aboutsummaryrefslogtreecommitdiff
path: root/sys/kern/kern_exec.c
blob: 763196a593e4a40e46ebfc377ab39dc70286dc97 (plain) (blame)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
/*-
 * SPDX-License-Identifier: BSD-2-Clause-FreeBSD
 *
 * Copyright (c) 1993, David Greenman
 * 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 "opt_capsicum.h"
#include "opt_hwpmc_hooks.h"
#include "opt_ktrace.h"
#include "opt_vm.h"

#include <sys/param.h>
#include <sys/systm.h>
#include <sys/acct.h>
#include <sys/asan.h>
#include <sys/capsicum.h>
#include <sys/compressor.h>
#include <sys/eventhandler.h>
#include <sys/exec.h>
#include <sys/fcntl.h>
#include <sys/filedesc.h>
#include <sys/imgact.h>
#include <sys/imgact_elf.h>
#include <sys/kernel.h>
#include <sys/lock.h>
#include <sys/malloc.h>
#include <sys/mman.h>
#include <sys/mount.h>
#include <sys/mutex.h>
#include <sys/namei.h>
#include <sys/priv.h>
#include <sys/proc.h>
#include <sys/ptrace.h>
#include <sys/resourcevar.h>
#include <sys/rwlock.h>
#include <sys/sched.h>
#include <sys/sdt.h>
#include <sys/sf_buf.h>
#include <sys/shm.h>
#include <sys/signalvar.h>
#include <sys/smp.h>
#include <sys/stat.h>
#include <sys/syscallsubr.h>
#include <sys/sysctl.h>
#include <sys/sysent.h>
#include <sys/sysproto.h>
#include <sys/timers.h>
#include <sys/umtx.h>
#include <sys/vnode.h>
#include <sys/wait.h>
#ifdef KTRACE
#include <sys/ktrace.h>
#endif

#include <vm/vm.h>
#include <vm/vm_param.h>
#include <vm/pmap.h>
#include <vm/vm_page.h>
#include <vm/vm_map.h>
#include <vm/vm_kern.h>
#include <vm/vm_extern.h>
#include <vm/vm_object.h>
#include <vm/vm_pager.h>

#ifdef	HWPMC_HOOKS
#include <sys/pmckern.h>
#endif

#include <machine/reg.h>

#include <security/audit/audit.h>
#include <security/mac/mac_framework.h>

#ifdef KDTRACE_HOOKS
#include <sys/dtrace_bsd.h>
dtrace_execexit_func_t	dtrace_fasttrap_exec;
#endif

SDT_PROVIDER_DECLARE(proc);
SDT_PROBE_DEFINE1(proc, , , exec, "char *");
SDT_PROBE_DEFINE1(proc, , , exec__failure, "int");
SDT_PROBE_DEFINE1(proc, , , exec__success, "char *");

MALLOC_DEFINE(M_PARGS, "proc-args", "Process arguments");

int coredump_pack_fileinfo = 1;
SYSCTL_INT(_kern, OID_AUTO, coredump_pack_fileinfo, CTLFLAG_RWTUN,
    &coredump_pack_fileinfo, 0,
    "Enable file path packing in 'procstat -f' coredump notes");

int coredump_pack_vmmapinfo = 1;
SYSCTL_INT(_kern, OID_AUTO, coredump_pack_vmmapinfo, CTLFLAG_RWTUN,
    &coredump_pack_vmmapinfo, 0,
    "Enable file path packing in 'procstat -v' coredump notes");

static int sysctl_kern_ps_strings(SYSCTL_HANDLER_ARGS);
static int sysctl_kern_usrstack(SYSCTL_HANDLER_ARGS);
static int sysctl_kern_stackprot(SYSCTL_HANDLER_ARGS);
static int do_execve(struct thread *td, struct image_args *args,
    struct mac *mac_p, struct vmspace *oldvmspace);

/* XXX This should be vm_size_t. */
SYSCTL_PROC(_kern, KERN_PS_STRINGS, ps_strings, CTLTYPE_ULONG|CTLFLAG_RD|
    CTLFLAG_CAPRD|CTLFLAG_MPSAFE, NULL, 0, sysctl_kern_ps_strings, "LU",
    "Location of process' ps_strings structure");

/* XXX This should be vm_size_t. */
SYSCTL_PROC(_kern, KERN_USRSTACK, usrstack, CTLTYPE_ULONG|CTLFLAG_RD|
    CTLFLAG_CAPRD|CTLFLAG_MPSAFE, NULL, 0, sysctl_kern_usrstack, "LU",
    "Top of process stack");

SYSCTL_PROC(_kern, OID_AUTO, stackprot, CTLTYPE_INT|CTLFLAG_RD|CTLFLAG_MPSAFE,
    NULL, 0, sysctl_kern_stackprot, "I",
    "Stack memory permissions");

u_long ps_arg_cache_limit = PAGE_SIZE / 16;
SYSCTL_ULONG(_kern, OID_AUTO, ps_arg_cache_limit, CTLFLAG_RW, 
    &ps_arg_cache_limit, 0,
    "Process' command line characters cache limit");

static int disallow_high_osrel;
SYSCTL_INT(_kern, OID_AUTO, disallow_high_osrel, CTLFLAG_RW,
    &disallow_high_osrel, 0,
    "Disallow execution of binaries built for higher version of the world");

static int map_at_zero = 0;
SYSCTL_INT(_security_bsd, OID_AUTO, map_at_zero, CTLFLAG_RWTUN, &map_at_zero, 0,
    "Permit processes to map an object at virtual address 0.");

static int
sysctl_kern_ps_strings(SYSCTL_HANDLER_ARGS)
{
	struct proc *p;
	int error;

	p = curproc;
#ifdef SCTL_MASK32
	if (req->flags & SCTL_MASK32) {
		unsigned int val;
		val = (unsigned int)p->p_sysent->sv_psstrings;
		error = SYSCTL_OUT(req, &val, sizeof(val));
	} else
#endif
		error = SYSCTL_OUT(req, &p->p_sysent->sv_psstrings,
		   sizeof(p->p_sysent->sv_psstrings));
	return error;
}

static int
sysctl_kern_usrstack(SYSCTL_HANDLER_ARGS)
{
	struct proc *p;
	int error;

	p = curproc;
#ifdef SCTL_MASK32
	if (req->flags & SCTL_MASK32) {
		unsigned int val;
		val = (unsigned int)p->p_sysent->sv_usrstack;
		error = SYSCTL_OUT(req, &val, sizeof(val));
	} else
#endif
		error = SYSCTL_OUT(req, &p->p_sysent->sv_usrstack,
		    sizeof(p->p_sysent->sv_usrstack));
	return error;
}

static int
sysctl_kern_stackprot(SYSCTL_HANDLER_ARGS)
{
	struct proc *p;

	p = curproc;
	return (SYSCTL_OUT(req, &p->p_sysent->sv_stackprot,
	    sizeof(p->p_sysent->sv_stackprot)));
}

/*
 * Each of the items is a pointer to a `const struct execsw', hence the
 * double pointer here.
 */
static const struct execsw **execsw;

#ifndef _SYS_SYSPROTO_H_
struct execve_args {
	char    *fname; 
	char    **argv;
	char    **envv; 
};
#endif

int
sys_execve(struct thread *td, struct execve_args *uap)
{
	struct image_args args;
	struct vmspace *oldvmspace;
	int error;

	error = pre_execve(td, &oldvmspace);
	if (error != 0)
		return (error);
	error = exec_copyin_args(&args, uap->fname, UIO_USERSPACE,
	    uap->argv, uap->envv);
	if (error == 0)
		error = kern_execve(td, &args, NULL, oldvmspace);
	post_execve(td, error, oldvmspace);
	AUDIT_SYSCALL_EXIT(error == EJUSTRETURN ? 0 : error, td);
	return (error);
}

#ifndef _SYS_SYSPROTO_H_
struct fexecve_args {
	int	fd;
	char	**argv;
	char	**envv;
};
#endif
int
sys_fexecve(struct thread *td, struct fexecve_args *uap)
{
	struct image_args args;
	struct vmspace *oldvmspace;
	int error;

	error = pre_execve(td, &oldvmspace);
	if (error != 0)
		return (error);
	error = exec_copyin_args(&args, NULL, UIO_SYSSPACE,
	    uap->argv, uap->envv);
	if (error == 0) {
		args.fd = uap->fd;
		error = kern_execve(td, &args, NULL, oldvmspace);
	}
	post_execve(td, error, oldvmspace);
	AUDIT_SYSCALL_EXIT(error == EJUSTRETURN ? 0 : error, td);
	return (error);
}

#ifndef _SYS_SYSPROTO_H_
struct __mac_execve_args {
	char	*fname;
	char	**argv;
	char	**envv;
	struct mac	*mac_p;
};
#endif

int
sys___mac_execve(struct thread *td, struct __mac_execve_args *uap)
{
#ifdef MAC
	struct image_args args;
	struct vmspace *oldvmspace;
	int error;

	error = pre_execve(td, &oldvmspace);
	if (error != 0)
		return (error);
	error = exec_copyin_args(&args, uap->fname, UIO_USERSPACE,
	    uap->argv, uap->envv);
	if (error == 0)
		error = kern_execve(td, &args, uap->mac_p, oldvmspace);
	post_execve(td, error, oldvmspace);
	AUDIT_SYSCALL_EXIT(error == EJUSTRETURN ? 0 : error, td);
	return (error);
#else
	return (ENOSYS);
#endif
}

int
pre_execve(struct thread *td, struct vmspace **oldvmspace)
{
	struct proc *p;
	int error;

	KASSERT(td == curthread, ("non-current thread %p", td));
	error = 0;
	p = td->td_proc;
	if ((p->p_flag & P_HADTHREADS) != 0) {
		PROC_LOCK(p);
		if (thread_single(p, SINGLE_BOUNDARY) != 0)
			error = ERESTART;
		PROC_UNLOCK(p);
	}
	KASSERT(error != 0 || (td->td_pflags & TDP_EXECVMSPC) == 0,
	    ("nested execve"));
	*oldvmspace = p->p_vmspace;
	return (error);
}

void
post_execve(struct thread *td, int error, struct vmspace *oldvmspace)
{
	struct proc *p;

	KASSERT(td == curthread, ("non-current thread %p", td));
	p = td->td_proc;
	if ((p->p_flag & P_HADTHREADS) != 0) {
		PROC_LOCK(p);
		/*
		 * If success, we upgrade to SINGLE_EXIT state to
		 * force other threads to suicide.
		 */
		if (error == EJUSTRETURN)
			thread_single(p, SINGLE_EXIT);
		else
			thread_single_end(p, SINGLE_BOUNDARY);
		PROC_UNLOCK(p);
	}
	exec_cleanup(td, oldvmspace);
}

/*
 * kern_execve() has the astonishing property of not always returning to
 * the caller.  If sufficiently bad things happen during the call to
 * do_execve(), it can end up calling exit1(); as a result, callers must
 * avoid doing anything which they might need to undo (e.g., allocating
 * memory).
 */
int
kern_execve(struct thread *td, struct image_args *args, struct mac *mac_p,
    struct vmspace *oldvmspace)
{

	AUDIT_ARG_ARGV(args->begin_argv, args->argc,
	    exec_args_get_begin_envv(args) - args->begin_argv);
	AUDIT_ARG_ENVV(exec_args_get_begin_envv(args), args->envc,
	    args->endp - exec_args_get_begin_envv(args));
	return (do_execve(td, args, mac_p, oldvmspace));
}

static void
execve_nosetid(struct image_params *imgp)
{
	imgp->credential_setid = false;
	if (imgp->newcred != NULL) {
		crfree(imgp->newcred);
		imgp->newcred = NULL;
	}
}

/*
 * In-kernel implementation of execve().  All arguments are assumed to be
 * userspace pointers from the passed thread.
 */
static int
do_execve(struct thread *td, struct image_args *args, struct mac *mac_p,
    struct vmspace *oldvmspace)
{
	struct proc *p = td->td_proc;
	struct nameidata nd;
	struct ucred *oldcred;
	struct uidinfo *euip = NULL;
	uintptr_t stack_base;
	struct image_params image_params, *imgp;
	struct vattr attr;
	int (*img_first)(struct image_params *);
	struct pargs *oldargs = NULL, *newargs = NULL;
	struct sigacts *oldsigacts = NULL, *newsigacts = NULL;
#ifdef KTRACE
	struct ktr_io_params *kiop;
#endif
	struct vnode *oldtextvp = NULL, *newtextvp;
	int credential_changing;
#ifdef MAC
	struct label *interpvplabel = NULL;
	int will_transition;
#endif
#ifdef HWPMC_HOOKS
	struct pmckern_procexec pe;
#endif
	int error, i, orig_osrel;
	uint32_t orig_fctl0;
	Elf_Brandinfo *orig_brandinfo;
	static const char fexecv_proc_title[] = "(fexecv)";

	imgp = &image_params;
#ifdef KTRACE
	kiop = NULL;
#endif

	/*
	 * Lock the process and set the P_INEXEC flag to indicate that
	 * it should be left alone until we're done here.  This is
	 * necessary to avoid race conditions - e.g. in ptrace() -
	 * that might allow a local user to illicitly obtain elevated
	 * privileges.
	 */
	PROC_LOCK(p);
	KASSERT((p->p_flag & P_INEXEC) == 0,
	    ("%s(): process already has P_INEXEC flag", __func__));
	p->p_flag |= P_INEXEC;
	PROC_UNLOCK(p);

	/*
	 * Initialize part of the common data
	 */
	bzero(imgp, sizeof(*imgp));
	imgp->proc = p;
	imgp->attr = &attr;
	imgp->args = args;
	oldcred = p->p_ucred;
	orig_osrel = p->p_osrel;
	orig_fctl0 = p->p_fctl0;
	orig_brandinfo = p->p_elf_brandinfo;

#ifdef MAC
	error = mac_execve_enter(imgp, mac_p);
	if (error)
		goto exec_fail;
#endif

	/*
	 * Translate the file name. namei() returns a vnode pointer
	 *	in ni_vp among other things.
	 *
	 * XXXAUDIT: It would be desirable to also audit the name of the
	 * interpreter if this is an interpreted binary.
	 */
	if (args->fname != NULL) {
		NDINIT(&nd, LOOKUP, ISOPEN | LOCKLEAF | LOCKSHARED | FOLLOW |
		    SAVENAME | AUDITVNODE1, UIO_SYSSPACE, args->fname, td);
	}

	SDT_PROBE1(proc, , , exec, args->fname);

interpret:
	if (args->fname != NULL) {
#ifdef CAPABILITY_MODE
		/*
		 * While capability mode can't reach this point via direct
		 * path arguments to execve(), we also don't allow
		 * interpreters to be used in capability mode (for now).
		 * Catch indirect lookups and return a permissions error.
		 */
		if (IN_CAPABILITY_MODE(td)) {
			error = ECAPMODE;
			goto exec_fail;
		}
#endif
		error = namei(&nd);
		if (error)
			goto exec_fail;

		newtextvp = nd.ni_vp;
		imgp->vp = newtextvp;
	} else {
		AUDIT_ARG_FD(args->fd);
		/*
		 * Descriptors opened only with O_EXEC or O_RDONLY are allowed.
		 */
		error = fgetvp_exec(td, args->fd, &cap_fexecve_rights, &newtextvp);
		if (error)
			goto exec_fail;
		vn_lock(newtextvp, LK_SHARED | LK_RETRY);
		AUDIT_ARG_VNODE1(newtextvp);
		imgp->vp = newtextvp;
	}

	/*
	 * Check file permissions.  Also 'opens' file and sets its vnode to
	 * text mode.
	 */
	error = exec_check_permissions(imgp);
	if (error)
		goto exec_fail_dealloc;

	imgp->object = imgp->vp->v_object;
	if (imgp->object != NULL)
		vm_object_reference(imgp->object);

	error = exec_map_first_page(imgp);
	if (error)
		goto exec_fail_dealloc;

	imgp->proc->p_osrel = 0;
	imgp->proc->p_fctl0 = 0;
	imgp->proc->p_elf_brandinfo = NULL;

	/*
	 * Implement image setuid/setgid.
	 *
	 * Determine new credentials before attempting image activators
	 * so that it can be used by process_exec handlers to determine
	 * credential/setid changes.
	 *
	 * Don't honor setuid/setgid if the filesystem prohibits it or if
	 * the process is being traced.
	 *
	 * We disable setuid/setgid/etc in capability mode on the basis
	 * that most setugid applications are not written with that
	 * environment in mind, and will therefore almost certainly operate
	 * incorrectly. In principle there's no reason that setugid
	 * applications might not be useful in capability mode, so we may want
	 * to reconsider this conservative design choice in the future.
	 *
	 * XXXMAC: For the time being, use NOSUID to also prohibit
	 * transitions on the file system.
	 */
	credential_changing = 0;
	credential_changing |= (attr.va_mode & S_ISUID) &&
	    oldcred->cr_uid != attr.va_uid;
	credential_changing |= (attr.va_mode & S_ISGID) &&
	    oldcred->cr_gid != attr.va_gid;
#ifdef MAC
	will_transition = mac_vnode_execve_will_transition(oldcred, imgp->vp,
	    interpvplabel, imgp);
	credential_changing |= will_transition;
#endif

	/* Don't inherit PROC_PDEATHSIG_CTL value if setuid/setgid. */
	if (credential_changing)
		imgp->proc->p_pdeathsig = 0;

	if (credential_changing &&
#ifdef CAPABILITY_MODE
	    ((oldcred->cr_flags & CRED_FLAG_CAPMODE) == 0) &&
#endif
	    (imgp->vp->v_mount->mnt_flag & MNT_NOSUID) == 0 &&
	    (p->p_flag & P_TRACED) == 0) {
		imgp->credential_setid = true;
		VOP_UNLOCK(imgp->vp);
		imgp->newcred = crdup(oldcred);
		if (attr.va_mode & S_ISUID) {
			euip = uifind(attr.va_uid);
			change_euid(imgp->newcred, euip);
		}
		vn_lock(imgp->vp, LK_SHARED | LK_RETRY);
		if (attr.va_mode & S_ISGID)
			change_egid(imgp->newcred, attr.va_gid);
		/*
		 * Implement correct POSIX saved-id behavior.
		 *
		 * XXXMAC: Note that the current logic will save the
		 * uid and gid if a MAC domain transition occurs, even
		 * though maybe it shouldn't.
		 */
		change_svuid(imgp->newcred, imgp->newcred->cr_uid);
		change_svgid(imgp->newcred, imgp->newcred->cr_gid);
	} else {
		/*
		 * Implement correct POSIX saved-id behavior.
		 *
		 * XXX: It's not clear that the existing behavior is
		 * POSIX-compliant.  A number of sources indicate that the
		 * saved uid/gid should only be updated if the new ruid is
		 * not equal to the old ruid, or the new euid is not equal
		 * to the old euid and the new euid is not equal to the old
		 * ruid.  The FreeBSD code always updates the saved uid/gid.
		 * Also, this code uses the new (replaced) euid and egid as
		 * the source, which may or may not be the right ones to use.
		 */
		if (oldcred->cr_svuid != oldcred->cr_uid ||
		    oldcred->cr_svgid != oldcred->cr_gid) {
			VOP_UNLOCK(imgp->vp);
			imgp->newcred = crdup(oldcred);
			vn_lock(imgp->vp, LK_SHARED | LK_RETRY);
			change_svuid(imgp->newcred, imgp->newcred->cr_uid);
			change_svgid(imgp->newcred, imgp->newcred->cr_gid);
		}
	}
	/* The new credentials are installed into the process later. */

	/*
	 * Do the best to calculate the full path to the image file.
	 */
	if (args->fname != NULL && args->fname[0] == '/')
		imgp->execpath = args->fname;
	else {
		VOP_UNLOCK(imgp->vp);
		if (vn_fullpath(imgp->vp, &imgp->execpath, &imgp->freepath) != 0)
			imgp->execpath = args->fname;
		vn_lock(imgp->vp, LK_SHARED | LK_RETRY);
	}

	/*
	 *	If the current process has a special image activator it
	 *	wants to try first, call it.   For example, emulating shell
	 *	scripts differently.
	 */
	error = -1;
	if ((img_first = imgp->proc->p_sysent->sv_imgact_try) != NULL)
		error = img_first(imgp);

	/*
	 *	Loop through the list of image activators, calling each one.
	 *	An activator returns -1 if there is no match, 0 on success,
	 *	and an error otherwise.
	 */
	for (i = 0; error == -1 && execsw[i]; ++i) {
		if (execsw[i]->ex_imgact == NULL ||
		    execsw[i]->ex_imgact == img_first) {
			continue;
		}
		error = (*execsw[i]->ex_imgact)(imgp);
	}

	if (error) {
		if (error == -1)
			error = ENOEXEC;
		goto exec_fail_dealloc;
	}

	/*
	 * Special interpreter operation, cleanup and loop up to try to
	 * activate the interpreter.
	 */
	if (imgp->interpreted) {
		exec_unmap_first_page(imgp);
		/*
		 * The text reference needs to be removed for scripts.
		 * There is a short period before we determine that
		 * something is a script where text reference is active.
		 * The vnode lock is held over this entire period
		 * so nothing should illegitimately be blocked.
		 */
		MPASS(imgp->textset);
		VOP_UNSET_TEXT_CHECKED(newtextvp);
		imgp->textset = false;
		/* free name buffer and old vnode */
		if (args->fname != NULL)
			NDFREE(&nd, NDF_ONLY_PNBUF);
#ifdef MAC
		mac_execve_interpreter_enter(newtextvp, &interpvplabel);
#endif
		if (imgp->opened) {
			VOP_CLOSE(newtextvp, FREAD, td->td_ucred, td);
			imgp->opened = 0;
		}
		vput(newtextvp);
		vm_object_deallocate(imgp->object);
		imgp->object = NULL;
		execve_nosetid(imgp);
		imgp->execpath = NULL;
		free(imgp->freepath, M_TEMP);
		imgp->freepath = NULL;
		/* set new name to that of the interpreter */
		NDINIT(&nd, LOOKUP, ISOPEN | LOCKLEAF | LOCKSHARED | FOLLOW |
		    SAVENAME, UIO_SYSSPACE, imgp->interpreter_name, td);
		args->fname = imgp->interpreter_name;
		goto interpret;
	}

	/*
	 * NB: We unlock the vnode here because it is believed that none
	 * of the sv_copyout_strings/sv_fixup operations require the vnode.
	 */
	VOP_UNLOCK(imgp->vp);

	if (disallow_high_osrel &&
	    P_OSREL_MAJOR(p->p_osrel) > P_OSREL_MAJOR(__FreeBSD_version)) {
		error = ENOEXEC;
		uprintf("Osrel %d for image %s too high\n", p->p_osrel,
		    imgp->execpath != NULL ? imgp->execpath : "<unresolved>");
		vn_lock(imgp->vp, LK_SHARED | LK_RETRY);
		goto exec_fail_dealloc;
	}

	/* ABI enforces the use of Capsicum. Switch into capabilities mode. */
	if (SV_PROC_FLAG(p, SV_CAPSICUM))
		sys_cap_enter(td, NULL);

	/*
	 * Copy out strings (args and env) and initialize stack base.
	 */
	error = (*p->p_sysent->sv_copyout_strings)(imgp, &stack_base);
	if (error != 0) {
		vn_lock(imgp->vp, LK_SHARED | LK_RETRY);
		goto exec_fail_dealloc;
	}

	/*
	 * Stack setup.
	 */
	error = (*p->p_sysent->sv_fixup)(&stack_base, imgp);
	if (error != 0) {
		vn_lock(imgp->vp, LK_SHARED | LK_RETRY);
		goto exec_fail_dealloc;
	}

	if (args->fdp != NULL) {
		/* Install a brand new file descriptor table. */
		fdinstall_remapped(td, args->fdp);
		args->fdp = NULL;
	} else {
		/*
		 * Keep on using the existing file descriptor table. For
		 * security and other reasons, the file descriptor table
		 * cannot be shared after an exec.
		 */
		fdunshare(td);
		pdunshare(td);
		/* close files on exec */
		fdcloseexec(td);
	}

	/*
	 * Malloc things before we need locks.
	 */
	i = exec_args_get_begin_envv(imgp->args) - imgp->args->begin_argv;
	/* Cache arguments if they fit inside our allowance */
	if (ps_arg_cache_limit >= i + sizeof(struct pargs)) {
		newargs = pargs_alloc(i);
		bcopy(imgp->args->begin_argv, newargs->ar_args, i);
	}

	/*
	 * For security and other reasons, signal handlers cannot
	 * be shared after an exec. The new process gets a copy of the old
	 * handlers. In execsigs(), the new process will have its signals
	 * reset.
	 */
	if (sigacts_shared(p->p_sigacts)) {
		oldsigacts = p->p_sigacts;
		newsigacts = sigacts_alloc();
		sigacts_copy(newsigacts, oldsigacts);
	}

	vn_lock(imgp->vp, LK_SHARED | LK_RETRY);

	PROC_LOCK(p);
	if (oldsigacts)
		p->p_sigacts = newsigacts;
	/* Stop profiling */
	stopprofclock(p);

	/* reset caught signals */
	execsigs(p);

	/* name this process - nameiexec(p, ndp) */
	bzero(p->p_comm, sizeof(p->p_comm));
	if (args->fname)
		bcopy(nd.ni_cnd.cn_nameptr, p->p_comm,
		    min(nd.ni_cnd.cn_namelen, MAXCOMLEN));
	else if (vn_commname(newtextvp, p->p_comm, sizeof(p->p_comm)) != 0)
		bcopy(fexecv_proc_title, p->p_comm, sizeof(fexecv_proc_title));
	bcopy(p->p_comm, td->td_name, sizeof(td->td_name));
#ifdef KTR
	sched_clear_tdname(td);
#endif

	/*
	 * mark as execed, wakeup the process that vforked (if any) and tell
	 * it that it now has its own resources back
	 */
	p->p_flag |= P_EXEC;
	if ((p->p_flag2 & P2_NOTRACE_EXEC) == 0)
		p->p_flag2 &= ~P2_NOTRACE;
	if ((p->p_flag2 & P2_STKGAP_DISABLE_EXEC) == 0)
		p->p_flag2 &= ~P2_STKGAP_DISABLE;
	if (p->p_flag & P_PPWAIT) {
		p->p_flag &= ~(P_PPWAIT | P_PPTRACE);
		cv_broadcast(&p->p_pwait);
		/* STOPs are no longer ignored, arrange for AST */
		signotify(td);
	}

	if ((imgp->sysent->sv_setid_allowed != NULL &&
	    !(*imgp->sysent->sv_setid_allowed)(td, imgp)) ||
	    (p->p_flag2 & P2_NO_NEW_PRIVS) != 0)
		execve_nosetid(imgp);

	/*
	 * Implement image setuid/setgid installation.
	 */
	if (imgp->credential_setid) {
		/*
		 * Turn off syscall tracing for set-id programs, except for
		 * root.  Record any set-id flags first to make sure that
		 * we do not regain any tracing during a possible block.
		 */
		setsugid(p);
#ifdef KTRACE
		kiop = ktrprocexec(p);
#endif
		/*
		 * Close any file descriptors 0..2 that reference procfs,
		 * then make sure file descriptors 0..2 are in use.
		 *
		 * Both fdsetugidsafety() and fdcheckstd() may call functions
		 * taking sleepable locks, so temporarily drop our locks.
		 */
		PROC_UNLOCK(p);
		VOP_UNLOCK(imgp->vp);
		fdsetugidsafety(td);
		error = fdcheckstd(td);
		vn_lock(imgp->vp, LK_SHARED | LK_RETRY);
		if (error != 0)
			goto exec_fail_dealloc;
		PROC_LOCK(p);
#ifdef MAC
		if (will_transition) {
			mac_vnode_execve_transition(oldcred, imgp->newcred,
			    imgp->vp, interpvplabel, imgp);
		}
#endif
	} else {
		if (oldcred->cr_uid == oldcred->cr_ruid &&
		    oldcred->cr_gid == oldcred->cr_rgid)
			p->p_flag &= ~P_SUGID;
	}
	/*
	 * Set the new credentials.
	 */
	if (imgp->newcred != NULL) {
		proc_set_cred(p, imgp->newcred);
		crfree(oldcred);
		oldcred = NULL;
	}

	/*
	 * Store the vp for use in procfs.  This vnode was referenced by namei
	 * or fgetvp_exec.
	 */
	oldtextvp = p->p_textvp;
	p->p_textvp = newtextvp;

#ifdef KDTRACE_HOOKS
	/*
	 * Tell the DTrace fasttrap provider about the exec if it
	 * has declared an interest.
	 */
	if (dtrace_fasttrap_exec)
		dtrace_fasttrap_exec(p);
#endif

	/*
	 * Notify others that we exec'd, and clear the P_INEXEC flag
	 * as we're now a bona fide freshly-execed process.
	 */
	KNOTE_LOCKED(p->p_klist, NOTE_EXEC);
	p->p_flag &= ~P_INEXEC;

	/* clear "fork but no exec" flag, as we _are_ execing */
	p->p_acflag &= ~AFORK;

	/*
	 * Free any previous argument cache and replace it with
	 * the new argument cache, if any.
	 */
	oldargs = p->p_args;
	p->p_args = newargs;
	newargs = NULL;

	PROC_UNLOCK(p);

#ifdef	HWPMC_HOOKS
	/*
	 * Check if system-wide sampling is in effect or if the
	 * current process is using PMCs.  If so, do exec() time
	 * processing.  This processing needs to happen AFTER the
	 * P_INEXEC flag is cleared.
	 */
	if (PMC_SYSTEM_SAMPLING_ACTIVE() || PMC_PROC_IS_USING_PMCS(p)) {
		VOP_UNLOCK(imgp->vp);
		pe.pm_credentialschanged = credential_changing;
		pe.pm_entryaddr = imgp->entry_addr;

		PMC_CALL_HOOK_X(td, PMC_FN_PROCESS_EXEC, (void *) &pe);
		vn_lock(imgp->vp, LK_SHARED | LK_RETRY);
	}
#endif

	/* Set values passed into the program in registers. */
	(*p->p_sysent->sv_setregs)(td, imgp, stack_base);

	VOP_MMAPPED(imgp->vp);

	SDT_PROBE1(proc, , , exec__success, args->fname);

exec_fail_dealloc:
	if (error != 0) {
		p->p_osrel = orig_osrel;
		p->p_fctl0 = orig_fctl0;
		p->p_elf_brandinfo = orig_brandinfo;
	}

	if (imgp->firstpage != NULL)
		exec_unmap_first_page(imgp);

	if (imgp->vp != NULL) {
		if (args->fname)
			NDFREE(&nd, NDF_ONLY_PNBUF);
		if (imgp->opened)
			VOP_CLOSE(imgp->vp, FREAD, td->td_ucred, td);
		if (imgp->textset)
			VOP_UNSET_TEXT_CHECKED(imgp->vp);
		if (error != 0)
			vput(imgp->vp);
		else
			VOP_UNLOCK(imgp->vp);
	}

	if (imgp->object != NULL)
		vm_object_deallocate(imgp->object);

	free(imgp->freepath, M_TEMP);

	if (error == 0) {
		if (p->p_ptevents & PTRACE_EXEC) {
			PROC_LOCK(p);
			if (p->p_ptevents & PTRACE_EXEC)
				td->td_dbgflags |= TDB_EXEC;
			PROC_UNLOCK(p);
		}
	} else {
exec_fail:
		/* we're done here, clear P_INEXEC */
		PROC_LOCK(p);
		p->p_flag &= ~P_INEXEC;
		PROC_UNLOCK(p);

		SDT_PROBE1(proc, , , exec__failure, error);
	}

	if (imgp->newcred != NULL && oldcred != NULL)
		crfree(imgp->newcred);

#ifdef MAC
	mac_execve_exit(imgp);
	mac_execve_interpreter_exit(interpvplabel);
#endif
	exec_free_args(args);

	/*
	 * Handle deferred decrement of ref counts.
	 */
	if (oldtextvp != NULL)
		vrele(oldtextvp);
#ifdef KTRACE
	ktr_io_params_free(kiop);
#endif
	pargs_drop(oldargs);
	pargs_drop(newargs);
	if (oldsigacts != NULL)
		sigacts_free(oldsigacts);
	if (euip != NULL)
		uifree(euip);

	if (error && imgp->vmspace_destroyed) {
		/* sorry, no more process anymore. exit gracefully */
		exec_cleanup(td, oldvmspace);
		exit1(td, 0, SIGABRT);
		/* NOT REACHED */
	}

#ifdef KTRACE
	if (error == 0)
		ktrprocctor(p);
#endif

	/*
	 * We don't want cpu_set_syscall_retval() to overwrite any of
	 * the register values put in place by exec_setregs().
	 * Implementations of cpu_set_syscall_retval() will leave
	 * registers unmodified when returning EJUSTRETURN.
	 */
	return (error == 0 ? EJUSTRETURN : error);
}

void
exec_cleanup(struct thread *td, struct vmspace *oldvmspace)
{
	if ((td->td_pflags & TDP_EXECVMSPC) != 0) {
		KASSERT(td->td_proc->p_vmspace != oldvmspace,
		    ("oldvmspace still used"));
		vmspace_free(oldvmspace);
		td->td_pflags &= ~TDP_EXECVMSPC;
	}
}

int
exec_map_first_page(struct image_params *imgp)
{
	vm_object_t object;
	vm_page_t m;
	int error;

	if (imgp->firstpage != NULL)
		exec_unmap_first_page(imgp);

	object = imgp->vp->v_object;
	if (object == NULL)
		return (EACCES);
#if VM_NRESERVLEVEL > 0
	if ((object->flags & OBJ_COLORED) == 0) {
		VM_OBJECT_WLOCK(object);
		vm_object_color(object, 0);
		VM_OBJECT_WUNLOCK(object);
	}
#endif
	error = vm_page_grab_valid_unlocked(&m, object, 0,
	    VM_ALLOC_COUNT(VM_INITIAL_PAGEIN) |
            VM_ALLOC_NORMAL | VM_ALLOC_NOBUSY | VM_ALLOC_WIRED);

	if (error != VM_PAGER_OK)
		return (EIO);
	imgp->firstpage = sf_buf_alloc(m, 0);
	imgp->image_header = (char *)sf_buf_kva(imgp->firstpage);

	return (0);
}

void
exec_unmap_first_page(struct image_params *imgp)
{
	vm_page_t m;

	if (imgp->firstpage != NULL) {
		m = sf_buf_page(imgp->firstpage);
		sf_buf_free(imgp->firstpage);
		imgp->firstpage = NULL;
		vm_page_unwire(m, PQ_ACTIVE);
	}
}

/*
 * Destroy old address space, and allocate a new stack.
 *	The new stack is only sgrowsiz large because it is grown
 *	automatically on a page fault.
 */
int
exec_new_vmspace(struct image_params *imgp, struct sysentvec *sv)
{
	int error;
	struct proc *p = imgp->proc;
	struct vmspace *vmspace = p->p_vmspace;
	struct thread *td = curthread;
	vm_object_t obj;
	struct rlimit rlim_stack;
	vm_offset_t sv_minuser, stack_addr;
	vm_map_t map;
	vm_prot_t stack_prot;
	u_long ssiz;

	imgp->vmspace_destroyed = 1;
	imgp->sysent = sv;

	sigfastblock_clear(td);
	umtx_exec(p);
	if (p->p_sysent->sv_onexec_old != NULL)
		p->p_sysent->sv_onexec_old(td);
	itimers_exec(p);
	if (sv->sv_onexec != NULL)
		sv->sv_onexec(p, imgp);

	EVENTHANDLER_DIRECT_INVOKE(process_exec, p, imgp);

	/*
	 * Blow away entire process VM, if address space not shared,
	 * otherwise, create a new VM space so that other threads are
	 * not disrupted
	 */
	map = &vmspace->vm_map;
	if (map_at_zero)
		sv_minuser = sv->sv_minuser;
	else
		sv_minuser = MAX(sv->sv_minuser, PAGE_SIZE);
	if (refcount_load(&vmspace->vm_refcnt) == 1 &&
	    vm_map_min(map) == sv_minuser &&
	    vm_map_max(map) == sv->sv_maxuser &&
	    cpu_exec_vmspace_reuse(p, map)) {
		shmexit(vmspace);
		pmap_remove_pages(vmspace_pmap(vmspace));
		vm_map_remove(map, vm_map_min(map), vm_map_max(map));
		/*
		 * An exec terminates mlockall(MCL_FUTURE).
		 * ASLR and W^X states must be re-evaluated.
		 */
		vm_map_lock(map);
		vm_map_modflags(map, 0, MAP_WIREFUTURE | MAP_ASLR |
		    MAP_ASLR_IGNSTART | MAP_WXORX);
		vm_map_unlock(map);
	} else {
		error = vmspace_exec(p, sv_minuser, sv->sv_maxuser);
		if (error)
			return (error);
		vmspace = p->p_vmspace;
		map = &vmspace->vm_map;
	}
	map->flags |= imgp->map_flags;

	/* Map a shared page */
	obj = sv->sv_shared_page_obj;
	if (obj != NULL) {
		vm_object_reference(obj);
		error = vm_map_fixed(map, obj, 0,
		    sv->sv_shared_page_base, sv->sv_shared_page_len,
		    VM_PROT_READ | VM_PROT_EXECUTE,
		    VM_PROT_READ | VM_PROT_EXECUTE,
		    MAP_INHERIT_SHARE | MAP_ACC_NO_CHARGE);
		if (error != KERN_SUCCESS) {
			vm_object_deallocate(obj);
			return (vm_mmap_to_errno(error));
		}
	}

	/* Allocate a new stack */
	if (imgp->stack_sz != 0) {
		ssiz = trunc_page(imgp->stack_sz);
		PROC_LOCK(p);
		lim_rlimit_proc(p, RLIMIT_STACK, &rlim_stack);
		PROC_UNLOCK(p);
		if (ssiz > rlim_stack.rlim_max)
			ssiz = rlim_stack.rlim_max;
		if (ssiz > rlim_stack.rlim_cur) {
			rlim_stack.rlim_cur = ssiz;
			kern_setrlimit(curthread, RLIMIT_STACK, &rlim_stack);
		}
	} else if (sv->sv_maxssiz != NULL) {
		ssiz = *sv->sv_maxssiz;
	} else {
		ssiz = maxssiz;
	}
	imgp->eff_stack_sz = lim_cur(curthread, RLIMIT_STACK);
	if (ssiz < imgp->eff_stack_sz)
		imgp->eff_stack_sz = ssiz;
	stack_addr = sv->sv_usrstack - ssiz;
	stack_prot = obj != NULL && imgp->stack_prot != 0 ?
	    imgp->stack_prot : sv->sv_stackprot;
	error = vm_map_stack(map, stack_addr, (vm_size_t)ssiz, stack_prot,
	    VM_PROT_ALL, MAP_STACK_GROWS_DOWN);
	if (error != KERN_SUCCESS) {
		uprintf("exec_new_vmspace: mapping stack size %#jx prot %#x "
		    "failed mach error %d errno %d\n", (uintmax_t)ssiz,
		    stack_prot, error, vm_mmap_to_errno(error));
		return (vm_mmap_to_errno(error));
	}

	/*
	 * vm_ssize and vm_maxsaddr are somewhat antiquated concepts, but they
	 * are still used to enforce the stack rlimit on the process stack.
	 */
	vmspace->vm_ssize = sgrowsiz >> PAGE_SHIFT;
	vmspace->vm_maxsaddr = (char *)stack_addr;

	return (0);
}

/*
 * Copy out argument and environment strings from the old process address
 * space into the temporary string buffer.
 */
int
exec_copyin_args(struct image_args *args, const char *fname,
    enum uio_seg segflg, char **argv, char **envv)
{
	u_long arg, env;
	int error;

	bzero(args, sizeof(*args));
	if (argv == NULL)
		return (EFAULT);

	/*
	 * Allocate demand-paged memory for the file name, argument, and
	 * environment strings.
	 */
	error = exec_alloc_args(args);
	if (error != 0)
		return (error);

	/*
	 * Copy the file name.
	 */
	error = exec_args_add_fname(args, fname, segflg);
	if (error != 0)
		goto err_exit;

	/*
	 * extract arguments first
	 */
	for (;;) {
		error = fueword(argv++, &arg);
		if (error == -1) {
			error = EFAULT;
			goto err_exit;
		}
		if (arg == 0)
			break;
		error = exec_args_add_arg(args, (char *)(uintptr_t)arg,
		    UIO_USERSPACE);
		if (error != 0)
			goto err_exit;
	}

	/*
	 * extract environment strings
	 */
	if (envv) {
		for (;;) {
			error = fueword(envv++, &env);
			if (error == -1) {
				error = EFAULT;
				goto err_exit;
			}
			if (env == 0)
				break;
			error = exec_args_add_env(args,
			    (char *)(uintptr_t)env, UIO_USERSPACE);
			if (error != 0)
				goto err_exit;
		}
	}

	return (0);

err_exit:
	exec_free_args(args);
	return (error);
}

int
exec_copyin_data_fds(struct thread *td, struct image_args *args,
    const void *data, size_t datalen, const int *fds, size_t fdslen)
{
	struct filedesc *ofdp;
	const char *p;
	int *kfds;
	int error;

	memset(args, '\0', sizeof(*args));
	ofdp = td->td_proc->p_fd;
	if (datalen >= ARG_MAX || fdslen >= ofdp->fd_nfiles)
		return (E2BIG);
	error = exec_alloc_args(args);
	if (error != 0)
		return (error);

	args->begin_argv = args->buf;
	args->stringspace = ARG_MAX;

	if (datalen > 0) {
		/*
		 * Argument buffer has been provided. Copy it into the
		 * kernel as a single string and add a terminating null
		 * byte.
		 */
		error = copyin(data, args->begin_argv, datalen);
		if (error != 0)
			goto err_exit;
		args->begin_argv[datalen] = '\0';
		args->endp = args->begin_argv + datalen + 1;
		args->stringspace -= datalen + 1;

		/*
		 * Traditional argument counting. Count the number of
		 * null bytes.
		 */
		for (p = args->begin_argv; p < args->endp; ++p)
			if (*p == '\0')
				++args->argc;
	} else {
		/* No argument buffer provided. */
		args->endp = args->begin_argv;
	}

	/* Create new file descriptor table. */
	kfds = malloc(fdslen * sizeof(int), M_TEMP, M_WAITOK);
	error = copyin(fds, kfds, fdslen * sizeof(int));
	if (error != 0) {
		free(kfds, M_TEMP);
		goto err_exit;
	}
	error = fdcopy_remapped(ofdp, kfds, fdslen, &args->fdp);
	free(kfds, M_TEMP);
	if (error != 0)
		goto err_exit;

	return (0);
err_exit:
	exec_free_args(args);
	return (error);
}

struct exec_args_kva {
	vm_offset_t addr;
	u_int gen;
	SLIST_ENTRY(exec_args_kva) next;
};

DPCPU_DEFINE_STATIC(struct exec_args_kva *, exec_args_kva);

static SLIST_HEAD(, exec_args_kva) exec_args_kva_freelist;
static struct mtx exec_args_kva_mtx;
static u_int exec_args_gen;

static void
exec_prealloc_args_kva(void *arg __unused)
{
	struct exec_args_kva *argkva;
	u_int i;

	SLIST_INIT(&exec_args_kva_freelist);
	mtx_init(&exec_args_kva_mtx, "exec args kva", NULL, MTX_DEF);
	for (i = 0; i < exec_map_entries; i++) {
		argkva = malloc(sizeof(*argkva), M_PARGS, M_WAITOK);
		argkva->addr = kmap_alloc_wait(exec_map, exec_map_entry_size);
		argkva->gen = exec_args_gen;
		SLIST_INSERT_HEAD(&exec_args_kva_freelist, argkva, next);
	}
}
SYSINIT(exec_args_kva, SI_SUB_EXEC, SI_ORDER_ANY, exec_prealloc_args_kva, NULL);

static vm_offset_t
exec_alloc_args_kva(void **cookie)
{
	struct exec_args_kva *argkva;

	argkva = (void *)atomic_readandclear_ptr(
	    (uintptr_t *)DPCPU_PTR(exec_args_kva));
	if (argkva == NULL) {
		mtx_lock(&exec_args_kva_mtx);
		while ((argkva = SLIST_FIRST(&exec_args_kva_freelist)) == NULL)
			(void)mtx_sleep(&exec_args_kva_freelist,
			    &exec_args_kva_mtx, 0, "execkva", 0);
		SLIST_REMOVE_HEAD(&exec_args_kva_freelist, next);
		mtx_unlock(&exec_args_kva_mtx);
	}
	kasan_mark((void *)argkva->addr, exec_map_entry_size,
	    exec_map_entry_size, 0);
	*(struct exec_args_kva **)cookie = argkva;
	return (argkva->addr);
}

static void
exec_release_args_kva(struct exec_args_kva *argkva, u_int gen)
{
	vm_offset_t base;

	base = argkva->addr;
	kasan_mark((void *)argkva->addr, 0, exec_map_entry_size,
	    KASAN_EXEC_ARGS_FREED);
	if (argkva->gen != gen) {
		(void)vm_map_madvise(exec_map, base, base + exec_map_entry_size,
		    MADV_FREE);
		argkva->gen = gen;
	}
	if (!atomic_cmpset_ptr((uintptr_t *)DPCPU_PTR(exec_args_kva),
	    (uintptr_t)NULL, (uintptr_t)argkva)) {
		mtx_lock(&exec_args_kva_mtx);
		SLIST_INSERT_HEAD(&exec_args_kva_freelist, argkva, next);
		wakeup_one(&exec_args_kva_freelist);
		mtx_unlock(&exec_args_kva_mtx);
	}
}

static void
exec_free_args_kva(void *cookie)
{

	exec_release_args_kva(cookie, exec_args_gen);
}

static void
exec_args_kva_lowmem(void *arg __unused)
{
	SLIST_HEAD(, exec_args_kva) head;
	struct exec_args_kva *argkva;
	u_int gen;
	int i;

	gen = atomic_fetchadd_int(&exec_args_gen, 1) + 1;

	/*
	 * Force an madvise of each KVA range. Any currently allocated ranges
	 * will have MADV_FREE applied once they are freed.
	 */
	SLIST_INIT(&head);
	mtx_lock(&exec_args_kva_mtx);
	SLIST_SWAP(&head, &exec_args_kva_freelist, exec_args_kva);
	mtx_unlock(&exec_args_kva_mtx);
	while ((argkva = SLIST_FIRST(&head)) != NULL) {
		SLIST_REMOVE_HEAD(&head, next);
		exec_release_args_kva(argkva, gen);
	}

	CPU_FOREACH(i) {
		argkva = (void *)atomic_readandclear_ptr(
		    (uintptr_t *)DPCPU_ID_PTR(i, exec_args_kva));
		if (argkva != NULL)
			exec_release_args_kva(argkva, gen);
	}
}
EVENTHANDLER_DEFINE(vm_lowmem, exec_args_kva_lowmem, NULL,
    EVENTHANDLER_PRI_ANY);

/*
 * Allocate temporary demand-paged, zero-filled memory for the file name,
 * argument, and environment strings.
 */
int
exec_alloc_args(struct image_args *args)
{

	args->buf = (char *)exec_alloc_args_kva(&args->bufkva);
	return (0);
}

void
exec_free_args(struct image_args *args)
{

	if (args->buf != NULL) {
		exec_free_args_kva(args->bufkva);
		args->buf = NULL;
	}
	if (args->fname_buf != NULL) {
		free(args->fname_buf, M_TEMP);
		args->fname_buf = NULL;
	}
	if (args->fdp != NULL)
		fdescfree_remapped(args->fdp);
}

/*
 * A set to functions to fill struct image args.
 *
 * NOTE: exec_args_add_fname() must be called (possibly with a NULL
 * fname) before the other functions.  All exec_args_add_arg() calls must
 * be made before any exec_args_add_env() calls.  exec_args_adjust_args()
 * may be called any time after exec_args_add_fname().
 *
 * exec_args_add_fname() - install path to be executed
 * exec_args_add_arg() - append an argument string
 * exec_args_add_env() - append an env string
 * exec_args_adjust_args() - adjust location of the argument list to
 *                           allow new arguments to be prepended
 */
int
exec_args_add_fname(struct image_args *args, const char *fname,
    enum uio_seg segflg)
{
	int error;
	size_t length;

	KASSERT(args->fname == NULL, ("fname already appended"));
	KASSERT(args->endp == NULL, ("already appending to args"));

	if (fname != NULL) {
		args->fname = args->buf;
		error = segflg == UIO_SYSSPACE ?
		    copystr(fname, args->fname, PATH_MAX, &length) :
		    copyinstr(fname, args->fname, PATH_MAX, &length);
		if (error != 0)
			return (error == ENAMETOOLONG ? E2BIG : error);
	} else
		length = 0;

	/* Set up for _arg_*()/_env_*() */
	args->endp = args->buf + length;
	/* begin_argv must be set and kept updated */
	args->begin_argv = args->endp;
	KASSERT(exec_map_entry_size - length >= ARG_MAX,
	    ("too little space remaining for arguments %zu < %zu",
	    exec_map_entry_size - length, (size_t)ARG_MAX));
	args->stringspace = ARG_MAX;

	return (0);
}

static int
exec_args_add_str(struct image_args *args, const char *str,
    enum uio_seg segflg, int *countp)
{
	int error;
	size_t length;

	KASSERT(args->endp != NULL, ("endp not initialized"));
	KASSERT(args->begin_argv != NULL, ("begin_argp not initialized"));

	error = (segflg == UIO_SYSSPACE) ?
	    copystr(str, args->endp, args->stringspace, &length) :
	    copyinstr(str, args->endp, args->stringspace, &length);
	if (error != 0)
		return (error == ENAMETOOLONG ? E2BIG : error);
	args->stringspace -= length;
	args->endp += length;
	(*countp)++;

	return (0);
}

int
exec_args_add_arg(struct image_args *args, const char *argp,
    enum uio_seg segflg)
{

	KASSERT(args->envc == 0, ("appending args after env"));

	return (exec_args_add_str(args, argp, segflg, &args->argc));
}

int
exec_args_add_env(struct image_args *args, const char *envp,
    enum uio_seg segflg)
{

	if (args->envc == 0)
		args->begin_envv = args->endp;

	return (exec_args_add_str(args, envp, segflg, &args->envc));
}

int
exec_args_adjust_args(struct image_args *args, size_t consume, ssize_t extend)
{
	ssize_t offset;

	KASSERT(args->endp != NULL, ("endp not initialized"));
	KASSERT(args->begin_argv != NULL, ("begin_argp not initialized"));

	offset = extend - consume;
	if (args->stringspace < offset)
		return (E2BIG);
	memmove(args->begin_argv + extend, args->begin_argv + consume,
	    args->endp - args->begin_argv + consume);
	if (args->envc > 0)
		args->begin_envv += offset;
	args->endp += offset;
	args->stringspace -= offset;
	return (0);
}

char *
exec_args_get_begin_envv(struct image_args *args)
{

	KASSERT(args->endp != NULL, ("endp not initialized"));

	if (args->envc > 0)
		return (args->begin_envv);
	return (args->endp);
}

void
exec_stackgap(struct image_params *imgp, uintptr_t *dp)
{
	if (imgp->sysent->sv_stackgap == NULL ||
	    (imgp->proc->p_fctl0 & (NT_FREEBSD_FCTL_ASLR_DISABLE |
	    NT_FREEBSD_FCTL_ASG_DISABLE)) != 0 ||
	    (imgp->map_flags & MAP_ASLR) == 0)
		return;
	imgp->sysent->sv_stackgap(imgp, dp);
}

/*
 * Copy strings out to the new process address space, constructing new arg
 * and env vector tables. Return a pointer to the base so that it can be used
 * as the initial stack pointer.
 */
int
exec_copyout_strings(struct image_params *imgp, uintptr_t *stack_base)
{
	int argc, envc;
	char **vectp;
	char *stringp;
	uintptr_t destp, ustringp;
	struct ps_strings *arginfo;
	struct proc *p;
	size_t execpath_len;
	int error, szsigcode, szps;
	char canary[sizeof(long) * 8];

	szps = sizeof(pagesizes[0]) * MAXPAGESIZES;
	/*
	 * Calculate string base and vector table pointers.
	 * Also deal with signal trampoline code for this exec type.
	 */
	if (imgp->execpath != NULL && imgp->auxargs != NULL)
		execpath_len = strlen(imgp->execpath) + 1;
	else
		execpath_len = 0;
	p = imgp->proc;
	szsigcode = 0;
	arginfo = (struct ps_strings *)p->p_sysent->sv_psstrings;
	imgp->ps_strings = arginfo;
	if (p->p_sysent->sv_sigcode_base == 0) {
		if (p->p_sysent->sv_szsigcode != NULL)
			szsigcode = *(p->p_sysent->sv_szsigcode);
	}
	destp =	(uintptr_t)arginfo;

	/*
	 * install sigcode
	 */
	if (szsigcode != 0) {
		destp -= szsigcode;
		destp = rounddown2(destp, sizeof(void *));
		error = copyout(p->p_sysent->sv_sigcode, (void *)destp,
		    szsigcode);
		if (error != 0)
			return (error);
	}

	/*
	 * Copy the image path for the rtld.
	 */
	if (execpath_len != 0) {
		destp -= execpath_len;
		destp = rounddown2(destp, sizeof(void *));
		imgp->execpathp = (void *)destp;
		error = copyout(imgp->execpath, imgp->execpathp, execpath_len);
		if (error != 0)
			return (error);
	}

	/*
	 * Prepare the canary for SSP.
	 */
	arc4rand(canary, sizeof(canary), 0);
	destp -= sizeof(canary);
	imgp->canary = (void *)destp;
	error = copyout(canary, imgp->canary, sizeof(canary));
	if (error != 0)
		return (error);
	imgp->canarylen = sizeof(canary);

	/*
	 * Prepare the pagesizes array.
	 */
	destp -= szps;
	destp = rounddown2(destp, sizeof(void *));
	imgp->pagesizes = (void *)destp;
	error = copyout(pagesizes, imgp->pagesizes, szps);
	if (error != 0)
		return (error);
	imgp->pagesizeslen = szps;

	/*
	 * Allocate room for the argument and environment strings.
	 */
	destp -= ARG_MAX - imgp->args->stringspace;
	destp = rounddown2(destp, sizeof(void *));
	ustringp = destp;

	exec_stackgap(imgp, &destp);

	if (imgp->auxargs) {
		/*
		 * Allocate room on the stack for the ELF auxargs
		 * array.  It has up to AT_COUNT entries.
		 */
		destp -= AT_COUNT * sizeof(Elf_Auxinfo);
		destp = rounddown2(destp, sizeof(void *));
	}

	vectp = (char **)destp;

	/*
	 * Allocate room for the argv[] and env vectors including the
	 * terminating NULL pointers.
	 */
	vectp -= imgp->args->argc + 1 + imgp->args->envc + 1;

	/*
	 * vectp also becomes our initial stack base
	 */
	*stack_base = (uintptr_t)vectp;

	stringp = imgp->args->begin_argv;
	argc = imgp->args->argc;
	envc = imgp->args->envc;

	/*
	 * Copy out strings - arguments and environment.
	 */
	error = copyout(stringp, (void *)ustringp,
	    ARG_MAX - imgp->args->stringspace);
	if (error != 0)
		return (error);

	/*
	 * Fill in "ps_strings" struct for ps, w, etc.
	 */
	imgp->argv = vectp;
	if (suword(&arginfo->ps_argvstr, (long)(intptr_t)vectp) != 0 ||
	    suword32(&arginfo->ps_nargvstr, argc) != 0)
		return (EFAULT);

	/*
	 * Fill in argument portion of vector table.
	 */
	for (; argc > 0; --argc) {
		if (suword(vectp++, ustringp) != 0)
			return (EFAULT);
		while (*stringp++ != 0)
			ustringp++;
		ustringp++;
	}

	/* a null vector table pointer separates the argp's from the envp's */
	if (suword(vectp++, 0) != 0)
		return (EFAULT);

	imgp->envv = vectp;
	if (suword(&arginfo->ps_envstr, (long)(intptr_t)vectp) != 0 ||
	    suword32(&arginfo->ps_nenvstr, envc) != 0)
		return (EFAULT);

	/*
	 * Fill in environment portion of vector table.
	 */
	for (; envc > 0; --envc) {
		if (suword(vectp++, ustringp) != 0)
			return (EFAULT);
		while (*stringp++ != 0)
			ustringp++;
		ustringp++;
	}

	/* end of vector table is a null pointer */
	if (suword(vectp, 0) != 0)
		return (EFAULT);

	if (imgp->auxargs) {
		vectp++;
		error = imgp->sysent->sv_copyout_auxargs(imgp,
		    (uintptr_t)vectp);
		if (error != 0)
			return (error);
	}

	return (0);
}

/*
 * Check permissions of file to execute.
 *	Called with imgp->vp locked.
 *	Return 0 for success or error code on failure.
 */
int
exec_check_permissions(struct image_params *imgp)
{
	struct vnode *vp = imgp->vp;
	struct vattr *attr = imgp->attr;
	struct thread *td;
	int error;

	td = curthread;

	/* Get file attributes */
	error = VOP_GETATTR(vp, attr, td->td_ucred);
	if (error)
		return (error);

#ifdef MAC
	error = mac_vnode_check_exec(td->td_ucred, imgp->vp, imgp);
	if (error)
		return (error);
#endif

	/*
	 * 1) Check if file execution is disabled for the filesystem that
	 *    this file resides on.
	 * 2) Ensure that at least one execute bit is on. Otherwise, a
	 *    privileged user will always succeed, and we don't want this
	 *    to happen unless the file really is executable.
	 * 3) Ensure that the file is a regular file.
	 */
	if ((vp->v_mount->mnt_flag & MNT_NOEXEC) ||
	    (attr->va_mode & (S_IXUSR | S_IXGRP | S_IXOTH)) == 0 ||
	    (attr->va_type != VREG))
		return (EACCES);

	/*
	 * Zero length files can't be exec'd
	 */
	if (attr->va_size == 0)
		return (ENOEXEC);

	/*
	 *  Check for execute permission to file based on current credentials.
	 */
	error = VOP_ACCESS(vp, VEXEC, td->td_ucred, td);
	if (error)
		return (error);

	/*
	 * Check number of open-for-writes on the file and deny execution
	 * if there are any.
	 *
	 * Add a text reference now so no one can write to the
	 * executable while we're activating it.
	 *
	 * Remember if this was set before and unset it in case this is not
	 * actually an executable image.
	 */
	error = VOP_SET_TEXT(vp);
	if (error != 0)
		return (error);
	imgp->textset = true;

	/*
	 * Call filesystem specific open routine (which does nothing in the
	 * general case).
	 */
	error = VOP_OPEN(vp, FREAD, td->td_ucred, td, NULL);
	if (error == 0)
		imgp->opened = 1;
	return (error);
}

/*
 * Exec handler registration
 */
int
exec_register(const struct execsw *execsw_arg)
{
	const struct execsw **es, **xs, **newexecsw;
	u_int count = 2;	/* New slot and trailing NULL */

	if (execsw)
		for (es = execsw; *es; es++)
			count++;
	newexecsw = malloc(count * sizeof(*es), M_TEMP, M_WAITOK);
	xs = newexecsw;
	if (execsw)
		for (es = execsw; *es; es++)
			*xs++ = *es;
	*xs++ = execsw_arg;
	*xs = NULL;
	if (execsw)
		free(execsw, M_TEMP);
	execsw = newexecsw;
	return (0);
}

int
exec_unregister(const struct execsw *execsw_arg)
{
	const struct execsw **es, **xs, **newexecsw;
	int count = 1;

	if (execsw == NULL)
		panic("unregister with no handlers left?\n");

	for (es = execsw; *es; es++) {
		if (*es == execsw_arg)
			break;
	}
	if (*es == NULL)
		return (ENOENT);
	for (es = execsw; *es; es++)
		if (*es != execsw_arg)
			count++;
	newexecsw = malloc(count * sizeof(*es), M_TEMP, M_WAITOK);
	xs = newexecsw;
	for (es = execsw; *es; es++)
		if (*es != execsw_arg)
			*xs++ = *es;
	*xs = NULL;
	if (execsw)
		free(execsw, M_TEMP);
	execsw = newexecsw;
	return (0);
}

/*
 * Write out a core segment to the compression stream.
 */
static int
compress_chunk(struct coredump_params *cp, char *base, char *buf, u_int len)
{
	u_int chunk_len;
	int error;

	while (len > 0) {
		chunk_len = MIN(len, CORE_BUF_SIZE);

		/*
		 * We can get EFAULT error here.
		 * In that case zero out the current chunk of the segment.
		 */
		error = copyin(base, buf, chunk_len);
		if (error != 0)
			bzero(buf, chunk_len);
		error = compressor_write(cp->comp, buf, chunk_len);
		if (error != 0)
			break;
		base += chunk_len;
		len -= chunk_len;
	}
	return (error);
}

int
core_write(struct coredump_params *cp, const void *base, size_t len,
    off_t offset, enum uio_seg seg, size_t *resid)
{

	return (vn_rdwr_inchunks(UIO_WRITE, cp->vp, __DECONST(void *, base),
	    len, offset, seg, IO_UNIT | IO_DIRECT | IO_RANGELOCKED,
	    cp->active_cred, cp->file_cred, resid, cp->td));
}

int
core_output(char *base, size_t len, off_t offset, struct coredump_params *cp,
    void *tmpbuf)
{
	vm_map_t map;
	struct mount *mp;
	size_t resid, runlen;
	int error;
	bool success;

	KASSERT((uintptr_t)base % PAGE_SIZE == 0,
	    ("%s: user address %p is not page-aligned", __func__, base));

	if (cp->comp != NULL)
		return (compress_chunk(cp, base, tmpbuf, len));

	map = &cp->td->td_proc->p_vmspace->vm_map;
	for (; len > 0; base += runlen, offset += runlen, len -= runlen) {
		/*
		 * Attempt to page in all virtual pages in the range.  If a
		 * virtual page is not backed by the pager, it is represented as
		 * a hole in the file.  This can occur with zero-filled
		 * anonymous memory or truncated files, for example.
		 */
		for (runlen = 0; runlen < len; runlen += PAGE_SIZE) {
			error = vm_fault(map, (uintptr_t)base + runlen,
			    VM_PROT_READ, VM_FAULT_NOFILL, NULL);
			if (runlen == 0)
				success = error == KERN_SUCCESS;
			else if ((error == KERN_SUCCESS) != success)
				break;
		}

		if (success) {
			error = core_write(cp, base, runlen, offset,
			    UIO_USERSPACE, &resid);
			if (error != 0) {
				if (error != EFAULT)
					break;

				/*
				 * EFAULT may be returned if the user mapping
				 * could not be accessed, e.g., because a mapped
				 * file has been truncated.  Skip the page if no
				 * progress was made, to protect against a
				 * hypothetical scenario where vm_fault() was
				 * successful but core_write() returns EFAULT
				 * anyway.
				 */
				runlen -= resid;
				if (runlen == 0) {
					success = false;
					runlen = PAGE_SIZE;
				}
			}
		}
		if (!success) {
			error = vn_start_write(cp->vp, &mp, V_WAIT);
			if (error != 0)
				break;
			vn_lock(cp->vp, LK_EXCLUSIVE | LK_RETRY);
			error = vn_truncate_locked(cp->vp, offset + runlen,
			    false, cp->td->td_ucred);
			VOP_UNLOCK(cp->vp);
			vn_finished_write(mp);
			if (error != 0)
				break;
		}
	}
	return (error);
}

/*
 * Drain into a core file.
 */
int
sbuf_drain_core_output(void *arg, const char *data, int len)
{
	struct coredump_params *cp;
	struct proc *p;
	int error, locked;

	cp = arg;
	p = cp->td->td_proc;

	/*
	 * Some kern_proc out routines that print to this sbuf may
	 * call us with the process lock held. Draining with the
	 * non-sleepable lock held is unsafe. The lock is needed for
	 * those routines when dumping a live process. In our case we
	 * can safely release the lock before draining and acquire
	 * again after.
	 */
	locked = PROC_LOCKED(p);
	if (locked)
		PROC_UNLOCK(p);
	if (cp->comp != NULL)
		error = compressor_write(cp->comp, __DECONST(char *, data), len);
	else
		error = core_write(cp, __DECONST(void *, data), len, cp->offset,
		    UIO_SYSSPACE, NULL);
	if (locked)
		PROC_LOCK(p);
	if (error != 0)
		return (-error);
	cp->offset += len;
	return (len);
}