summaryrefslogtreecommitdiffstats
path: root/sys/amd64/linux32/linux32_sysvec.c
blob: c06ce1198f86a1e6c9c62ca7768f4b9cc094a109 (plain)
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
/*-
 * Copyright (c) 2004 Tim J. Robbins
 * Copyright (c) 2003 Peter Wemm
 * Copyright (c) 2002 Doug Rabson
 * Copyright (c) 1998-1999 Andrew Gallatin
 * Copyright (c) 1994-1996 Søren Schmidt
 * 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
 *    in this position and unchanged.
 * 2. Redistributions in binary form must reproduce the above copyright
 *    notice, this list of conditions and the following disclaimer in the
 *    documentation and/or other materials provided with the distribution.
 * 3. The name of the author may not be used to endorse or promote products
 *    derived from this software without specific prior written permission
 *
 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``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 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_compat.h"

#ifndef COMPAT_FREEBSD32
#error "Unable to compile Linux-emulator due to missing COMPAT_FREEBSD32 option!"
#endif

#define	__ELF_WORD_SIZE	32

#include <sys/param.h>
#include <sys/systm.h>
#include <sys/exec.h>
#include <sys/fcntl.h>
#include <sys/imgact.h>
#include <sys/imgact_elf.h>
#include <sys/kernel.h>
#include <sys/lock.h>
#include <sys/malloc.h>
#include <sys/module.h>
#include <sys/mutex.h>
#include <sys/proc.h>
#include <sys/resourcevar.h>
#include <sys/signalvar.h>
#include <sys/sysctl.h>
#include <sys/syscallsubr.h>
#include <sys/sysent.h>
#include <sys/sysproto.h>
#include <sys/vnode.h>
#include <sys/eventhandler.h>

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

#include <machine/cpu.h>
#include <machine/md_var.h>
#include <machine/pcb.h>
#include <machine/specialreg.h>

#include <amd64/linux32/linux.h>
#include <amd64/linux32/linux32_proto.h>
#include <compat/linux/linux_emul.h>
#include <compat/linux/linux_futex.h>
#include <compat/linux/linux_ioctl.h>
#include <compat/linux/linux_mib.h>
#include <compat/linux/linux_misc.h>
#include <compat/linux/linux_signal.h>
#include <compat/linux/linux_util.h>

MODULE_VERSION(linux, 1);

MALLOC_DEFINE(M_LINUX, "linux", "Linux mode structures");

#define	AUXARGS_ENTRY_32(pos, id, val)	\
	do {				\
		suword32(pos++, id);	\
		suword32(pos++, val);	\
	} while (0)

#if BYTE_ORDER == LITTLE_ENDIAN
#define SHELLMAGIC      0x2123 /* #! */
#else
#define SHELLMAGIC      0x2321
#endif

/*
 * Allow the sendsig functions to use the ldebug() facility
 * even though they are not syscalls themselves. Map them
 * to syscall 0. This is slightly less bogus than using
 * ldebug(sigreturn).
 */
#define	LINUX_SYS_linux_rt_sendsig	0
#define	LINUX_SYS_linux_sendsig		0

const char *linux_platform = "i686";
static int linux_szplatform;
extern char linux_sigcode[];
extern int linux_szsigcode;

extern struct sysent linux_sysent[LINUX_SYS_MAXSYSCALL];

SET_DECLARE(linux_ioctl_handler_set, struct linux_ioctl_handler);
SET_DECLARE(linux_device_handler_set, struct linux_device_handler);

static int	elf_linux_fixup(register_t **stack_base,
		    struct image_params *iparams);
static register_t *linux_copyout_strings(struct image_params *imgp);
static void     linux_sendsig(sig_t catcher, ksiginfo_t *ksi, sigset_t *mask);
static void	exec_linux_setregs(struct thread *td, 
				   struct image_params *imgp, u_long stack);
static void	linux32_fixlimit(struct rlimit *rl, int which);
static boolean_t linux32_trans_osrel(const Elf_Note *note, int32_t *osrel);

static eventhandler_tag linux_exit_tag;
static eventhandler_tag linux_exec_tag;

/*
 * Linux syscalls return negative errno's, we do positive and map them
 * Reference:
 *   FreeBSD: src/sys/sys/errno.h
 *   Linux:   linux-2.6.17.8/include/asm-generic/errno-base.h
 *            linux-2.6.17.8/include/asm-generic/errno.h
 */
static int bsd_to_linux_errno[ELAST + 1] = {
	-0,  -1,  -2,  -3,  -4,  -5,  -6,  -7,  -8,  -9,
	-10, -35, -12, -13, -14, -15, -16, -17, -18, -19,
	-20, -21, -22, -23, -24, -25, -26, -27, -28, -29,
	-30, -31, -32, -33, -34, -11,-115,-114, -88, -89,
	-90, -91, -92, -93, -94, -95, -96, -97, -98, -99,
	-100,-101,-102,-103,-104,-105,-106,-107,-108,-109,
	-110,-111, -40, -36,-112,-113, -39, -11, -87,-122,
	-116, -66,  -6,  -6,  -6,  -6,  -6, -37, -38,  -9,
	  -6,  -6, -43, -42, -75,-125, -84, -95, -16, -74,
	 -72, -67, -71
};

int bsd_to_linux_signal[LINUX_SIGTBLSZ] = {
	LINUX_SIGHUP, LINUX_SIGINT, LINUX_SIGQUIT, LINUX_SIGILL,
	LINUX_SIGTRAP, LINUX_SIGABRT, 0, LINUX_SIGFPE,
	LINUX_SIGKILL, LINUX_SIGBUS, LINUX_SIGSEGV, LINUX_SIGSYS,
	LINUX_SIGPIPE, LINUX_SIGALRM, LINUX_SIGTERM, LINUX_SIGURG,
	LINUX_SIGSTOP, LINUX_SIGTSTP, LINUX_SIGCONT, LINUX_SIGCHLD,
	LINUX_SIGTTIN, LINUX_SIGTTOU, LINUX_SIGIO, LINUX_SIGXCPU,
	LINUX_SIGXFSZ, LINUX_SIGVTALRM, LINUX_SIGPROF, LINUX_SIGWINCH,
	0, LINUX_SIGUSR1, LINUX_SIGUSR2
};

int linux_to_bsd_signal[LINUX_SIGTBLSZ] = {
	SIGHUP, SIGINT, SIGQUIT, SIGILL,
	SIGTRAP, SIGABRT, SIGBUS, SIGFPE,
	SIGKILL, SIGUSR1, SIGSEGV, SIGUSR2,
	SIGPIPE, SIGALRM, SIGTERM, SIGBUS,
	SIGCHLD, SIGCONT, SIGSTOP, SIGTSTP,
	SIGTTIN, SIGTTOU, SIGURG, SIGXCPU,
	SIGXFSZ, SIGVTALRM, SIGPROF, SIGWINCH,
	SIGIO, SIGURG, SIGSYS
};

#define LINUX_T_UNKNOWN  255
static int _bsd_to_linux_trapcode[] = {
	LINUX_T_UNKNOWN,	/* 0 */
	6,			/* 1  T_PRIVINFLT */
	LINUX_T_UNKNOWN,	/* 2 */
	3,			/* 3  T_BPTFLT */
	LINUX_T_UNKNOWN,	/* 4 */
	LINUX_T_UNKNOWN,	/* 5 */
	16,			/* 6  T_ARITHTRAP */
	254,			/* 7  T_ASTFLT */
	LINUX_T_UNKNOWN,	/* 8 */
	13,			/* 9  T_PROTFLT */
	1,			/* 10 T_TRCTRAP */
	LINUX_T_UNKNOWN,	/* 11 */
	14,			/* 12 T_PAGEFLT */
	LINUX_T_UNKNOWN,	/* 13 */
	17,			/* 14 T_ALIGNFLT */
	LINUX_T_UNKNOWN,	/* 15 */
	LINUX_T_UNKNOWN,	/* 16 */
	LINUX_T_UNKNOWN,	/* 17 */
	0,			/* 18 T_DIVIDE */
	2,			/* 19 T_NMI */
	4,			/* 20 T_OFLOW */
	5,			/* 21 T_BOUND */
	7,			/* 22 T_DNA */
	8,			/* 23 T_DOUBLEFLT */
	9,			/* 24 T_FPOPFLT */
	10,			/* 25 T_TSSFLT */
	11,			/* 26 T_SEGNPFLT */
	12,			/* 27 T_STKFLT */
	18,			/* 28 T_MCHK */
	19,			/* 29 T_XMMFLT */
	15			/* 30 T_RESERVED */
};
#define bsd_to_linux_trapcode(code) \
    ((code)<sizeof(_bsd_to_linux_trapcode)/sizeof(*_bsd_to_linux_trapcode)? \
     _bsd_to_linux_trapcode[(code)]: \
     LINUX_T_UNKNOWN)

struct linux32_ps_strings {
	u_int32_t ps_argvstr;	/* first of 0 or more argument strings */
	u_int ps_nargvstr;	/* the number of argument strings */
	u_int32_t ps_envstr;	/* first of 0 or more environment strings */
	u_int ps_nenvstr;	/* the number of environment strings */
};

/*
 * If FreeBSD & Linux have a difference of opinion about what a trap
 * means, deal with it here.
 *
 * MPSAFE
 */
static int
translate_traps(int signal, int trap_code)
{
	if (signal != SIGBUS)
		return signal;
	switch (trap_code) {
	case T_PROTFLT:
	case T_TSSFLT:
	case T_DOUBLEFLT:
	case T_PAGEFLT:
		return SIGSEGV;
	default:
		return signal;
	}
}

static int
elf_linux_fixup(register_t **stack_base, struct image_params *imgp)
{
	Elf32_Auxargs *args;
	Elf32_Addr *base;
	Elf32_Addr *pos, *uplatform;
	struct linux32_ps_strings *arginfo;

	arginfo = (struct linux32_ps_strings *)LINUX32_PS_STRINGS;
	uplatform = (Elf32_Addr *)((caddr_t)arginfo - linux_szplatform);

	KASSERT(curthread->td_proc == imgp->proc,
	    ("unsafe elf_linux_fixup(), should be curproc"));
	base = (Elf32_Addr *)*stack_base;
	args = (Elf32_Auxargs *)imgp->auxargs;
	pos = base + (imgp->args->argc + imgp->args->envc + 2);

	AUXARGS_ENTRY_32(pos, LINUX_AT_HWCAP, cpu_feature);

	/*
	 * Do not export AT_CLKTCK when emulating Linux kernel prior to 2.4.0,
	 * as it has appeared in the 2.4.0-rc7 first time.
	 * Being exported, AT_CLKTCK is returned by sysconf(_SC_CLK_TCK),
	 * glibc falls back to the hard-coded CLK_TCK value when aux entry
	 * is not present.
	 * Also see linux_times() implementation.
	 */
	if (linux_kernver(curthread) >= LINUX_KERNVER_2004000)
		AUXARGS_ENTRY_32(pos, LINUX_AT_CLKTCK, stclohz);
	AUXARGS_ENTRY_32(pos, AT_PHDR, args->phdr);
	AUXARGS_ENTRY_32(pos, AT_PHENT, args->phent);
	AUXARGS_ENTRY_32(pos, AT_PHNUM, args->phnum);
	AUXARGS_ENTRY_32(pos, AT_PAGESZ, args->pagesz);
	AUXARGS_ENTRY_32(pos, AT_FLAGS, args->flags);
	AUXARGS_ENTRY_32(pos, AT_ENTRY, args->entry);
	AUXARGS_ENTRY_32(pos, AT_BASE, args->base);
	AUXARGS_ENTRY_32(pos, LINUX_AT_SECURE, 0);
	AUXARGS_ENTRY_32(pos, AT_UID, imgp->proc->p_ucred->cr_ruid);
	AUXARGS_ENTRY_32(pos, AT_EUID, imgp->proc->p_ucred->cr_svuid);
	AUXARGS_ENTRY_32(pos, AT_GID, imgp->proc->p_ucred->cr_rgid);
	AUXARGS_ENTRY_32(pos, AT_EGID, imgp->proc->p_ucred->cr_svgid);
	AUXARGS_ENTRY_32(pos, LINUX_AT_PLATFORM, PTROUT(uplatform));
	if (args->execfd != -1)
		AUXARGS_ENTRY_32(pos, AT_EXECFD, args->execfd);
	AUXARGS_ENTRY_32(pos, AT_NULL, 0);

	free(imgp->auxargs, M_TEMP);
	imgp->auxargs = NULL;

	base--;
	suword32(base, (uint32_t)imgp->args->argc);
	*stack_base = (register_t *)base;
	return 0;
}

extern unsigned long linux_sznonrtsigcode;

static void
linux_rt_sendsig(sig_t catcher, ksiginfo_t *ksi, sigset_t *mask)
{
	struct thread *td = curthread;
	struct proc *p = td->td_proc;
	struct sigacts *psp;
	struct trapframe *regs;
	struct l_rt_sigframe *fp, frame;
	int oonstack;
	int sig;
	int code;
	
	sig = ksi->ksi_signo;
	code = ksi->ksi_code;
	PROC_LOCK_ASSERT(p, MA_OWNED);
	psp = p->p_sigacts;
	mtx_assert(&psp->ps_mtx, MA_OWNED);
	regs = td->td_frame;
	oonstack = sigonstack(regs->tf_rsp);

#ifdef DEBUG
	if (ldebug(rt_sendsig))
		printf(ARGS(rt_sendsig, "%p, %d, %p, %u"),
		    catcher, sig, (void*)mask, code);
#endif
	/*
	 * Allocate space for the signal handler context.
	 */
	if ((td->td_pflags & TDP_ALTSTACK) && !oonstack &&
	    SIGISMEMBER(psp->ps_sigonstack, sig)) {
		fp = (struct l_rt_sigframe *)(td->td_sigstk.ss_sp +
		    td->td_sigstk.ss_size - sizeof(struct l_rt_sigframe));
	} else
		fp = (struct l_rt_sigframe *)regs->tf_rsp - 1;
	mtx_unlock(&psp->ps_mtx);

	/*
	 * Build the argument list for the signal handler.
	 */
	if (p->p_sysent->sv_sigtbl)
		if (sig <= p->p_sysent->sv_sigsize)
			sig = p->p_sysent->sv_sigtbl[_SIG_IDX(sig)];

	bzero(&frame, sizeof(frame));

	frame.sf_handler = PTROUT(catcher);
	frame.sf_sig = sig;
	frame.sf_siginfo = PTROUT(&fp->sf_si);
	frame.sf_ucontext = PTROUT(&fp->sf_sc);

	/* Fill in POSIX parts */
	ksiginfo_to_lsiginfo(ksi, &frame.sf_si, sig);

	/*
	 * Build the signal context to be used by sigreturn.
	 */
	frame.sf_sc.uc_flags = 0;		/* XXX ??? */
	frame.sf_sc.uc_link = 0;		/* XXX ??? */

	frame.sf_sc.uc_stack.ss_sp = PTROUT(td->td_sigstk.ss_sp);
	frame.sf_sc.uc_stack.ss_size = td->td_sigstk.ss_size;
	frame.sf_sc.uc_stack.ss_flags = (td->td_pflags & TDP_ALTSTACK)
	    ? ((oonstack) ? LINUX_SS_ONSTACK : 0) : LINUX_SS_DISABLE;
	PROC_UNLOCK(p);

	bsd_to_linux_sigset(mask, &frame.sf_sc.uc_sigmask);

	frame.sf_sc.uc_mcontext.sc_mask   = frame.sf_sc.uc_sigmask.__bits[0];
	frame.sf_sc.uc_mcontext.sc_edi    = regs->tf_rdi;
	frame.sf_sc.uc_mcontext.sc_esi    = regs->tf_rsi;
	frame.sf_sc.uc_mcontext.sc_ebp    = regs->tf_rbp;
	frame.sf_sc.uc_mcontext.sc_ebx    = regs->tf_rbx;
	frame.sf_sc.uc_mcontext.sc_edx    = regs->tf_rdx;
	frame.sf_sc.uc_mcontext.sc_ecx    = regs->tf_rcx;
	frame.sf_sc.uc_mcontext.sc_eax    = regs->tf_rax;
	frame.sf_sc.uc_mcontext.sc_eip    = regs->tf_rip;
	frame.sf_sc.uc_mcontext.sc_cs     = regs->tf_cs;
	frame.sf_sc.uc_mcontext.sc_gs     = regs->tf_gs;
	frame.sf_sc.uc_mcontext.sc_fs     = regs->tf_fs;
	frame.sf_sc.uc_mcontext.sc_es     = regs->tf_es;
	frame.sf_sc.uc_mcontext.sc_ds     = regs->tf_ds;
	frame.sf_sc.uc_mcontext.sc_eflags = regs->tf_rflags;
	frame.sf_sc.uc_mcontext.sc_esp_at_signal = regs->tf_rsp;
	frame.sf_sc.uc_mcontext.sc_ss     = regs->tf_ss;
	frame.sf_sc.uc_mcontext.sc_err    = regs->tf_err;
	frame.sf_sc.uc_mcontext.sc_cr2    = (u_int32_t)(uintptr_t)ksi->ksi_addr;
	frame.sf_sc.uc_mcontext.sc_trapno = bsd_to_linux_trapcode(code);

#ifdef DEBUG
	if (ldebug(rt_sendsig))
		printf(LMSG("rt_sendsig flags: 0x%x, sp: %p, ss: 0x%lx, mask: 0x%x"),
		    frame.sf_sc.uc_stack.ss_flags, td->td_sigstk.ss_sp,
		    td->td_sigstk.ss_size, frame.sf_sc.uc_mcontext.sc_mask);
#endif

	if (copyout(&frame, fp, sizeof(frame)) != 0) {
		/*
		 * Process has trashed its stack; give it an illegal
		 * instruction to halt it in its tracks.
		 */
#ifdef DEBUG
		if (ldebug(rt_sendsig))
			printf(LMSG("rt_sendsig: bad stack %p, oonstack=%x"),
			    fp, oonstack);
#endif
		PROC_LOCK(p);
		sigexit(td, SIGILL);
	}

	/*
	 * Build context to run handler in.
	 */
	regs->tf_rsp = PTROUT(fp);
	regs->tf_rip = p->p_sysent->sv_sigcode_base + linux_sznonrtsigcode;
	regs->tf_rflags &= ~(PSL_T | PSL_D);
	regs->tf_cs = _ucode32sel;
	regs->tf_ss = _udatasel;
	regs->tf_ds = _udatasel;
	regs->tf_es = _udatasel;
	regs->tf_fs = _ufssel;
	regs->tf_gs = _ugssel;
	regs->tf_flags = TF_HASSEGS;
	set_pcb_flags(td->td_pcb, PCB_FULL_IRET);
	PROC_LOCK(p);
	mtx_lock(&psp->ps_mtx);
}


/*
 * Send an interrupt to process.
 *
 * Stack is set up to allow sigcode stored
 * in u. to call routine, followed by kcall
 * to sigreturn routine below.  After sigreturn
 * resets the signal mask, the stack, and the
 * frame pointer, it returns to the user
 * specified pc, psl.
 */
static void
linux_sendsig(sig_t catcher, ksiginfo_t *ksi, sigset_t *mask)
{
	struct thread *td = curthread;
	struct proc *p = td->td_proc;
	struct sigacts *psp;
	struct trapframe *regs;
	struct l_sigframe *fp, frame;
	l_sigset_t lmask;
	int oonstack, i;
	int sig, code;

	sig = ksi->ksi_signo;
	code = ksi->ksi_code;
	PROC_LOCK_ASSERT(p, MA_OWNED);
	psp = p->p_sigacts;
	mtx_assert(&psp->ps_mtx, MA_OWNED);
	if (SIGISMEMBER(psp->ps_siginfo, sig)) {
		/* Signal handler installed with SA_SIGINFO. */
		linux_rt_sendsig(catcher, ksi, mask);
		return;
	}

	regs = td->td_frame;
	oonstack = sigonstack(regs->tf_rsp);

#ifdef DEBUG
	if (ldebug(sendsig))
		printf(ARGS(sendsig, "%p, %d, %p, %u"),
		    catcher, sig, (void*)mask, code);
#endif

	/*
	 * Allocate space for the signal handler context.
	 */
	if ((td->td_pflags & TDP_ALTSTACK) && !oonstack &&
	    SIGISMEMBER(psp->ps_sigonstack, sig)) {
		fp = (struct l_sigframe *)(td->td_sigstk.ss_sp +
		    td->td_sigstk.ss_size - sizeof(struct l_sigframe));
	} else
		fp = (struct l_sigframe *)regs->tf_rsp - 1;
	mtx_unlock(&psp->ps_mtx);
	PROC_UNLOCK(p);

	/*
	 * Build the argument list for the signal handler.
	 */
	if (p->p_sysent->sv_sigtbl)
		if (sig <= p->p_sysent->sv_sigsize)
			sig = p->p_sysent->sv_sigtbl[_SIG_IDX(sig)];

	bzero(&frame, sizeof(frame));

	frame.sf_handler = PTROUT(catcher);
	frame.sf_sig = sig;

	bsd_to_linux_sigset(mask, &lmask);

	/*
	 * Build the signal context to be used by sigreturn.
	 */
	frame.sf_sc.sc_mask   = lmask.__bits[0];
	frame.sf_sc.sc_gs     = regs->tf_gs;
	frame.sf_sc.sc_fs     = regs->tf_fs;
	frame.sf_sc.sc_es     = regs->tf_es;
	frame.sf_sc.sc_ds     = regs->tf_ds;
	frame.sf_sc.sc_edi    = regs->tf_rdi;
	frame.sf_sc.sc_esi    = regs->tf_rsi;
	frame.sf_sc.sc_ebp    = regs->tf_rbp;
	frame.sf_sc.sc_ebx    = regs->tf_rbx;
	frame.sf_sc.sc_edx    = regs->tf_rdx;
	frame.sf_sc.sc_ecx    = regs->tf_rcx;
	frame.sf_sc.sc_eax    = regs->tf_rax;
	frame.sf_sc.sc_eip    = regs->tf_rip;
	frame.sf_sc.sc_cs     = regs->tf_cs;
	frame.sf_sc.sc_eflags = regs->tf_rflags;
	frame.sf_sc.sc_esp_at_signal = regs->tf_rsp;
	frame.sf_sc.sc_ss     = regs->tf_ss;
	frame.sf_sc.sc_err    = regs->tf_err;
	frame.sf_sc.sc_cr2    = (u_int32_t)(uintptr_t)ksi->ksi_addr;
	frame.sf_sc.sc_trapno = bsd_to_linux_trapcode(code);

	for (i = 0; i < (LINUX_NSIG_WORDS-1); i++)
		frame.sf_extramask[i] = lmask.__bits[i+1];

	if (copyout(&frame, fp, sizeof(frame)) != 0) {
		/*
		 * Process has trashed its stack; give it an illegal
		 * instruction to halt it in its tracks.
		 */
		PROC_LOCK(p);
		sigexit(td, SIGILL);
	}

	/*
	 * Build context to run handler in.
	 */
	regs->tf_rsp = PTROUT(fp);
	regs->tf_rip = p->p_sysent->sv_sigcode_base;
	regs->tf_rflags &= ~(PSL_T | PSL_D);
	regs->tf_cs = _ucode32sel;
	regs->tf_ss = _udatasel;
	regs->tf_ds = _udatasel;
	regs->tf_es = _udatasel;
	regs->tf_fs = _ufssel;
	regs->tf_gs = _ugssel;
	regs->tf_flags = TF_HASSEGS;
	set_pcb_flags(td->td_pcb, PCB_FULL_IRET);
	PROC_LOCK(p);
	mtx_lock(&psp->ps_mtx);
}

/*
 * System call to cleanup state after a signal
 * has been taken.  Reset signal mask and
 * stack state from context left by sendsig (above).
 * Return to previous pc and psl as specified by
 * context left by sendsig. Check carefully to
 * make sure that the user has not modified the
 * psl to gain improper privileges or to cause
 * a machine fault.
 */
int
linux_sigreturn(struct thread *td, struct linux_sigreturn_args *args)
{
	struct l_sigframe frame;
	struct trapframe *regs;
	sigset_t bmask;
	l_sigset_t lmask;
	int eflags, i;
	ksiginfo_t ksi;

	regs = td->td_frame;

#ifdef DEBUG
	if (ldebug(sigreturn))
		printf(ARGS(sigreturn, "%p"), (void *)args->sfp);
#endif
	/*
	 * The trampoline code hands us the sigframe.
	 * It is unsafe to keep track of it ourselves, in the event that a
	 * program jumps out of a signal handler.
	 */
	if (copyin(args->sfp, &frame, sizeof(frame)) != 0)
		return (EFAULT);

	/*
	 * Check for security violations.
	 */
#define	EFLAGS_SECURE(ef, oef)	((((ef) ^ (oef)) & ~PSL_USERCHANGE) == 0)
	eflags = frame.sf_sc.sc_eflags;
	if (!EFLAGS_SECURE(eflags, regs->tf_rflags))
		return(EINVAL);

	/*
	 * Don't allow users to load a valid privileged %cs.  Let the
	 * hardware check for invalid selectors, excess privilege in
	 * other selectors, invalid %eip's and invalid %esp's.
	 */
#define	CS_SECURE(cs)	(ISPL(cs) == SEL_UPL)
	if (!CS_SECURE(frame.sf_sc.sc_cs)) {
		ksiginfo_init_trap(&ksi);
		ksi.ksi_signo = SIGBUS;
		ksi.ksi_code = BUS_OBJERR;
		ksi.ksi_trapno = T_PROTFLT;
		ksi.ksi_addr = (void *)regs->tf_rip;
		trapsignal(td, &ksi);
		return(EINVAL);
	}

	lmask.__bits[0] = frame.sf_sc.sc_mask;
	for (i = 0; i < (LINUX_NSIG_WORDS-1); i++)
		lmask.__bits[i+1] = frame.sf_extramask[i];
	linux_to_bsd_sigset(&lmask, &bmask);
	kern_sigprocmask(td, SIG_SETMASK, &bmask, NULL, 0);

	/*
	 * Restore signal context.
	 */
	regs->tf_rdi    = frame.sf_sc.sc_edi;
	regs->tf_rsi    = frame.sf_sc.sc_esi;
	regs->tf_rbp    = frame.sf_sc.sc_ebp;
	regs->tf_rbx    = frame.sf_sc.sc_ebx;
	regs->tf_rdx    = frame.sf_sc.sc_edx;
	regs->tf_rcx    = frame.sf_sc.sc_ecx;
	regs->tf_rax    = frame.sf_sc.sc_eax;
	regs->tf_rip    = frame.sf_sc.sc_eip;
	regs->tf_cs     = frame.sf_sc.sc_cs;
	regs->tf_ds     = frame.sf_sc.sc_ds;
	regs->tf_es     = frame.sf_sc.sc_es;
	regs->tf_fs     = frame.sf_sc.sc_fs;
	regs->tf_gs     = frame.sf_sc.sc_gs;
	regs->tf_rflags = eflags;
	regs->tf_rsp    = frame.sf_sc.sc_esp_at_signal;
	regs->tf_ss     = frame.sf_sc.sc_ss;
	set_pcb_flags(td->td_pcb, PCB_FULL_IRET);

	return (EJUSTRETURN);
}

/*
 * System call to cleanup state after a signal
 * has been taken.  Reset signal mask and
 * stack state from context left by rt_sendsig (above).
 * Return to previous pc and psl as specified by
 * context left by sendsig. Check carefully to
 * make sure that the user has not modified the
 * psl to gain improper privileges or to cause
 * a machine fault.
 */
int
linux_rt_sigreturn(struct thread *td, struct linux_rt_sigreturn_args *args)
{
	struct l_ucontext uc;
	struct l_sigcontext *context;
	sigset_t bmask;
	l_stack_t *lss;
	stack_t ss;
	struct trapframe *regs;
	int eflags;
	ksiginfo_t ksi;

	regs = td->td_frame;

#ifdef DEBUG
	if (ldebug(rt_sigreturn))
		printf(ARGS(rt_sigreturn, "%p"), (void *)args->ucp);
#endif
	/*
	 * The trampoline code hands us the ucontext.
	 * It is unsafe to keep track of it ourselves, in the event that a
	 * program jumps out of a signal handler.
	 */
	if (copyin(args->ucp, &uc, sizeof(uc)) != 0)
		return (EFAULT);

	context = &uc.uc_mcontext;

	/*
	 * Check for security violations.
	 */
#define	EFLAGS_SECURE(ef, oef)	((((ef) ^ (oef)) & ~PSL_USERCHANGE) == 0)
	eflags = context->sc_eflags;
	if (!EFLAGS_SECURE(eflags, regs->tf_rflags))
		return(EINVAL);

	/*
	 * Don't allow users to load a valid privileged %cs.  Let the
	 * hardware check for invalid selectors, excess privilege in
	 * other selectors, invalid %eip's and invalid %esp's.
	 */
#define	CS_SECURE(cs)	(ISPL(cs) == SEL_UPL)
	if (!CS_SECURE(context->sc_cs)) {
		ksiginfo_init_trap(&ksi);
		ksi.ksi_signo = SIGBUS;
		ksi.ksi_code = BUS_OBJERR;
		ksi.ksi_trapno = T_PROTFLT;
		ksi.ksi_addr = (void *)regs->tf_rip;
		trapsignal(td, &ksi);
		return(EINVAL);
	}

	linux_to_bsd_sigset(&uc.uc_sigmask, &bmask);
	kern_sigprocmask(td, SIG_SETMASK, &bmask, NULL, 0);

	/*
	 * Restore signal context
	 */
	regs->tf_gs	= context->sc_gs;
	regs->tf_fs	= context->sc_fs;
	regs->tf_es	= context->sc_es;
	regs->tf_ds	= context->sc_ds;
	regs->tf_rdi    = context->sc_edi;
	regs->tf_rsi    = context->sc_esi;
	regs->tf_rbp    = context->sc_ebp;
	regs->tf_rbx    = context->sc_ebx;
	regs->tf_rdx    = context->sc_edx;
	regs->tf_rcx    = context->sc_ecx;
	regs->tf_rax    = context->sc_eax;
	regs->tf_rip    = context->sc_eip;
	regs->tf_cs     = context->sc_cs;
	regs->tf_rflags = eflags;
	regs->tf_rsp    = context->sc_esp_at_signal;
	regs->tf_ss     = context->sc_ss;
	set_pcb_flags(td->td_pcb, PCB_FULL_IRET);

	/*
	 * call sigaltstack & ignore results..
	 */
	lss = &uc.uc_stack;
	ss.ss_sp = PTRIN(lss->ss_sp);
	ss.ss_size = lss->ss_size;
	ss.ss_flags = linux_to_bsd_sigaltstack(lss->ss_flags);

#ifdef DEBUG
	if (ldebug(rt_sigreturn))
		printf(LMSG("rt_sigret flags: 0x%x, sp: %p, ss: 0x%lx, mask: 0x%x"),
		    ss.ss_flags, ss.ss_sp, ss.ss_size, context->sc_mask);
#endif
	(void)kern_sigaltstack(td, &ss, NULL);

	return (EJUSTRETURN);
}

static int
linux32_fetch_syscall_args(struct thread *td, struct syscall_args *sa)
{
	struct proc *p;
	struct trapframe *frame;

	p = td->td_proc;
	frame = td->td_frame;

	sa->args[0] = frame->tf_rbx;
	sa->args[1] = frame->tf_rcx;
	sa->args[2] = frame->tf_rdx;
	sa->args[3] = frame->tf_rsi;
	sa->args[4] = frame->tf_rdi;
	sa->args[5] = frame->tf_rbp;	/* Unconfirmed */
	sa->code = frame->tf_rax;

	if (sa->code >= p->p_sysent->sv_size)
		sa->callp = &p->p_sysent->sv_table[0];
	else
		sa->callp = &p->p_sysent->sv_table[sa->code];
	sa->narg = sa->callp->sy_narg;

	td->td_retval[0] = 0;
	td->td_retval[1] = frame->tf_rdx;

	return (0);
}

/*
 * If a linux binary is exec'ing something, try this image activator
 * first.  We override standard shell script execution in order to
 * be able to modify the interpreter path.  We only do this if a linux
 * binary is doing the exec, so we do not create an EXEC module for it.
 */
static int	exec_linux_imgact_try(struct image_params *iparams);

static int
exec_linux_imgact_try(struct image_params *imgp)
{
	const char *head = (const char *)imgp->image_header;
	char *rpath;
	int error = -1;

	/*
	* The interpreter for shell scripts run from a linux binary needs
	* to be located in /compat/linux if possible in order to recursively
	* maintain linux path emulation.
	*/
	if (((const short *)head)[0] == SHELLMAGIC) {
		/*
		* Run our normal shell image activator.  If it succeeds attempt
		* to use the alternate path for the interpreter.  If an
		* alternate * path is found, use our stringspace to store it.
		*/
		if ((error = exec_shell_imgact(imgp)) == 0) {
			linux_emul_convpath(FIRST_THREAD_IN_PROC(imgp->proc),
			    imgp->interpreter_name, UIO_SYSSPACE, &rpath, 0,
			    AT_FDCWD);
			if (rpath != NULL)
				imgp->args->fname_buf =
				    imgp->interpreter_name = rpath;
		}
	}
	return (error);
}

/*
 * Clear registers on exec
 * XXX copied from ia32_signal.c.
 */
static void
exec_linux_setregs(struct thread *td, struct image_params *imgp, u_long stack)
{
	struct trapframe *regs = td->td_frame;
	struct pcb *pcb = td->td_pcb;

	mtx_lock(&dt_lock);
	if (td->td_proc->p_md.md_ldt != NULL)
		user_ldt_free(td);
	else
		mtx_unlock(&dt_lock);

	critical_enter();
	wrmsr(MSR_FSBASE, 0);
	wrmsr(MSR_KGSBASE, 0);	/* User value while we're in the kernel */
	pcb->pcb_fsbase = 0;
	pcb->pcb_gsbase = 0;
	critical_exit();
	pcb->pcb_initial_fpucw = __LINUX_NPXCW__;

	bzero((char *)regs, sizeof(struct trapframe));
	regs->tf_rip = imgp->entry_addr;
	regs->tf_rsp = stack;
	regs->tf_rflags = PSL_USER | (regs->tf_rflags & PSL_T);
	regs->tf_gs = _ugssel;
	regs->tf_fs = _ufssel;
	regs->tf_es = _udatasel;
	regs->tf_ds = _udatasel;
	regs->tf_ss = _udatasel;
	regs->tf_flags = TF_HASSEGS;
	regs->tf_cs = _ucode32sel;
	regs->tf_rbx = imgp->ps_strings;

	fpstate_drop(td);

	/* Do full restore on return so that we can change to a different %cs */
	set_pcb_flags(pcb, PCB_32BIT | PCB_FULL_IRET);
	td->td_retval[1] = 0;
}

/*
 * XXX copied from ia32_sysvec.c.
 */
static register_t *
linux_copyout_strings(struct image_params *imgp)
{
	int argc, envc;
	u_int32_t *vectp;
	char *stringp, *destp;
	u_int32_t *stack_base;
	struct linux32_ps_strings *arginfo;

	/*
	 * Calculate string base and vector table pointers.
	 * Also deal with signal trampoline code for this exec type.
	 */
	arginfo = (struct linux32_ps_strings *)LINUX32_PS_STRINGS;
	destp =	(caddr_t)arginfo - SPARE_USRSPACE - linux_szplatform -
	    roundup((ARG_MAX - imgp->args->stringspace),
	    sizeof(char *));

	/*
	 * Install LINUX_PLATFORM
	 */
	copyout(linux_platform, ((caddr_t)arginfo - linux_szplatform),
	    linux_szplatform);

	/*
	 * If we have a valid auxargs ptr, prepare some room
	 * on the stack.
	 */
	if (imgp->auxargs) {
		/*
		 * 'AT_COUNT*2' is size for the ELF Auxargs data. This is for
		 * lower compatibility.
		 */
		imgp->auxarg_size = (imgp->auxarg_size) ? imgp->auxarg_size :
		    (LINUX_AT_COUNT * 2);
		/*
		 * The '+ 2' is for the null pointers at the end of each of
		 * the arg and env vector sets,and imgp->auxarg_size is room
		 * for argument of Runtime loader.
		 */
		vectp = (u_int32_t *) (destp - (imgp->args->argc +
		    imgp->args->envc + 2 + imgp->auxarg_size) *
		    sizeof(u_int32_t));

	} else
		/*
		 * The '+ 2' is for the null pointers at the end of each of
		 * the arg and env vector sets
		 */
		vectp = (u_int32_t *)(destp - (imgp->args->argc +
		    imgp->args->envc + 2) * sizeof(u_int32_t));

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

	stringp = imgp->args->begin_argv;
	argc = imgp->args->argc;
	envc = imgp->args->envc;
	/*
	 * Copy out strings - arguments and environment.
	 */
	copyout(stringp, destp, ARG_MAX - imgp->args->stringspace);

	/*
	 * Fill in "ps_strings" struct for ps, w, etc.
	 */
	suword32(&arginfo->ps_argvstr, (uint32_t)(intptr_t)vectp);
	suword32(&arginfo->ps_nargvstr, argc);

	/*
	 * Fill in argument portion of vector table.
	 */
	for (; argc > 0; --argc) {
		suword32(vectp++, (uint32_t)(intptr_t)destp);
		while (*stringp++ != 0)
			destp++;
		destp++;
	}

	/* a null vector table pointer separates the argp's from the envp's */
	suword32(vectp++, 0);

	suword32(&arginfo->ps_envstr, (uint32_t)(intptr_t)vectp);
	suword32(&arginfo->ps_nenvstr, envc);

	/*
	 * Fill in environment portion of vector table.
	 */
	for (; envc > 0; --envc) {
		suword32(vectp++, (uint32_t)(intptr_t)destp);
		while (*stringp++ != 0)
			destp++;
		destp++;
	}

	/* end of vector table is a null pointer */
	suword32(vectp, 0);

	return ((register_t *)stack_base);
}

static SYSCTL_NODE(_compat, OID_AUTO, linux32, CTLFLAG_RW, 0,
    "32-bit Linux emulation");

static u_long	linux32_maxdsiz = LINUX32_MAXDSIZ;
SYSCTL_ULONG(_compat_linux32, OID_AUTO, maxdsiz, CTLFLAG_RW,
    &linux32_maxdsiz, 0, "");
static u_long	linux32_maxssiz = LINUX32_MAXSSIZ;
SYSCTL_ULONG(_compat_linux32, OID_AUTO, maxssiz, CTLFLAG_RW,
    &linux32_maxssiz, 0, "");
static u_long	linux32_maxvmem = LINUX32_MAXVMEM;
SYSCTL_ULONG(_compat_linux32, OID_AUTO, maxvmem, CTLFLAG_RW,
    &linux32_maxvmem, 0, "");

static void
linux32_fixlimit(struct rlimit *rl, int which)
{

	switch (which) {
	case RLIMIT_DATA:
		if (linux32_maxdsiz != 0) {
			if (rl->rlim_cur > linux32_maxdsiz)
				rl->rlim_cur = linux32_maxdsiz;
			if (rl->rlim_max > linux32_maxdsiz)
				rl->rlim_max = linux32_maxdsiz;
		}
		break;
	case RLIMIT_STACK:
		if (linux32_maxssiz != 0) {
			if (rl->rlim_cur > linux32_maxssiz)
				rl->rlim_cur = linux32_maxssiz;
			if (rl->rlim_max > linux32_maxssiz)
				rl->rlim_max = linux32_maxssiz;
		}
		break;
	case RLIMIT_VMEM:
		if (linux32_maxvmem != 0) {
			if (rl->rlim_cur > linux32_maxvmem)
				rl->rlim_cur = linux32_maxvmem;
			if (rl->rlim_max > linux32_maxvmem)
				rl->rlim_max = linux32_maxvmem;
		}
		break;
	}
}

struct sysentvec elf_linux_sysvec = {
	.sv_size	= LINUX_SYS_MAXSYSCALL,
	.sv_table	= linux_sysent,
	.sv_mask	= 0,
	.sv_sigsize	= LINUX_SIGTBLSZ,
	.sv_sigtbl	= bsd_to_linux_signal,
	.sv_errsize	= ELAST + 1,
	.sv_errtbl	= bsd_to_linux_errno,
	.sv_transtrap	= translate_traps,
	.sv_fixup	= elf_linux_fixup,
	.sv_sendsig	= linux_sendsig,
	.sv_sigcode	= linux_sigcode,
	.sv_szsigcode	= &linux_szsigcode,
	.sv_prepsyscall	= NULL,
	.sv_name	= "Linux ELF32",
	.sv_coredump	= elf32_coredump,
	.sv_imgact_try	= exec_linux_imgact_try,
	.sv_minsigstksz	= LINUX_MINSIGSTKSZ,
	.sv_pagesize	= PAGE_SIZE,
	.sv_minuser	= VM_MIN_ADDRESS,
	.sv_maxuser	= LINUX32_MAXUSER,
	.sv_usrstack	= LINUX32_USRSTACK,
	.sv_psstrings	= LINUX32_PS_STRINGS,
	.sv_stackprot	= VM_PROT_ALL,
	.sv_copyout_strings = linux_copyout_strings,
	.sv_setregs	= exec_linux_setregs,
	.sv_fixlimit	= linux32_fixlimit,
	.sv_maxssiz	= &linux32_maxssiz,
	.sv_flags	= SV_ABI_LINUX | SV_ILP32 | SV_IA32 | SV_SHP,
	.sv_set_syscall_retval = cpu_set_syscall_retval,
	.sv_fetch_syscall_args = linux32_fetch_syscall_args,
	.sv_syscallnames = NULL,
	.sv_shared_page_base = LINUX32_SHAREDPAGE,
	.sv_shared_page_len = PAGE_SIZE,
	.sv_schedtail	= linux_schedtail,
};
INIT_SYSENTVEC(elf_sysvec, &elf_linux_sysvec);

static char GNU_ABI_VENDOR[] = "GNU";
static int GNULINUX_ABI_DESC = 0;

static boolean_t
linux32_trans_osrel(const Elf_Note *note, int32_t *osrel)
{
	const Elf32_Word *desc;
	uintptr_t p;

	p = (uintptr_t)(note + 1);
	p += roundup2(note->n_namesz, sizeof(Elf32_Addr));

	desc = (const Elf32_Word *)p;
	if (desc[0] != GNULINUX_ABI_DESC)
		return (FALSE);

	/*
	 * For linux we encode osrel as follows (see linux_mib.c):
	 * VVVMMMIII (version, major, minor), see linux_mib.c.
	 */
	*osrel = desc[1] * 1000000 + desc[2] * 1000 + desc[3];

	return (TRUE);
}

static Elf_Brandnote linux32_brandnote = {
	.hdr.n_namesz	= sizeof(GNU_ABI_VENDOR),
	.hdr.n_descsz	= 16,	/* XXX at least 16 */
	.hdr.n_type	= 1,
	.vendor		= GNU_ABI_VENDOR,
	.flags		= BN_TRANSLATE_OSREL,
	.trans_osrel	= linux32_trans_osrel
};

static Elf32_Brandinfo linux_brand = {
	.brand		= ELFOSABI_LINUX,
	.machine	= EM_386,
	.compat_3_brand	= "Linux",
	.emul_path	= "/compat/linux",
	.interp_path	= "/lib/ld-linux.so.1",
	.sysvec		= &elf_linux_sysvec,
	.interp_newpath	= NULL,
	.brand_note	= &linux32_brandnote,
	.flags		= BI_CAN_EXEC_DYN | BI_BRAND_NOTE
};

static Elf32_Brandinfo linux_glibc2brand = {
	.brand		= ELFOSABI_LINUX,
	.machine	= EM_386,
	.compat_3_brand	= "Linux",
	.emul_path	= "/compat/linux",
	.interp_path	= "/lib/ld-linux.so.2",
	.sysvec		= &elf_linux_sysvec,
	.interp_newpath	= NULL,
	.brand_note	= &linux32_brandnote,
	.flags		= BI_CAN_EXEC_DYN | BI_BRAND_NOTE
};

Elf32_Brandinfo *linux_brandlist[] = {
	&linux_brand,
	&linux_glibc2brand,
	NULL
};

static int
linux_elf_modevent(module_t mod, int type, void *data)
{
	Elf32_Brandinfo **brandinfo;
	int error;
	struct linux_ioctl_handler **lihp;
	struct linux_device_handler **ldhp;

	error = 0;

	switch(type) {
	case MOD_LOAD:
		for (brandinfo = &linux_brandlist[0]; *brandinfo != NULL;
		     ++brandinfo)
			if (elf32_insert_brand_entry(*brandinfo) < 0)
				error = EINVAL;
		if (error == 0) {
			SET_FOREACH(lihp, linux_ioctl_handler_set)
				linux_ioctl_register_handler(*lihp);
			SET_FOREACH(ldhp, linux_device_handler_set)
				linux_device_register_handler(*ldhp);
			mtx_init(&emul_lock, "emuldata lock", NULL, MTX_DEF);
			sx_init(&emul_shared_lock, "emuldata->shared lock");
			LIST_INIT(&futex_list);
			mtx_init(&futex_mtx, "ftllk", NULL, MTX_DEF);
			linux_exit_tag = EVENTHANDLER_REGISTER(process_exit,
			    linux_proc_exit, NULL, 1000);
			linux_exec_tag = EVENTHANDLER_REGISTER(process_exec,
			    linux_proc_exec, NULL, 1000);
			linux_szplatform = roundup(strlen(linux_platform) + 1,
			    sizeof(char *));
			linux_osd_jail_register();
			stclohz = (stathz ? stathz : hz);
			if (bootverbose)
				printf("Linux ELF exec handler installed\n");
		} else
			printf("cannot insert Linux ELF brand handler\n");
		break;
	case MOD_UNLOAD:
		for (brandinfo = &linux_brandlist[0]; *brandinfo != NULL;
		     ++brandinfo)
			if (elf32_brand_inuse(*brandinfo))
				error = EBUSY;
		if (error == 0) {
			for (brandinfo = &linux_brandlist[0];
			     *brandinfo != NULL; ++brandinfo)
				if (elf32_remove_brand_entry(*brandinfo) < 0)
					error = EINVAL;
		}
		if (error == 0) {
			SET_FOREACH(lihp, linux_ioctl_handler_set)
				linux_ioctl_unregister_handler(*lihp);
			SET_FOREACH(ldhp, linux_device_handler_set)
				linux_device_unregister_handler(*ldhp);
			mtx_destroy(&emul_lock);
			sx_destroy(&emul_shared_lock);
			mtx_destroy(&futex_mtx);
			EVENTHANDLER_DEREGISTER(process_exit, linux_exit_tag);
			EVENTHANDLER_DEREGISTER(process_exec, linux_exec_tag);
			linux_osd_jail_deregister();
			if (bootverbose)
				printf("Linux ELF exec handler removed\n");
		} else
			printf("Could not deinstall ELF interpreter entry\n");
		break;
	default:
		return EOPNOTSUPP;
	}
	return error;
}

static moduledata_t linux_elf_mod = {
	"linuxelf",
	linux_elf_modevent,
	0
};

DECLARE_MODULE_TIED(linuxelf, linux_elf_mod, SI_SUB_EXEC, SI_ORDER_ANY);
OpenPOWER on IntegriCloud