linux_machdep.c revision 1.51.6.3
1 /* $NetBSD: linux_machdep.c,v 1.51.6.3 2020/01/21 18:12:54 martin Exp $ */ 2 3 /*- 4 * Copyright (c) 2005 Emmanuel Dreyfus, all rights reserved. 5 * 6 * Redistribution and use in source and binary forms, with or without 7 * modification, are permitted provided that the following conditions 8 * are met: 9 * 1. Redistributions of source code must retain the above copyright 10 * notice, this list of conditions and the following disclaimer. 11 * 2. Redistributions in binary form must reproduce the above copyright 12 * notice, this list of conditions and the following disclaimer in the 13 * documentation and/or other materials provided with the distribution. 14 * 3. All advertising materials mentioning features or use of this software 15 * must display the following acknowledgement: 16 * This product includes software developed by Emmanuel Dreyfus 17 * 4. The name of the author may not be used to endorse or promote 18 * products derived from this software without specific prior written 19 * permission. 20 * 21 * THIS SOFTWARE IS PROVIDED BY THE THE AUTHOR AND CONTRIBUTORS ``AS IS'' 22 * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, 23 * THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR 24 * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS 25 * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR 26 * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF 27 * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS 28 * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN 29 * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) 30 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE 31 * POSSIBILITY OF SUCH DAMAGE. 32 */ 33 34 #include <sys/cdefs.h> 35 36 __KERNEL_RCSID(0, "$NetBSD: linux_machdep.c,v 1.51.6.3 2020/01/21 18:12:54 martin Exp $"); 37 38 #include <sys/param.h> 39 #include <sys/types.h> 40 #include <sys/systm.h> 41 #include <sys/signal.h> 42 #include <sys/exec.h> 43 #include <sys/proc.h> 44 #include <sys/ptrace.h> /* for process_read_fpregs() */ 45 #include <sys/ucontext.h> 46 #include <sys/conf.h> 47 48 #include <machine/reg.h> 49 #include <machine/pcb.h> 50 #include <machine/mcontext.h> 51 #include <machine/specialreg.h> 52 #include <machine/vmparam.h> 53 #include <machine/cpufunc.h> 54 #include <x86/include/sysarch.h> 55 56 /* 57 * To see whether wscons is configured (for virtual console ioctl calls). 58 */ 59 #if defined(_KERNEL_OPT) 60 #include "wsdisplay.h" 61 #endif 62 #if (NWSDISPLAY > 0) 63 #include <dev/wscons/wsconsio.h> 64 #include <dev/wscons/wsdisplay_usl_io.h> 65 #endif 66 67 68 #include <compat/linux/common/linux_signal.h> 69 #include <compat/linux/common/linux_errno.h> 70 #include <compat/linux/common/linux_exec.h> 71 #include <compat/linux/common/linux_ioctl.h> 72 #include <compat/linux/common/linux_prctl.h> 73 #include <compat/linux/common/linux_machdep.h> 74 #include <compat/linux/common/linux_ipc.h> 75 #include <compat/linux/common/linux_sem.h> 76 #include <compat/linux/linux_syscall.h> 77 #include <compat/linux/linux_syscallargs.h> 78 79 static void linux_buildcontext(struct lwp *, void *, void *); 80 81 void 82 linux_setregs(struct lwp *l, struct exec_package *epp, vaddr_t stack) 83 { 84 struct pcb *pcb = lwp_getpcb(l); 85 struct trapframe *tf; 86 87 #ifdef USER_LDT 88 pmap_ldt_cleanup(l); 89 #endif 90 91 fpu_save_area_clear(l, __NetBSD_NPXCW__); 92 93 kpreempt_disable(); 94 pcb->pcb_flags = 0; 95 l->l_proc->p_flag &= ~PK_32; 96 l->l_md.md_flags = MDL_IRET; 97 cpu_fsgs_zero(l); 98 kpreempt_enable(); 99 100 tf = l->l_md.md_regs; 101 tf->tf_rax = 0; 102 tf->tf_rbx = 0; 103 tf->tf_rcx = epp->ep_entry; 104 tf->tf_rdx = 0; 105 tf->tf_rsi = 0; 106 tf->tf_rdi = 0; 107 tf->tf_rbp = 0; 108 tf->tf_rsp = stack; 109 tf->tf_r8 = 0; 110 tf->tf_r9 = 0; 111 tf->tf_r10 = 0; 112 tf->tf_r11 = 0; 113 tf->tf_r12 = 0; 114 tf->tf_r13 = 0; 115 tf->tf_r14 = 0; 116 tf->tf_r15 = 0; 117 tf->tf_rip = epp->ep_entry; 118 tf->tf_rflags = PSL_USERSET; 119 tf->tf_cs = GSEL(GUCODE_SEL, SEL_UPL); 120 tf->tf_ss = GSEL(GUDATA_SEL, SEL_UPL); 121 tf->tf_ds = GSEL(GUDATA_SEL, SEL_UPL); 122 tf->tf_es = 0; 123 124 return; 125 } 126 127 void 128 linux_sendsig(const ksiginfo_t *ksi, const sigset_t *mask) 129 { 130 struct lwp *l = curlwp; 131 struct proc *p = l->l_proc; 132 struct pcb *pcb = lwp_getpcb(l); 133 struct sigacts *ps = p->p_sigacts; 134 int onstack, error; 135 int sig = ksi->ksi_signo; 136 struct linux_rt_sigframe *sfp, sigframe; 137 struct linux__fpstate *fpsp; 138 struct fpreg fpregs; 139 struct trapframe *tf = l->l_md.md_regs; 140 sig_t catcher = SIGACTION(p, sig).sa_handler; 141 linux_sigset_t lmask; 142 char *sp; 143 144 /* Do we need to jump onto the signal stack? */ 145 onstack = 146 (l->l_sigstk.ss_flags & (SS_DISABLE | SS_ONSTACK)) == 0 && 147 (SIGACTION(p, sig).sa_flags & SA_ONSTACK) != 0; 148 149 /* Allocate space for the signal handler context. */ 150 if (onstack) 151 sp = ((char *)l->l_sigstk.ss_sp + 152 l->l_sigstk.ss_size); 153 else 154 sp = (char *)tf->tf_rsp - 128; 155 156 /* Save FPU state */ 157 sp = (char *) (((long)sp - sizeof (*fpsp)) & ~0xfUL); 158 fpsp = (struct linux__fpstate *)sp; 159 160 /* 161 * Populate the rt_sigframe 162 */ 163 sp = (char *) 164 ((((long)sp - sizeof(struct linux_rt_sigframe)) & ~0xfUL) - 8); 165 sfp = (struct linux_rt_sigframe *)sp; 166 167 memset(&sigframe, 0, sizeof(sigframe)); 168 if (ps->sa_sigdesc[sig].sd_vers != 0) 169 sigframe.pretcode = 170 (char *)(u_long)ps->sa_sigdesc[sig].sd_tramp; 171 else 172 sigframe.pretcode = NULL; 173 174 /* 175 * The user context 176 */ 177 sigframe.uc.luc_flags = 0; 178 sigframe.uc.luc_link = NULL; 179 180 /* This is used regardless of SA_ONSTACK in Linux */ 181 sigframe.uc.luc_stack.ss_sp = l->l_sigstk.ss_sp; 182 sigframe.uc.luc_stack.ss_size = l->l_sigstk.ss_size; 183 sigframe.uc.luc_stack.ss_flags = 0; 184 if (l->l_sigstk.ss_flags & SS_ONSTACK) 185 sigframe.uc.luc_stack.ss_flags |= LINUX_SS_ONSTACK; 186 if (l->l_sigstk.ss_flags & SS_DISABLE) 187 sigframe.uc.luc_stack.ss_flags |= LINUX_SS_DISABLE; 188 189 sigframe.uc.luc_mcontext.r8 = tf->tf_r8; 190 sigframe.uc.luc_mcontext.r9 = tf->tf_r9; 191 sigframe.uc.luc_mcontext.r10 = tf->tf_r10; 192 sigframe.uc.luc_mcontext.r11 = tf->tf_r11; 193 sigframe.uc.luc_mcontext.r12 = tf->tf_r12; 194 sigframe.uc.luc_mcontext.r13 = tf->tf_r13; 195 sigframe.uc.luc_mcontext.r14 = tf->tf_r14; 196 sigframe.uc.luc_mcontext.r15 = tf->tf_r15; 197 sigframe.uc.luc_mcontext.rdi = tf->tf_rdi; 198 sigframe.uc.luc_mcontext.rsi = tf->tf_rsi; 199 sigframe.uc.luc_mcontext.rbp = tf->tf_rbp; 200 sigframe.uc.luc_mcontext.rbx = tf->tf_rbx; 201 sigframe.uc.luc_mcontext.rdx = tf->tf_rdx; 202 sigframe.uc.luc_mcontext.rax = tf->tf_rax; 203 sigframe.uc.luc_mcontext.rcx = tf->tf_rcx; 204 sigframe.uc.luc_mcontext.rsp = tf->tf_rsp; 205 sigframe.uc.luc_mcontext.rip = tf->tf_rip; 206 sigframe.uc.luc_mcontext.eflags = tf->tf_rflags; 207 sigframe.uc.luc_mcontext.cs = tf->tf_cs; 208 sigframe.uc.luc_mcontext.gs = tf->tf_gs; 209 sigframe.uc.luc_mcontext.fs = tf->tf_fs; 210 sigframe.uc.luc_mcontext.err = tf->tf_err; 211 sigframe.uc.luc_mcontext.trapno = tf->tf_trapno; 212 native_to_linux_sigset(&lmask, mask); 213 sigframe.uc.luc_mcontext.oldmask = lmask.sig[0]; 214 sigframe.uc.luc_mcontext.cr2 = (long)pcb->pcb_onfault; 215 sigframe.uc.luc_mcontext.fpstate = fpsp; 216 native_to_linux_sigset(&sigframe.uc.luc_sigmask, mask); 217 native_to_linux_siginfo(&sigframe.info, &ksi->ksi_info); 218 sendsig_reset(l, sig); 219 mutex_exit(p->p_lock); 220 error = 0; 221 222 /* 223 * Save FPU state, if any 224 */ 225 if (fpsp != NULL) { 226 size_t fp_size = sizeof fpregs; 227 /* The netbsd and linux structures both match the fxsave data */ 228 (void)process_read_fpregs(l, &fpregs, &fp_size); 229 error = copyout(&fpregs, fpsp, sizeof(*fpsp)); 230 } 231 232 if (error == 0) 233 error = copyout(&sigframe, sp, sizeof(sigframe)); 234 235 mutex_enter(p->p_lock); 236 237 if (error != 0) { 238 sigexit(l, SIGILL); 239 return; 240 } 241 242 if ((vaddr_t)catcher >= VM_MAXUSER_ADDRESS) { 243 sigexit(l, SIGILL); 244 return; 245 } 246 247 linux_buildcontext(l, catcher, sp); 248 tf->tf_rdi = sigframe.info.lsi_signo; 249 tf->tf_rax = 0; 250 tf->tf_rsi = (long)&sfp->info; 251 tf->tf_rdx = (long)&sfp->uc; 252 253 /* 254 * Remember we use signal stack 255 */ 256 if (onstack) 257 l->l_sigstk.ss_flags |= SS_ONSTACK; 258 return; 259 } 260 261 int 262 linux_sys_modify_ldt(struct lwp *l, const struct linux_sys_modify_ldt_args *v, register_t *retval) 263 { 264 return 0; 265 } 266 267 int 268 linux_sys_iopl(struct lwp *l, const struct linux_sys_iopl_args *v, register_t *retval) 269 { 270 return 0; 271 } 272 273 int 274 linux_sys_ioperm(struct lwp *l, const struct linux_sys_ioperm_args *v, register_t *retval) 275 { 276 return 0; 277 } 278 279 dev_t 280 linux_fakedev(dev_t dev, int raw) 281 { 282 283 extern const struct cdevsw ptc_cdevsw, pts_cdevsw; 284 const struct cdevsw *cd = cdevsw_lookup(dev); 285 286 if (raw) { 287 #if (NWSDISPLAY > 0) 288 extern const struct cdevsw wsdisplay_cdevsw; 289 if (cd == &wsdisplay_cdevsw) 290 return makedev(LINUX_CONS_MAJOR, (minor(dev) + 1)); 291 #endif 292 } 293 294 if (cd == &ptc_cdevsw) 295 return makedev(LINUX_PTC_MAJOR, minor(dev)); 296 if (cd == &pts_cdevsw) 297 return makedev(LINUX_PTS_MAJOR, minor(dev)); 298 299 return ((minor(dev) & 0xff) | ((major(dev) & 0xfff) << 8) 300 | (((unsigned long long int) (minor(dev) & ~0xff)) << 12) 301 | (((unsigned long long int) (major(dev) & ~0xfff)) << 32)); 302 } 303 304 int 305 linux_machdepioctl(struct lwp *l, const struct linux_sys_ioctl_args *v, register_t *retval) 306 { 307 return 0; 308 } 309 310 int 311 linux_sys_rt_sigreturn(struct lwp *l, const void *v, register_t *retval) 312 { 313 struct linux_ucontext *luctx; 314 struct trapframe *tf = l->l_md.md_regs; 315 struct linux_sigcontext *lsigctx; 316 struct linux_rt_sigframe frame, *fp; 317 ucontext_t uctx; 318 mcontext_t *mctx; 319 struct fxsave *fxarea; 320 int error; 321 322 fp = (struct linux_rt_sigframe *)(tf->tf_rsp - 8); 323 if ((error = copyin(fp, &frame, sizeof(frame))) != 0) { 324 mutex_enter(l->l_proc->p_lock); 325 sigexit(l, SIGILL); 326 return error; 327 } 328 luctx = &frame.uc; 329 lsigctx = &luctx->luc_mcontext; 330 331 memset(&uctx, 0, sizeof(uctx)); 332 mctx = (mcontext_t *)&uctx.uc_mcontext; 333 fxarea = (struct fxsave *)&mctx->__fpregs; 334 335 /* 336 * Set the flags. Linux always have CPU, stack and signal state, 337 * FPU is optional. uc_flags is not used to tell what we have. 338 */ 339 uctx.uc_flags = (_UC_SIGMASK|_UC_CPU|_UC_STACK|_UC_CLRSTACK); 340 if (lsigctx->fpstate != NULL) 341 uctx.uc_flags |= _UC_FPU; 342 uctx.uc_link = NULL; 343 344 /* 345 * Signal set 346 */ 347 linux_to_native_sigset(&uctx.uc_sigmask, &luctx->luc_sigmask); 348 349 /* 350 * CPU state 351 */ 352 mctx->__gregs[_REG_R8] = lsigctx->r8; 353 mctx->__gregs[_REG_R9] = lsigctx->r9; 354 mctx->__gregs[_REG_R10] = lsigctx->r10; 355 mctx->__gregs[_REG_R11] = lsigctx->r11; 356 mctx->__gregs[_REG_R12] = lsigctx->r12; 357 mctx->__gregs[_REG_R13] = lsigctx->r13; 358 mctx->__gregs[_REG_R14] = lsigctx->r14; 359 mctx->__gregs[_REG_R15] = lsigctx->r15; 360 mctx->__gregs[_REG_RDI] = lsigctx->rdi; 361 mctx->__gregs[_REG_RSI] = lsigctx->rsi; 362 mctx->__gregs[_REG_RBP] = lsigctx->rbp; 363 mctx->__gregs[_REG_RBX] = lsigctx->rbx; 364 mctx->__gregs[_REG_RAX] = lsigctx->rax; 365 mctx->__gregs[_REG_RDX] = lsigctx->rdx; 366 mctx->__gregs[_REG_RCX] = lsigctx->rcx; 367 mctx->__gregs[_REG_RIP] = lsigctx->rip; 368 mctx->__gregs[_REG_RFLAGS] = lsigctx->eflags; 369 mctx->__gregs[_REG_CS] = lsigctx->cs; 370 mctx->__gregs[_REG_GS] = lsigctx->gs; 371 mctx->__gregs[_REG_FS] = lsigctx->fs; 372 mctx->__gregs[_REG_ERR] = lsigctx->err; 373 mctx->__gregs[_REG_TRAPNO] = lsigctx->trapno; 374 mctx->__gregs[_REG_ES] = tf->tf_es; 375 mctx->__gregs[_REG_DS] = tf->tf_ds; 376 mctx->__gregs[_REG_RSP] = lsigctx->rsp; /* XXX */ 377 mctx->__gregs[_REG_SS] = tf->tf_ss; 378 379 /* 380 * FPU state 381 */ 382 if (lsigctx->fpstate != NULL) { 383 /* Both structures match the fxstate data */ 384 error = copyin(lsigctx->fpstate, fxarea, sizeof(*fxarea)); 385 if (error != 0) { 386 mutex_enter(l->l_proc->p_lock); 387 sigexit(l, SIGILL); 388 return error; 389 } 390 } 391 392 /* 393 * And the stack 394 */ 395 uctx.uc_stack.ss_flags = 0; 396 if (luctx->luc_stack.ss_flags & LINUX_SS_ONSTACK) 397 uctx.uc_stack.ss_flags |= SS_ONSTACK; 398 399 if (luctx->luc_stack.ss_flags & LINUX_SS_DISABLE) 400 uctx.uc_stack.ss_flags |= SS_DISABLE; 401 402 uctx.uc_stack.ss_sp = luctx->luc_stack.ss_sp; 403 uctx.uc_stack.ss_size = luctx->luc_stack.ss_size; 404 405 /* 406 * And let setucontext deal with that. 407 */ 408 mutex_enter(l->l_proc->p_lock); 409 error = setucontext(l, &uctx); 410 mutex_exit(l->l_proc->p_lock); 411 if (error) 412 return error; 413 414 return EJUSTRETURN; 415 } 416 417 int 418 linux_sys_arch_prctl(struct lwp *l, 419 const struct linux_sys_arch_prctl_args *uap, register_t *retval) 420 { 421 /* { 422 syscallarg(int) code; 423 syscallarg(unsigned long) addr; 424 } */ 425 void *addr = (void *)SCARG(uap, addr); 426 427 switch(SCARG(uap, code)) { 428 case LINUX_ARCH_SET_GS: 429 return x86_set_sdbase(addr, 'g', l, true); 430 431 case LINUX_ARCH_GET_GS: 432 return x86_get_sdbase(addr, 'g'); 433 434 case LINUX_ARCH_SET_FS: 435 return x86_set_sdbase(addr, 'f', l, true); 436 437 case LINUX_ARCH_GET_FS: 438 return x86_get_sdbase(addr, 'f'); 439 440 default: 441 #ifdef DEBUG_LINUX 442 printf("linux_sys_arch_prctl: unexpected code %d\n", 443 SCARG(uap, code)); 444 #endif 445 return EINVAL; 446 } 447 /* NOTREACHED */ 448 } 449 450 const int linux_vsyscall_to_syscall[] = { 451 LINUX_SYS_gettimeofday, 452 LINUX_SYS_time, 453 LINUX_SYS_nosys, /* nosys */ 454 LINUX_SYS_nosys, /* nosys */ 455 }; 456 457 int 458 linux_usertrap(struct lwp *l, vaddr_t trapaddr, void *arg) 459 { 460 struct trapframe *tf = arg; 461 uint64_t retaddr; 462 size_t vsyscallnr; 463 464 /* 465 * Check for a vsyscall. %rip must be the fault address, 466 * and the address must be in the Linux vsyscall area. 467 * Also, vsyscalls are only done at 1024-byte boundaries. 468 */ 469 470 if (__predict_true(trapaddr < LINUX_VSYSCALL_START)) 471 return 0; 472 473 if (trapaddr != tf->tf_rip) 474 return 0; 475 476 if ((tf->tf_rip & (LINUX_VSYSCALL_SIZE - 1)) != 0) 477 return 0; 478 479 vsyscallnr = (tf->tf_rip - LINUX_VSYSCALL_START) / LINUX_VSYSCALL_SIZE; 480 481 if (vsyscallnr > LINUX_VSYSCALL_MAXNR) 482 return 0; 483 484 /* 485 * Get the return address from the top of the stack, 486 * and fix up the return address. 487 * This assumes the faulting instruction was callq *reg, 488 * which is the only way that vsyscalls are ever entered. 489 */ 490 if (copyin((void *)tf->tf_rsp, &retaddr, sizeof retaddr) != 0) 491 return 0; 492 if ((vaddr_t)retaddr >= VM_MAXUSER_ADDRESS) 493 return 0; 494 tf->tf_rip = retaddr; 495 tf->tf_rax = linux_vsyscall_to_syscall[vsyscallnr]; 496 tf->tf_rsp += 8; /* "pop" the return address */ 497 498 #if 0 499 printf("usertrap: rip %p rsp %p retaddr %p vsys %d sys %d\n", 500 (void *)tf->tf_rip, (void *)tf->tf_rsp, (void *)retaddr, 501 vsyscallnr, (int)tf->tf_rax); 502 #endif 503 504 (*l->l_proc->p_md.md_syscall)(tf); 505 506 return 1; 507 } 508 509 static void 510 linux_buildcontext(struct lwp *l, void *catcher, void *f) 511 { 512 struct trapframe *tf = l->l_md.md_regs; 513 514 tf->tf_ds = GSEL(GUDATA_SEL, SEL_UPL); 515 tf->tf_rip = (u_int64_t)catcher; 516 tf->tf_cs = GSEL(GUCODE_SEL, SEL_UPL); 517 tf->tf_rflags &= ~PSL_CLEARSIG; 518 tf->tf_rsp = (u_int64_t)f; 519 tf->tf_ss = GSEL(GUDATA_SEL, SEL_UPL); 520 } 521
Indexes created Tue Sep 30 20:09:53 GMT 2025