freebsd-dev/sys/kern/kern_prot.c

1343 lines
29 KiB
C
Raw Normal View History

1994-05-24 10:09:53 +00:00
/*
* Copyright (c) 1982, 1986, 1989, 1990, 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. 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.
*
* @(#)kern_prot.c 8.6 (Berkeley) 1/21/94
1999-08-28 01:08:13 +00:00
* $FreeBSD$
1994-05-24 10:09:53 +00:00
*/
/*
* System calls related to processes and protection
*/
#include "opt_compat.h"
1994-05-24 10:09:53 +00:00
#include <sys/param.h>
#include <sys/acct.h>
#include <sys/systm.h>
#include <sys/sysproto.h>
#include <sys/kernel.h>
#include <sys/lock.h>
1994-05-24 10:09:53 +00:00
#include <sys/proc.h>
#include <sys/malloc.h>
#include <sys/pioctl.h>
#include <sys/resourcevar.h>
#include <sys/sysctl.h>
#include <sys/jail.h>
1994-05-24 10:09:53 +00:00
static MALLOC_DEFINE(M_CRED, "cred", "credentials");
#ifndef _SYS_SYSPROTO_H_
struct getpid_args {
1994-05-24 10:09:53 +00:00
int dummy;
};
#endif
1994-05-24 10:09:53 +00:00
Commit major SMP cleanups and move the BGL (big giant lock) in the syscall path inward. A system call may select whether it needs the MP lock or not (the default being that it does need it). A great deal of conditional SMP code for various deadended experiments has been removed. 'cil' and 'cml' have been removed entirely, and the locking around the cpl has been removed. The conditional separately-locked fast-interrupt code has been removed, meaning that interrupts must hold the CPL now (but they pretty much had to anyway). Another reason for doing this is that the original separate-lock for interrupts just doesn't apply to the interrupt thread mechanism being contemplated. Modifications to the cpl may now ONLY occur while holding the MP lock. For example, if an otherwise MP safe syscall needs to mess with the cpl, it must hold the MP lock for the duration and must (as usual) save/restore the cpl in a nested fashion. This is precursor work for the real meat coming later: avoiding having to hold the MP lock for common syscalls and I/O's and interrupt threads. It is expected that the spl mechanisms and new interrupt threading mechanisms will be able to run in tandem, allowing a slow piecemeal transition to occur. This patch should result in a moderate performance improvement due to the considerable amount of code that has been removed from the critical path, especially the simplification of the spl*() calls. The real performance gains will come later. Approved by: jkh Reviewed by: current, bde (exception.s) Some work taken from: luoqi's patch
2000-03-28 07:16:37 +00:00
/*
* getpid - MP SAFE
Commit major SMP cleanups and move the BGL (big giant lock) in the syscall path inward. A system call may select whether it needs the MP lock or not (the default being that it does need it). A great deal of conditional SMP code for various deadended experiments has been removed. 'cil' and 'cml' have been removed entirely, and the locking around the cpl has been removed. The conditional separately-locked fast-interrupt code has been removed, meaning that interrupts must hold the CPL now (but they pretty much had to anyway). Another reason for doing this is that the original separate-lock for interrupts just doesn't apply to the interrupt thread mechanism being contemplated. Modifications to the cpl may now ONLY occur while holding the MP lock. For example, if an otherwise MP safe syscall needs to mess with the cpl, it must hold the MP lock for the duration and must (as usual) save/restore the cpl in a nested fashion. This is precursor work for the real meat coming later: avoiding having to hold the MP lock for common syscalls and I/O's and interrupt threads. It is expected that the spl mechanisms and new interrupt threading mechanisms will be able to run in tandem, allowing a slow piecemeal transition to occur. This patch should result in a moderate performance improvement due to the considerable amount of code that has been removed from the critical path, especially the simplification of the spl*() calls. The real performance gains will come later. Approved by: jkh Reviewed by: current, bde (exception.s) Some work taken from: luoqi's patch
2000-03-28 07:16:37 +00:00
*/
1994-05-24 10:09:53 +00:00
/* ARGSUSED */
int
getpid(p, uap)
1994-05-24 10:09:53 +00:00
struct proc *p;
struct getpid_args *uap;
1994-05-24 10:09:53 +00:00
{
p->p_retval[0] = p->p_pid;
1994-05-24 10:09:53 +00:00
#if defined(COMPAT_43) || defined(COMPAT_SUNOS)
PROCTREE_LOCK(PT_SHARED);
p->p_retval[1] = p->p_pptr->p_pid;
PROCTREE_LOCK(PT_RELEASE);
1994-05-24 10:09:53 +00:00
#endif
return (0);
}
/*
* getppid - MP SAFE
*/
#ifndef _SYS_SYSPROTO_H_
struct getppid_args {
int dummy;
};
#endif
1994-05-24 10:09:53 +00:00
/* ARGSUSED */
int
getppid(p, uap)
1994-05-24 10:09:53 +00:00
struct proc *p;
struct getppid_args *uap;
1994-05-24 10:09:53 +00:00
{
PROCTREE_LOCK(PT_SHARED);
p->p_retval[0] = p->p_pptr->p_pid;
PROCTREE_LOCK(PT_RELEASE);
1994-05-24 10:09:53 +00:00
return (0);
}
Commit major SMP cleanups and move the BGL (big giant lock) in the syscall path inward. A system call may select whether it needs the MP lock or not (the default being that it does need it). A great deal of conditional SMP code for various deadended experiments has been removed. 'cil' and 'cml' have been removed entirely, and the locking around the cpl has been removed. The conditional separately-locked fast-interrupt code has been removed, meaning that interrupts must hold the CPL now (but they pretty much had to anyway). Another reason for doing this is that the original separate-lock for interrupts just doesn't apply to the interrupt thread mechanism being contemplated. Modifications to the cpl may now ONLY occur while holding the MP lock. For example, if an otherwise MP safe syscall needs to mess with the cpl, it must hold the MP lock for the duration and must (as usual) save/restore the cpl in a nested fashion. This is precursor work for the real meat coming later: avoiding having to hold the MP lock for common syscalls and I/O's and interrupt threads. It is expected that the spl mechanisms and new interrupt threading mechanisms will be able to run in tandem, allowing a slow piecemeal transition to occur. This patch should result in a moderate performance improvement due to the considerable amount of code that has been removed from the critical path, especially the simplification of the spl*() calls. The real performance gains will come later. Approved by: jkh Reviewed by: current, bde (exception.s) Some work taken from: luoqi's patch
2000-03-28 07:16:37 +00:00
/*
* Get process group ID; note that POSIX getpgrp takes no parameter
*
* MP SAFE
*/
#ifndef _SYS_SYSPROTO_H_
struct getpgrp_args {
int dummy;
};
#endif
int
getpgrp(p, uap)
1994-05-24 10:09:53 +00:00
struct proc *p;
struct getpgrp_args *uap;
1994-05-24 10:09:53 +00:00
{
p->p_retval[0] = p->p_pgrp->pg_id;
1994-05-24 10:09:53 +00:00
return (0);
}
/* Get an arbitary pid's process group id */
#ifndef _SYS_SYSPROTO_H_
struct getpgid_args {
pid_t pid;
};
#endif
int
getpgid(p, uap)
struct proc *p;
struct getpgid_args *uap;
{
struct proc *pt;
pt = p;
if (uap->pid == 0)
goto found;
if ((pt = pfind(uap->pid)) == 0)
return ESRCH;
found:
p->p_retval[0] = pt->p_pgrp->pg_id;
return 0;
}
/*
* Get an arbitary pid's session id.
*/
#ifndef _SYS_SYSPROTO_H_
struct getsid_args {
pid_t pid;
};
#endif
int
getsid(p, uap)
struct proc *p;
struct getsid_args *uap;
{
struct proc *pt;
pt = p;
if (uap->pid == 0)
goto found;
if ((pt = pfind(uap->pid)) == 0)
return ESRCH;
found:
p->p_retval[0] = pt->p_session->s_sid;
return 0;
}
/*
* getuid() - MP SAFE
*/
#ifndef _SYS_SYSPROTO_H_
struct getuid_args {
int dummy;
};
#endif
1994-05-24 10:09:53 +00:00
/* ARGSUSED */
int
getuid(p, uap)
1994-05-24 10:09:53 +00:00
struct proc *p;
struct getuid_args *uap;
1994-05-24 10:09:53 +00:00
{
p->p_retval[0] = p->p_cred->p_ruid;
1994-05-24 10:09:53 +00:00
#if defined(COMPAT_43) || defined(COMPAT_SUNOS)
p->p_retval[1] = p->p_ucred->cr_uid;
1994-05-24 10:09:53 +00:00
#endif
return (0);
}
/*
* geteuid() - MP SAFE
*/
#ifndef _SYS_SYSPROTO_H_
struct geteuid_args {
int dummy;
};
#endif
1994-05-24 10:09:53 +00:00
/* ARGSUSED */
int
geteuid(p, uap)
1994-05-24 10:09:53 +00:00
struct proc *p;
struct geteuid_args *uap;
1994-05-24 10:09:53 +00:00
{
p->p_retval[0] = p->p_ucred->cr_uid;
1994-05-24 10:09:53 +00:00
return (0);
}
/*
* getgid() - MP SAFE
*/
#ifndef _SYS_SYSPROTO_H_
struct getgid_args {
int dummy;
};
#endif
1994-05-24 10:09:53 +00:00
/* ARGSUSED */
int
getgid(p, uap)
1994-05-24 10:09:53 +00:00
struct proc *p;
struct getgid_args *uap;
1994-05-24 10:09:53 +00:00
{
p->p_retval[0] = p->p_cred->p_rgid;
1994-05-24 10:09:53 +00:00
#if defined(COMPAT_43) || defined(COMPAT_SUNOS)
p->p_retval[1] = p->p_ucred->cr_groups[0];
1994-05-24 10:09:53 +00:00
#endif
return (0);
}
/*
* Get effective group ID. The "egid" is groups[0], and could be obtained
* via getgroups. This syscall exists because it is somewhat painful to do
* correctly in a library function.
*/
#ifndef _SYS_SYSPROTO_H_
struct getegid_args {
int dummy;
};
#endif
1994-05-24 10:09:53 +00:00
/* ARGSUSED */
int
getegid(p, uap)
1994-05-24 10:09:53 +00:00
struct proc *p;
struct getegid_args *uap;
1994-05-24 10:09:53 +00:00
{
p->p_retval[0] = p->p_ucred->cr_groups[0];
1994-05-24 10:09:53 +00:00
return (0);
}
#ifndef _SYS_SYSPROTO_H_
1994-05-24 10:09:53 +00:00
struct getgroups_args {
u_int gidsetsize;
gid_t *gidset;
};
#endif
int
getgroups(p, uap)
1994-05-24 10:09:53 +00:00
struct proc *p;
register struct getgroups_args *uap;
{
register struct pcred *pc = p->p_cred;
register u_int ngrp;
int error;
if ((ngrp = uap->gidsetsize) == 0) {
p->p_retval[0] = pc->pc_ucred->cr_ngroups;
1994-05-24 10:09:53 +00:00
return (0);
}
if (ngrp < pc->pc_ucred->cr_ngroups)
return (EINVAL);
ngrp = pc->pc_ucred->cr_ngroups;
if ((error = copyout((caddr_t)pc->pc_ucred->cr_groups,
(caddr_t)uap->gidset, ngrp * sizeof(gid_t))))
1994-05-24 10:09:53 +00:00
return (error);
p->p_retval[0] = ngrp;
1994-05-24 10:09:53 +00:00
return (0);
}
#ifndef _SYS_SYSPROTO_H_
struct setsid_args {
int dummy;
};
#endif
1994-05-24 10:09:53 +00:00
/* ARGSUSED */
int
setsid(p, uap)
1994-05-24 10:09:53 +00:00
register struct proc *p;
struct setsid_args *uap;
1994-05-24 10:09:53 +00:00
{
if (p->p_pgid == p->p_pid || pgfind(p->p_pid)) {
return (EPERM);
} else {
(void)enterpgrp(p, p->p_pid, 1);
p->p_retval[0] = p->p_pid;
1994-05-24 10:09:53 +00:00
return (0);
}
}
/*
* set process group (setpgid/old setpgrp)
*
* caller does setpgid(targpid, targpgid)
*
* pid must be caller or child of caller (ESRCH)
* if a child
* pid must be in same session (EPERM)
* pid can't have done an exec (EACCES)
* if pgid != pid
* there must exist some pid in same session having pgid (EPERM)
* pid must not be session leader (EPERM)
*/
#ifndef _SYS_SYSPROTO_H_
1994-05-24 10:09:53 +00:00
struct setpgid_args {
int pid; /* target process id */
int pgid; /* target pgrp id */
};
#endif
1994-05-24 10:09:53 +00:00
/* ARGSUSED */
int
setpgid(curp, uap)
1994-05-24 10:09:53 +00:00
struct proc *curp;
register struct setpgid_args *uap;
{
register struct proc *targp; /* target process */
register struct pgrp *pgrp; /* target pgrp */
if (uap->pgid < 0)
return (EINVAL);
1994-05-24 10:09:53 +00:00
if (uap->pid != 0 && uap->pid != curp->p_pid) {
if ((targp = pfind(uap->pid)) == 0 || !inferior(targp))
return (ESRCH);
if (targp->p_pgrp == NULL || targp->p_session != curp->p_session)
1994-05-24 10:09:53 +00:00
return (EPERM);
if (targp->p_flag & P_EXEC)
return (EACCES);
} else
targp = curp;
if (SESS_LEADER(targp))
return (EPERM);
if (uap->pgid == 0)
uap->pgid = targp->p_pid;
else if (uap->pgid != targp->p_pid)
if ((pgrp = pgfind(uap->pgid)) == 0 ||
pgrp->pg_session != curp->p_session)
return (EPERM);
return (enterpgrp(targp, uap->pgid, 0));
}
/*
* Use the clause in B.4.2.2 that allows setuid/setgid to be 4.2/4.3BSD
* compatible. It says that setting the uid/gid to euid/egid is a special
* case of "appropriate privilege". Once the rules are expanded out, this
* basically means that setuid(nnn) sets all three id's, in all permitted
* cases unless _POSIX_SAVED_IDS is enabled. In that case, setuid(getuid())
* does not set the saved id - this is dangerous for traditional BSD
* programs. For this reason, we *really* do not want to set
* _POSIX_SAVED_IDS and do not want to clear POSIX_APPENDIX_B_4_2_2.
*/
#define POSIX_APPENDIX_B_4_2_2
#ifndef _SYS_SYSPROTO_H_
1994-05-24 10:09:53 +00:00
struct setuid_args {
uid_t uid;
};
#endif
1994-05-24 10:09:53 +00:00
/* ARGSUSED */
int
setuid(p, uap)
1994-05-24 10:09:53 +00:00
struct proc *p;
struct setuid_args *uap;
{
register struct pcred *pc = p->p_cred;
register uid_t uid;
int error;
/*
* See if we have "permission" by POSIX 1003.1 rules.
*
* Note that setuid(geteuid()) is a special case of
* "appropriate privileges" in appendix B.4.2.2. We need
* to use this clause to be compatible with traditional BSD
* semantics. Basically, it means that "setuid(xx)" sets all
* three id's (assuming you have privs).
*
* Notes on the logic. We do things in three steps.
* 1: We determine if the euid is going to change, and do EPERM
* right away. We unconditionally change the euid later if this
* test is satisfied, simplifying that part of the logic.
* 2: We determine if the real and/or saved uid's are going to
* change. Determined by compile options.
* 3: Change euid last. (after tests in #2 for "appropriate privs")
*/
1994-05-24 10:09:53 +00:00
uid = uap->uid;
if (uid != pc->p_ruid && /* allow setuid(getuid()) */
#ifdef _POSIX_SAVED_IDS
uid != pc->p_svuid && /* allow setuid(saved gid) */
#endif
#ifdef POSIX_APPENDIX_B_4_2_2 /* Use BSD-compat clause from B.4.2.2 */
uid != pc->pc_ucred->cr_uid && /* allow setuid(geteuid()) */
#endif
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
(error = suser_xxx(0, p, PRISON_ROOT)))
1994-05-24 10:09:53 +00:00
return (error);
#ifdef _POSIX_SAVED_IDS
1994-05-24 10:09:53 +00:00
/*
* Do we have "appropriate privileges" (are we root or uid == euid)
* If so, we are changing the real uid and/or saved uid.
1994-05-24 10:09:53 +00:00
*/
if (
#ifdef POSIX_APPENDIX_B_4_2_2 /* Use the clause from B.4.2.2 */
uid == pc->pc_ucred->cr_uid ||
#endif
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
suser_xxx(0, p, PRISON_ROOT) == 0) /* we are using privs */
#endif
{
/*
* Set the real uid and transfer proc count to new user.
*/
if (uid != pc->p_ruid) {
change_ruid(p, uid);
setsugid(p);
}
/*
* Set saved uid
*
* XXX always set saved uid even if not _POSIX_SAVED_IDS, as
* the security of seteuid() depends on it. B.4.2.2 says it
* is important that we should do this.
*/
if (pc->p_svuid != uid) {
pc->p_svuid = uid;
setsugid(p);
}
}
/*
* In all permitted cases, we are changing the euid.
* Copy credentials so other references do not see our changes.
*/
if (pc->pc_ucred->cr_uid != uid) {
change_euid(p, uid);
setsugid(p);
}
1994-05-24 10:09:53 +00:00
return (0);
}
#ifndef _SYS_SYSPROTO_H_
1994-05-24 10:09:53 +00:00
struct seteuid_args {
uid_t euid;
};
#endif
1994-05-24 10:09:53 +00:00
/* ARGSUSED */
int
seteuid(p, uap)
1994-05-24 10:09:53 +00:00
struct proc *p;
struct seteuid_args *uap;
{
register struct pcred *pc = p->p_cred;
register uid_t euid;
int error;
euid = uap->euid;
if (euid != pc->p_ruid && /* allow seteuid(getuid()) */
euid != pc->p_svuid && /* allow seteuid(saved uid) */
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
(error = suser_xxx(0, p, PRISON_ROOT)))
1994-05-24 10:09:53 +00:00
return (error);
/*
* Everything's okay, do it. Copy credentials so other references do
* not see our changes.
*/
if (pc->pc_ucred->cr_uid != euid) {
change_euid(p, euid);
setsugid(p);
}
1994-05-24 10:09:53 +00:00
return (0);
}
#ifndef _SYS_SYSPROTO_H_
1994-05-24 10:09:53 +00:00
struct setgid_args {
gid_t gid;
};
#endif
1994-05-24 10:09:53 +00:00
/* ARGSUSED */
int
setgid(p, uap)
1994-05-24 10:09:53 +00:00
struct proc *p;
struct setgid_args *uap;
{
register struct pcred *pc = p->p_cred;
register gid_t gid;
int error;
/*
* See if we have "permission" by POSIX 1003.1 rules.
*
* Note that setgid(getegid()) is a special case of
* "appropriate privileges" in appendix B.4.2.2. We need
* to use this clause to be compatible with traditional BSD
* semantics. Basically, it means that "setgid(xx)" sets all
* three id's (assuming you have privs).
*
* For notes on the logic here, see setuid() above.
*/
1994-05-24 10:09:53 +00:00
gid = uap->gid;
if (gid != pc->p_rgid && /* allow setgid(getgid()) */
#ifdef _POSIX_SAVED_IDS
gid != pc->p_svgid && /* allow setgid(saved gid) */
#endif
#ifdef POSIX_APPENDIX_B_4_2_2 /* Use BSD-compat clause from B.4.2.2 */
gid != pc->pc_ucred->cr_groups[0] && /* allow setgid(getegid()) */
#endif
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
(error = suser_xxx(0, p, PRISON_ROOT)))
1994-05-24 10:09:53 +00:00
return (error);
#ifdef _POSIX_SAVED_IDS
/*
* Do we have "appropriate privileges" (are we root or gid == egid)
* If so, we are changing the real uid and saved gid.
*/
if (
#ifdef POSIX_APPENDIX_B_4_2_2 /* use the clause from B.4.2.2 */
gid == pc->pc_ucred->cr_groups[0] ||
#endif
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
suser_xxx(0, p, PRISON_ROOT) == 0) /* we are using privs */
#endif
{
/*
* Set real gid
*/
if (pc->p_rgid != gid) {
pc->p_rgid = gid;
setsugid(p);
}
/*
* Set saved gid
*
* XXX always set saved gid even if not _POSIX_SAVED_IDS, as
* the security of setegid() depends on it. B.4.2.2 says it
* is important that we should do this.
*/
if (pc->p_svgid != gid) {
pc->p_svgid = gid;
setsugid(p);
}
}
/*
* In all cases permitted cases, we are changing the egid.
* Copy credentials so other references do not see our changes.
*/
if (pc->pc_ucred->cr_groups[0] != gid) {
pc->pc_ucred = crcopy(pc->pc_ucred);
pc->pc_ucred->cr_groups[0] = gid;
setsugid(p);
}
1994-05-24 10:09:53 +00:00
return (0);
}
#ifndef _SYS_SYSPROTO_H_
1994-05-24 10:09:53 +00:00
struct setegid_args {
gid_t egid;
};
#endif
1994-05-24 10:09:53 +00:00
/* ARGSUSED */
int
setegid(p, uap)
1994-05-24 10:09:53 +00:00
struct proc *p;
struct setegid_args *uap;
{
register struct pcred *pc = p->p_cred;
register gid_t egid;
int error;
egid = uap->egid;
if (egid != pc->p_rgid && /* allow setegid(getgid()) */
egid != pc->p_svgid && /* allow setegid(saved gid) */
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
(error = suser_xxx(0, p, PRISON_ROOT)))
1994-05-24 10:09:53 +00:00
return (error);
if (pc->pc_ucred->cr_groups[0] != egid) {
pc->pc_ucred = crcopy(pc->pc_ucred);
pc->pc_ucred->cr_groups[0] = egid;
setsugid(p);
}
1994-05-24 10:09:53 +00:00
return (0);
}
#ifndef _SYS_SYSPROTO_H_
1994-05-24 10:09:53 +00:00
struct setgroups_args {
u_int gidsetsize;
gid_t *gidset;
};
#endif
1994-05-24 10:09:53 +00:00
/* ARGSUSED */
int
setgroups(p, uap)
1994-05-24 10:09:53 +00:00
struct proc *p;
struct setgroups_args *uap;
{
register struct pcred *pc = p->p_cred;
register u_int ngrp;
int error;
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
if ((error = suser_xxx(0, p, PRISON_ROOT)))
1994-05-24 10:09:53 +00:00
return (error);
ngrp = uap->gidsetsize;
if (ngrp > NGROUPS)
1994-05-24 10:09:53 +00:00
return (EINVAL);
/*
* XXX A little bit lazy here. We could test if anything has
* changed before crcopy() and setting P_SUGID.
*/
1994-05-24 10:09:53 +00:00
pc->pc_ucred = crcopy(pc->pc_ucred);
if (ngrp < 1) {
/*
* setgroups(0, NULL) is a legitimate way of clearing the
* groups vector on non-BSD systems (which generally do not
* have the egid in the groups[0]). We risk security holes
* when running non-BSD software if we do not do the same.
*/
pc->pc_ucred->cr_ngroups = 1;
} else {
if ((error = copyin((caddr_t)uap->gidset,
(caddr_t)pc->pc_ucred->cr_groups, ngrp * sizeof(gid_t))))
return (error);
pc->pc_ucred->cr_ngroups = ngrp;
}
setsugid(p);
1994-05-24 10:09:53 +00:00
return (0);
}
#ifndef _SYS_SYSPROTO_H_
1994-05-24 10:09:53 +00:00
struct setreuid_args {
uid_t ruid;
uid_t euid;
1994-05-24 10:09:53 +00:00
};
#endif
1994-05-24 10:09:53 +00:00
/* ARGSUSED */
int
setreuid(p, uap)
1994-05-24 10:09:53 +00:00
register struct proc *p;
struct setreuid_args *uap;
{
register struct pcred *pc = p->p_cred;
register uid_t ruid, euid;
int error;
1994-05-24 10:09:53 +00:00
ruid = uap->ruid;
euid = uap->euid;
if (((ruid != (uid_t)-1 && ruid != pc->p_ruid && ruid != pc->p_svuid) ||
(euid != (uid_t)-1 && euid != pc->pc_ucred->cr_uid &&
euid != pc->p_ruid && euid != pc->p_svuid)) &&
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
(error = suser_xxx(0, p, PRISON_ROOT)) != 0)
return (error);
if (euid != (uid_t)-1 && pc->pc_ucred->cr_uid != euid) {
change_euid(p, euid);
setsugid(p);
}
if (ruid != (uid_t)-1 && pc->p_ruid != ruid) {
change_ruid(p, ruid);
setsugid(p);
}
if ((ruid != (uid_t)-1 || pc->pc_ucred->cr_uid != pc->p_ruid) &&
pc->p_svuid != pc->pc_ucred->cr_uid) {
pc->p_svuid = pc->pc_ucred->cr_uid;
setsugid(p);
}
return (0);
1994-05-24 10:09:53 +00:00
}
#ifndef _SYS_SYSPROTO_H_
1994-05-24 10:09:53 +00:00
struct setregid_args {
gid_t rgid;
gid_t egid;
1994-05-24 10:09:53 +00:00
};
#endif
1994-05-24 10:09:53 +00:00
/* ARGSUSED */
int
setregid(p, uap)
1994-05-24 10:09:53 +00:00
register struct proc *p;
struct setregid_args *uap;
{
register struct pcred *pc = p->p_cred;
register gid_t rgid, egid;
int error;
1994-05-24 10:09:53 +00:00
rgid = uap->rgid;
egid = uap->egid;
if (((rgid != (gid_t)-1 && rgid != pc->p_rgid && rgid != pc->p_svgid) ||
(egid != (gid_t)-1 && egid != pc->pc_ucred->cr_groups[0] &&
egid != pc->p_rgid && egid != pc->p_svgid)) &&
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
(error = suser_xxx(0, p, PRISON_ROOT)) != 0)
return (error);
if (egid != (gid_t)-1 && pc->pc_ucred->cr_groups[0] != egid) {
pc->pc_ucred = crcopy(pc->pc_ucred);
pc->pc_ucred->cr_groups[0] = egid;
setsugid(p);
}
if (rgid != (gid_t)-1 && pc->p_rgid != rgid) {
pc->p_rgid = rgid;
setsugid(p);
}
if ((rgid != (gid_t)-1 || pc->pc_ucred->cr_groups[0] != pc->p_rgid) &&
pc->p_svgid != pc->pc_ucred->cr_groups[0]) {
pc->p_svgid = pc->pc_ucred->cr_groups[0];
setsugid(p);
}
return (0);
1994-05-24 10:09:53 +00:00
}
/*
* setresuid(ruid, euid, suid) is like setreuid except control over the
* saved uid is explicit.
*/
#ifndef _SYS_SYSPROTO_H_
struct setresuid_args {
uid_t ruid;
uid_t euid;
uid_t suid;
};
#endif
/* ARGSUSED */
int
setresuid(p, uap)
register struct proc *p;
struct setresuid_args *uap;
{
register struct pcred *pc = p->p_cred;
register uid_t ruid, euid, suid;
int error;
ruid = uap->ruid;
euid = uap->euid;
suid = uap->suid;
if (((ruid != (uid_t)-1 && ruid != pc->p_ruid && ruid != pc->p_svuid &&
ruid != pc->pc_ucred->cr_uid) ||
(euid != (uid_t)-1 && euid != pc->p_ruid && euid != pc->p_svuid &&
euid != pc->pc_ucred->cr_uid) ||
(suid != (uid_t)-1 && suid != pc->p_ruid && suid != pc->p_svuid &&
suid != pc->pc_ucred->cr_uid)) &&
(error = suser_xxx(0, p, PRISON_ROOT)) != 0)
return (error);
if (euid != (uid_t)-1 && pc->pc_ucred->cr_uid != euid) {
change_euid(p, euid);
setsugid(p);
}
if (ruid != (uid_t)-1 && pc->p_ruid != ruid) {
change_ruid(p, ruid);
setsugid(p);
}
if (suid != (uid_t)-1 && pc->p_svuid != suid) {
pc->p_svuid = suid;
setsugid(p);
}
return (0);
}
/*
* setresgid(rgid, egid, sgid) is like setregid except control over the
* saved gid is explicit.
*/
#ifndef _SYS_SYSPROTO_H_
struct setresgid_args {
gid_t rgid;
gid_t egid;
gid_t sgid;
};
#endif
/* ARGSUSED */
int
setresgid(p, uap)
register struct proc *p;
struct setresgid_args *uap;
{
register struct pcred *pc = p->p_cred;
register gid_t rgid, egid, sgid;
int error;
rgid = uap->rgid;
egid = uap->egid;
sgid = uap->sgid;
if (((rgid != (gid_t)-1 && rgid != pc->p_rgid && rgid != pc->p_svgid &&
rgid != pc->pc_ucred->cr_groups[0]) ||
(egid != (gid_t)-1 && egid != pc->p_rgid && egid != pc->p_svgid &&
egid != pc->pc_ucred->cr_groups[0]) ||
(sgid != (gid_t)-1 && sgid != pc->p_rgid && sgid != pc->p_svgid &&
sgid != pc->pc_ucred->cr_groups[0])) &&
(error = suser_xxx(0, p, PRISON_ROOT)) != 0)
return (error);
if (egid != (gid_t)-1 && pc->pc_ucred->cr_groups[0] != egid) {
pc->pc_ucred = crcopy(pc->pc_ucred);
pc->pc_ucred->cr_groups[0] = egid;
setsugid(p);
}
if (rgid != (gid_t)-1 && pc->p_rgid != rgid) {
pc->p_rgid = rgid;
setsugid(p);
}
if (sgid != (gid_t)-1 && pc->p_svgid != sgid) {
pc->p_svgid = sgid;
setsugid(p);
}
return (0);
}
#ifndef _SYS_SYSPROTO_H_
struct getresuid_args {
uid_t *ruid;
uid_t *euid;
uid_t *suid;
};
#endif
/* ARGSUSED */
int
getresuid(p, uap)
register struct proc *p;
struct getresuid_args *uap;
{
struct pcred *pc = p->p_cred;
int error1 = 0, error2 = 0, error3 = 0;
if (uap->ruid)
error1 = copyout((caddr_t)&pc->p_ruid,
(caddr_t)uap->ruid, sizeof(pc->p_ruid));
if (uap->euid)
error2 = copyout((caddr_t)&pc->pc_ucred->cr_uid,
(caddr_t)uap->euid, sizeof(pc->pc_ucred->cr_uid));
if (uap->suid)
error3 = copyout((caddr_t)&pc->p_svuid,
(caddr_t)uap->suid, sizeof(pc->p_svuid));
return error1 ? error1 : (error2 ? error2 : error3);
}
#ifndef _SYS_SYSPROTO_H_
struct getresgid_args {
gid_t *rgid;
gid_t *egid;
gid_t *sgid;
};
#endif
/* ARGSUSED */
int
getresgid(p, uap)
register struct proc *p;
struct getresgid_args *uap;
{
struct pcred *pc = p->p_cred;
int error1 = 0, error2 = 0, error3 = 0;
if (uap->rgid)
error1 = copyout((caddr_t)&pc->p_rgid,
(caddr_t)uap->rgid, sizeof(pc->p_rgid));
if (uap->egid)
error2 = copyout((caddr_t)&pc->pc_ucred->cr_groups[0],
(caddr_t)uap->egid, sizeof(pc->pc_ucred->cr_groups[0]));
if (uap->sgid)
error3 = copyout((caddr_t)&pc->p_svgid,
(caddr_t)uap->sgid, sizeof(pc->p_svgid));
return error1 ? error1 : (error2 ? error2 : error3);
}
#ifndef _SYS_SYSPROTO_H_
struct issetugid_args {
int dummy;
};
#endif
/* ARGSUSED */
int
issetugid(p, uap)
register struct proc *p;
struct issetugid_args *uap;
{
/*
* Note: OpenBSD sets a P_SUGIDEXEC flag set at execve() time,
* we use P_SUGID because we consider changing the owners as
* "tainting" as well.
* This is significant for procs that start as root and "become"
* a user without an exec - programs cannot know *everything*
* that libc *might* have put in their data segment.
*/
p->p_retval[0] = (p->p_flag & P_SUGID) ? 1 : 0;
return (0);
}
1994-05-24 10:09:53 +00:00
/*
* Check if gid is a member of the group set.
*/
int
1994-05-24 10:09:53 +00:00
groupmember(gid, cred)
gid_t gid;
register struct ucred *cred;
{
register gid_t *gp;
gid_t *egp;
egp = &(cred->cr_groups[cred->cr_ngroups]);
for (gp = cred->cr_groups; gp < egp; gp++)
if (*gp == gid)
return (1);
return (0);
}
static int suser_permitted = 1;
SYSCTL_INT(_kern, OID_AUTO, suser_permitted, CTLFLAG_RW, &suser_permitted, 0,
"processes with uid 0 have privilege");
1994-05-24 10:09:53 +00:00
/*
* Test whether the specified credentials imply "super-user"
* privilege; if so, and we have accounting info, set the flag
* indicating use of super-powers.
* Returns 0 or error.
*/
int
suser(p)
struct proc *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 suser_xxx(0, p, 0);
}
int
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
suser_xxx(cred, proc, flag)
struct ucred *cred;
struct proc *proc;
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
int flag;
1994-05-24 10:09:53 +00:00
{
if (!suser_permitted)
return (EPERM);
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
if (!cred && !proc) {
printf("suser_xxx(): THINK!\n");
return (EPERM);
1994-05-24 10:09:53 +00:00
}
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
if (!cred)
cred = proc->p_ucred;
if (cred->cr_uid != 0)
return (EPERM);
if (jailed(cred) && !(flag & PRISON_ROOT))
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 (EPERM);
return (0);
1994-05-24 10:09:53 +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
static int
p_cansee(struct proc *p1, struct proc *p2, int *privused)
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
{
int error;
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
if (privused != NULL)
*privused = 0;
if ((error = prison_check(p1->p_ucred, p2->p_ucred)))
return (error);
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
if (!ps_showallprocs && p1->p_ucred->cr_uid != p2->p_ucred->cr_uid) {
if (suser_xxx(NULL, p1, PRISON_ROOT) == 0) {
if (privused != NULL)
*privused = 1;
return (0);
}
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
return (ESRCH);
}
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
return (0);
}
static int
p_cankill(struct proc *p1, struct proc *p2, int *privused)
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
{
int error;
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
if (privused != NULL)
*privused = 0;
if (p1 == p2)
return (0);
if ((error = prison_check(p1->p_ucred, p2->p_ucred)))
return (error);
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
if (p1->p_cred->p_ruid == p2->p_cred->p_ruid)
return (0);
if (p1->p_ucred->cr_uid == p2->p_cred->p_ruid)
return (0);
/*
* XXX should a process be able to affect another process
* acting as the same uid (i.e., a userland nfsd or the like?)
*/
if (p1->p_cred->p_ruid == p2->p_ucred->cr_uid)
return (0);
if (p1->p_ucred->cr_uid == p2->p_ucred->cr_uid)
return (0);
if (!suser_xxx(0, p1, PRISON_ROOT)) {
if (privused != NULL)
*privused = 1;
return (0);
}
#ifdef CAPABILITIES
if (!cap_check_xxx(0, p1, CAP_KILL, PRISON_ROOT)) {
if (privused != NULL)
*privused = 1;
return (0);
}
#endif
return (EPERM);
}
static int
p_cansched(struct proc *p1, struct proc *p2, int *privused)
{
int error;
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
if (privused != NULL)
*privused = 0;
if (p1 == p2)
return (0);
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
if ((error = prison_check(p1->p_ucred, p2->p_ucred)))
return (error);
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
if (p1->p_cred->p_ruid == p2->p_cred->p_ruid)
return (0);
if (p1->p_ucred->cr_uid == p2->p_cred->p_ruid)
return (0);
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
/*
* XXX should a process be able to affect another process
* acting as the same uid (i.e., a userland nfsd or the like?)
*/
if (p1->p_cred->p_ruid == p2->p_ucred->cr_uid)
return (0);
if (p1->p_ucred->cr_uid == p2->p_ucred->cr_uid)
return (0);
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
if (!suser_xxx(0, p1, PRISON_ROOT)) {
if (privused != NULL)
*privused = 1;
return (0);
}
#ifdef CAPABILITIES
if (!cap_check_xxx(0, p1, CAP_SYS_NICE, PRISON_ROOT)) {
if (privused != NULL)
*privused = 1;
return (0);
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
}
#endif
return (EPERM);
}
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
static int
p_candebug(struct proc *p1, struct proc *p2, int *privused)
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
{
int error;
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
if (privused != NULL)
*privused = 0;
/* XXX it is authorized, but semantics don't permit it */
if (p1 == p2)
return (0);
if ((error = prison_check(p1->p_ucred, p2->p_ucred)))
return (error);
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
/* not owned by you, has done setuid (unless you're root) */
/* add a CAP_SYS_PTRACE here? */
if (p1->p_cred->pc_ucred->cr_uid != p2->p_cred->p_ruid ||
p1->p_cred->p_ruid != p2->p_cred->p_ruid ||
p1->p_cred->p_svuid != p2->p_cred->p_ruid ||
p2->p_flag & P_SUGID) {
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
if ((error = suser_xxx(0, p1, PRISON_ROOT)))
return (error);
if (privused != NULL)
*privused = 1;
}
/* can't trace init when securelevel > 0 */
if (securelevel > 0 && p2->p_pid == 1)
return (EPERM);
return (0);
}
int
p_can(struct proc *p1, struct proc *p2, int operation,
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
int *privused)
{
switch(operation) {
case P_CAN_SEE:
return (p_cansee(p1, p2, privused));
case P_CAN_KILL:
return (p_cankill(p1, p2, privused));
case P_CAN_SCHED:
return (p_cansched(p1, p2, privused));
case P_CAN_DEBUG:
return (p_candebug(p1, p2, privused));
default:
panic("p_can: invalid operation");
}
}
1994-05-24 10:09:53 +00:00
/*
* Allocate a zeroed cred structure.
*/
struct ucred *
crget()
{
register struct ucred *cr;
MALLOC(cr, struct ucred *, sizeof(*cr), M_CRED, M_WAITOK|M_ZERO);
1994-05-24 10:09:53 +00:00
cr->cr_ref = 1;
mtx_init(&cr->cr_mtx, "ucred", MTX_DEF);
1994-05-24 10:09:53 +00:00
return (cr);
}
/*
* Claim another reference to a ucred structure
*/
void
crhold(cr)
struct ucred *cr;
{
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(&cr->cr_mtx);
cr->cr_ref++;
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(&(cr)->cr_mtx);
}
1994-05-24 10:09:53 +00:00
/*
* Free a cred structure.
* Throws away space when ref count gets to 0.
*/
void
1994-05-24 10:09:53 +00:00
crfree(cr)
struct ucred *cr;
{
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(&cr->cr_mtx);
if (--cr->cr_ref == 0) {
mtx_destroy(&cr->cr_mtx);
/*
* Some callers of crget(), such as nfs_statfs(),
* allocate a temporary credential, but don't
* allocate a uidinfo structure.
*/
if (cr->cr_uidinfo != NULL)
uifree(cr->cr_uidinfo);
/*
* Free a prison, if any.
*/
if (jailed(cr))
prison_free(cr->cr_prison);
1994-05-24 10:09:53 +00:00
FREE((caddr_t)cr, M_CRED);
} else {
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(&cr->cr_mtx);
}
1994-05-24 10:09:53 +00:00
}
/*
* Copy cred structure to a new one and free the old one.
*/
struct ucred *
crcopy(cr)
struct ucred *cr;
{
struct ucred *newcr;
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(&cr->cr_mtx);
if (cr->cr_ref == 1) {
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(&cr->cr_mtx);
1994-05-24 10:09:53 +00:00
return (cr);
}
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(&cr->cr_mtx);
newcr = crdup(cr);
1994-05-24 10:09:53 +00:00
crfree(cr);
return (newcr);
}
/*
* Dup cred struct to a new held one.
*/
struct ucred *
crdup(cr)
struct ucred *cr;
{
struct ucred *newcr;
MALLOC(newcr, struct ucred *, sizeof(*cr), M_CRED, M_WAITOK);
1994-05-24 10:09:53 +00:00
*newcr = *cr;
mtx_init(&newcr->cr_mtx, "ucred", MTX_DEF);
uihold(newcr->cr_uidinfo);
if (jailed(newcr))
prison_hold(newcr->cr_prison);
1994-05-24 10:09:53 +00:00
newcr->cr_ref = 1;
return (newcr);
}
/*
* Get login name, if available.
*/
#ifndef _SYS_SYSPROTO_H_
1994-05-24 10:09:53 +00:00
struct getlogin_args {
char *namebuf;
u_int namelen;
};
#endif
1994-05-24 10:09:53 +00:00
/* ARGSUSED */
int
getlogin(p, uap)
1994-05-24 10:09:53 +00:00
struct proc *p;
struct getlogin_args *uap;
{
if (uap->namelen > MAXLOGNAME)
1997-03-03 23:02:43 +00:00
uap->namelen = MAXLOGNAME;
1994-05-24 10:09:53 +00:00
return (copyout((caddr_t) p->p_pgrp->pg_session->s_login,
(caddr_t) uap->namebuf, uap->namelen));
}
/*
* Set login name.
*/
#ifndef _SYS_SYSPROTO_H_
1994-05-24 10:09:53 +00:00
struct setlogin_args {
char *namebuf;
};
#endif
1994-05-24 10:09:53 +00:00
/* ARGSUSED */
int
setlogin(p, uap)
1994-05-24 10:09:53 +00:00
struct proc *p;
struct setlogin_args *uap;
{
int error;
char logintmp[MAXLOGNAME];
1994-05-24 10:09:53 +00:00
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
if ((error = suser_xxx(0, p, PRISON_ROOT)))
1994-05-24 10:09:53 +00:00
return (error);
error = copyinstr((caddr_t) uap->namebuf, (caddr_t) logintmp,
sizeof(logintmp), (size_t *)0);
1994-05-24 10:09:53 +00:00
if (error == ENAMETOOLONG)
error = EINVAL;
else if (!error)
(void) memcpy(p->p_pgrp->pg_session->s_login, logintmp,
sizeof(logintmp));
1994-05-24 10:09:53 +00:00
return (error);
}
void
setsugid(p)
struct proc *p;
{
p->p_flag |= P_SUGID;
if (!(p->p_pfsflags & PF_ISUGID))
p->p_stops = 0;
}
/*
* Helper function to change the effective uid of a process
*/
void
change_euid(p, euid)
struct proc *p;
uid_t euid;
{
struct pcred *pc;
struct uidinfo *uip;
pc = p->p_cred;
/*
* crcopy is essentially a NOP if ucred has a reference count
* of 1, which is true if it has already been copied.
*/
pc->pc_ucred = crcopy(pc->pc_ucred);
uip = pc->pc_ucred->cr_uidinfo;
pc->pc_ucred->cr_uid = euid;
pc->pc_ucred->cr_uidinfo = uifind(euid);
uifree(uip);
}
/*
* Helper function to change the real uid of a process
*
* The per-uid process count for this process is transfered from
* the old uid to the new uid.
*/
void
change_ruid(p, ruid)
struct proc *p;
uid_t ruid;
{
struct pcred *pc;
struct uidinfo *uip;
pc = p->p_cred;
(void)chgproccnt(pc->p_uidinfo, -1, 0);
uip = pc->p_uidinfo;
/* It is assumed that pcred is not shared between processes */
pc->p_ruid = ruid;
pc->p_uidinfo = uifind(ruid);
(void)chgproccnt(pc->p_uidinfo, 1, 0);
uifree(uip);
}