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
|
/*
* CDDL HEADER START
*
* The contents of this file are subject to the terms of the
* Common Development and Distribution License (the "License").
* You may not use this file except in compliance with the License.
*
* You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
* or http://www.opensolaris.org/os/licensing.
* See the License for the specific language governing permissions
* and limitations under the License.
*
* When distributing Covered Code, include this CDDL HEADER in each
* file and include the License file at usr/src/OPENSOLARIS.LICENSE.
* If applicable, add the following below this CDDL HEADER, with the
* fields enclosed by brackets "[]" replaced with your own identifying
* information: Portions Copyright [yyyy] [name of copyright owner]
*
* CDDL HEADER END
*
* Portions Copyright 2006-2008 John Birrell jb@freebsd.org
*
* $FreeBSD$
*
*/
/*
* Copyright 2006 Sun Microsystems, Inc. All rights reserved.
* Use is subject to license terms.
*/
#include <sys/cdefs.h>
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/conf.h>
#include <sys/cpuvar.h>
#include <sys/fcntl.h>
#include <sys/filio.h>
#include <sys/kdb.h>
#include <sys/kernel.h>
#include <sys/kmem.h>
#include <sys/kthread.h>
#include <sys/limits.h>
#include <sys/linker.h>
#include <sys/lock.h>
#include <sys/malloc.h>
#include <sys/module.h>
#include <sys/mutex.h>
#include <sys/poll.h>
#include <sys/proc.h>
#include <sys/selinfo.h>
#include <sys/smp.h>
#include <sys/sysproto.h>
#include <sys/sysent.h>
#include <sys/uio.h>
#include <sys/unistd.h>
#include <machine/stdarg.h>
#include <sys/dtrace.h>
#ifdef LINUX_SYSTRACE
#if defined(__amd64__)
#include <amd64/linux32/linux.h>
#include <amd64/linux32/linux32_proto.h>
#include <amd64/linux32/linux32_syscalls.c>
#include <amd64/linux32/linux32_systrace_args.c>
#define MODNAME "linux32"
#elif defined(__i386__)
#include <i386/linux/linux.h>
#include <i386/linux/linux_proto.h>
#include <i386/linux/linux_syscalls.c>
#include <i386/linux/linux_systrace_args.c>
#define MODNAME "linux"
#else
#error Only i386 and amd64 are supported.
#endif
extern struct sysent linux_sysent[];
#define MAXSYSCALL LINUX_SYS_MAXSYSCALL
#define SYSCALLNAMES linux_syscallnames
#define SYSENT linux_sysent
#elif defined(FREEBSD32_SYSTRACE)
/*
* The syscall arguments are processed into a DTrace argument array
* using a generated function. See sys/kern/makesyscalls.sh.
*/
#include <compat/freebsd32/freebsd32_proto.h>
#include <compat/freebsd32/freebsd32_util.h>
#include <compat/freebsd32/freebsd32_syscall.h>
#include <compat/freebsd32/freebsd32_systrace_args.c>
extern const char *freebsd32_syscallnames[];
#define MODNAME "freebsd32"
#define MAXSYSCALL FREEBSD32_SYS_MAXSYSCALL
#define SYSCALLNAMES freebsd32_syscallnames
#define SYSENT freebsd32_sysent
#else
/*
* The syscall arguments are processed into a DTrace argument array
* using a generated function. See sys/kern/makesyscalls.sh.
*/
#include <sys/syscall.h>
#include <kern/systrace_args.c>
#define MODNAME "freebsd"
#define MAXSYSCALL SYS_MAXSYSCALL
#define SYSCALLNAMES syscallnames
#define SYSENT sysent
#endif
#define PROVNAME "syscall"
#define DEVNAME "dtrace/systrace/" MODNAME
#define SYSTRACE_ARTIFICIAL_FRAMES 1
#define SYSTRACE_SHIFT 16
#define SYSTRACE_ISENTRY(x) ((int)(x) >> SYSTRACE_SHIFT)
#define SYSTRACE_SYSNUM(x) ((int)(x) & ((1 << SYSTRACE_SHIFT) - 1))
#define SYSTRACE_ENTRY(id) ((1 << SYSTRACE_SHIFT) | (id))
#define SYSTRACE_RETURN(id) (id)
#if ((1 << SYSTRACE_SHIFT) <= MAXSYSCALL)
#error 1 << SYSTRACE_SHIFT must exceed number of system calls
#endif
static d_open_t systrace_open;
static int systrace_unload(void);
static void systrace_getargdesc(void *, dtrace_id_t, void *, dtrace_argdesc_t *);
static void systrace_provide(void *, dtrace_probedesc_t *);
static void systrace_destroy(void *, dtrace_id_t, void *);
static void systrace_enable(void *, dtrace_id_t, void *);
static void systrace_disable(void *, dtrace_id_t, void *);
static void systrace_load(void *);
static struct cdevsw systrace_cdevsw = {
.d_version = D_VERSION,
.d_open = systrace_open,
#ifdef LINUX_SYSTRACE
.d_name = "systrace_" MODNAME,
#else
.d_name = "systrace",
#endif
};
static union {
const char **p_constnames;
char **pp_syscallnames;
} uglyhack = { SYSCALLNAMES };
static dtrace_pattr_t systrace_attr = {
{ DTRACE_STABILITY_EVOLVING, DTRACE_STABILITY_EVOLVING, DTRACE_CLASS_COMMON },
{ DTRACE_STABILITY_PRIVATE, DTRACE_STABILITY_PRIVATE, DTRACE_CLASS_UNKNOWN },
{ DTRACE_STABILITY_PRIVATE, DTRACE_STABILITY_PRIVATE, DTRACE_CLASS_ISA },
{ DTRACE_STABILITY_EVOLVING, DTRACE_STABILITY_EVOLVING, DTRACE_CLASS_COMMON },
{ DTRACE_STABILITY_PRIVATE, DTRACE_STABILITY_PRIVATE, DTRACE_CLASS_ISA },
};
static dtrace_pops_t systrace_pops = {
systrace_provide,
NULL,
systrace_enable,
systrace_disable,
NULL,
NULL,
systrace_getargdesc,
NULL,
NULL,
systrace_destroy
};
static struct cdev *systrace_cdev;
static dtrace_provider_id_t systrace_id;
typedef void (*systrace_dtrace_probe_t)(dtrace_id_t, uintptr_t, uintptr_t,
uintptr_t, uintptr_t, uintptr_t, uintptr_t, uintptr_t, uintptr_t);
#if !defined(LINUX_SYSTRACE)
/*
* Probe callback function.
*
* Note: This function is called for _all_ syscalls, regardless of which sysent
* array the syscall comes from. It could be a standard syscall or a
* compat syscall from something like Linux.
*/
static void
systrace_probe(u_int32_t id, int sysnum, struct sysent *sysent, void *params,
int ret)
{
systrace_dtrace_probe_t probe;
int n_args = 0;
u_int64_t uargs[8];
memset(uargs, 0, sizeof(uargs));
/*
* Check if this syscall has an argument conversion function
* registered.
*/
if (params && sysent->sy_systrace_args_func != NULL) {
/*
* Convert the syscall parameters using the registered
* function.
*/
(*sysent->sy_systrace_args_func)(sysnum, params, uargs, &n_args);
} else if (params) {
/*
* Use the built-in system call argument conversion
* function to translate the syscall structure fields
* into the array of 64-bit values that DTrace
* expects.
*/
systrace_args(sysnum, params, uargs, &n_args);
} else {
/*
* Since params is NULL, this is a 'return' probe.
* Set arg0 and arg1 as the return value of this syscall.
*/
uargs[0] = uargs[1] = ret;
}
/* Process the probe using the converted argments. */
probe = (systrace_dtrace_probe_t)dtrace_probe;
probe(id, uargs[0], uargs[1], uargs[2], uargs[3], uargs[4], uargs[5],
uargs[6], uargs[7]);
}
#endif
static void
systrace_getargdesc(void *arg, dtrace_id_t id, void *parg, dtrace_argdesc_t *desc)
{
int sysnum = SYSTRACE_SYSNUM((uintptr_t)parg);
if (SYSTRACE_ISENTRY((uintptr_t)parg))
systrace_entry_setargdesc(sysnum, desc->dtargd_ndx,
desc->dtargd_native, sizeof(desc->dtargd_native));
else
systrace_return_setargdesc(sysnum, desc->dtargd_ndx,
desc->dtargd_native, sizeof(desc->dtargd_native));
if (desc->dtargd_native[0] == '\0')
desc->dtargd_ndx = DTRACE_ARGNONE;
return;
}
static void
systrace_provide(void *arg, dtrace_probedesc_t *desc)
{
int i;
if (desc != NULL)
return;
for (i = 0; i < MAXSYSCALL; i++) {
if (dtrace_probe_lookup(systrace_id, MODNAME,
uglyhack.pp_syscallnames[i], "entry") != 0)
continue;
(void) dtrace_probe_create(systrace_id, MODNAME, uglyhack.pp_syscallnames[i],
"entry", SYSTRACE_ARTIFICIAL_FRAMES,
(void *)((uintptr_t)SYSTRACE_ENTRY(i)));
(void) dtrace_probe_create(systrace_id, MODNAME, uglyhack.pp_syscallnames[i],
"return", SYSTRACE_ARTIFICIAL_FRAMES,
(void *)((uintptr_t)SYSTRACE_RETURN(i)));
}
}
static void
systrace_destroy(void *arg, dtrace_id_t id, void *parg)
{
#ifdef DEBUG
int sysnum = SYSTRACE_SYSNUM((uintptr_t)parg);
/*
* There's nothing to do here but assert that we have actually been
* disabled.
*/
if (SYSTRACE_ISENTRY((uintptr_t)parg)) {
ASSERT(sysent[sysnum].sy_entry == 0);
} else {
ASSERT(sysent[sysnum].sy_return == 0);
}
#endif
}
static void
systrace_enable(void *arg, dtrace_id_t id, void *parg)
{
int sysnum = SYSTRACE_SYSNUM((uintptr_t)parg);
if (SYSENT[sysnum].sy_systrace_args_func == NULL)
SYSENT[sysnum].sy_systrace_args_func = systrace_args;
if (SYSTRACE_ISENTRY((uintptr_t)parg))
SYSENT[sysnum].sy_entry = id;
else
SYSENT[sysnum].sy_return = id;
}
static void
systrace_disable(void *arg, dtrace_id_t id, void *parg)
{
int sysnum = SYSTRACE_SYSNUM((uintptr_t)parg);
SYSENT[sysnum].sy_entry = 0;
SYSENT[sysnum].sy_return = 0;
}
static void
systrace_load(void *dummy)
{
/* Create the /dev/dtrace/systrace entry. */
systrace_cdev = make_dev(&systrace_cdevsw, 0, UID_ROOT, GID_WHEEL, 0600,
DEVNAME);
if (dtrace_register(PROVNAME, &systrace_attr, DTRACE_PRIV_USER,
NULL, &systrace_pops, NULL, &systrace_id) != 0)
return;
#if !defined(LINUX_SYSTRACE)
systrace_probe_func = systrace_probe;
#endif
}
static int
systrace_unload()
{
int error = 0;
if ((error = dtrace_unregister(systrace_id)) != 0)
return (error);
#if !defined(LINUX_SYSTRACE)
systrace_probe_func = NULL;
#endif
destroy_dev(systrace_cdev);
return (error);
}
static int
systrace_modevent(module_t mod __unused, int type, void *data __unused)
{
int error = 0;
switch (type) {
case MOD_LOAD:
break;
case MOD_UNLOAD:
break;
case MOD_SHUTDOWN:
break;
default:
error = EOPNOTSUPP;
break;
}
return (error);
}
static int
systrace_open(struct cdev *dev __unused, int oflags __unused, int devtype __unused, struct thread *td __unused)
{
return (0);
}
SYSINIT(systrace_load, SI_SUB_DTRACE_PROVIDER, SI_ORDER_ANY, systrace_load, NULL);
SYSUNINIT(systrace_unload, SI_SUB_DTRACE_PROVIDER, SI_ORDER_ANY, systrace_unload, NULL);
#ifdef LINUX_SYSTRACE
DEV_MODULE(systrace_linux32, systrace_modevent, NULL);
MODULE_VERSION(systrace_linux32, 1);
MODULE_DEPEND(systrace_linux32, linux, 1, 1, 1);
MODULE_DEPEND(systrace_linux32, dtrace, 1, 1, 1);
MODULE_DEPEND(systrace_linux32, opensolaris, 1, 1, 1);
#elif defined(FREEBSD32_SYSTRACE)
DEV_MODULE(systrace_freebsd32, systrace_modevent, NULL);
MODULE_VERSION(systrace_freebsd32, 1);
MODULE_DEPEND(systrace_freebsd32, dtrace, 1, 1, 1);
MODULE_DEPEND(systrace_freebsd32, opensolaris, 1, 1, 1);
#else
DEV_MODULE(systrace, systrace_modevent, NULL);
MODULE_VERSION(systrace, 1);
MODULE_DEPEND(systrace, dtrace, 1, 1, 1);
MODULE_DEPEND(systrace, opensolaris, 1, 1, 1);
#endif
|