2 * Copyright (C) 1994, David Greenman
3 * Copyright (c) 1990, 1993
4 * The Regents of the University of California. All rights reserved.
5 * Copyright (C) 2010 Konstantin Belousov <kib@freebsd.org>
7 * This code is derived from software contributed to Berkeley by
8 * the University of Utah, and William Jolitz.
10 * Redistribution and use in source and binary forms, with or without
11 * modification, are permitted provided that the following conditions
13 * 1. Redistributions of source code must retain the above copyright
14 * notice, this list of conditions and the following disclaimer.
15 * 2. Redistributions in binary form must reproduce the above copyright
16 * notice, this list of conditions and the following disclaimer in the
17 * documentation and/or other materials provided with the distribution.
18 * 3. All advertising materials mentioning features or use of this software
19 * must display the following acknowledgement:
20 * This product includes software developed by the University of
21 * California, Berkeley and its contributors.
22 * 4. Neither the name of the University nor the names of its contributors
23 * may be used to endorse or promote products derived from this software
24 * without specific prior written permission.
26 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
27 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
28 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
29 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
30 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
31 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
32 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
33 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
34 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
35 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
38 * from: @(#)trap.c 7.4 (Berkeley) 5/13/91
41 #include "opt_capsicum.h"
42 #include "opt_ktrace.h"
43 #include "opt_kdtrace.h"
45 __FBSDID("$FreeBSD$");
47 #include <sys/capability.h>
51 #include <sys/ktrace.h>
53 #include <security/audit/audit.h>
56 syscallenter(struct thread *td, struct syscall_args *sa)
61 PCPU_INC(cnt.v_syscall);
65 if (td->td_ucred != p->p_ucred)
66 cred_update_thread(td);
67 traced = (p->p_flag & P_TRACED) != 0;
68 if (traced || td->td_dbgflags & TDB_USERWR) {
70 td->td_dbgflags &= ~TDB_USERWR;
72 td->td_dbgflags |= TDB_SCE;
75 error = (p->p_sysent->sv_fetch_syscall_args)(td, sa);
77 if (KTRPOINT(td, KTR_SYSCALL))
78 ktrsyscall(sa->code, sa->narg, sa->args);
82 "syscall: td=%p pid %d %s (%#lx, %#lx, %#lx)",
83 td, td->td_proc->p_pid, syscallname(p, sa->code),
84 sa->args[0], sa->args[1], sa->args[2]);
87 STOPEVENT(p, S_SCE, sa->narg);
88 if (p->p_flag & P_TRACED) {
90 td->td_dbg_sc_code = sa->code;
91 td->td_dbg_sc_narg = sa->narg;
92 if (p->p_stops & S_PT_SCE)
93 ptracestop((td), SIGTRAP);
96 if (td->td_dbgflags & TDB_USERWR) {
98 * Reread syscall number and arguments if
99 * debugger modified registers or memory.
101 error = (p->p_sysent->sv_fetch_syscall_args)(td, sa);
103 td->td_dbg_sc_code = sa->code;
104 td->td_dbg_sc_narg = sa->narg;
107 if (KTRPOINT(td, KTR_SYSCALL))
108 ktrsyscall(sa->code, sa->narg, sa->args);
114 #ifdef CAPABILITY_MODE
116 * In capability mode, we only allow access to system calls
117 * flagged with SYF_CAPENABLED.
119 if (IN_CAPABILITY_MODE(td) &&
120 !(sa->callp->sy_flags & SYF_CAPENABLED)) {
126 error = syscall_thread_enter(td, sa->callp);
132 * If the systrace module has registered it's probe
133 * callback and if there is a probe active for the
134 * syscall 'entry', process the probe.
136 if (systrace_probe_func != NULL && sa->callp->sy_entry != 0)
137 (*systrace_probe_func)(sa->callp->sy_entry, sa->code,
138 sa->callp, sa->args, 0);
141 AUDIT_SYSCALL_ENTER(sa->code, td);
142 error = (sa->callp->sy_call)(td, sa->args);
143 AUDIT_SYSCALL_EXIT(error, td);
145 /* Save the latest error return value. */
146 if ((td->td_pflags & TDP_NERRNO) == 0)
147 td->td_errno = error;
151 * If the systrace module has registered it's probe
152 * callback and if there is a probe active for the
153 * syscall 'return', process the probe.
155 if (systrace_probe_func != NULL && sa->callp->sy_return != 0)
156 (*systrace_probe_func)(sa->callp->sy_return, sa->code,
157 sa->callp, NULL, (error) ? -1 : td->td_retval[0]);
159 syscall_thread_exit(td, sa->callp);
160 CTR4(KTR_SYSC, "syscall: p=%p error=%d return %#lx %#lx",
161 p, error, td->td_retval[0], td->td_retval[1]);
166 td->td_dbgflags &= ~TDB_SCE;
169 (p->p_sysent->sv_set_syscall_retval)(td, error);
174 syscallret(struct thread *td, int error, struct syscall_args *sa __unused)
182 * Check for misbehavior.
184 WITNESS_WARN(WARN_PANIC, NULL, "System call %s returning",
185 syscallname(p, sa->code));
186 KASSERT(td->td_critnest == 0,
187 ("System call %s returning in a critical section",
188 syscallname(p, sa->code)));
189 KASSERT(td->td_locks == 0,
190 ("System call %s returning with %d locks held",
191 syscallname(p, sa->code), td->td_locks));
192 KASSERT((td->td_pflags & TDP_NOFAULTING) == 0,
193 ("System call %s returning with pagefaults disabled",
194 syscallname(p, sa->code)));
195 KASSERT((td->td_pflags & TDP_NOSLEEPING) == 0,
196 ("System call %s returning with sleep disabled",
197 syscallname(p, sa->code)));
200 * Handle reschedule and other end-of-syscall issues
202 userret(td, td->td_frame);
204 CTR4(KTR_SYSC, "syscall %s exit thread %p pid %d proc %s",
205 syscallname(p, sa->code), td, td->td_proc->p_pid, td->td_name);
208 if (KTRPOINT(td, KTR_SYSRET)) {
209 ktrsysret(sa->code, (td->td_pflags & TDP_NERRNO) == 0 ?
210 error : td->td_errno, td->td_retval[0]);
213 td->td_pflags &= ~TDP_NERRNO;
215 if (p->p_flag & P_TRACED) {
218 td->td_dbgflags |= TDB_SCX;
223 * This works because errno is findable through the
224 * register set. If we ever support an emulation where this
225 * is not the case, this code will need to be revisited.
227 STOPEVENT(p, S_SCX, sa->code);
228 if (traced || (td->td_dbgflags & (TDB_EXEC | TDB_FORK)) != 0) {
231 * If tracing the execed process, trap to the debugger
232 * so that breakpoints can be set before the program
233 * executes. If debugger requested tracing of syscall
234 * returns, do it now too.
237 ((td->td_dbgflags & (TDB_FORK | TDB_EXEC)) != 0 ||
238 (p->p_stops & S_PT_SCX) != 0))
239 ptracestop(td, SIGTRAP);
240 td->td_dbgflags &= ~(TDB_SCX | TDB_EXEC | TDB_FORK);
244 if (td->td_pflags & TDP_RFPPWAIT) {
246 * Preserve synchronization semantics of vfork. If
247 * waiting for child to exec or exit, fork set
248 * P_PPWAIT on child, and there we sleep on our proc
251 * Do it after the ptracestop() above is finished, to
252 * not block our debugger until child execs or exits
253 * to finish vfork wait.
255 td->td_pflags &= ~TDP_RFPPWAIT;
256 p2 = td->td_rfppwait_p;
258 while (p2->p_flag & P_PPWAIT)
259 cv_wait(&p2->p_pwait, &p2->p_mtx);