1/*
2 * Copyright (c) 2000-2013 Apple Inc. All rights reserved.
3 *
4 * @APPLE_OSREFERENCE_LICENSE_HEADER_START@
5 *
6 * This file contains Original Code and/or Modifications of Original Code
7 * as defined in and that are subject to the Apple Public Source License
8 * Version 2.0 (the 'License'). You may not use this file except in
9 * compliance with the License. The rights granted to you under the License
10 * may not be used to create, or enable the creation or redistribution of,
11 * unlawful or unlicensed copies of an Apple operating system, or to
12 * circumvent, violate, or enable the circumvention or violation of, any
13 * terms of an Apple operating system software license agreement.
14 *
15 * Please obtain a copy of the License at
16 * http://www.opensource.apple.com/apsl/ and read it before using this file.
17 *
18 * The Original Code and all software distributed under the License are
19 * distributed on an 'AS IS' basis, WITHOUT WARRANTY OF ANY KIND, EITHER
20 * EXPRESS OR IMPLIED, AND APPLE HEREBY DISCLAIMS ALL SUCH WARRANTIES,
21 * INCLUDING WITHOUT LIMITATION, ANY WARRANTIES OF MERCHANTABILITY,
22 * FITNESS FOR A PARTICULAR PURPOSE, QUIET ENJOYMENT OR NON-INFRINGEMENT.
23 * Please see the License for the specific language governing rights and
24 * limitations under the License.
25 *
26 * @APPLE_OSREFERENCE_LICENSE_HEADER_END@
27 *
28 *
29 * Copyright (c) 1982, 1986, 1989, 1991, 1992, 1993
30 *	The Regents of the University of California.  All rights reserved.
31 * (c) UNIX System Laboratories, Inc.
32 * All or some portions of this file are derived from material licensed
33 * to the University of California by American Telephone and Telegraph
34 * Co. or Unix System Laboratories, Inc. and are reproduced herein with
35 * the permission of UNIX System Laboratories, Inc.
36 *
37 * Redistribution and use in source and binary forms, with or without
38 * modification, are permitted provided that the following conditions
39 * are met:
40 * 1. Redistributions of source code must retain the above copyright
41 *    notice, this list of conditions and the following disclaimer.
42 * 2. Redistributions in binary form must reproduce the above copyright
43 *    notice, this list of conditions and the following disclaimer in the
44 *    documentation and/or other materials provided with the distribution.
45 * 3. All advertising materials mentioning features or use of this software
46 *    must display the following acknowledgement:
47 *	This product includes software developed by the University of
48 *	California, Berkeley and its contributors.
49 * 4. Neither the name of the University nor the names of its contributors
50 *    may be used to endorse or promote products derived from this software
51 *    without specific prior written permission.
52 *
53 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
54 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
55 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
56 * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
57 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
58 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
59 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
60 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
61 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
62 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
63 * SUCH DAMAGE.
64 *
65 *	@(#)init_main.c	8.16 (Berkeley) 5/14/95
66 */
67
68/*
69 *
70 * Mach Operating System
71 * Copyright (c) 1987 Carnegie-Mellon University
72 * All rights reserved.  The CMU software License Agreement specifies
73 * the terms and conditions for use and redistribution.
74 */
75/*
76 * NOTICE: This file was modified by McAfee Research in 2004 to introduce
77 * support for mandatory and extensible security protections.  This notice
78 * is included in support of clause 2.2 (b) of the Apple Public License,
79 * Version 2.0.
80 */
81
82#include <sys/param.h>
83#include <sys/filedesc.h>
84#include <sys/kernel.h>
85#include <sys/mount_internal.h>
86#include <sys/proc_internal.h>
87#include <sys/kauth.h>
88#include <sys/systm.h>
89#include <sys/vnode_internal.h>
90#include <sys/conf.h>
91#include <sys/buf_internal.h>
92#include <sys/clist.h>
93#include <sys/user.h>
94#include <sys/time.h>
95#include <sys/systm.h>
96#include <sys/mman.h>
97
98#include <security/audit/audit.h>
99
100#include <sys/malloc.h>
101#include <sys/dkstat.h>
102#include <sys/codesign.h>
103
104#include <kern/startup.h>
105#include <kern/thread.h>
106#include <kern/task.h>
107#include <kern/ast.h>
108#include <kern/kalloc.h>
109#include <mach/mach_host.h>
110
111#include <mach/vm_param.h>
112
113#include <vm/vm_map.h>
114#include <vm/vm_kern.h>
115
116#include <sys/ux_exception.h>	/* for ux_exception_port */
117
118#include <sys/reboot.h>
119#include <mach/exception_types.h>
120#include <dev/busvar.h>			/* for pseudo_inits */
121#include <sys/kdebug.h>
122
123#include <mach/mach_types.h>
124#include <mach/vm_prot.h>
125#include <mach/semaphore.h>
126#include <mach/sync_policy.h>
127#include <kern/clock.h>
128#include <mach/kern_return.h>
129#include <mach/thread_act.h>		/* for thread_resume() */
130#include <mach/task.h>			/* for task_set_exception_ports() */
131#include <sys/ux_exception.h>		/* for ux_handler() */
132#include <sys/ubc_internal.h>		/* for ubc_init() */
133#include <sys/mcache.h>			/* for mcache_init() */
134#include <sys/mbuf.h>			/* for mbinit() */
135#include <sys/event.h>			/* for knote_init() */
136#include <sys/kern_memorystatus.h>	/* for memorystatus_init() */
137#include <sys/aio_kern.h>		/* for aio_init() */
138#include <sys/semaphore.h>		/* for psem_cache_init() */
139#include <net/dlil.h>			/* for dlil_init() */
140#include <net/kpi_protocol.h>		/* for proto_kpi_init() */
141#include <net/iptap.h>			/* for iptap_init() */
142#include <sys/pipe.h>			/* for pipeinit() */
143#include <sys/socketvar.h>		/* for socketinit() */
144#include <sys/protosw.h>		/* for domaininit() */
145#include <kern/sched_prim.h>		/* for thread_wakeup() */
146#include <net/if_ether.h>		/* for ether_family_init() */
147#include <net/if_gif.h>			/* for gif_init() */
148#include <vm/vm_protos.h>		/* for vnode_pager_bootstrap() */
149#include <miscfs/devfs/devfsdefs.h>	/* for devfs_kernel_mount() */
150#include <mach/host_priv.h>		/* for host_set_exception_ports() */
151#include <kern/host.h>			/* for host_priv_self() */
152#include <vm/vm_kern.h>			/* for kmem_suballoc() */
153#include <sys/semaphore.h>		/* for psem_lock_init() */
154#include <sys/msgbuf.h>			/* for log_setsize() */
155#include <sys/tty.h>			/* for tty_init() */
156#include <sys/proc_uuid_policy.h>	/* proc_uuid_policy_init() */
157#include <netinet/flow_divert.h>	/* flow_divert_init() */
158#include <net/content_filter.h>		/* for cfil_init() */
159#include <net/necp.h>			/* for necp_init() */
160#include <net/packet_mangler.h>		/* for pkt_mnglr_init() */
161#include <net/if_utun.h>		/* for utun_register_control() */
162#include <net/if_ipsec.h>       /* for ipsec_register_control() */
163#include <net/net_str_id.h>		/* for net_str_id_init() */
164#include <net/netsrc.h>			/* for netsrc_init() */
165#include <net/ntstat.h>			/* for nstat_init() */
166#include <netinet/tcp_cc.h>			/* for tcp_cc_init() */
167#include <kern/assert.h>		/* for assert() */
168#include <sys/kern_overrides.h>		/* for init_system_override() */
169
170#include <net/init.h>
171
172#if CONFIG_MACF
173#include <security/mac_framework.h>
174#include <security/mac_internal.h>	/* mac_init_bsd() */
175#include <security/mac_mach_internal.h>	/* mac_update_task_label() */
176#endif
177
178#include <machine/exec.h>
179
180#if NFSCLIENT
181#include <sys/netboot.h>
182#endif
183
184#if CONFIG_IMAGEBOOT
185#include <sys/imageboot.h>
186#endif
187
188#if PFLOG
189#include <net/if_pflog.h>
190#endif
191
192#include <pexpert/pexpert.h>
193#include <machine/pal_routines.h>
194#include <console/video_console.h>
195
196void * get_user_regs(thread_t);		/* XXX kludge for <machine/thread.h> */
197void IOKitInitializeTime(void);		/* XXX */
198void IOSleep(unsigned int);		/* XXX */
199void loopattach(void);			/* XXX */
200
201const char    copyright[] =
202"Copyright (c) 1982, 1986, 1989, 1991, 1993\n\t"
203"The Regents of the University of California. "
204"All rights reserved.\n\n";
205
206/* Components of the first process -- never freed. */
207struct	proc proc0;
208struct	session session0;
209struct	pgrp pgrp0;
210struct	filedesc filedesc0;
211struct	plimit limit0;
212struct	pstats pstats0;
213struct	sigacts sigacts0;
214proc_t kernproc;
215proc_t initproc;
216
217long tk_cancc;
218long tk_nin;
219long tk_nout;
220long tk_rawcc;
221
222int lock_trace = 0;
223/* Global variables to make pstat happy. We do swapping differently */
224int nswdev, nswap;
225int nswapmap;
226void *swapmap;
227struct swdevt swdevt[1];
228
229dev_t	rootdev;		/* device of the root */
230dev_t	dumpdev;		/* device to take dumps on */
231long	dumplo;			/* offset into dumpdev */
232long	hostid;
233char	hostname[MAXHOSTNAMELEN];
234int		hostnamelen;
235char	domainname[MAXDOMNAMELEN];
236int		domainnamelen;
237
238char rootdevice[16]; 	/* hfs device names have at least 9 chars */
239
240#if  KMEMSTATS
241struct	kmemstats kmemstats[M_LAST];
242#endif
243
244struct	vnode *rootvp;
245int boothowto = RB_DEBUG;
246
247extern kern_return_t IOFindBSDRoot(char *, unsigned int, dev_t *, u_int32_t *);
248extern void IOSecureBSDRoot(const char * rootName);
249extern kern_return_t IOKitBSDInit(void );
250extern void kminit(void);
251extern void file_lock_init(void);
252extern void kmeminit(void);
253extern void bsd_bufferinit(void);
254extern void throttle_init(void);
255extern void macx_init(void);
256extern void acct_init(void);
257
258extern int serverperfmode;
259extern int ncl;
260
261vm_map_t	bsd_pageable_map;
262vm_map_t	mb_map;
263
264static  int bsd_simul_execs;
265static int bsd_pageable_map_size;
266__private_extern__ int execargs_cache_size = 0;
267__private_extern__ int execargs_free_count = 0;
268__private_extern__ vm_offset_t * execargs_cache = NULL;
269
270void bsd_exec_setup(int);
271
272__private_extern__ int bootarg_vnode_cache_defeat = 0;
273
274/*
275 * Prevent kernel-based ASLR from being used, for testing.
276 */
277#if DEVELOPMENT || DEBUG
278__private_extern__ int bootarg_disable_aslr = 0;
279#endif
280
281int	cmask = CMASK;
282extern int customnbuf;
283
284void bsd_init(void);
285kern_return_t bsd_autoconf(void);
286void bsd_utaskbootstrap(void);
287
288static void parse_bsd_args(void);
289extern task_t bsd_init_task;
290extern char    init_task_failure_data[];
291#if CONFIG_DEV_KMEM
292extern void dev_kmem_init(void);
293#endif
294extern void time_zone_slock_init(void);
295extern void select_wait_queue_init(void);
296static void process_name(const char *, proc_t);
297
298static void setconf(void);
299
300#if SYSV_SHM
301extern void sysv_shm_lock_init(void);
302#endif
303#if SYSV_SEM
304extern void sysv_sem_lock_init(void);
305#endif
306#if SYSV_MSG
307extern void sysv_msg_lock_init(void);
308#endif
309
310#if CONFIG_MACF
311#if defined (__i386__) || defined (__x86_64__)
312/* MACF policy_check configuration flags; see policy_check.c for details */
313int policy_check_flags = 0;
314
315extern int check_policy_init(int);
316#endif
317#endif	/* CONFIG_MACF */
318
319/* If we are using CONFIG_DTRACE */
320#if CONFIG_DTRACE
321	extern void dtrace_postinit(void);
322#endif
323
324/*
325 * Initialization code.
326 * Called from cold start routine as
327 * soon as a stack and segmentation
328 * have been established.
329 * Functions:
330 *	turn on clock
331 *	hand craft 0th process
332 *	call all initialization routines
333 *  hand craft 1st user process
334 */
335
336/*
337 *	Sets the name for the given task.
338 */
339static void
340process_name(const char *s, proc_t p)
341{
342	size_t length = strlen(s);
343
344	bcopy(s, p->p_comm,
345		length >= sizeof(p->p_comm) ? sizeof(p->p_comm) :
346			length + 1);
347}
348
349/* To allow these values to be patched, they're globals here */
350#include <machine/vmparam.h>
351struct rlimit vm_initial_limit_stack = { DFLSSIZ, MAXSSIZ - PAGE_MAX_SIZE };
352struct rlimit vm_initial_limit_data = { DFLDSIZ, MAXDSIZ };
353struct rlimit vm_initial_limit_core = { DFLCSIZ, MAXCSIZ };
354
355extern thread_t	cloneproc(task_t, coalition_t, proc_t, int, int);
356extern int 	(*mountroot)(void);
357
358lck_grp_t * proc_lck_grp;
359lck_grp_t * proc_slock_grp;
360lck_grp_t * proc_fdmlock_grp;
361lck_grp_t * proc_mlock_grp;
362lck_grp_attr_t * proc_lck_grp_attr;
363lck_attr_t * proc_lck_attr;
364lck_mtx_t * proc_list_mlock;
365lck_mtx_t * proc_klist_mlock;
366
367extern lck_mtx_t * execargs_cache_lock;
368
369/* hook called after root is mounted XXX temporary hack */
370void (*mountroot_post_hook)(void);
371void (*unmountroot_pre_hook)(void);
372
373/*
374 * This function is called very early on in the Mach startup, from the
375 * function start_kernel_threads() in osfmk/kern/startup.c.  It's called
376 * in the context of the current (startup) task using a call to the
377 * function kernel_thread_create() to jump into start_kernel_threads().
378 * Internally, kernel_thread_create() calls thread_create_internal(),
379 * which calls uthread_alloc().  The function of uthread_alloc() is
380 * normally to allocate a uthread structure, and fill out the uu_sigmask,
381 * uu_context fields.  It skips filling these out in the case of the "task"
382 * being "kernel_task", because the order of operation is inverted.  To
383 * account for that, we need to manually fill in at least the contents
384 * of the uu_context.vc_ucred field so that the uthread structure can be
385 * used like any other.
386 */
387
388void
389bsd_init(void)
390{
391	struct uthread *ut;
392	unsigned int i;
393	struct vfs_context context;
394	kern_return_t	ret;
395	struct ucred temp_cred;
396	struct posix_cred temp_pcred;
397#if NFSCLIENT || CONFIG_IMAGEBOOT
398	boolean_t       netboot = FALSE;
399#endif
400
401#define bsd_init_kprintf(x...) /* kprintf("bsd_init: " x) */
402
403	throttle_init();
404
405	printf(copyright);
406
407	bsd_init_kprintf("calling kmeminit\n");
408	kmeminit();
409
410	bsd_init_kprintf("calling parse_bsd_args\n");
411	parse_bsd_args();
412
413#if CONFIG_DEV_KMEM
414	bsd_init_kprintf("calling dev_kmem_init\n");
415	dev_kmem_init();
416#endif
417
418	/* Initialize kauth subsystem before instancing the first credential */
419	bsd_init_kprintf("calling kauth_init\n");
420	kauth_init();
421
422	/* Initialize process and pgrp structures. */
423	bsd_init_kprintf("calling procinit\n");
424	procinit();
425
426	/* Initialize the ttys (MUST be before kminit()/bsd_autoconf()!)*/
427	tty_init();
428
429	kernproc = &proc0;	/* implicitly bzero'ed */
430
431	/* kernel_task->proc = kernproc; */
432	set_bsdtask_info(kernel_task,(void *)kernproc);
433
434	/* give kernproc a name */
435	bsd_init_kprintf("calling process_name\n");
436	process_name("kernel_task", kernproc);
437
438	/* allocate proc lock group attribute and group */
439	bsd_init_kprintf("calling lck_grp_attr_alloc_init\n");
440	proc_lck_grp_attr= lck_grp_attr_alloc_init();
441
442	proc_lck_grp = lck_grp_alloc_init("proc",  proc_lck_grp_attr);
443#if CONFIG_FINE_LOCK_GROUPS
444	proc_slock_grp = lck_grp_alloc_init("proc-slock",  proc_lck_grp_attr);
445	proc_fdmlock_grp = lck_grp_alloc_init("proc-fdmlock",  proc_lck_grp_attr);
446	proc_mlock_grp = lck_grp_alloc_init("proc-mlock",  proc_lck_grp_attr);
447#endif
448	/* Allocate proc lock attribute */
449	proc_lck_attr = lck_attr_alloc_init();
450#if 0
451#if __PROC_INTERNAL_DEBUG
452	lck_attr_setdebug(proc_lck_attr);
453#endif
454#endif
455
456#if CONFIG_FINE_LOCK_GROUPS
457	proc_list_mlock = lck_mtx_alloc_init(proc_mlock_grp, proc_lck_attr);
458	proc_klist_mlock = lck_mtx_alloc_init(proc_mlock_grp, proc_lck_attr);
459	lck_mtx_init(&kernproc->p_mlock, proc_mlock_grp, proc_lck_attr);
460	lck_mtx_init(&kernproc->p_fdmlock, proc_fdmlock_grp, proc_lck_attr);
461	lck_spin_init(&kernproc->p_slock, proc_slock_grp, proc_lck_attr);
462#else
463	proc_list_mlock = lck_mtx_alloc_init(proc_lck_grp, proc_lck_attr);
464	proc_klist_mlock = lck_mtx_alloc_init(proc_lck_grp, proc_lck_attr);
465	lck_mtx_init(&kernproc->p_mlock, proc_lck_grp, proc_lck_attr);
466	lck_mtx_init(&kernproc->p_fdmlock, proc_lck_grp, proc_lck_attr);
467	lck_spin_init(&kernproc->p_slock, proc_lck_grp, proc_lck_attr);
468#endif
469
470	assert(bsd_simul_execs != 0);
471	execargs_cache_lock = lck_mtx_alloc_init(proc_lck_grp, proc_lck_attr);
472	execargs_cache_size = bsd_simul_execs;
473	execargs_free_count = bsd_simul_execs;
474	execargs_cache = (vm_offset_t *)kalloc(bsd_simul_execs * sizeof(vm_offset_t));
475	bzero(execargs_cache, bsd_simul_execs * sizeof(vm_offset_t));
476
477	if (current_task() != kernel_task)
478		printf("bsd_init: We have a problem, "
479				"current task is not kernel task\n");
480
481	bsd_init_kprintf("calling get_bsdthread_info\n");
482	ut = (uthread_t)get_bsdthread_info(current_thread());
483
484#if CONFIG_MACF
485	/*
486	 * Initialize the MAC Framework
487	 */
488	mac_policy_initbsd();
489	kernproc->p_mac_enforce = 0;
490
491#if defined (__i386__) || defined (__x86_64__)
492	/*
493	 * We currently only support this on i386/x86_64, as that is the
494	 * only lock code we have instrumented so far.
495	 */
496	check_policy_init(policy_check_flags);
497#endif
498#endif /* MAC */
499
500	/* Initialize System Override call */
501	init_system_override();
502
503	/*
504	 * Create process 0.
505	 */
506	proc_list_lock();
507	LIST_INSERT_HEAD(&allproc, kernproc, p_list);
508	kernproc->p_pgrp = &pgrp0;
509	LIST_INSERT_HEAD(PGRPHASH(0), &pgrp0, pg_hash);
510	LIST_INIT(&pgrp0.pg_members);
511#ifdef CONFIG_FINE_LOCK_GROUPS
512	lck_mtx_init(&pgrp0.pg_mlock, proc_mlock_grp, proc_lck_attr);
513#else
514	lck_mtx_init(&pgrp0.pg_mlock, proc_lck_grp, proc_lck_attr);
515#endif
516	/* There is no other bsd thread this point and is safe without pgrp lock */
517	LIST_INSERT_HEAD(&pgrp0.pg_members, kernproc, p_pglist);
518	kernproc->p_listflag |= P_LIST_INPGRP;
519	kernproc->p_pgrpid = 0;
520	kernproc->p_uniqueid = 0;
521
522	pgrp0.pg_session = &session0;
523	pgrp0.pg_membercnt = 1;
524
525	session0.s_count = 1;
526	session0.s_leader = kernproc;
527	session0.s_listflags = 0;
528#ifdef CONFIG_FINE_LOCK_GROUPS
529	lck_mtx_init(&session0.s_mlock, proc_mlock_grp, proc_lck_attr);
530#else
531	lck_mtx_init(&session0.s_mlock, proc_lck_grp, proc_lck_attr);
532#endif
533	LIST_INSERT_HEAD(SESSHASH(0), &session0, s_hash);
534	proc_list_unlock();
535
536#if CONFIG_LCTX
537	kernproc->p_lctx = NULL;
538#endif
539
540	kernproc->task = kernel_task;
541
542	kernproc->p_stat = SRUN;
543	kernproc->p_flag = P_SYSTEM;
544	kernproc->p_lflag = 0;
545	kernproc->p_ladvflag = 0;
546
547#if DEVELOPMENT || DEBUG
548	if (bootarg_disable_aslr)
549		kernproc->p_flag |= P_DISABLE_ASLR;
550#endif
551
552	kernproc->p_nice = NZERO;
553	kernproc->p_pptr = kernproc;
554
555	TAILQ_INIT(&kernproc->p_uthlist);
556	TAILQ_INSERT_TAIL(&kernproc->p_uthlist, ut, uu_list);
557
558	kernproc->sigwait = FALSE;
559	kernproc->sigwait_thread = THREAD_NULL;
560	kernproc->exit_thread = THREAD_NULL;
561	kernproc->p_csflags = CS_VALID;
562
563	/*
564	 * Create credential.  This also Initializes the audit information.
565	 */
566	bsd_init_kprintf("calling bzero\n");
567	bzero(&temp_cred, sizeof(temp_cred));
568	bzero(&temp_pcred, sizeof(temp_pcred));
569	temp_pcred.cr_ngroups = 1;
570	/* kern_proc, shouldn't call up to DS for group membership */
571	temp_pcred.cr_flags = CRF_NOMEMBERD;
572	temp_cred.cr_audit.as_aia_p = audit_default_aia_p;
573
574	bsd_init_kprintf("calling kauth_cred_create\n");
575	/*
576	 * We have to label the temp cred before we create from it to
577	 * properly set cr_ngroups, or the create will fail.
578	 */
579	posix_cred_label(&temp_cred, &temp_pcred);
580	kernproc->p_ucred = kauth_cred_create(&temp_cred);
581
582	/* update cred on proc */
583	PROC_UPDATE_CREDS_ONPROC(kernproc);
584
585	/* give the (already exisiting) initial thread a reference on it */
586	bsd_init_kprintf("calling kauth_cred_ref\n");
587	kauth_cred_ref(kernproc->p_ucred);
588	ut->uu_context.vc_ucred = kernproc->p_ucred;
589	ut->uu_context.vc_thread = current_thread();
590
591	TAILQ_INIT(&kernproc->p_aio_activeq);
592	TAILQ_INIT(&kernproc->p_aio_doneq);
593	kernproc->p_aio_total_count = 0;
594	kernproc->p_aio_active_count = 0;
595
596	bsd_init_kprintf("calling file_lock_init\n");
597	file_lock_init();
598
599#if CONFIG_MACF
600	mac_cred_label_associate_kernel(kernproc->p_ucred);
601#endif
602
603	/* Create the file descriptor table. */
604	kernproc->p_fd = &filedesc0;
605	filedesc0.fd_cmask = cmask;
606	filedesc0.fd_knlistsize = -1;
607	filedesc0.fd_knlist = NULL;
608	filedesc0.fd_knhash = NULL;
609	filedesc0.fd_knhashmask = 0;
610
611	/* Create the limits structures. */
612	kernproc->p_limit = &limit0;
613	for (i = 0; i < sizeof(kernproc->p_rlimit)/sizeof(kernproc->p_rlimit[0]); i++)
614		limit0.pl_rlimit[i].rlim_cur =
615			limit0.pl_rlimit[i].rlim_max = RLIM_INFINITY;
616	limit0.pl_rlimit[RLIMIT_NOFILE].rlim_cur = NOFILE;
617	limit0.pl_rlimit[RLIMIT_NPROC].rlim_cur = maxprocperuid;
618	limit0.pl_rlimit[RLIMIT_NPROC].rlim_max = maxproc;
619	limit0.pl_rlimit[RLIMIT_STACK] = vm_initial_limit_stack;
620	limit0.pl_rlimit[RLIMIT_DATA] = vm_initial_limit_data;
621	limit0.pl_rlimit[RLIMIT_CORE] = vm_initial_limit_core;
622	limit0.pl_refcnt = 1;
623
624	kernproc->p_stats = &pstats0;
625	kernproc->p_sigacts = &sigacts0;
626
627	/*
628	 * Charge root for one process: launchd.
629	 */
630	bsd_init_kprintf("calling chgproccnt\n");
631	(void)chgproccnt(0, 1);
632
633	/*
634	 *	Allocate a kernel submap for pageable memory
635	 *	for temporary copying (execve()).
636	 */
637	{
638		vm_offset_t	minimum;
639
640		bsd_init_kprintf("calling kmem_suballoc\n");
641		assert(bsd_pageable_map_size != 0);
642		ret = kmem_suballoc(kernel_map,
643				&minimum,
644				(vm_size_t)bsd_pageable_map_size,
645				TRUE,
646				VM_FLAGS_ANYWHERE,
647				&bsd_pageable_map);
648		if (ret != KERN_SUCCESS)
649			panic("bsd_init: Failed to allocate bsd pageable map");
650	}
651
652	/*
653	 * Initialize buffers and hash links for buffers
654	 *
655	 * SIDE EFFECT: Starts a thread for bcleanbuf_thread(), so must
656	 *		happen after a credential has been associated with
657	 *		the kernel task.
658	 */
659	bsd_init_kprintf("calling bsd_bufferinit\n");
660	bsd_bufferinit();
661
662	/* Initialize the execve() semaphore */
663	bsd_init_kprintf("calling semaphore_create\n");
664
665	if (ret != KERN_SUCCESS)
666		panic("bsd_init: Failed to create execve semaphore");
667
668	/*
669	 * Initialize the calendar.
670	 */
671	bsd_init_kprintf("calling IOKitInitializeTime\n");
672	IOKitInitializeTime();
673
674	bsd_init_kprintf("calling ubc_init\n");
675	ubc_init();
676
677	/*
678	 * Initialize device-switches.
679	 */
680	bsd_init_kprintf("calling devsw_init() \n");
681	devsw_init();
682
683	/* Initialize the file systems. */
684	bsd_init_kprintf("calling vfsinit\n");
685	vfsinit();
686
687#if CONFIG_PROC_UUID_POLICY
688	/* Initial proc_uuid_policy subsystem */
689	bsd_init_kprintf("calling proc_uuid_policy_init()\n");
690	proc_uuid_policy_init();
691#endif
692
693#if SOCKETS
694	/* Initialize per-CPU cache allocator */
695	mcache_init();
696
697	/* Initialize mbuf's. */
698	bsd_init_kprintf("calling mbinit\n");
699	mbinit();
700	net_str_id_init(); /* for mbuf tags */
701#endif /* SOCKETS */
702
703	/*
704	 * Initializes security event auditing.
705	 * XXX: Should/could this occur later?
706	 */
707#if CONFIG_AUDIT
708	bsd_init_kprintf("calling audit_init\n");
709 	audit_init();
710#endif
711
712	/* Initialize kqueues */
713	bsd_init_kprintf("calling knote_init\n");
714	knote_init();
715
716	/* Initialize for async IO */
717	bsd_init_kprintf("calling aio_init\n");
718	aio_init();
719
720	/* Initialize pipes */
721	bsd_init_kprintf("calling pipeinit\n");
722	pipeinit();
723
724	/* Initialize SysV shm subsystem locks; the subsystem proper is
725	 * initialized through a sysctl.
726	 */
727#if SYSV_SHM
728	bsd_init_kprintf("calling sysv_shm_lock_init\n");
729	sysv_shm_lock_init();
730#endif
731#if SYSV_SEM
732	bsd_init_kprintf("calling sysv_sem_lock_init\n");
733	sysv_sem_lock_init();
734#endif
735#if SYSV_MSG
736	bsd_init_kprintf("sysv_msg_lock_init\n");
737	sysv_msg_lock_init();
738#endif
739	bsd_init_kprintf("calling pshm_lock_init\n");
740	pshm_lock_init();
741	bsd_init_kprintf("calling psem_lock_init\n");
742	psem_lock_init();
743
744	pthread_init();
745	/* POSIX Shm and Sem */
746	bsd_init_kprintf("calling pshm_cache_init\n");
747	pshm_cache_init();
748	bsd_init_kprintf("calling psem_cache_init\n");
749	psem_cache_init();
750	bsd_init_kprintf("calling time_zone_slock_init\n");
751	time_zone_slock_init();
752	bsd_init_kprintf("calling select_wait_queue_init\n");
753	select_wait_queue_init();
754
755	/*
756	 * Initialize protocols.  Block reception of incoming packets
757	 * until everything is ready.
758	 */
759	bsd_init_kprintf("calling sysctl_register_fixed\n");
760	sysctl_register_fixed();
761	bsd_init_kprintf("calling sysctl_mib_init\n");
762	sysctl_mib_init();
763#if NETWORKING
764	bsd_init_kprintf("calling dlil_init\n");
765	dlil_init();
766	bsd_init_kprintf("calling proto_kpi_init\n");
767	proto_kpi_init();
768#endif /* NETWORKING */
769#if SOCKETS
770	bsd_init_kprintf("calling socketinit\n");
771	socketinit();
772	bsd_init_kprintf("calling domaininit\n");
773	domaininit();
774	iptap_init();
775#if FLOW_DIVERT
776	flow_divert_init();
777#endif	/* FLOW_DIVERT */
778#endif /* SOCKETS */
779
780	kernproc->p_fd->fd_cdir = NULL;
781	kernproc->p_fd->fd_rdir = NULL;
782
783#if CONFIG_FREEZE
784#ifndef CONFIG_MEMORYSTATUS
785    #error "CONFIG_FREEZE defined without matching CONFIG_MEMORYSTATUS"
786#endif
787	/* Initialise background freezing */
788	bsd_init_kprintf("calling memorystatus_freeze_init\n");
789	memorystatus_freeze_init();
790#endif
791
792#if CONFIG_MEMORYSTATUS
793	/* Initialize kernel memory status notifications */
794	bsd_init_kprintf("calling memorystatus_init\n");
795	memorystatus_init();
796#endif /* CONFIG_MEMORYSTATUS */
797
798	bsd_init_kprintf("calling macx_init\n");
799	macx_init();
800
801	bsd_init_kprintf("calling acct_init\n");
802	acct_init();
803
804#ifdef GPROF
805	/* Initialize kernel profiling. */
806	kmstartup();
807#endif
808
809	bsd_init_kprintf("calling bsd_autoconf\n");
810	bsd_autoconf();
811
812#if CONFIG_DTRACE
813	dtrace_postinit();
814#endif
815
816	/*
817	 * We attach the loopback interface *way* down here to ensure
818	 * it happens after autoconf(), otherwise it becomes the
819	 * "primary" interface.
820	 */
821#include <loop.h>
822#if NLOOP > 0
823	bsd_init_kprintf("calling loopattach\n");
824	loopattach();			/* XXX */
825#endif
826#if NGIF
827	/* Initialize gif interface (after lo0) */
828	gif_init();
829#endif
830
831#if PFLOG
832	/* Initialize packet filter log interface */
833	pfloginit();
834#endif /* PFLOG */
835
836#if NETHER > 0
837	/* Register the built-in dlil ethernet interface family */
838	bsd_init_kprintf("calling ether_family_init\n");
839	ether_family_init();
840#endif /* ETHER */
841
842#if NETWORKING
843	/* Call any kext code that wants to run just after network init */
844	bsd_init_kprintf("calling net_init_run\n");
845	net_init_run();
846
847#if CONTENT_FILTER
848	cfil_init();
849#endif
850
851#if PACKET_MANGLER
852	pkt_mnglr_init();
853#endif
854
855#if NECP
856	/* Initialize Network Extension Control Policies */
857	necp_init();
858#endif
859
860	/* register user tunnel kernel control handler */
861	utun_register_control();
862#if IPSEC
863	ipsec_register_control();
864#endif /* IPSEC */
865	netsrc_init();
866	nstat_init();
867	tcp_cc_init();
868#endif /* NETWORKING */
869
870	bsd_init_kprintf("calling vnode_pager_bootstrap\n");
871	vnode_pager_bootstrap();
872
873	bsd_init_kprintf("calling inittodr\n");
874	inittodr(0);
875
876	/* Mount the root file system. */
877	while( TRUE) {
878		int err;
879
880		bsd_init_kprintf("calling setconf\n");
881		setconf();
882#if NFSCLIENT
883		netboot = (mountroot == netboot_mountroot);
884#endif
885
886		bsd_init_kprintf("vfs_mountroot\n");
887		if (0 == (err = vfs_mountroot()))
888			break;
889		rootdevice[0] = '\0';
890#if NFSCLIENT
891		if (netboot) {
892			PE_display_icon( 0, "noroot");  /* XXX a netboot-specific icon would be nicer */
893			vc_progress_set(FALSE, 0);
894			for (i=1; 1; i*=2) {
895				printf("bsd_init: failed to mount network root, error %d, %s\n",
896					err, PE_boot_args());
897				printf("We are hanging here...\n");
898				IOSleep(i*60*1000);
899			}
900			/*NOTREACHED*/
901		}
902#endif
903		printf("cannot mount root, errno = %d\n", err);
904		boothowto |= RB_ASKNAME;
905	}
906
907	IOSecureBSDRoot(rootdevice);
908
909	context.vc_thread = current_thread();
910	context.vc_ucred = kernproc->p_ucred;
911	mountlist.tqh_first->mnt_flag |= MNT_ROOTFS;
912
913	bsd_init_kprintf("calling VFS_ROOT\n");
914	/* Get the vnode for '/'.  Set fdp->fd_fd.fd_cdir to reference it. */
915	if (VFS_ROOT(mountlist.tqh_first, &rootvnode, &context))
916		panic("bsd_init: cannot find root vnode: %s", PE_boot_args());
917	rootvnode->v_flag |= VROOT;
918	(void)vnode_ref(rootvnode);
919	(void)vnode_put(rootvnode);
920	filedesc0.fd_cdir = rootvnode;
921
922#if NFSCLIENT
923	if (netboot) {
924		int err;
925
926		netboot = TRUE;
927		/* post mount setup */
928		if ((err = netboot_setup()) != 0) {
929			PE_display_icon( 0, "noroot");  /* XXX a netboot-specific icon would be nicer */
930			vc_progress_set(FALSE, 0);
931			for (i=1; 1; i*=2) {
932				printf("bsd_init: NetBoot could not find root, error %d: %s\n",
933					err, PE_boot_args());
934				printf("We are hanging here...\n");
935				IOSleep(i*60*1000);
936			}
937			/*NOTREACHED*/
938		}
939	}
940#endif
941
942
943#if CONFIG_IMAGEBOOT
944	/*
945	 * See if a system disk image is present. If so, mount it and
946	 * switch the root vnode to point to it
947	 */
948	if (netboot == FALSE && imageboot_needed()) {
949		/*
950		 * An image was found.  No turning back: we're booted
951		 * with a kernel from the disk image.
952		 */
953		imageboot_setup();
954	}
955#endif /* CONFIG_IMAGEBOOT */
956
957	/* set initial time; all other resource data is  already zero'ed */
958	microtime_with_abstime(&kernproc->p_start, &kernproc->p_stats->ps_start);
959
960#if DEVFS
961	{
962	    char mounthere[] = "/dev";	/* !const because of internal casting */
963
964	    bsd_init_kprintf("calling devfs_kernel_mount\n");
965	    devfs_kernel_mount(mounthere);
966	}
967#endif /* DEVFS */
968
969	/* Initialize signal state for process 0. */
970	bsd_init_kprintf("calling siginit\n");
971	siginit(kernproc);
972
973	bsd_init_kprintf("calling bsd_utaskbootstrap\n");
974	bsd_utaskbootstrap();
975
976#if defined(__LP64__)
977	kernproc->p_flag |= P_LP64;
978#endif
979
980	pal_kernel_announce();
981
982	bsd_init_kprintf("calling mountroot_post_hook\n");
983
984	/* invoke post-root-mount hook */
985	if (mountroot_post_hook != NULL)
986		mountroot_post_hook();
987
988#if 0 /* not yet */
989	consider_zone_gc(FALSE);
990#endif
991
992	bsd_init_kprintf("done\n");
993}
994
995void
996bsdinit_task(void)
997{
998	proc_t p = current_proc();
999	struct uthread *ut;
1000	thread_t thread;
1001
1002	process_name("init", p);
1003
1004	ux_handler_init();
1005
1006	thread = current_thread();
1007	(void) host_set_exception_ports(host_priv_self(),
1008					EXC_MASK_ALL & ~(EXC_MASK_RPC_ALERT),//pilotfish (shark) needs this port
1009					(mach_port_t) ux_exception_port,
1010					EXCEPTION_DEFAULT| MACH_EXCEPTION_CODES,
1011					0);
1012
1013	ut = (uthread_t)get_bsdthread_info(thread);
1014
1015	bsd_init_task = get_threadtask(thread);
1016	init_task_failure_data[0] = 0;
1017
1018#if CONFIG_MACF
1019	mac_cred_label_associate_user(p->p_ucred);
1020#endif
1021	load_init_program(p);
1022	lock_trace = 1;
1023}
1024
1025kern_return_t
1026bsd_autoconf(void)
1027{
1028	kprintf("bsd_autoconf: calling kminit\n");
1029	kminit();
1030
1031	/*
1032	 * Early startup for bsd pseudodevices.
1033	 */
1034	{
1035	    struct pseudo_init *pi;
1036
1037	    for (pi = pseudo_inits; pi->ps_func; pi++)
1038		(*pi->ps_func) (pi->ps_count);
1039	}
1040
1041	return( IOKitBSDInit());
1042}
1043
1044
1045#include <sys/disklabel.h>  /* for MAXPARTITIONS */
1046
1047static void
1048setconf(void)
1049{
1050	u_int32_t	flags;
1051	kern_return_t	err;
1052
1053	err = IOFindBSDRoot(rootdevice, sizeof(rootdevice), &rootdev, &flags);
1054	if( err) {
1055		printf("setconf: IOFindBSDRoot returned an error (%d);"
1056			"setting rootdevice to 'sd0a'.\n", err); /* XXX DEBUG TEMP */
1057		rootdev = makedev( 6, 0 );
1058		strlcpy(rootdevice, "sd0a", sizeof(rootdevice));
1059		flags = 0;
1060	}
1061
1062#if NFSCLIENT
1063	if( flags & 1 ) {
1064		/* network device */
1065		mountroot = netboot_mountroot;
1066	} else {
1067#endif
1068		/* otherwise have vfs determine root filesystem */
1069		mountroot = NULL;
1070#if NFSCLIENT
1071	}
1072#endif
1073
1074}
1075
1076void
1077bsd_utaskbootstrap(void)
1078{
1079	thread_t thread;
1080	struct uthread *ut;
1081
1082	/*
1083	 * Clone the bootstrap process from the kernel process, without
1084	 * inheriting either task characteristics or memory from the kernel;
1085	 */
1086	thread = cloneproc(TASK_NULL, COALITION_NULL, kernproc, FALSE, TRUE);
1087
1088	/* Hold the reference as it will be dropped during shutdown */
1089	initproc = proc_find(1);
1090#if __PROC_INTERNAL_DEBUG
1091	if (initproc == PROC_NULL)
1092		panic("bsd_utaskbootstrap: initproc not set\n");
1093#endif
1094	/*
1095	 * Since we aren't going back out the normal way to our parent,
1096	 * we have to drop the transition locks explicitly.
1097	 */
1098	proc_signalend(initproc, 0);
1099	proc_transend(initproc, 0);
1100
1101	ut = (struct uthread *)get_bsdthread_info(thread);
1102	ut->uu_sigmask = 0;
1103	act_set_astbsd(thread);
1104	(void) thread_resume(thread);
1105}
1106
1107static void
1108parse_bsd_args(void)
1109{
1110	char namep[16];
1111	int msgbuf;
1112
1113	if ( PE_parse_boot_argn("-s", namep, sizeof (namep)))
1114		boothowto |= RB_SINGLE;
1115
1116	if (PE_parse_boot_argn("-b", namep, sizeof (namep)))
1117		boothowto |= RB_NOBOOTRC;
1118
1119	if (PE_parse_boot_argn("-x", namep, sizeof (namep))) /* safe boot */
1120		boothowto |= RB_SAFEBOOT;
1121
1122
1123	/* disable vnode_cache_is_authorized() by setting vnode_cache_defeat */
1124	if (PE_parse_boot_argn("-vnode_cache_defeat", namep, sizeof (namep)))
1125		bootarg_vnode_cache_defeat = 1;
1126
1127#if DEVELOPMENT || DEBUG
1128	if (PE_parse_boot_argn("-disable_aslr", namep, sizeof (namep)))
1129		bootarg_disable_aslr = 1;
1130#endif
1131
1132	PE_parse_boot_argn("ncl", &ncl, sizeof (ncl));
1133	if (PE_parse_boot_argn("nbuf", &max_nbuf_headers,
1134				sizeof (max_nbuf_headers))) {
1135		customnbuf = 1;
1136	}
1137
1138#if CONFIG_MACF
1139#if defined (__i386__) || defined (__x86_64__)
1140	PE_parse_boot_argn("policy_check", &policy_check_flags, sizeof (policy_check_flags));
1141#endif
1142#endif	/* CONFIG_MACF */
1143
1144	if (PE_parse_boot_argn("msgbuf", &msgbuf, sizeof (msgbuf))) {
1145		log_setsize(msgbuf);
1146	}
1147
1148	if (PE_parse_boot_argn("-novfscache", namep, sizeof(namep))) {
1149		nc_disabled = 1;
1150	}
1151}
1152
1153void
1154bsd_exec_setup(int scale)
1155{
1156
1157	switch (scale) {
1158		case 0:
1159		case 1:
1160			bsd_simul_execs = BSD_SIMUL_EXECS;
1161			break;
1162		case 2:
1163		case 3:
1164			bsd_simul_execs = 65;
1165			break;
1166		case 4:
1167		case 5:
1168			bsd_simul_execs = 129;
1169			break;
1170		case 6:
1171		case 7:
1172			bsd_simul_execs = 257;
1173			break;
1174		default:
1175			bsd_simul_execs = 513;
1176			break;
1177
1178	}
1179	bsd_pageable_map_size = (bsd_simul_execs * BSD_PAGEABLE_SIZE_PER_EXEC);
1180}
1181
1182#if !NFSCLIENT
1183int
1184netboot_root(void);
1185
1186int
1187netboot_root(void)
1188{
1189	return(0);
1190}
1191#endif
1192