Author: jhb
Date: Thu Sep 19 18:53:42 2013
New Revision: 255708
URL: http://svnweb.freebsd.org/changeset/base/255708
Log:
Extend the support for exempting processes from being killed when swap is
exhausted.
- Add a new protect(1) command that can be used to set or revoke protection
from arbitrary processes. Similar to ktrace it can apply a change to all
existing descendants of a process as well as future descendants.
- Add a new procctl(2) system call that provides a generic interface for
control operations on processes (as opposed to the debugger-specific
operations provided by ptrace(2)). procctl(2) uses a combination of
idtype_t and an id to identify the set of processes on which to operate
similar to wait6().
- Add a PROC_SPROTECT control operation to manage the protection status
of a set of processes. MADV_PROTECT still works for backwards
compatability.
- Add a p_flag2 to struct proc (and a corresponding ki_flag2 to kinfo_proc)
the first bit of which is used to track if P_PROTECT should be inherited
by new child processes.
Reviewed by: kib, jilles (earlier version)
Approved by: re (delphij)
MFC after: 1 month
Added:
head/lib/libc/sys/procctl.2 (contents, props changed)
head/sys/sys/procctl.h (contents, props changed)
head/usr.bin/protect/
head/usr.bin/protect/Makefile (contents, props changed)
head/usr.bin/protect/protect.1 (contents, props changed)
head/usr.bin/protect/protect.c (contents, props changed)
Modified:
head/lib/libc/sys/Makefile.inc
head/lib/libc/sys/Symbol.map
head/sys/compat/freebsd32/freebsd32.h
head/sys/compat/freebsd32/freebsd32_misc.c
head/sys/compat/freebsd32/syscalls.master
head/sys/kern/init_main.c
head/sys/kern/kern_fork.c
head/sys/kern/kern_proc.c
head/sys/kern/sys_process.c
head/sys/kern/syscalls.master
head/sys/sys/proc.h
head/sys/sys/syscallsubr.h
head/sys/sys/user.h
head/sys/vm/vm_mmap.c
head/usr.bin/Makefile
head/usr.bin/kdump/kdump.c
head/usr.bin/kdump/mksubr
head/usr.bin/truss/syscall.h
head/usr.bin/truss/syscalls.c
Modified: head/lib/libc/sys/Makefile.inc
==============================================================================
--- head/lib/libc/sys/Makefile.inc Thu Sep 19 18:00:05 2013
(r255707)
+++ head/lib/libc/sys/Makefile.inc Thu Sep 19 18:53:42 2013
(r255708)
@@ -197,6 +197,7 @@ MAN+= abort2.2 \
posix_fadvise.2 \
posix_fallocate.2 \
posix_openpt.2 \
+ procctl.2 \
profil.2 \
pselect.2 \
ptrace.2 \
Modified: head/lib/libc/sys/Symbol.map
==============================================================================
--- head/lib/libc/sys/Symbol.map Thu Sep 19 18:00:05 2013
(r255707)
+++ head/lib/libc/sys/Symbol.map Thu Sep 19 18:53:42 2013
(r255708)
@@ -395,6 +395,7 @@ FBSD_1.3 {
ffclock_setestimate;
pipe2;
posix_fadvise;
+ procctl;
wait6;
};
@@ -822,6 +823,8 @@ FBSDprivate_1.0 {
__sys_poll;
_preadv;
__sys_preadv;
+ _procctl;
+ __sys_procctl;
_profil;
__sys_profil;
_pselect;
Added: head/lib/libc/sys/procctl.2
==============================================================================
--- /dev/null 00:00:00 1970 (empty, because file is newly added)
+++ head/lib/libc/sys/procctl.2 Thu Sep 19 18:53:42 2013 (r255708)
@@ -0,0 +1,142 @@
+.\" Copyright (c) 2013 Advanced Computing Technologies LLC
+.\" Written by: John H. Baldwin <j...@freebsd.org>
+.\" All rights reserved.
+.\"
+.\" Redistribution and use in source and binary forms, with or without
+.\" modification, are permitted provided that the following conditions
+.\" are met:
+.\" 1. Redistributions of source code must retain the above copyright
+.\" notice, this list of conditions and the following disclaimer.
+.\" 2. Redistributions in binary form must reproduce the above copyright
+.\" notice, this list of conditions and the following disclaimer in the
+.\" documentation and/or other materials provided with the distribution.
+.\"
+.\" THIS SOFTWARE IS PROVIDED BY THE AUTHOR 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 AUTHOR 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.
+.\"
+.\" $FreeBSD$
+.\"
+.Dd September 19, 2013
+.Dt PROCCTL 2
+.Os
+.Sh NAME
+.Nm procctl
+.Nd control processes
+.Sh LIBRARY
+.Lb libc
+.Sh SYNOPSIS
+.In sys/procctl.h
+.Ft int
+.Fn procctl "idtype_t idtype" "id_t id" "int cmd" "void *arg"
+.Sh DESCRIPTION
+The
+.Fn procctl
+system call provides for control over processes.
+The
+.Fa idtype
+and
+.Fa id
+arguments specify the set of processes to control.
+If multiple processes match the identifier,
+.Nm
+will make a
+.Dq best effort
+to control as many of the selected possibles as possible.
+An error is only returned if no selected processes successfully complete
+the request.
+The following identifier types are supported:
+.Bl -tag -width Dv P_PGID
+.It Dv P_PID
+Control the process with the process ID
+.Fa id .
+.It Dv P_PGID
+Control processes belonging to the process group with the ID
+.Fa id .
+.El
+.Pp
+The control request to perform is specified by the
+.Fa cmd
+argument.
+The following commands are supported:
+.Bl -tag -width Dv PROC_SPROTECT
+.It Dv PROC_SPROTECT
+Set process protection state.
+This is used to mark a process as protected from being killed if the system
+exhausts available memory and swap.
+The
+.Fa arg
+parameter must point to an integer containing an operation and zero or more
+optional flags.
+The following operations are supported:
+.Bl -tag -width Dv PPROT_CLEAR
+.It Dv PPROT_SET
+Mark the selected processes as protected.
+.It Dv PPROT_CLEAR
+Clear the protected state of selected processes.
+.El
+.Pp
+The following optional flags are supported:
+.Bl -tag -width Dv PPROT_DESCEND
+.It Dv PPROT_DESCEND
+Apply the requested operation to all child processes of each selected process
+in addition to each selected process.
+.It Dv PPROT_INHERIT
+When used with
+.Dv PPROT_SET ,
+mark all future child processes of each selected process as protected.
+Future child processes will also mark all of their future child processes.
+.El
+.El
+.Sh RETURN VALUES
+If an error occurs, a value of -1 is returned and
+.Va errno
+is set to indicate the error.
+.Sh ERRORS
+The
+.Fn procctl
+system call
+will fail if:
+.Bl -tag -width Er
+.It Bq Er EFAULT
+The
+.Fa arg
+points outside the process's allocated address space.
+.It Bq Er EINVAL
+The
+.Fa cmd
+argument specifies an unsupported command.
+.Pp
+The
+.Fa idtype
+argument specifies an unsupported identifier type.
+.It Bq Er EPERM
+The calling process does not have permission to perform the requested
+operation on any of the selected processes.
+.It Bq Er ESRCH
+No processes matched the requested
+.Fa idtype
+and
+.Fa id .
+.It Bq Er EINVAL
+An invalid operation or flag was passed in
+.Fa arg
+for a
+.Dv PROC_SPROTECT
+command.
+.El
+.Sh SEE ALSO
+.Xr ptrace 2
+.Sh HISTORY
+The
+.Fn procctl
+function appeared in
+.Fx 10 .
Modified: head/sys/compat/freebsd32/freebsd32.h
==============================================================================
--- head/sys/compat/freebsd32/freebsd32.h Thu Sep 19 18:00:05 2013
(r255707)
+++ head/sys/compat/freebsd32/freebsd32.h Thu Sep 19 18:53:42 2013
(r255708)
@@ -342,6 +342,7 @@ struct kinfo_proc32 {
char ki_loginclass[LOGINCLASSLEN+1];
char ki_sparestrings[50];
int ki_spareints[KI_NSPARE_INT];
+ int ki_flag2;
int ki_fibnum;
u_int ki_cr_flags;
int ki_jid;
Modified: head/sys/compat/freebsd32/freebsd32_misc.c
==============================================================================
--- head/sys/compat/freebsd32/freebsd32_misc.c Thu Sep 19 18:00:05 2013
(r255707)
+++ head/sys/compat/freebsd32/freebsd32_misc.c Thu Sep 19 18:53:42 2013
(r255708)
@@ -56,6 +56,7 @@ __FBSDID("$FreeBSD$");
#include <sys/mutex.h>
#include <sys/namei.h>
#include <sys/proc.h>
+#include <sys/procctl.h>
#include <sys/reboot.h>
#include <sys/resource.h>
#include <sys/resourcevar.h>
@@ -3000,3 +3001,23 @@ convert_sigevent32(struct sigevent32 *si
}
return (0);
}
+
+int
+freebsd32_procctl(struct thread *td, struct freebsd32_procctl_args *uap)
+{
+ void *data;
+ int error, flags;
+
+ switch (uap->com) {
+ case PROC_SPROTECT:
+ error = copyin(PTRIN(uap->data), &flags, sizeof(flags));
+ if (error)
+ return (error);
+ data = &flags;
+ break;
+ default:
+ return (EINVAL);
+ }
+ return (kern_procctl(td, uap->idtype, PAIR32TO64(id_t, uap->id),
+ uap->com, data));
+}
Modified: head/sys/compat/freebsd32/syscalls.master
==============================================================================
--- head/sys/compat/freebsd32/syscalls.master Thu Sep 19 18:00:05 2013
(r255707)
+++ head/sys/compat/freebsd32/syscalls.master Thu Sep 19 18:53:42 2013
(r255708)
@@ -1056,3 +1056,12 @@
542 AUE_PIPE NOPROTO { int pipe2(int *fildes, int flags); }
543 AUE_NULL NOSTD { int freebsd32_aio_mlock( \
struct aiocb32 *aiocbp); }
+#ifdef PAD64_REQUIRED
+544 AUE_NULL STD { int freebsd32_procctl(int idtype, int pad, \
+ uint32_t id1, uint32_t id2, int com, \
+ void *data); }
+#else
+544 AUE_NULL STD { int freebsd32_procctl(int idtype, \
+ uint32_t id1, uint32_t id2, int com, \
+ void *data); }
+#endif
Modified: head/sys/kern/init_main.c
==============================================================================
--- head/sys/kern/init_main.c Thu Sep 19 18:00:05 2013 (r255707)
+++ head/sys/kern/init_main.c Thu Sep 19 18:53:42 2013 (r255708)
@@ -474,6 +474,7 @@ proc0_init(void *dummy __unused)
p->p_sysent = &null_sysvec;
p->p_flag = P_SYSTEM | P_INMEM;
+ p->p_flag2 = 0;
p->p_state = PRS_NORMAL;
knlist_init_mtx(&p->p_klist, &p->p_mtx);
STAILQ_INIT(&p->p_ktr);
Modified: head/sys/kern/kern_fork.c
==============================================================================
--- head/sys/kern/kern_fork.c Thu Sep 19 18:00:05 2013 (r255707)
+++ head/sys/kern/kern_fork.c Thu Sep 19 18:53:42 2013 (r255708)
@@ -489,6 +489,7 @@ do_fork(struct thread *td, int flags, st
* Increase reference counts on shared objects.
*/
p2->p_flag = P_INMEM;
+ p2->p_flag2 = 0;
p2->p_swtick = ticks;
if (p1->p_flag & P_PROFIL)
startprofclock(p2);
@@ -512,6 +513,11 @@ do_fork(struct thread *td, int flags, st
p2->p_fd = fd;
p2->p_fdtol = fdtol;
+ if (p1->p_flag2 & P2_INHERIT_PROTECTED) {
+ p2->p_flag |= P_PROTECTED;
+ p2->p_flag2 |= P2_INHERIT_PROTECTED;
+ }
+
/*
* p_limit is copy-on-write. Bump its refcount.
*/
Modified: head/sys/kern/kern_proc.c
==============================================================================
--- head/sys/kern/kern_proc.c Thu Sep 19 18:00:05 2013 (r255707)
+++ head/sys/kern/kern_proc.c Thu Sep 19 18:53:42 2013 (r255708)
@@ -802,6 +802,7 @@ fill_kinfo_proc_only(struct proc *p, str
kp->ki_fd = p->p_fd;
kp->ki_vmspace = p->p_vmspace;
kp->ki_flag = p->p_flag;
+ kp->ki_flag2 = p->p_flag2;
cred = p->p_ucred;
if (cred) {
kp->ki_uid = cred->cr_uid;
@@ -1161,6 +1162,7 @@ freebsd32_kinfo_proc_out(const struct ki
bcopy(ki->ki_comm, ki32->ki_comm, COMMLEN + 1);
bcopy(ki->ki_emul, ki32->ki_emul, KI_EMULNAMELEN + 1);
bcopy(ki->ki_loginclass, ki32->ki_loginclass, LOGINCLASSLEN + 1);
+ CP(*ki, *ki32, ki_flag2);
CP(*ki, *ki32, ki_fibnum);
CP(*ki, *ki32, ki_cr_flags);
CP(*ki, *ki32, ki_jid);
Modified: head/sys/kern/sys_process.c
==============================================================================
--- head/sys/kern/sys_process.c Thu Sep 19 18:00:05 2013 (r255707)
+++ head/sys/kern/sys_process.c Thu Sep 19 18:53:42 2013 (r255708)
@@ -41,7 +41,9 @@ __FBSDID("$FreeBSD$");
#include <sys/syscallsubr.h>
#include <sys/sysent.h>
#include <sys/sysproto.h>
+#include <sys/priv.h>
#include <sys/proc.h>
+#include <sys/procctl.h>
#include <sys/vnode.h>
#include <sys/ptrace.h>
#include <sys/rwlock.h>
@@ -1240,3 +1242,196 @@ stopevent(struct proc *p, unsigned int e
msleep(&p->p_step, &p->p_mtx, PWAIT, "stopevent", 0);
} while (p->p_step);
}
+
+static int
+protect_setchild(struct thread *td, struct proc *p, int flags)
+{
+
+ PROC_LOCK_ASSERT(p, MA_OWNED);
+ if (p->p_flag & P_SYSTEM || p_cansee(td, p) != 0)
+ return (0);
+ if (flags & PPROT_SET) {
+ p->p_flag |= P_PROTECTED;
+ if (flags & PPROT_INHERIT)
+ p->p_flag2 |= P2_INHERIT_PROTECTED;
+ } else {
+ p->p_flag &= ~P_PROTECTED;
+ p->p_flag2 &= ~P2_INHERIT_PROTECTED;
+ }
+ return (1);
+}
+
+static int
+protect_setchildren(struct thread *td, struct proc *top, int flags)
+{
+ struct proc *p;
+ int ret;
+
+ p = top;
+ ret = 0;
+ sx_assert(&proctree_lock, SX_LOCKED);
+ for (;;) {
+ ret |= protect_setchild(td, p, flags);
+ PROC_UNLOCK(p);
+ /*
+ * If this process has children, descend to them next,
+ * otherwise do any siblings, and if done with this level,
+ * follow back up the tree (but not past top).
+ */
+ if (!LIST_EMPTY(&p->p_children))
+ p = LIST_FIRST(&p->p_children);
+ else for (;;) {
+ if (p == top) {
+ PROC_LOCK(p);
+ return (ret);
+ }
+ if (LIST_NEXT(p, p_sibling)) {
+ p = LIST_NEXT(p, p_sibling);
+ break;
+ }
+ p = p->p_pptr;
+ }
+ PROC_LOCK(p);
+ }
+}
+
+static int
+protect_set(struct thread *td, struct proc *p, int flags)
+{
+ int error, ret;
+
+ switch (PPROT_OP(flags)) {
+ case PPROT_SET:
+ case PPROT_CLEAR:
+ break;
+ default:
+ return (EINVAL);
+ }
+
+ if ((PPROT_FLAGS(flags) & ~(PPROT_DESCEND | PPROT_INHERIT)) != 0)
+ return (EINVAL);
+
+ error = priv_check(td, PRIV_VM_MADV_PROTECT);
+ if (error)
+ return (error);
+
+ if (flags & PPROT_DESCEND)
+ ret = protect_setchildren(td, p, flags);
+ else
+ ret = protect_setchild(td, p, flags);
+ if (ret == 0)
+ return (EPERM);
+ return (0);
+}
+
+#ifndef _SYS_SYSPROTO_H_
+struct procctl_args {
+ idtype_t idtype;
+ id_t id;
+ int com;
+ void *data;
+};
+#endif
+/* ARGSUSED */
+int
+sys_procctl(struct thread *td, struct procctl_args *uap)
+{
+ int error, flags;
+ void *data;
+
+ switch (uap->com) {
+ case PROC_SPROTECT:
+ error = copyin(uap->data, &flags, sizeof(flags));
+ if (error)
+ return (error);
+ data = &flags;
+ break;
+ default:
+ return (EINVAL);
+ }
+
+ return (kern_procctl(td, uap->idtype, uap->id, uap->com, data));
+}
+
+static int
+kern_procctl_single(struct thread *td, struct proc *p, int com, void *data)
+{
+
+ PROC_LOCK_ASSERT(p, MA_OWNED);
+ switch (com) {
+ case PROC_SPROTECT:
+ return (protect_set(td, p, *(int *)data));
+ default:
+ return (EINVAL);
+ }
+}
+
+int
+kern_procctl(struct thread *td, idtype_t idtype, id_t id, int com, void *data)
+{
+ struct pgrp *pg;
+ struct proc *p;
+ int error, first_error, ok;
+
+ sx_slock(&proctree_lock);
+ switch (idtype) {
+ case P_PID:
+ p = pfind(id);
+ if (p == NULL) {
+ error = ESRCH;
+ break;
+ }
+ if (p->p_state == PRS_NEW)
+ error = ESRCH;
+ else
+ error = p_cansee(td, p);
+ if (error == 0)
+ error = kern_procctl_single(td, p, com, data);
+ PROC_UNLOCK(p);
+ break;
+ case P_PGID:
+ /*
+ * Attempt to apply the operation to all members of the
+ * group. Ignore processes in the group that can't be
+ * seen. Ignore errors so long as at least one process is
+ * able to complete the request successfully.
+ */
+ pg = pgfind(id);
+ if (pg == NULL) {
+ error = ESRCH;
+ break;
+ }
+ PGRP_UNLOCK(pg);
+ ok = 0;
+ first_error = 0;
+ LIST_FOREACH(p, &pg->pg_members, p_pglist) {
+ PROC_LOCK(p);
+ if (p->p_state == PRS_NEW || p_cansee(td, p) != 0) {
+ PROC_UNLOCK(p);
+ continue;
+ }
+ error = kern_procctl_single(td, p, com, data);
+ PROC_UNLOCK(p);
+ if (error == 0)
+ ok = 1;
+ else if (first_error == 0)
+ first_error = error;
+ }
+ if (ok)
+ error = 0;
+ else if (first_error != 0)
+ error = first_error;
+ else
+ /*
+ * Was not able to see any processes in the
+ * process group.
+ */
+ error = ESRCH;
+ break;
+ default:
+ error = EINVAL;
+ break;
+ }
+ sx_sunlock(&proctree_lock);
+ return (error);
+}
Modified: head/sys/kern/syscalls.master
==============================================================================
--- head/sys/kern/syscalls.master Thu Sep 19 18:00:05 2013
(r255707)
+++ head/sys/kern/syscalls.master Thu Sep 19 18:53:42 2013
(r255708)
@@ -978,5 +978,7 @@
int flags); }
542 AUE_PIPE STD { int pipe2(int *fildes, int flags); }
543 AUE_NULL NOSTD { int aio_mlock(struct aiocb *aiocbp); }
+544 AUE_NULL STD { int procctl(idtype_t idtype, id_t id, \
+ int com, void *data); }
; Please copy any additions and changes to the following compatability tables:
; sys/compat/freebsd32/syscalls.master
Modified: head/sys/sys/proc.h
==============================================================================
--- head/sys/sys/proc.h Thu Sep 19 18:00:05 2013 (r255707)
+++ head/sys/sys/proc.h Thu Sep 19 18:53:42 2013 (r255708)
@@ -492,11 +492,8 @@ struct proc {
struct callout p_limco; /* (c) Limit callout handle */
struct sigacts *p_sigacts; /* (x) Signal actions, state (CPU). */
- /*
- * The following don't make too much sense.
- * See the td_ or ke_ versions of the same flags.
- */
int p_flag; /* (c) P_* flags. */
+ int p_flag2; /* (c) P2_* flags. */
enum {
PRS_NEW = 0, /* In creation */
PRS_NORMAL, /* threads can be run. */
@@ -641,6 +638,9 @@ struct proc {
#define P_SHOULDSTOP(p) ((p)->p_flag & P_STOPPED)
#define P_KILLED(p) ((p)->p_flag & P_WKILLED)
+/* These flags are kept in p_flag2. */
+#define P2_INHERIT_PROTECTED 0x00000001 /* New children get
P_PROTECTED. */
+
/*
* These were process status values (p_stat), now they are only used in
* legacy conversion code.
Added: head/sys/sys/procctl.h
==============================================================================
--- /dev/null 00:00:00 1970 (empty, because file is newly added)
+++ head/sys/sys/procctl.h Thu Sep 19 18:53:42 2013 (r255708)
@@ -0,0 +1,55 @@
+/*-
+ * Copyright (c) 2013 Advanced Computing Technologies LLC
+ * Written by: John H. Baldwin <j...@freebsd.org>
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in the
+ * documentation and/or other materials provided with the distribution.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHOR 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 AUTHOR 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.
+ *
+ * $FreeBSD$
+ */
+
+#ifndef _SYS_PROCCTL_H_
+#define _SYS_PROCCTL_H_
+
+#define PROC_SPROTECT 1 /* set protected state */
+
+/* Operations for PROC_SPROTECT (passed in integer arg). */
+#define PPROT_OP(x) ((x) & 0xf)
+#define PPROT_SET 1
+#define PPROT_CLEAR 2
+
+/* Flags for PROC_SPROTECT (ORed in with operation). */
+#define PPROT_FLAGS(x) ((x) & ~0xf)
+#define PPROT_DESCEND 0x10
+#define PPROT_INHERIT 0x20
+
+#ifndef _KERNEL
+#include <sys/types.h>
+#include <sys/wait.h>
+
+__BEGIN_DECLS
+int procctl(idtype_t, id_t, int, void *);
+__END_DECLS
+
+#endif
+
+#endif /* !_SYS_PROCCTL_H_ */
Modified: head/sys/sys/syscallsubr.h
==============================================================================
--- head/sys/sys/syscallsubr.h Thu Sep 19 18:00:05 2013 (r255707)
+++ head/sys/sys/syscallsubr.h Thu Sep 19 18:53:42 2013 (r255708)
@@ -167,6 +167,8 @@ int kern_posix_fadvise(struct thread *td
int advice);
int kern_posix_fallocate(struct thread *td, int fd, off_t offset,
off_t len);
+int kern_procctl(struct thread *td, enum idtype idtype, id_t id, int com,
+ void *data);
int kern_preadv(struct thread *td, int fd, struct uio *auio, off_t offset);
int kern_pselect(struct thread *td, int nd, fd_set *in, fd_set *ou,
fd_set *ex, struct timeval *tvp, sigset_t *uset, int abi_nfdbits);
Modified: head/sys/sys/user.h
==============================================================================
--- head/sys/sys/user.h Thu Sep 19 18:00:05 2013 (r255707)
+++ head/sys/sys/user.h Thu Sep 19 18:53:42 2013 (r255708)
@@ -84,7 +84,7 @@
* it in two places: function fill_kinfo_proc in sys/kern/kern_proc.c and
* function kvm_proclist in lib/libkvm/kvm_proc.c .
*/
-#define KI_NSPARE_INT 8
+#define KI_NSPARE_INT 7
#define KI_NSPARE_LONG 12
#define KI_NSPARE_PTR 6
@@ -187,6 +187,7 @@ struct kinfo_proc {
*/
char ki_sparestrings[50]; /* spare string space */
int ki_spareints[KI_NSPARE_INT]; /* spare room for growth */
+ int ki_flag2; /* P2_* flags */
int ki_fibnum; /* Default FIB number */
u_int ki_cr_flags; /* Credential flags */
int ki_jid; /* Process jail ID */
Modified: head/sys/vm/vm_mmap.c
==============================================================================
--- head/sys/vm/vm_mmap.c Thu Sep 19 18:00:05 2013 (r255707)
+++ head/sys/vm/vm_mmap.c Thu Sep 19 18:53:42 2013 (r255708)
@@ -56,6 +56,7 @@ __FBSDID("$FreeBSD$");
#include <sys/filedesc.h>
#include <sys/priv.h>
#include <sys/proc.h>
+#include <sys/procctl.h>
#include <sys/racct.h>
#include <sys/resource.h>
#include <sys/resourcevar.h>
@@ -68,6 +69,7 @@ __FBSDID("$FreeBSD$");
#include <sys/mount.h>
#include <sys/conf.h>
#include <sys/stat.h>
+#include <sys/syscallsubr.h>
#include <sys/sysent.h>
#include <sys/vmmeter.h>
@@ -739,23 +741,18 @@ sys_madvise(td, uap)
{
vm_offset_t start, end;
vm_map_t map;
- struct proc *p;
- int error;
+ int flags;
/*
* Check for our special case, advising the swap pager we are
* "immortal."
*/
if (uap->behav == MADV_PROTECT) {
- error = priv_check(td, PRIV_VM_MADV_PROTECT);
- if (error == 0) {
- p = td->td_proc;
- PROC_LOCK(p);
- p->p_flag |= P_PROTECTED;
- PROC_UNLOCK(p);
- }
- return (error);
+ flags = PPROT_SET;
+ return (kern_procctl(td, P_PID, td->td_proc->p_pid,
+ PROC_SPROTECT, &flags));
}
+
/*
* Check for illegal behavior
*/
Modified: head/usr.bin/Makefile
==============================================================================
--- head/usr.bin/Makefile Thu Sep 19 18:00:05 2013 (r255707)
+++ head/usr.bin/Makefile Thu Sep 19 18:53:42 2013 (r255708)
@@ -132,6 +132,7 @@ SUBDIR= alias \
printenv \
printf \
procstat \
+ protect \
rctl \
renice \
rev \
Modified: head/usr.bin/kdump/kdump.c
==============================================================================
--- head/usr.bin/kdump/kdump.c Thu Sep 19 18:00:05 2013 (r255707)
+++ head/usr.bin/kdump/kdump.c Thu Sep 19 18:53:42 2013 (r255708)
@@ -1161,6 +1161,18 @@ ktrsyscall(struct ktr_syscall *ktr, u_in
ip++;
narg--;
break;
+ case SYS_procctl:
+ putchar('(');
+ idtypename(*ip, decimal);
+ c = ',';
+ ip++;
+ narg--;
+ print_number(ip, narg, c);
+ putchar(',');
+ procctlcmdname(*ip);
+ ip++;
+ narg--;
+ break;
}
}
while (narg > 0) {
Modified: head/usr.bin/kdump/mksubr
==============================================================================
--- head/usr.bin/kdump/mksubr Thu Sep 19 18:00:05 2013 (r255707)
+++ head/usr.bin/kdump/mksubr Thu Sep 19 18:53:42 2013 (r255708)
@@ -169,6 +169,7 @@ cat <<_EOF_
#include <netinet/in.h>
#include <sys/param.h>
#include <sys/mount.h>
+#include <sys/procctl.h>
#include <sys/ptrace.h>
#include <sys/resource.h>
#include <sys/reboot.h>
@@ -465,6 +466,7 @@ auto_or_type "mountflagsname" "
auto_switch_type "msyncflagsname" "MS_[A-Z]+[[:space:]]+0x[0-9]+"
"sys/mman.h"
auto_or_type "nfssvcname" "NFSSVC_[A-Z0-9]+[[:space:]]+0x[0-9]+"
"nfs/nfssvc.h"
auto_switch_type "prioname" "PRIO_[A-Z]+[[:space:]]+[0-9]"
"sys/resource.h"
+auto_switch_type "procctlcmdname" "PROC_[A-Z]+[[:space:]]+[0-9]"
"sys/procctl.h"
auto_switch_type "ptraceopname" "PT_[[:alnum:]_]+[[:space:]]+[0-9]+"
"sys/ptrace.h"
auto_switch_type "quotactlname" "Q_[A-Z]+[[:space:]]+0x[0-9]+"
"ufs/ufs/quota.h"
auto_or_type "rebootoptname" "RB_[A-Z]+[[:space:]]+0x[0-9]+"
"sys/reboot.h"
Added: head/usr.bin/protect/Makefile
==============================================================================
--- /dev/null 00:00:00 1970 (empty, because file is newly added)
+++ head/usr.bin/protect/Makefile Thu Sep 19 18:53:42 2013
(r255708)
@@ -0,0 +1,6 @@
+# $FreeBSD$
+
+PROG= protect
+WARNS?= 6
+
+.include <bsd.prog.mk>
Added: head/usr.bin/protect/protect.1
==============================================================================
--- /dev/null 00:00:00 1970 (empty, because file is newly added)
+++ head/usr.bin/protect/protect.1 Thu Sep 19 18:53:42 2013
(r255708)
@@ -0,0 +1,89 @@
+.\" Copyright (c) 2013 Advanced Computing Technologies LLC
+.\" Written by: John H. Baldwin <j...@freebsd.org>
+.\" All rights reserved.
+.\"
+.\" Redistribution and use in source and binary forms, with or without
+.\" modification, are permitted provided that the following conditions
+.\" are met:
+.\" 1. Redistributions of source code must retain the above copyright
+.\" notice, this list of conditions and the following disclaimer.
+.\" 2. Redistributions in binary form must reproduce the above copyright
+.\" notice, this list of conditions and the following disclaimer in the
+.\" documentation and/or other materials provided with the distribution.
+.\"
+.\" THIS SOFTWARE IS PROVIDED BY THE AUTHOR 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 AUTHOR 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.
+.\"
+.\" $FreeBSD$
+.\"
+.Dd September 19, 2013
+.Dt PROTECT 1
+.Os
+.Sh NAME
+.Nm protect
+.Nd "protect processes from being killed when swap space is exhausted"
+.Sh SYNOPSIS
+.Nm
+.Op Fl i
+.Ar command
+.Nm
+.Op Fl cdi
+.Fl g Ar pgrp | Fl p Ar pid
+.Sh DESCRIPTION
+The
+.Nm
+command is used to mark processes as protected.
+The kernel does not kill protected processes when swap space is exhausted.
+Note that this protected state is not inherited by child processes by default.
+.Pp
+The options are:
+.Bl -tag -width indent
+.It Fl c
+Remove protection from the specified processes.
+.It Fl d
+Apply the operation to all current children of the specified processes.
+.It Fl i
+Apply the operation to all future children of the specified processes.
+.It Fl g Ar pgrp
+Apply the operation to all processes in the specified process group.
+.It Fl p Ar pid
+Apply the operation to the specified process.
+.It Ar command
+Execute
+.Ar command
+as a protected process.
+.El
+.Pp
+Note that only one of the
+.Fl p
+or
+.Fl g
+flags may be specified when adjusting the state of existing processes.
+.Sh EXIT STATUS
+.Ex -std
+.Sh EXAMPLES
+Mark the Xorg server as protected:
+.Pp
+.Dl "pgrep Xorg | xargs protect -p"
+Protect all ssh sessions and their child processes:
+.Pp
+.Dl "pgrep sshd | xargs protect -dip"
+Remove protection from all current and future processes:
+.Pp
+.Dl "protect -cdi -p 1"
+.Sh SEE ALSO
+.Xr pprotect 2
+.Sh BUGS
+If you protect a runaway process that allocates all memory the system will
+deadlock.
+.Pp
+Inheritance of the protected state is not yet implemented.
Added: head/usr.bin/protect/protect.c
==============================================================================
--- /dev/null 00:00:00 1970 (empty, because file is newly added)
+++ head/usr.bin/protect/protect.c Thu Sep 19 18:53:42 2013
(r255708)
@@ -0,0 +1,122 @@
+/*-
+ * Copyright (c) 2013 Advanced Computing Technologies LLC
+ * Written by: John H. Baldwin <j...@freebsd.org>
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in the
+ * documentation and/or other materials provided with the distribution.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHOR 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 AUTHOR 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.
+ */
+
+#include <sys/cdefs.h>
+__FBSDID("$FreeBSD$");
+
+#include <sys/procctl.h>
+#include <sys/types.h>
+#include <sys/mman.h>
+#include <err.h>
+#include <errno.h>
+#include <stdbool.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <unistd.h>
+
+static void
+usage(void)
+{
+
+ fprintf(stderr, "usage: protect [-i] command\n");
+ fprintf(stderr, " protect [-cdi] -g pgrp | -p pid\n");
+ exit(1);
+}
+
+static id_t
+parse_id(char *id)
+{
+ static bool first = true;
+ long value;
+ char *ch;
+
+ if (!first) {
+ warnx("only one -g or -p flag is permitted");
+ usage();
+ }
+ value = strtol(id, &ch, 0);
+ if (*ch != '\0') {
+ warnx("invalid process id");
+ usage();
+ }
+ return (value);
+}
+
+int
+main(int argc, char *argv[])
+{
+ idtype_t idtype;
+ id_t id;
+ int ch, flags;
+ bool descend, inherit, idset;
+
+ idtype = P_PID;
+ id = getpid();
+ flags = PPROT_SET;
+ descend = inherit = idset = false;
+ while ((ch = getopt(argc, argv, "cdig:p:")) != -1)
+ switch (ch) {
+ case 'c':
+ flags = PPROT_CLEAR;
+ break;
+ case 'd':
+ descend = true;
+ break;
+ case 'i':
+ inherit = true;
+ break;
+ case 'g':
+ idtype = P_PGID;
+ id = parse_id(optarg);
+ idset = true;
+ break;
+ case 'p':
+ idtype = P_PID;
+ id = parse_id(optarg);
+ idset = true;
+ break;
+ }
+ argc -= optind;
+ argv += optind;
+
+ if ((idset && argc != 0) || (!idset && (argc == 0 || descend)))
+ usage();
+
+ if (descend)
+ flags |= PPROT_DESCEND;
+ if (inherit)
+ flags |= PPROT_INHERIT;
+ if (procctl(idtype, id, PROC_SPROTECT, &flags) == -1)
+ err(1, "procctl");
+
+ if (argc != 0) {
+ errno = 0;
+ execvp(*argv, argv);
+ err(errno == ENOENT ? 127 : 126, "%s", *argv);
+ }
+ return (0);
+}
Modified: head/usr.bin/truss/syscall.h
==============================================================================
--- head/usr.bin/truss/syscall.h Thu Sep 19 18:00:05 2013
(r255707)
+++ head/usr.bin/truss/syscall.h Thu Sep 19 18:53:42 2013
(r255708)
@@ -40,7 +40,7 @@ enum Argtype { None = 1, Hex, Octal, Int
Fd_set, Sigaction, Fcntl, Mprot, Mmapflags, Whence, Readlinkres,
Umtx, Sigset, Sigprocmask, Kevent, Sockdomain, Socktype, Open,
Fcntlflag, Rusage, BinString, Shutdown, Resource, Rlimit, Timeval2,
- Pathconf, Rforkflags, ExitStatus, Waitoptions, Idtype };
+ Pathconf, Rforkflags, ExitStatus, Waitoptions, Idtype, Procctl };
#define ARG_MASK 0xff
#define OUT 0x100
Modified: head/usr.bin/truss/syscalls.c
==============================================================================
--- head/usr.bin/truss/syscalls.c Thu Sep 19 18:00:05 2013
(r255707)
+++ head/usr.bin/truss/syscalls.c Thu Sep 19 18:53:42 2013
(r255708)
@@ -41,6 +41,7 @@ static const char rcsid[] =
#include <sys/types.h>
#include <sys/mman.h>
+#include <sys/procctl.h>
#include <sys/ptrace.h>
#include <sys/socket.h>
#include <sys/time.h>
@@ -270,6 +271,8 @@ static struct syscall syscalls[] = {
{ .name = "wait6", .ret_type = 1, .nargs = 6,
.args = { { Idtype, 0 }, { Int, 1 }, { ExitStatus | OUT, 2 },
{ Waitoptions, 3 }, { Rusage | OUT, 4 }, { Ptr, 5 } } },
+ { .name = "procctl", .ret_type = 1, .nargs = 4,
+ .args = { { Idtype, 0 }, { Int, 1 }, { Procctl, 2 }, { Ptr, 3 } } },
{ .name = 0 },
};
@@ -399,6 +402,10 @@ static struct xlat idtype_arg[] = {
X(P_CTID) X(P_CPUID) X(P_PSETID) XEND
};
+static struct xlat procctl_arg[] = {
+ X(PROC_SPROTECT) XEND
+};
+
#undef X
#undef XEND
@@ -1198,6 +1205,9 @@ print_arg(struct syscall_args *sc, unsig
case Idtype:
tmp = strdup(xlookup(idtype_arg, args[sc->offset]));
break;
+ case Procctl:
+ tmp = strdup(xlookup(procctl_arg, args[sc->offset]));
+ break;
default:
errx(1, "Invalid argument type %d\n", sc->type & ARG_MASK);
}
_______________________________________________
svn-src-head@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
