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"
44 __FBSDID("$FreeBSD$");
46 #include <sys/capsicum.h>
48 #include <sys/vmmeter.h>
51 #include <sys/ktrace.h>
53 #include <security/audit/audit.h>
56 syscallenter(struct thread *td)
59 struct syscall_args *sa;
62 PCPU_INC(cnt.v_syscall);
67 if (td->td_cowgen != p->p_cowgen)
68 thread_cow_update(td);
69 traced = (p->p_flag & P_TRACED) != 0;
70 if (traced || td->td_dbgflags & TDB_USERWR) {
72 td->td_dbgflags &= ~TDB_USERWR;
74 td->td_dbgflags |= TDB_SCE;
77 error = (p->p_sysent->sv_fetch_syscall_args)(td);
79 if (KTRPOINT(td, KTR_SYSCALL))
80 ktrsyscall(sa->code, sa->narg, sa->args);
82 KTR_START4(KTR_SYSC, "syscall", syscallname(p, sa->code),
83 (uintptr_t)td, "pid:%d", td->td_proc->p_pid, "arg0:%p", sa->args[0],
84 "arg1:%p", sa->args[1], "arg2:%p", sa->args[2]);
88 STOPEVENT(p, S_SCE, sa->narg);
89 if (p->p_flag & P_TRACED) {
91 if (p->p_ptevents & PTRACE_SCE)
92 ptracestop((td), SIGTRAP, NULL);
95 if (td->td_dbgflags & TDB_USERWR) {
97 * Reread syscall number and arguments if
98 * debugger modified registers or memory.
100 error = (p->p_sysent->sv_fetch_syscall_args)(td);
102 if (KTRPOINT(td, KTR_SYSCALL))
103 ktrsyscall(sa->code, sa->narg, sa->args);
109 #ifdef CAPABILITY_MODE
111 * In capability mode, we only allow access to system calls
112 * flagged with SYF_CAPENABLED.
114 if (IN_CAPABILITY_MODE(td) &&
115 !(sa->callp->sy_flags & SYF_CAPENABLED)) {
121 error = syscall_thread_enter(td, sa->callp);
126 /* Give the syscall:::entry DTrace probe a chance to fire. */
127 if (systrace_probe_func != NULL && sa->callp->sy_entry != 0)
128 (*systrace_probe_func)(sa, SYSTRACE_ENTRY, 0);
131 AUDIT_SYSCALL_ENTER(sa->code, td);
132 error = (sa->callp->sy_call)(td, sa->args);
133 AUDIT_SYSCALL_EXIT(error, td);
135 /* Save the latest error return value. */
136 if ((td->td_pflags & TDP_NERRNO) == 0)
137 td->td_errno = error;
140 /* Give the syscall:::return DTrace probe a chance to fire. */
141 if (systrace_probe_func != NULL && sa->callp->sy_return != 0)
142 (*systrace_probe_func)(sa, SYSTRACE_RETURN,
143 error ? -1 : td->td_retval[0]);
145 syscall_thread_exit(td, sa->callp);
148 KTR_STOP4(KTR_SYSC, "syscall", syscallname(p, sa->code),
149 (uintptr_t)td, "pid:%d", td->td_proc->p_pid, "error:%d", error,
150 "retval0:%#lx", td->td_retval[0], "retval1:%#lx",
154 td->td_dbgflags &= ~TDB_SCE;
157 (p->p_sysent->sv_set_syscall_retval)(td, error);
162 syscallret(struct thread *td, int error)
165 struct syscall_args *sa;
169 KASSERT((td->td_pflags & TDP_FORKING) == 0,
170 ("fork() did not clear TDP_FORKING upon completion"));
174 if ((trap_enotcap || (p->p_flag2 & P2_TRAPCAP) != 0) &&
175 IN_CAPABILITY_MODE(td)) {
176 error1 = (td->td_pflags & TDP_NERRNO) == 0 ? error :
178 if (error1 == ENOTCAPABLE || error1 == ECAPMODE) {
179 ksiginfo_init_trap(&ksi);
180 ksi.ksi_signo = SIGTRAP;
181 ksi.ksi_errno = error1;
182 ksi.ksi_code = TRAP_CAP;
183 trapsignal(td, &ksi);
188 * Handle reschedule and other end-of-syscall issues
190 userret(td, td->td_frame);
193 if (KTRPOINT(td, KTR_SYSRET)) {
194 ktrsysret(sa->code, (td->td_pflags & TDP_NERRNO) == 0 ?
195 error : td->td_errno, td->td_retval[0]);
198 td->td_pflags &= ~TDP_NERRNO;
200 if (p->p_flag & P_TRACED) {
203 td->td_dbgflags |= TDB_SCX;
208 * This works because errno is findable through the
209 * register set. If we ever support an emulation where this
210 * is not the case, this code will need to be revisited.
212 STOPEVENT(p, S_SCX, sa->code);
213 if (traced || (td->td_dbgflags & (TDB_EXEC | TDB_FORK)) != 0) {
216 * If tracing the execed process, trap to the debugger
217 * so that breakpoints can be set before the program
218 * executes. If debugger requested tracing of syscall
219 * returns, do it now too.
222 ((td->td_dbgflags & (TDB_FORK | TDB_EXEC)) != 0 ||
223 (p->p_ptevents & PTRACE_SCX) != 0))
224 ptracestop(td, SIGTRAP, NULL);
225 td->td_dbgflags &= ~(TDB_SCX | TDB_EXEC | TDB_FORK);
229 if (td->td_pflags & TDP_RFPPWAIT) {
231 * Preserve synchronization semantics of vfork. If
232 * waiting for child to exec or exit, fork set
233 * P_PPWAIT on child, and there we sleep on our proc
236 * Do it after the ptracestop() above is finished, to
237 * not block our debugger until child execs or exits
238 * to finish vfork wait.
240 td->td_pflags &= ~TDP_RFPPWAIT;
241 p2 = td->td_rfppwait_p;
244 while (p2->p_flag & P_PPWAIT) {
246 if (thread_suspend_check_needed()) {
248 thread_suspend_check(0);
254 cv_timedwait(&p2->p_pwait, &p2->p_mtx, hz);
258 if (td->td_dbgflags & TDB_VFORK) {
260 if (p->p_ptevents & PTRACE_VFORK)
261 ptracestop(td, SIGTRAP, NULL);
262 td->td_dbgflags &= ~TDB_VFORK;