linux_machdep.c revision 1.14
1 /* $NetBSD: linux_machdep.c,v 1.14 2006/08/23 19:49:09 manu 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 1.14 2006/08/23 19:49:09 manu 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/user.h> 46 #include <sys/wait.h> 47 #include <sys/ucontext.h> 48 49 #include <machine/reg.h> 50 #include <machine/pcb.h> 51 #include <machine/fpu.h> 52 #include <machine/mcontext.h> 53 #include <machine/specialreg.h> 54 #include <machine/vmparam.h> 55 56 #include <compat/linux/common/linux_signal.h> 57 #include <compat/linux/common/linux_errno.h> 58 #include <compat/linux/common/linux_exec.h> 59 #include <compat/linux/common/linux_ioctl.h> 60 #include <compat/linux/common/linux_prctl.h> 61 #include <compat/linux/common/linux_machdep.h> 62 #include <compat/linux/linux_syscall.h> 63 #include <compat/linux/linux_syscallargs.h> 64 65 static void linux_buildcontext(struct lwp *, void *, void *); 66 67 void 68 linux_setregs(l, epp, stack) 69 struct lwp *l; 70 struct exec_package *epp; 71 u_long stack; 72 { 73 struct pcb *pcb = &l->l_addr->u_pcb; 74 struct trapframe *tf; 75 76 /* If we were using the FPU, forget about it. */ 77 if (l->l_addr->u_pcb.pcb_fpcpu != NULL) 78 fpusave_lwp(l, 0); 79 80 l->l_md.md_flags &= ~MDP_USEDFPU; 81 pcb->pcb_flags = 0; 82 pcb->pcb_savefpu.fp_fxsave.fx_fcw = __NetBSD_NPXCW__; 83 pcb->pcb_savefpu.fp_fxsave.fx_mxcsr = __INITIAL_MXCSR__; 84 pcb->pcb_savefpu.fp_fxsave.fx_mxcsr_mask = __INITIAL_MXCSR_MASK__; 85 pcb->pcb_fs = 0; 86 pcb->pcb_gs = 0; 87 88 l->l_proc->p_flag &= ~P_32; 89 90 tf = l->l_md.md_regs; 91 tf->tf_rax = 0; 92 tf->tf_rbx = 0; 93 tf->tf_rcx = epp->ep_entry; 94 tf->tf_rdx = 0; 95 tf->tf_rsi = 0; 96 tf->tf_rdi = 0; 97 tf->tf_rbp = 0; 98 tf->tf_rsp = stack; 99 tf->tf_r8 = 0; 100 tf->tf_r9 = 0; 101 tf->tf_r10 = 0; 102 tf->tf_r11 = 0; 103 tf->tf_r12 = 0; 104 tf->tf_r13 = 0; 105 tf->tf_r14 = 0; 106 tf->tf_r15 = 0; 107 tf->tf_rip = epp->ep_entry; 108 tf->tf_rflags = PSL_USERSET; 109 tf->tf_cs = GSEL(GUCODE_SEL, SEL_UPL); 110 tf->tf_ss = GSEL(GUDATA_SEL, SEL_UPL); 111 tf->tf_ds = 0; 112 tf->tf_es = 0; 113 tf->tf_fs = 0; 114 tf->tf_gs = 0; 115 116 return; 117 } 118 119 void 120 linux_sendsig(ksi, mask) 121 const ksiginfo_t *ksi; 122 const sigset_t *mask; 123 { 124 struct lwp *l = curlwp; 125 struct proc *p = l->l_proc; 126 struct sigacts *ps = p->p_sigacts; 127 int onstack; 128 int sig = ksi->ksi_signo; 129 struct linux_rt_sigframe *sfp, sigframe; 130 struct linux__fpstate *fpsp, fpstate; 131 struct fpreg fpregs; 132 struct trapframe *tf = l->l_md.md_regs; 133 sig_t catcher = SIGACTION(p, sig).sa_handler; 134 linux_sigset_t lmask; 135 char *sp; 136 int error; 137 138 /* Do we need to jump onto the signal stack? */ 139 onstack = 140 (p->p_sigctx.ps_sigstk.ss_flags & (SS_DISABLE | SS_ONSTACK)) == 0 && 141 (SIGACTION(p, sig).sa_flags & SA_ONSTACK) != 0; 142 143 /* Allocate space for the signal handler context. */ 144 if (onstack) 145 sp = ((caddr_t)p->p_sigctx.ps_sigstk.ss_sp + 146 p->p_sigctx.ps_sigstk.ss_size); 147 else 148 sp = (caddr_t)tf->tf_rsp - 128; 149 150 151 /* 152 * Save FPU state, if any 153 */ 154 if (l->l_md.md_flags & MDP_USEDFPU) { 155 sp = (char *) 156 (((long)sp - sizeof(struct linux__fpstate)) & ~0xfUL); 157 fpsp = (struct linux__fpstate *)sp; 158 159 (void)process_read_fpregs(l, &fpregs); 160 bzero(&fpstate, sizeof(fpstate)); 161 162 fpstate.cwd = fpregs.fp_fcw; 163 fpstate.swd = fpregs.fp_fsw; 164 fpstate.twd = fpregs.fp_ftw; 165 fpstate.fop = fpregs.fp_fop; 166 fpstate.rip = fpregs.fp_rip; 167 fpstate.rdp = fpregs.fp_rdp; 168 fpstate.mxcsr = fpregs.fp_mxcsr; 169 fpstate.mxcsr_mask = fpregs.fp_mxcsr_mask; 170 memcpy(&fpstate.st_space, &fpregs.fp_st, 171 sizeof(fpstate.st_space)); 172 memcpy(&fpstate.xmm_space, &fpregs.fp_xmm, 173 sizeof(fpstate.xmm_space)); 174 175 if ((error = copyout(&fpstate, fpsp, sizeof(fpstate))) != 0) { 176 sigexit(l, SIGILL); 177 return; 178 } 179 } else { 180 fpsp = NULL; 181 } 182 183 /* 184 * Populate the rt_sigframe 185 */ 186 sp = (char *) 187 ((((long)sp - sizeof(struct linux_rt_sigframe)) & ~0xfUL) - 8); 188 sfp = (struct linux_rt_sigframe *)sp; 189 190 bzero(&sigframe, sizeof(sigframe)); 191 if (ps->sa_sigdesc[sig].sd_vers != 0) 192 sigframe.pretcode = 193 (char *)(u_long)ps->sa_sigdesc[sig].sd_tramp; 194 else 195 sigframe.pretcode = NULL; 196 197 /* 198 * The user context 199 */ 200 sigframe.uc.luc_flags = 0; 201 sigframe.uc.luc_link = NULL; 202 203 /* This is used regardless of SA_ONSTACK in Linux */ 204 sigframe.uc.luc_stack.ss_sp = p->p_sigctx.ps_sigstk.ss_sp; 205 sigframe.uc.luc_stack.ss_size = p->p_sigctx.ps_sigstk.ss_size; 206 sigframe.uc.luc_stack.ss_flags = 0; 207 if (p->p_sigctx.ps_sigstk.ss_flags & SS_ONSTACK) 208 sigframe.uc.luc_stack.ss_flags |= LINUX_SS_ONSTACK; 209 if (p->p_sigctx.ps_sigstk.ss_flags & SS_DISABLE) 210 sigframe.uc.luc_stack.ss_flags |= LINUX_SS_DISABLE; 211 212 sigframe.uc.luc_mcontext.r8 = tf->tf_r8; 213 sigframe.uc.luc_mcontext.r9 = tf->tf_r9; 214 sigframe.uc.luc_mcontext.r10 = tf->tf_r10; 215 sigframe.uc.luc_mcontext.r11 = tf->tf_r11; 216 sigframe.uc.luc_mcontext.r12 = tf->tf_r12; 217 sigframe.uc.luc_mcontext.r13 = tf->tf_r13; 218 sigframe.uc.luc_mcontext.r14 = tf->tf_r14; 219 sigframe.uc.luc_mcontext.r15 = tf->tf_r15; 220 sigframe.uc.luc_mcontext.rdi = tf->tf_rdi; 221 sigframe.uc.luc_mcontext.rsi = tf->tf_rsi; 222 sigframe.uc.luc_mcontext.rbp = tf->tf_rbp; 223 sigframe.uc.luc_mcontext.rbx = tf->tf_rbx; 224 sigframe.uc.luc_mcontext.rdx = tf->tf_rdx; 225 sigframe.uc.luc_mcontext.rcx = tf->tf_rcx; 226 sigframe.uc.luc_mcontext.rsp = tf->tf_rsp; 227 sigframe.uc.luc_mcontext.rip = tf->tf_rip; 228 sigframe.uc.luc_mcontext.eflags = tf->tf_rflags; 229 sigframe.uc.luc_mcontext.cs = tf->tf_cs; 230 sigframe.uc.luc_mcontext.gs = tf->tf_gs; 231 sigframe.uc.luc_mcontext.fs = tf->tf_fs; 232 sigframe.uc.luc_mcontext.err = tf->tf_err; 233 sigframe.uc.luc_mcontext.trapno = tf->tf_trapno; 234 native_to_linux_sigset(&lmask, mask); 235 sigframe.uc.luc_mcontext.oldmask = lmask.sig[0]; 236 sigframe.uc.luc_mcontext.cr2 = (long)l->l_addr->u_pcb.pcb_onfault; 237 sigframe.uc.luc_mcontext.fpstate = fpsp; 238 native_to_linux_sigset(&sigframe.uc.luc_sigmask, mask); 239 240 /* 241 * the siginfo structure 242 */ 243 sigframe.info.lsi_signo = native_to_linux_signo[sig]; 244 sigframe.info.lsi_errno = native_to_linux_errno[ksi->ksi_errno]; 245 sigframe.info.lsi_code = ksi->ksi_code; 246 247 /* XXX This is a rought conversion, taken from i386 code */ 248 switch (sigframe.info.lsi_signo) { 249 case LINUX_SIGILL: 250 case LINUX_SIGFPE: 251 case LINUX_SIGSEGV: 252 case LINUX_SIGBUS: 253 case LINUX_SIGTRAP: 254 sigframe.info._sifields._sigfault._addr = ksi->ksi_addr; 255 break; 256 case LINUX_SIGCHLD: 257 sigframe.info._sifields._sigchld._pid = ksi->ksi_pid; 258 sigframe.info._sifields._sigchld._uid = ksi->ksi_uid; 259 sigframe.info._sifields._sigchld._utime = ksi->ksi_utime; 260 sigframe.info._sifields._sigchld._stime = ksi->ksi_stime; 261 262 if (WCOREDUMP(ksi->ksi_status)) { 263 sigframe.info.lsi_code = LINUX_CLD_DUMPED; 264 sigframe.info._sifields._sigchld._status = 265 _WSTATUS(ksi->ksi_status); 266 } else if (_WSTATUS(ksi->ksi_status)) { 267 sigframe.info.lsi_code = LINUX_CLD_KILLED; 268 sigframe.info._sifields._sigchld._status = 269 _WSTATUS(ksi->ksi_status); 270 } else { 271 sigframe.info.lsi_code = LINUX_CLD_EXITED; 272 sigframe.info._sifields._sigchld._status = 273 ((ksi->ksi_status & 0xff00U) >> 8); 274 } 275 break; 276 case LINUX_SIGIO: 277 sigframe.info._sifields._sigpoll._band = ksi->ksi_band; 278 sigframe.info._sifields._sigpoll._fd = ksi->ksi_fd; 279 break; 280 default: 281 sigframe.info._sifields._sigchld._pid = ksi->ksi_pid; 282 sigframe.info._sifields._sigchld._uid = ksi->ksi_uid; 283 if ((sigframe.info.lsi_signo == LINUX_SIGALRM) || 284 (sigframe.info.lsi_signo >= LINUX_SIGRTMIN)) 285 sigframe.info._sifields._timer._sigval.sival_ptr = 286 ksi->ksi_sigval.sival_ptr; 287 break; 288 } 289 290 if ((error = copyout(&sigframe, sp, sizeof(sigframe))) != 0) { 291 sigexit(l, SIGILL); 292 return; 293 } 294 295 linux_buildcontext(l, catcher, sp); 296 tf->tf_rdi = sigframe.info.lsi_signo; 297 tf->tf_rax = 0; 298 tf->tf_rsi = (long)&sfp->info; 299 tf->tf_rdx = (long)&sfp->uc; 300 301 /* 302 * Remember we use signal stack 303 */ 304 if (onstack) 305 p->p_sigctx.ps_sigstk.ss_flags |= SS_ONSTACK; 306 return; 307 } 308 309 int 310 linux_sys_modify_ldt(l, v, retval) 311 struct lwp *l; 312 void *v; 313 register_t *retval; 314 { 315 printf("linux_sys_modify_ldt\n"); 316 return 0; 317 } 318 319 int 320 linux_sys_iopl(l, v, retval) 321 struct lwp *l; 322 void *v; 323 register_t *retval; 324 { 325 return 0; 326 } 327 328 int 329 linux_sys_ioperm(l, v, retval) 330 struct lwp *l; 331 void *v; 332 register_t *retval; 333 { 334 return 0; 335 } 336 337 dev_t 338 linux_fakedev(dev, raw) 339 dev_t dev; 340 int raw; 341 { 342 return 0; 343 } 344 345 int 346 linux_machdepioctl(l, v, retval) 347 struct lwp *l; 348 void *v; 349 register_t *retval; 350 { 351 return 0; 352 } 353 354 int 355 linux_sys_rt_sigreturn(l, v, retval) 356 struct lwp *l; 357 void *v; 358 register_t *retval; 359 { 360 struct linux_ucontext *luctx; 361 struct trapframe *tf = l->l_md.md_regs; 362 struct linux_sigcontext *lsigctx; 363 struct linux__fpstate fpstate; 364 struct linux_rt_sigframe frame, *fp; 365 ucontext_t uctx; 366 mcontext_t *mctx; 367 struct fxsave64 *fxsave; 368 int error; 369 370 fp = (struct linux_rt_sigframe *)(tf->tf_rsp - 8); 371 if ((error = copyin(fp, &frame, sizeof(frame))) != 0) { 372 sigexit(l, SIGILL); 373 return error; 374 } 375 luctx = &frame.uc; 376 lsigctx = &luctx->luc_mcontext; 377 378 bzero(&uctx, sizeof(uctx)); 379 mctx = (mcontext_t *)&uctx.uc_mcontext; 380 fxsave = (struct fxsave64 *)&mctx->__fpregs; 381 382 /* 383 * Set the flags. Linux always have CPU, stack and signal state, 384 * FPU is optional. uc_flags is not used to tell what we have. 385 */ 386 uctx.uc_flags = (_UC_SIGMASK|_UC_CPU|_UC_STACK|_UC_CLRSTACK); 387 if (lsigctx->fpstate != NULL) 388 uctx.uc_flags |= _UC_FPU; 389 uctx.uc_link = NULL; 390 391 /* 392 * Signal set 393 */ 394 linux_to_native_sigset(&uctx.uc_sigmask, &luctx->luc_sigmask); 395 396 /* 397 * CPU state 398 */ 399 mctx->__gregs[_REG_R8] = lsigctx->r8; 400 mctx->__gregs[_REG_R9] = lsigctx->r9; 401 mctx->__gregs[_REG_R10] = lsigctx->r10; 402 mctx->__gregs[_REG_R11] = lsigctx->r11; 403 mctx->__gregs[_REG_R12] = lsigctx->r12; 404 mctx->__gregs[_REG_R13] = lsigctx->r13; 405 mctx->__gregs[_REG_R14] = lsigctx->r14; 406 mctx->__gregs[_REG_R15] = lsigctx->r15; 407 mctx->__gregs[_REG_RDI] = lsigctx->rdi; 408 mctx->__gregs[_REG_RSI] = lsigctx->rsi; 409 mctx->__gregs[_REG_RBP] = lsigctx->rbp; 410 mctx->__gregs[_REG_RBX] = lsigctx->rbx; 411 mctx->__gregs[_REG_RAX] = tf->tf_rax; 412 mctx->__gregs[_REG_RDX] = lsigctx->rdx; 413 mctx->__gregs[_REG_RCX] = lsigctx->rcx; 414 mctx->__gregs[_REG_RIP] = lsigctx->rip; 415 mctx->__gregs[_REG_RFL] = lsigctx->eflags; 416 mctx->__gregs[_REG_CS] = lsigctx->cs; 417 mctx->__gregs[_REG_GS] = lsigctx->gs; 418 mctx->__gregs[_REG_FS] = lsigctx->fs; 419 mctx->__gregs[_REG_ERR] = lsigctx->err; 420 mctx->__gregs[_REG_TRAPNO] = lsigctx->trapno; 421 mctx->__gregs[_REG_ES] = tf->tf_es; 422 mctx->__gregs[_REG_DS] = tf->tf_ds; 423 mctx->__gregs[_REG_URSP] = lsigctx->rsp; /* XXX */ 424 mctx->__gregs[_REG_SS] = tf->tf_ss; 425 426 /* 427 * FPU state 428 */ 429 if (lsigctx->fpstate != NULL) { 430 error = copyin(lsigctx->fpstate, &fpstate, sizeof(fpstate)); 431 if (error != 0) { 432 sigexit(l, SIGILL); 433 return error; 434 } 435 436 fxsave->fx_fcw = fpstate.cwd; 437 fxsave->fx_fsw = fpstate.swd; 438 fxsave->fx_ftw = fpstate.twd; 439 fxsave->fx_fop = fpstate.fop; 440 fxsave->fx_rip = fpstate.rip; 441 fxsave->fx_rdp = fpstate.rdp; 442 fxsave->fx_mxcsr = fpstate.mxcsr; 443 fxsave->fx_mxcsr_mask = fpstate.mxcsr_mask; 444 memcpy(&fxsave->fx_st, &fpstate.st_space, 445 sizeof(fxsave->fx_st)); 446 memcpy(&fxsave->fx_xmm, &fpstate.xmm_space, 447 sizeof(fxsave->fx_xmm)); 448 } 449 450 /* 451 * And the stack 452 */ 453 uctx.uc_stack.ss_flags = 0; 454 if (luctx->luc_stack.ss_flags & LINUX_SS_ONSTACK); 455 uctx.uc_stack.ss_flags = SS_ONSTACK; 456 457 if (luctx->luc_stack.ss_flags & LINUX_SS_DISABLE); 458 uctx.uc_stack.ss_flags = SS_DISABLE; 459 460 uctx.uc_stack.ss_sp = luctx->luc_stack.ss_sp; 461 uctx.uc_stack.ss_size = luctx->luc_stack.ss_size; 462 463 /* 464 * And let setucontext deal with that. 465 */ 466 return setucontext(l, &uctx); 467 } 468 469 int 470 linux_sys_arch_prctl(l, v, retval) 471 struct lwp *l; 472 void *v; 473 register_t *retval; 474 { 475 struct linux_sys_arch_prctl_args /* { 476 syscallarg(int) code; 477 syscallarg(unsigned long) addr; 478 } */ *uap = v; 479 struct pcb *pcb = &l->l_addr->u_pcb; 480 struct trapframe *tf = l->l_md.md_regs; 481 int error; 482 uint64_t taddr; 483 484 switch(SCARG(uap, code)) { 485 case LINUX_ARCH_SET_GS: 486 taddr = SCARG(uap, addr); 487 if (taddr >= VM_MAXUSER_ADDRESS) 488 return EINVAL; 489 pcb->pcb_gs = taddr; 490 pcb->pcb_flags |= PCB_GS64; 491 if (l == curlwp) 492 wrmsr(MSR_KERNELGSBASE, taddr); 493 break; 494 495 case LINUX_ARCH_GET_GS: 496 if (pcb->pcb_flags & PCB_GS64) 497 taddr = pcb->pcb_gs; 498 else { 499 error = memseg_baseaddr(l, tf->tf_fs, NULL, 0, &taddr); 500 if (error != 0) 501 return error; 502 } 503 error = copyout(&taddr, (char *)SCARG(uap, addr), 8); 504 if (error != 0) 505 return error; 506 break; 507 508 case LINUX_ARCH_SET_FS: 509 taddr = SCARG(uap, addr); 510 if (taddr >= VM_MAXUSER_ADDRESS) 511 return EINVAL; 512 pcb->pcb_fs = taddr; 513 pcb->pcb_flags |= PCB_FS64; 514 if (l == curlwp) 515 wrmsr(MSR_FSBASE, taddr); 516 break; 517 518 case LINUX_ARCH_GET_FS: 519 if (pcb->pcb_flags & PCB_FS64) 520 taddr = pcb->pcb_fs; 521 else { 522 error = memseg_baseaddr(l, tf->tf_fs, NULL, 0, &taddr); 523 if (error != 0) 524 return error; 525 } 526 error = copyout(&taddr, (char *)SCARG(uap, addr), 8); 527 if (error != 0) 528 return error; 529 break; 530 531 default: 532 #ifdef DEBUG_LINUX 533 printf("linux_sys_arch_prctl: unexpected code %d\n", 534 SCARG(uap, code)); 535 #endif 536 return EINVAL; 537 } 538 539 return 0; 540 } 541 542 const int linux_vsyscall_to_syscall[] = { 543 LINUX_SYS_gettimeofday, 544 LINUX_SYS_time, 545 LINUX_SYS_nosys, 546 LINUX_SYS_nosys, 547 }; 548 549 int 550 linux_usertrap(struct lwp *l, vaddr_t trapaddr, void *arg) 551 { 552 struct trapframe *tf = arg; 553 uint64_t retaddr; 554 int vsyscallnr; 555 556 /* 557 * Check for a vsyscall. %rip must be the fault address, 558 * and the address must be in the Linux vsyscall area. 559 * Also, vsyscalls are only done at 1024-byte boundaries. 560 */ 561 562 if (__predict_true(trapaddr < LINUX_VSYSCALL_START)) 563 return 0; 564 565 if (trapaddr != tf->tf_rip) 566 return 0; 567 568 if ((tf->tf_rip & (LINUX_VSYSCALL_SIZE - 1)) != 0) 569 return 0; 570 571 vsyscallnr = (tf->tf_rip - LINUX_VSYSCALL_START) / LINUX_VSYSCALL_SIZE; 572 573 if (vsyscallnr > LINUX_VSYSCALL_MAXNR) 574 return 0; 575 576 /* 577 * Get the return address from the top of the stack, 578 * and fix up the return address. 579 * This assumes the faulting instruction was callq *reg, 580 * which is the only way that vsyscalls are ever entered. 581 */ 582 if (copyin((void *)tf->tf_rsp, &retaddr, sizeof retaddr) != 0) 583 return 0; 584 tf->tf_rip = retaddr; 585 tf->tf_rax = linux_vsyscall_to_syscall[vsyscallnr]; 586 tf->tf_rsp += 8; /* "pop" the return address */ 587 588 #if 0 589 printf("usertrap: rip %p rsp %p retaddr %p vsys %d sys %d\n", 590 (void *)tf->tf_rip, (void *)tf->tf_rsp, (void *)retaddr, 591 vsyscallnr, (int)tf->tf_rax); 592 #endif 593 594 (*l->l_proc->p_md.md_syscall)(tf); 595 596 return 1; 597 } 598 599 static void 600 linux_buildcontext(struct lwp *l, void *catcher, void *f) 601 { 602 struct trapframe *tf = l->l_md.md_regs; 603 604 tf->tf_ds = GSEL(GUDATA_SEL, SEL_UPL); 605 tf->tf_rip = (u_int64_t)catcher; 606 tf->tf_cs = GSEL(GUCODE_SEL, SEL_UPL); 607 tf->tf_rflags &= ~(PSL_T|PSL_VM|PSL_AC); 608 tf->tf_rsp = (u_int64_t)f; 609 tf->tf_ss = GSEL(GUDATA_SEL, SEL_UPL); 610 } 611 612 void * 613 linux_get_newtls(l) 614 struct lwp *l; 615 { 616 struct trapframe *tf = l->l_md.md_regs; 617 618 return (void *)tf->tf_r8; 619 } 620 621 int 622 linux_set_newtls(l, tls) 623 struct lwp *l; 624 void *tls; 625 { 626 struct linux_sys_arch_prctl_args cup; 627 register_t retval; 628 629 SCARG(&cup, code) = LINUX_ARCH_SET_FS; 630 SCARG(&cup, addr) = (unsigned long)tls; 631 632 return linux_sys_arch_prctl(l, &cup, &retval); 633 } 634
Indexes created Tue Sep 30 20:09:53 GMT 2025