->ki_ocomm, p->p_comm, sizeof(kp->ki_ocomm) - 1);
|
1995-11-14 09:16:27 +00:00
|
|
|
}
|
Change the proc information returned from the kernel so that it
no longer contains kernel specific data structures, but rather
only scalar values and structures that are already part of the
kernel/user interface, specifically rusage and rtprio. It no
longer contains proc, session, pcred, ucred, procsig, vmspace,
pstats, mtx, sigiolst, klist, callout, pasleep, or mdproc. If
any of these changed in size, ps, w, fstat, gcore, systat, and
top would all stop working. The new structure has over 200 bytes
of unassigned space for future values to be added, yet is nearly
100 bytes smaller per entry than the structure that it replaced.
2000-12-12 07:25:57 +00:00
|
|
|
kp->ki_siglist = p->p_siglist;
|
|
|
|
|
kp->ki_sigmask = p->p_sigmask;
|
|
|
|
|
kp->ki_xstat = p->p_xstat;
|
|
|
|
|
kp->ki_acflag = p->p_acflag;
|
|
|
|
|
kp->ki_flag = p->p_flag;
|
2001-02-21 06:39:57 +00:00
|
|
|
/* If jailed(p->p_ucred), emulate the old P_JAILED flag. */
|
|
|
|
|
if (jailed(p->p_ucred))
|
|
|
|
|
kp->ki_flag |= P_JAILED;
|
Change the proc information returned from the kernel so that it
no longer contains kernel specific data structures, but rather
only scalar values and structures that are already part of the
kernel/user interface, specifically rusage and rtprio. It no
longer contains proc, session, pcred, ucred, procsig, vmspace,
pstats, mtx, sigiolst, klist, callout, pasleep, or mdproc. If
any of these changed in size, ps, w, fstat, gcore, systat, and
top would all stop working. The new structure has over 200 bytes
of unassigned space for future values to be added, yet is nearly
100 bytes smaller per entry than the structure that it replaced.
2000-12-12 07:25:57 +00:00
|
|
|
kp->ki_lock = p->p_lock;
|
2001-01-24 11:05:50 +00:00
|
|
|
if (p->p_pptr)
|
|
|
|
|
kp->ki_ppid = p->p_pptr->p_pid;
|
2001-03-07 03:28:50 +00:00
|
|
|
PROC_UNLOCK(p);
|
1995-11-14 09:16:27 +00:00
|
|
|
}
|
|
|
|
|
|
2001-01-24 11:05:50 +00:00
|
|
|
/*
|
|
|
|
|
* Locate a zombie process by number
|
|
|
|
|
*/
|
|
|
|
|
struct proc *
|
1996-06-06 17:19:21 +00:00
|
|
|
zpfind(pid_t pid)
|
|
|
|
|
{
|
|
|
|
|
struct proc *p;
|
|
|
|
|
|
2001-03-28 11:52:56 +00:00
|
|
|
sx_slock(&allproc_lock);
|
1999-11-16 16:28:58 +00:00
|
|
|
LIST_FOREACH(p, &zombproc, p_list)
|
2001-04-24 00:51:53 +00:00
|
|
|
if (p->p_pid == pid) {
|
|
|
|
|
PROC_LOCK(p);
|
2000-12-13 00:17:05 +00:00
|
|
|
break;
|
2001-04-24 00:51:53 +00:00
|
|
|
}
|
2001-03-28 11:52:56 +00:00
|
|
|
sx_sunlock(&allproc_lock);
|
2000-12-13 00:17:05 +00:00
|
|
|
return (p);
|
1996-06-06 17:19:21 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
static int
|
|
|
|
|
sysctl_out_proc(struct proc *p, struct sysctl_req *req, int doingzomb)
|
|
|
|
|
{
|
Change the proc information returned from the kernel so that it
no longer contains kernel specific data structures, but rather
only scalar values and structures that are already part of the
kernel/user interface, specifically rusage and rtprio. It no
longer contains proc, session, pcred, ucred, procsig, vmspace,
pstats, mtx, sigiolst, klist, callout, pasleep, or mdproc. If
any of these changed in size, ps, w, fstat, gcore, systat, and
top would all stop working. The new structure has over 200 bytes
of unassigned space for future values to be added, yet is nearly
100 bytes smaller per entry than the structure that it replaced.
2000-12-12 07:25:57 +00:00
|
|
|
struct kinfo_proc kinfo_proc;
|
1996-06-06 17:19:21 +00:00
|
|
|
int error;
|
2001-04-24 00:51:53 +00:00
|
|
|
struct proc *np;
|
1996-06-06 17:19:21 +00:00
|
|
|
pid_t pid = p->p_pid;
|
|
|
|
|
|
Change the proc information returned from the kernel so that it
no longer contains kernel specific data structures, but rather
only scalar values and structures that are already part of the
kernel/user interface, specifically rusage and rtprio. It no
longer contains proc, session, pcred, ucred, procsig, vmspace,
pstats, mtx, sigiolst, klist, callout, pasleep, or mdproc. If
any of these changed in size, ps, w, fstat, gcore, systat, and
top would all stop working. The new structure has over 200 bytes
of unassigned space for future values to be added, yet is nearly
100 bytes smaller per entry than the structure that it replaced.
2000-12-12 07:25:57 +00:00
|
|
|
fill_kinfo_proc(p, &kinfo_proc);
|
|
|
|
|
error = SYSCTL_OUT(req, (caddr_t)&kinfo_proc, sizeof(kinfo_proc));
|
1996-06-06 17:19:21 +00:00
|
|
|
if (error)
|
|
|
|
|
return (error);
|
2001-04-24 00:51:53 +00:00
|
|
|
if (doingzomb)
|
|
|
|
|
np = zpfind(pid);
|
|
|
|
|
else {
|
|
|
|
|
if (pid == 0)
|
|
|
|
|
return (0);
|
|
|
|
|
np = pfind(pid);
|
|
|
|
|
}
|
|
|
|
|
if (np == NULL)
|
1996-06-06 17:19:21 +00:00
|
|
|
return EAGAIN;
|
2001-04-24 00:51:53 +00:00
|
|
|
if (np != p) {
|
|
|
|
|
PROC_UNLOCK(np);
|
1996-06-06 17:19:21 +00:00
|
|
|
return EAGAIN;
|
2001-04-24 00:51:53 +00:00
|
|
|
}
|
|
|
|
|
PROC_UNLOCK(np);
|
1996-06-06 17:19:21 +00:00
|
|
|
return (0);
|
|
|
|
|
}
|
|
|
|
|
|
1995-11-14 09:16:27 +00:00
|
|
|
static int
|
2000-07-04 11:25:35 +00:00
|
|
|
sysctl_kern_proc(SYSCTL_HANDLER_ARGS)
|
1995-11-14 09:16:27 +00:00
|
|
|
{
|
|
|
|
|
int *name = (int*) arg1;
|
|
|
|
|
u_int namelen = arg2;
|
|
|
|
|
struct proc *p;
|
|
|
|
|
int doingzomb;
|
|
|
|
|
int error = 0;
|
|
|
|
|
|
1996-06-06 17:19:21 +00:00
|
|
|
if (oidp->oid_number == KERN_PROC_PID) {
|
|
|
|
|
if (namelen != 1)
|
|
|
|
|
return (EINVAL);
|
|
|
|
|
p = pfind((pid_t)name[0]);
|
|
|
|
|
if (!p)
|
|
|
|
|
return (0);
|
2001-07-05 17:10:46 +00:00
|
|
|
if (p_cansee(curproc, p)) {
|
2001-04-24 00:51:53 +00:00
|
|
|
PROC_UNLOCK(p);
|
This Implements the mumbled about "Jail" feature.
This is a seriously beefed up chroot kind of thing. The process
is jailed along the same lines as a chroot does it, but with
additional tough restrictions imposed on what the superuser can do.
For all I know, it is safe to hand over the root bit inside a
prison to the customer living in that prison, this is what
it was developed for in fact: "real virtual servers".
Each prison has an ip number associated with it, which all IP
communications will be coerced to use and each prison has its own
hostname.
Needless to say, you need more RAM this way, but the advantage is
that each customer can run their own particular version of apache
and not stomp on the toes of their neighbors.
It generally does what one would expect, but setting up a jail
still takes a little knowledge.
A few notes:
I have no scripts for setting up a jail, don't ask me for them.
The IP number should be an alias on one of the interfaces.
mount a /proc in each jail, it will make ps more useable.
/proc/<pid>/status tells the hostname of the prison for
jailed processes.
Quotas are only sensible if you have a mountpoint per prison.
There are no privisions for stopping resource-hogging.
Some "#ifdef INET" and similar may be missing (send patches!)
If somebody wants to take it from here and develop it into
more of a "virtual machine" they should be most welcome!
Tools, comments, patches & documentation most welcome.
Have fun...
Sponsored by: http://www.rndassociates.com/
Run for almost a year by: http://www.servetheweb.com/
1999-04-28 11:38:52 +00:00
|
|
|
return (0);
|
2001-04-24 00:51:53 +00:00
|
|
|
}
|
|
|
|
|
PROC_UNLOCK(p);
|
1996-06-06 17:19:21 +00:00
|
|
|
error = sysctl_out_proc(p, req, 0);
|
|
|
|
|
return (error);
|
|
|
|
|
}
|
|
|
|
|
if (oidp->oid_number == KERN_PROC_ALL && !namelen)
|
|
|
|
|
;
|
|
|
|
|
else if (oidp->oid_number != KERN_PROC_ALL && namelen == 1)
|
|
|
|
|
;
|
|
|
|
|
else
|
1995-11-14 09:16:27 +00:00
|
|
|
return (EINVAL);
|
1996-06-06 17:19:21 +00:00
|
|
|
|
1995-11-14 09:16:27 +00:00
|
|
|
if (!req->oldptr) {
|
1996-06-06 17:19:21 +00:00
|
|
|
/* overestimate by 5 procs */
|
1995-11-14 09:16:27 +00:00
|
|
|
error = SYSCTL_OUT(req, 0, sizeof (struct kinfo_proc) * 5);
|
|
|
|
|
if (error)
|
|
|
|
|
return (error);
|
|
|
|
|
}
|
2001-03-28 11:52:56 +00:00
|
|
|
sx_slock(&allproc_lock);
|
1996-06-06 17:19:21 +00:00
|
|
|
for (doingzomb=0 ; doingzomb < 2 ; doingzomb++) {
|
|
|
|
|
if (!doingzomb)
|
1999-11-16 16:28:58 +00:00
|
|
|
p = LIST_FIRST(&allproc);
|
1996-06-06 17:19:21 +00:00
|
|
|
else
|
1999-11-16 16:28:58 +00:00
|
|
|
p = LIST_FIRST(&zombproc);
|
|
|
|
|
for (; p != 0; p = LIST_NEXT(p, p_list)) {
|
2000-08-23 21:41:25 +00:00
|
|
|
/*
|
o Centralize inter-process access control, introducing:
int p_can(p1, p2, operation, privused)
which allows specification of subject process, object process,
inter-process operation, and an optional call-by-reference privused
flag, allowing the caller to determine if privilege was required
for the call to succeed. This allows jail, kern.ps_showallprocs and
regular credential-based interaction checks to occur in one block of
code. Possible operations are P_CAN_SEE, P_CAN_SCHED, P_CAN_KILL,
and P_CAN_DEBUG. p_can currently breaks out as a wrapper to a
series of static function checks in kern_prot, which should not
be invoked directly.
o Commented out capabilities entries are included for some checks.
o Update most inter-process authorization to make use of p_can() instead
of manual checks, PRISON_CHECK(), P_TRESPASS(), and
kern.ps_showallprocs.
o Modify suser{,_xxx} to use const arguments, as it no longer modifies
process flags due to the disabling of ASU.
o Modify some checks/errors in procfs so that ENOENT is returned instead
of ESRCH, further improving concealment of processes that should not
be visible to other processes. Also introduce new access checks to
improve hiding of processes for procfs_lookup(), procfs_getattr(),
procfs_readdir(). Correct a bug reported by bp concerning not
handling the CREATE case in procfs_lookup(). Remove volatile flag in
procfs that caused apparently spurious qualifier warnigns (approved by
bde).
o Add comment noting that ktrace() has not been updated, as its access
control checks are different from ptrace(), whereas they should
probably be the same. Further discussion should happen on this topic.
Reviewed by: bde, green, phk, freebsd-security, others
Approved by: bde
Obtained from: TrustedBSD Project
2000-08-30 04:49:09 +00:00
|
|
|
* Show a user only appropriate processes.
|
2000-08-23 21:41:25 +00:00
|
|
|
*/
|
2001-07-05 17:10:46 +00:00
|
|
|
if (p_cansee(curproc, p))
|
2000-08-23 21:41:25 +00:00
|
|
|
continue;
|
1996-06-06 17:19:21 +00:00
|
|
|
/*
|
|
|
|
|
* Skip embryonic processes.
|
|
|
|
|
*/
|
|
|
|
|
if (p->p_stat == SIDL)
|
1995-11-14 09:16:27 +00:00
|
|
|
continue;
|
1996-06-06 17:19:21 +00:00
|
|
|
/*
|
|
|
|
|
* TODO - make more efficient (see notes below).
|
|
|
|
|
* do by session.
|
|
|
|
|
*/
|
|
|
|
|
switch (oidp->oid_number) {
|
|
|
|
|
|
|
|
|
|
case KERN_PROC_PGRP:
|
|
|
|
|
/* could do this by traversing pgrp */
|
|
|
|
|
if (p->p_pgrp == NULL ||
|
|
|
|
|
p->p_pgrp->pg_id != (pid_t)name[0])
|
|
|
|
|
continue;
|
|
|
|
|
break;
|
|
|
|
|
|
|
|
|
|
case KERN_PROC_TTY:
|
|
|
|
|
if ((p->p_flag & P_CONTROLT) == 0 ||
|
|
|
|
|
p->p_session == NULL ||
|
|
|
|
|
p->p_session->s_ttyp == NULL ||
|
1999-07-17 20:29:10 +00:00
|
|
|
dev2udev(p->p_session->s_ttyp->t_dev) !=
|
|
|
|
|
(udev_t)name[0])
|
1996-06-06 17:19:21 +00:00
|
|
|
continue;
|
|
|
|
|
break;
|
|
|
|
|
|
|
|
|
|
case KERN_PROC_UID:
|
|
|
|
|
if (p->p_ucred == NULL ||
|
|
|
|
|
p->p_ucred->cr_uid != (uid_t)name[0])
|
|
|
|
|
continue;
|
|
|
|
|
break;
|
|
|
|
|
|
|
|
|
|
case KERN_PROC_RUID:
|
|
|
|
|
if (p->p_ucred == NULL ||
|
o Merge contents of struct pcred into struct ucred. Specifically, add the
real uid, saved uid, real gid, and saved gid to ucred, as well as the
pcred->pc_uidinfo, which was associated with the real uid, only rename
it to cr_ruidinfo so as not to conflict with cr_uidinfo, which
corresponds to the effective uid.
o Remove p_cred from struct proc; add p_ucred to struct proc, replacing
original macro that pointed.
p->p_ucred to p->p_cred->pc_ucred.
o Universally update code so that it makes use of ucred instead of pcred,
p->p_ucred instead of p->p_pcred, cr_ruidinfo instead of p_uidinfo,
cr_{r,sv}{u,g}id instead of p_*, etc.
o Remove pcred0 and its initialization from init_main.c; initialize
cr_ruidinfo there.
o Restruction many credential modification chunks to always crdup while
we figure out locking and optimizations; generally speaking, this
means moving to a structure like this:
newcred = crdup(oldcred);
...
p->p_ucred = newcred;
crfree(oldcred);
It's not race-free, but better than nothing. There are also races
in sys_process.c, all inter-process authorization, fork, exec, and
exit.
o Remove sigio->sio_ruid since sigio->sio_ucred now contains the ruid;
remove comments indicating that the old arrangement was a problem.
o Restructure exec1() a little to use newcred/oldcred arrangement, and
use improved uid management primitives.
o Clean up exit1() so as to do less work in credential cleanup due to
pcred removal.
o Clean up fork1() so as to do less work in credential cleanup and
allocation.
o Clean up ktrcanset() to take into account changes, and move to using
suser_xxx() instead of performing a direct uid==0 comparision.
o Improve commenting in various kern_prot.c credential modification
calls to better document current behavior. In a couple of places,
current behavior is a little questionable and we need to check
POSIX.1 to make sure it's "right". More commenting work still
remains to be done.
o Update credential management calls, such as crfree(), to take into
account new ruidinfo reference.
o Modify or add the following uid and gid helper routines:
change_euid()
change_egid()
change_ruid()
change_rgid()
change_svuid()
change_svgid()
In each case, the call now acts on a credential not a process, and as
such no longer requires more complicated process locking/etc. They
now assume the caller will do any necessary allocation of an
exclusive credential reference. Each is commented to document its
reference requirements.
o CANSIGIO() is simplified to require only credentials, not processes
and pcreds.
o Remove lots of (p_pcred==NULL) checks.
o Add an XXX to authorization code in nfs_lock.c, since it's
questionable, and needs to be considered carefully.
o Simplify posix4 authorization code to require only credentials, not
processes and pcreds. Note that this authorization, as well as
CANSIGIO(), needs to be updated to use the p_cansignal() and
p_cansched() centralized authorization routines, as they currently
do not take into account some desirable restrictions that are handled
by the centralized routines, as well as being inconsistent with other
similar authorization instances.
o Update libkvm to take these changes into account.
Obtained from: TrustedBSD Project
Reviewed by: green, bde, jhb, freebsd-arch, freebsd-audit
2001-05-25 16:59:11 +00:00
|
|
|
p->p_ucred->cr_ruid != (uid_t)name[0])
|
1996-06-06 17:19:21 +00:00
|
|
|
continue;
|
|
|
|
|
break;
|
|
|
|
|
}
|
1995-11-14 09:16:27 +00:00
|
|
|
|
2001-07-05 17:10:46 +00:00
|
|
|
if (p_cansee(curproc, p))
|
This Implements the mumbled about "Jail" feature.
This is a seriously beefed up chroot kind of thing. The process
is jailed along the same lines as a chroot does it, but with
additional tough restrictions imposed on what the superuser can do.
For all I know, it is safe to hand over the root bit inside a
prison to the customer living in that prison, this is what
it was developed for in fact: "real virtual servers".
Each prison has an ip number associated with it, which all IP
communications will be coerced to use and each prison has its own
hostname.
Needless to say, you need more RAM this way, but the advantage is
that each customer can run their own particular version of apache
and not stomp on the toes of their neighbors.
It generally does what one would expect, but setting up a jail
still takes a little knowledge.
A few notes:
I have no scripts for setting up a jail, don't ask me for them.
The IP number should be an alias on one of the interfaces.
mount a /proc in each jail, it will make ps more useable.
/proc/<pid>/status tells the hostname of the prison for
jailed processes.
Quotas are only sensible if you have a mountpoint per prison.
There are no privisions for stopping resource-hogging.
Some "#ifdef INET" and similar may be missing (send patches!)
If somebody wants to take it from here and develop it into
more of a "virtual machine" they should be most welcome!
Tools, comments, patches & documentation most welcome.
Have fun...
Sponsored by: http://www.rndassociates.com/
Run for almost a year by: http://www.servetheweb.com/
1999-04-28 11:38:52 +00:00
|
|
|
continue;
|
|
|
|
|
|
1996-06-06 17:19:21 +00:00
|
|
|
error = sysctl_out_proc(p, req, doingzomb);
|
2000-11-22 07:42:04 +00:00
|
|
|
if (error) {
|
2001-03-28 11:52:56 +00:00
|
|
|
sx_sunlock(&allproc_lock);
|
1996-06-06 17:19:21 +00:00
|
|
|
return (error);
|
2000-11-22 07:42:04 +00:00
|
|
|
}
|
1995-11-14 09:16:27 +00:00
|
|
|
}
|
|
|
|
|
}
|
2001-03-28 11:52:56 +00:00
|
|
|
sx_sunlock(&allproc_lock);
|
1995-11-14 09:16:27 +00:00
|
|
|
return (0);
|
|
|
|
|
}
|
|
|
|
|
|
1999-11-16 20:31:58 +00:00
|
|
|
/*
|
|
|
|
|
* This sysctl allows a process to retrieve the argument list or process
|
|
|
|
|
* title for another process without groping around in the address space
|
|
|
|
|
* of the other process. It also allow a process to set its own "process
|
|
|
|
|
* title to a string of its own choice.
|
|
|
|
|
*/
|
|
|
|
|
static int
|
2000-07-04 11:25:35 +00:00
|
|
|
sysctl_kern_proc_args(SYSCTL_HANDLER_ARGS)
|
1999-11-16 20:31:58 +00:00
|
|
|
{
|
|
|
|
|
int *name = (int*) arg1;
|
|
|
|
|
u_int namelen = arg2;
|
|
|
|
|
struct proc *p;
|
|
|
|
|
struct pargs *pa;
|
|
|
|
|
int error = 0;
|
|
|
|
|
|
|
|
|
|
if (namelen != 1)
|
|
|
|
|
return (EINVAL);
|
|
|
|
|
|
|
|
|
|
p = pfind((pid_t)name[0]);
|
|
|
|
|
if (!p)
|
|
|
|
|
return (0);
|
|
|
|
|
|
2001-07-05 17:10:46 +00:00
|
|
|
if ((!ps_argsopen) && p_cansee(curproc, p)) {
|
2001-04-24 00:51:53 +00:00
|
|
|
PROC_UNLOCK(p);
|
1999-11-16 20:31:58 +00:00
|
|
|
return (0);
|
2001-04-24 00:51:53 +00:00
|
|
|
}
|
|
|
|
|
PROC_UNLOCK(p);
|
1999-11-16 20:31:58 +00:00
|
|
|
|
|
|
|
|
if (req->newptr && curproc != p)
|
|
|
|
|
return (EPERM);
|
|
|
|
|
|
|
|
|
|
if (req->oldptr && p->p_args != NULL)
|
|
|
|
|
error = SYSCTL_OUT(req, p->p_args->ar_args, p->p_args->ar_length);
|
|
|
|
|
if (req->newptr == NULL)
|
|
|
|
|
return (error);
|
|
|
|
|
|
2001-04-24 00:51:53 +00:00
|
|
|
PROC_LOCK(p);
|
2001-06-20 23:10:06 +00:00
|
|
|
pa = p->p_args;
|
1999-11-16 20:31:58 +00:00
|
|
|
p->p_args = NULL;
|
2001-03-07 03:28:50 +00:00
|
|
|
PROC_UNLOCK(p);
|
2001-06-20 23:10:06 +00:00
|
|
|
if (pa != NULL && --pa->ar_ref == 0)
|
|
|
|
|
FREE(pa, M_PARGS);
|
1999-11-16 20:31:58 +00:00
|
|
|
|
|
|
|
|
if (req->newlen + sizeof(struct pargs) > ps_arg_cache_limit)
|
|
|
|
|
return (error);
|
|
|
|
|
|
|
|
|
|
MALLOC(pa, struct pargs *, sizeof(struct pargs) + req->newlen,
|
|
|
|
|
M_PARGS, M_WAITOK);
|
|
|
|
|
pa->ar_ref = 1;
|
|
|
|
|
pa->ar_length = req->newlen;
|
|
|
|
|
error = SYSCTL_IN(req, pa->ar_args, req->newlen);
|
2001-03-07 03:28:50 +00:00
|
|
|
if (!error) {
|
|
|
|
|
PROC_LOCK(p);
|
1999-11-16 20:31:58 +00:00
|
|
|
p->p_args = pa;
|
2001-03-07 03:28:50 +00:00
|
|
|
PROC_UNLOCK(p);
|
|
|
|
|
} else
|
1999-11-16 20:31:58 +00:00
|
|
|
FREE(pa, M_PARGS);
|
|
|
|
|
return (error);
|
|
|
|
|
}
|
1996-06-06 17:19:21 +00:00
|
|
|
|
|
|
|
|
SYSCTL_NODE(_kern, KERN_PROC, proc, CTLFLAG_RD, 0, "Process table");
|
|
|
|
|
|
|
|
|
|
SYSCTL_PROC(_kern_proc, KERN_PROC_ALL, all, CTLFLAG_RD|CTLTYPE_STRUCT,
|
1999-05-03 23:57:32 +00:00
|
|
|
0, 0, sysctl_kern_proc, "S,proc", "Return entire process table");
|
1996-06-06 17:19:21 +00:00
|
|
|
|
|
|
|
|
SYSCTL_NODE(_kern_proc, KERN_PROC_PGRP, pgrp, CTLFLAG_RD,
|
|
|
|
|
sysctl_kern_proc, "Process table");
|
|
|
|
|
|
|
|
|
|
SYSCTL_NODE(_kern_proc, KERN_PROC_TTY, tty, CTLFLAG_RD,
|
|
|
|
|
sysctl_kern_proc, "Process table");
|
|
|
|
|
|
|
|
|
|
SYSCTL_NODE(_kern_proc, KERN_PROC_UID, uid, CTLFLAG_RD,
|
|
|
|
|
sysctl_kern_proc, "Process table");
|
|
|
|
|
|
|
|
|
|
SYSCTL_NODE(_kern_proc, KERN_PROC_RUID, ruid, CTLFLAG_RD,
|
|
|
|
|
sysctl_kern_proc, "Process table");
|
|
|
|
|
|
|
|
|
|
SYSCTL_NODE(_kern_proc, KERN_PROC_PID, pid, CTLFLAG_RD,
|
1995-11-14 09:16:27 +00:00
|
|
|
sysctl_kern_proc, "Process table");
|
1999-11-16 20:31:58 +00:00
|
|
|
|
2000-02-08 19:54:15 +00:00
|
|
|
SYSCTL_NODE(_kern_proc, KERN_PROC_ARGS, args, CTLFLAG_RW | CTLFLAG_ANYBODY,
|
|
|
|
|
sysctl_kern_proc_args, "Process argument list");
|
