1 /*-
2 * Copyright (c) 2013 Dmitry Chagin
3 * Copyright (c) 2004 Tim J. Robbins
4 * Copyright (c) 2002 Doug Rabson
5 * Copyright (c) 2000 Marcel Moolenaar
6 * All rights reserved.
7 *
8 * Redistribution and use in source and binary forms, with or without
9 * modification, are permitted provided that the following conditions
10 * are met:
11 * 1. Redistributions of source code must retain the above copyright
12 * notice, this list of conditions and the following disclaimer
13 * in this position and unchanged.
14 * 2. Redistributions in binary form must reproduce the above copyright
15 * notice, this list of conditions and the following disclaimer in the
16 * documentation and/or other materials provided with the distribution.
17 * 3. The name of the author may not be used to endorse or promote products
18 * derived from this software without specific prior written permission.
19 *
20 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
21 * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
22 * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
23 * IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
24 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
25 * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
26 * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
27 * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
28 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
29 * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
30 */
31
32 #include <sys/cdefs.h>
33 __FBSDID("$FreeBSD$");
34
35 #include <sys/param.h>
36 #include <sys/capsicum.h>
37 #include <sys/clock.h>
38 #include <sys/dirent.h>
39 #include <sys/fcntl.h>
40 #include <sys/file.h>
41 #include <sys/filedesc.h>
42 #include <sys/imgact.h>
43 #include <sys/kernel.h>
44 #include <sys/ktr.h>
45 #include <sys/limits.h>
46 #include <sys/lock.h>
47 #include <sys/malloc.h>
48 #include <sys/mman.h>
49 #include <sys/mutex.h>
50 #include <sys/priv.h>
51 #include <sys/proc.h>
52 #include <sys/resource.h>
53 #include <sys/resourcevar.h>
54 #include <sys/sched.h>
55 #include <sys/syscallsubr.h>
56 #include <sys/sysproto.h>
57 #include <sys/systm.h>
58 #include <sys/unistd.h>
59 #include <sys/vnode.h>
60 #include <sys/wait.h>
61
62 #include <security/mac/mac_framework.h>
63
64 #include <ufs/ufs/extattr.h>
65 #include <ufs/ufs/quota.h>
66 #include <ufs/ufs/ufsmount.h>
67
68 #include <machine/frame.h>
69 #include <machine/md_var.h>
70 #include <machine/pcb.h>
71 #include <machine/psl.h>
72 #include <machine/segments.h>
73 #include <machine/specialreg.h>
74
75 #include <vm/pmap.h>
76 #include <vm/vm.h>
77 #include <vm/vm_extern.h>
78 #include <vm/vm_kern.h>
79 #include <vm/vm_map.h>
80
81 #include <x86/ifunc.h>
82 #include <x86/sysarch.h>
83
84 #include <security/audit/audit.h>
85
86 #include <amd64/linux/linux.h>
87 #include <amd64/linux/linux_proto.h>
88 #include <compat/linux/linux_emul.h>
89 #include <compat/linux/linux_file.h>
90 #include <compat/linux/linux_ipc.h>
91 #include <compat/linux/linux_misc.h>
92 #include <compat/linux/linux_mmap.h>
93 #include <compat/linux/linux_signal.h>
94 #include <compat/linux/linux_util.h>
95
96 int
linux_execve(struct thread * td,struct linux_execve_args * args)97 linux_execve(struct thread *td, struct linux_execve_args *args)
98 {
99 struct image_args eargs;
100 char *path;
101 int error;
102
103 LINUX_CTR(execve);
104
105 if (!LUSECONVPATH(td)) {
106 error = exec_copyin_args(&eargs, args->path, UIO_USERSPACE,
107 args->argp, args->envp);
108 } else {
109 LCONVPATHEXIST(td, args->path, &path);
110 error = exec_copyin_args(&eargs, path, UIO_SYSSPACE, args->argp,
111 args->envp);
112 LFREEPATH(path);
113 }
114 if (error == 0)
115 error = linux_common_execve(td, &eargs);
116 AUDIT_SYSCALL_EXIT(error == EJUSTRETURN ? 0 : error, td);
117 return (error);
118 }
119
120 int
linux_set_upcall(struct thread * td,register_t stack)121 linux_set_upcall(struct thread *td, register_t stack)
122 {
123
124 if (stack)
125 td->td_frame->tf_rsp = stack;
126
127 /*
128 * The newly created Linux thread returns
129 * to the user space by the same path that a parent does.
130 */
131 td->td_frame->tf_rax = 0;
132 return (0);
133 }
134
135 int
linux_mmap2(struct thread * td,struct linux_mmap2_args * args)136 linux_mmap2(struct thread *td, struct linux_mmap2_args *args)
137 {
138
139 return (linux_mmap_common(td, PTROUT(args->addr), args->len, args->prot,
140 args->flags, args->fd, args->pgoff));
141 }
142
143 int
linux_mprotect(struct thread * td,struct linux_mprotect_args * uap)144 linux_mprotect(struct thread *td, struct linux_mprotect_args *uap)
145 {
146
147 return (linux_mprotect_common(td, PTROUT(uap->addr), uap->len, uap->prot));
148 }
149
150 int
linux_madvise(struct thread * td,struct linux_madvise_args * uap)151 linux_madvise(struct thread *td, struct linux_madvise_args *uap)
152 {
153
154 return (linux_madvise_common(td, PTROUT(uap->addr), uap->len, uap->behav));
155 }
156
157 int
linux_iopl(struct thread * td,struct linux_iopl_args * args)158 linux_iopl(struct thread *td, struct linux_iopl_args *args)
159 {
160 int error;
161
162 LINUX_CTR(iopl);
163
164 if (args->level > 3)
165 return (EINVAL);
166 if ((error = priv_check(td, PRIV_IO)) != 0)
167 return (error);
168 if ((error = securelevel_gt(td->td_ucred, 0)) != 0)
169 return (error);
170 td->td_frame->tf_rflags = (td->td_frame->tf_rflags & ~PSL_IOPL) |
171 (args->level * (PSL_IOPL / 3));
172
173 return (0);
174 }
175
176 int
linux_pause(struct thread * td,struct linux_pause_args * args)177 linux_pause(struct thread *td, struct linux_pause_args *args)
178 {
179 struct proc *p = td->td_proc;
180 sigset_t sigmask;
181
182 LINUX_CTR(pause);
183
184 PROC_LOCK(p);
185 sigmask = td->td_sigmask;
186 PROC_UNLOCK(p);
187 return (kern_sigsuspend(td, sigmask));
188 }
189
190 int
linux_sigaltstack(struct thread * td,struct linux_sigaltstack_args * uap)191 linux_sigaltstack(struct thread *td, struct linux_sigaltstack_args *uap)
192 {
193 stack_t ss, oss;
194 l_stack_t lss;
195 int error;
196
197 memset(&lss, 0, sizeof(lss));
198 LINUX_CTR2(sigaltstack, "%p, %p", uap->uss, uap->uoss);
199
200 if (uap->uss != NULL) {
201 error = copyin(uap->uss, &lss, sizeof(l_stack_t));
202 if (error)
203 return (error);
204
205 ss.ss_sp = PTRIN(lss.ss_sp);
206 ss.ss_size = lss.ss_size;
207 ss.ss_flags = linux_to_bsd_sigaltstack(lss.ss_flags);
208 }
209 error = kern_sigaltstack(td, (uap->uss != NULL) ? &ss : NULL,
210 (uap->uoss != NULL) ? &oss : NULL);
211 if (!error && uap->uoss != NULL) {
212 lss.ss_sp = PTROUT(oss.ss_sp);
213 lss.ss_size = oss.ss_size;
214 lss.ss_flags = bsd_to_linux_sigaltstack(oss.ss_flags);
215 error = copyout(&lss, uap->uoss, sizeof(l_stack_t));
216 }
217
218 return (error);
219 }
220
221 int
linux_arch_prctl(struct thread * td,struct linux_arch_prctl_args * args)222 linux_arch_prctl(struct thread *td, struct linux_arch_prctl_args *args)
223 {
224 unsigned long long cet[3];
225 struct pcb *pcb;
226 int error;
227
228 pcb = td->td_pcb;
229 LINUX_CTR2(arch_prctl, "0x%x, %p", args->code, args->addr);
230
231 switch (args->code) {
232 case LINUX_ARCH_SET_GS:
233 if (args->addr < VM_MAXUSER_ADDRESS) {
234 update_pcb_bases(pcb);
235 pcb->pcb_gsbase = args->addr;
236 td->td_frame->tf_gs = _ugssel;
237 error = 0;
238 } else
239 error = EPERM;
240 break;
241 case LINUX_ARCH_SET_FS:
242 if (args->addr < VM_MAXUSER_ADDRESS) {
243 update_pcb_bases(pcb);
244 pcb->pcb_fsbase = args->addr;
245 td->td_frame->tf_fs = _ufssel;
246 error = 0;
247 } else
248 error = EPERM;
249 break;
250 case LINUX_ARCH_GET_FS:
251 error = copyout(&pcb->pcb_fsbase, PTRIN(args->addr),
252 sizeof(args->addr));
253 break;
254 case LINUX_ARCH_GET_GS:
255 error = copyout(&pcb->pcb_gsbase, PTRIN(args->addr),
256 sizeof(args->addr));
257 break;
258 case LINUX_ARCH_CET_STATUS:
259 memset(cet, 0, sizeof(cet));
260 error = copyout(&cet, PTRIN(args->addr), sizeof(cet));
261 break;
262 default:
263 linux_msg(td, "unsupported arch_prctl code %#x", args->code);
264 error = EINVAL;
265 }
266 return (error);
267 }
268
269 int
linux_set_cloned_tls(struct thread * td,void * desc)270 linux_set_cloned_tls(struct thread *td, void *desc)
271 {
272 struct pcb *pcb;
273
274 if ((uint64_t)desc >= VM_MAXUSER_ADDRESS)
275 return (EPERM);
276
277 pcb = td->td_pcb;
278 update_pcb_bases(pcb);
279 pcb->pcb_fsbase = (register_t)desc;
280 td->td_frame->tf_fs = _ufssel;
281
282 return (0);
283 }
284
285 int futex_xchgl_nosmap(int oparg, uint32_t *uaddr, int *oldval);
286 int futex_xchgl_smap(int oparg, uint32_t *uaddr, int *oldval);
287 DEFINE_IFUNC(, int, futex_xchgl, (int, uint32_t *, int *))
288 {
289
290 return ((cpu_stdext_feature & CPUID_STDEXT_SMAP) != 0 ?
291 futex_xchgl_smap : futex_xchgl_nosmap);
292 }
293
294 int futex_addl_nosmap(int oparg, uint32_t *uaddr, int *oldval);
295 int futex_addl_smap(int oparg, uint32_t *uaddr, int *oldval);
296 DEFINE_IFUNC(, int, futex_addl, (int, uint32_t *, int *))
297 {
298
299 return ((cpu_stdext_feature & CPUID_STDEXT_SMAP) != 0 ?
300 futex_addl_smap : futex_addl_nosmap);
301 }
302
303 int futex_orl_nosmap(int oparg, uint32_t *uaddr, int *oldval);
304 int futex_orl_smap(int oparg, uint32_t *uaddr, int *oldval);
305 DEFINE_IFUNC(, int, futex_orl, (int, uint32_t *, int *))
306 {
307
308 return ((cpu_stdext_feature & CPUID_STDEXT_SMAP) != 0 ?
309 futex_orl_smap : futex_orl_nosmap);
310 }
311
312 int futex_andl_nosmap(int oparg, uint32_t *uaddr, int *oldval);
313 int futex_andl_smap(int oparg, uint32_t *uaddr, int *oldval);
314 DEFINE_IFUNC(, int, futex_andl, (int, uint32_t *, int *))
315 {
316
317 return ((cpu_stdext_feature & CPUID_STDEXT_SMAP) != 0 ?
318 futex_andl_smap : futex_andl_nosmap);
319 }
320
321 int futex_xorl_nosmap(int oparg, uint32_t *uaddr, int *oldval);
322 int futex_xorl_smap(int oparg, uint32_t *uaddr, int *oldval);
323 DEFINE_IFUNC(, int, futex_xorl, (int, uint32_t *, int *))
324 {
325
326 return ((cpu_stdext_feature & CPUID_STDEXT_SMAP) != 0 ?
327 futex_xorl_smap : futex_xorl_nosmap);
328 }
329