mirror of
https://github.com/Stichting-MINIX-Research-Foundation/netbsd.git
synced 2025-08-11 06:59:29 -04:00
519 lines
13 KiB
C
519 lines
13 KiB
C
/* $NetBSD: linux_machdep.c,v 1.42 2011/02/20 08:14:10 matt Exp $ */
|
|
|
|
/*-
|
|
* Copyright (c) 1995, 2000, 2001 The NetBSD Foundation, Inc.
|
|
* All rights reserved.
|
|
*
|
|
* This code is derived from software contributed to The NetBSD Foundation
|
|
* by Frank van der Linden and Emmanuel Dreyfus.
|
|
*
|
|
* Redistribution and use in source and binary forms, with or without
|
|
* modification, are permitted provided that the following conditions
|
|
* are met:
|
|
* 1. Redistributions of source code must retain the above copyright
|
|
* notice, this list of conditions and the following disclaimer.
|
|
* 2. Redistributions in binary form must reproduce the above copyright
|
|
* notice, this list of conditions and the following disclaimer in the
|
|
* documentation and/or other materials provided with the distribution.
|
|
*
|
|
* THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS
|
|
* ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
|
|
* TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
|
|
* PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS
|
|
* BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
|
|
* CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
|
|
* SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
|
|
* INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
|
|
* CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
|
|
* ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
|
|
* POSSIBILITY OF SUCH DAMAGE.
|
|
*/
|
|
|
|
#include <sys/cdefs.h>
|
|
__KERNEL_RCSID(0, "$NetBSD: linux_machdep.c,v 1.42 2011/02/20 08:14:10 matt Exp $");
|
|
|
|
#include <sys/param.h>
|
|
#include <sys/systm.h>
|
|
#include <sys/signalvar.h>
|
|
#include <sys/kernel.h>
|
|
#include <sys/proc.h>
|
|
#include <sys/buf.h>
|
|
#include <sys/reboot.h>
|
|
#include <sys/conf.h>
|
|
#include <sys/exec.h>
|
|
#include <sys/file.h>
|
|
#include <sys/callout.h>
|
|
#include <sys/malloc.h>
|
|
#include <sys/mbuf.h>
|
|
#include <sys/msgbuf.h>
|
|
#include <sys/mount.h>
|
|
#include <sys/vnode.h>
|
|
#include <sys/device.h>
|
|
#include <sys/syscallargs.h>
|
|
#include <sys/filedesc.h>
|
|
#include <sys/exec_elf.h>
|
|
#include <sys/disklabel.h>
|
|
#include <sys/ioctl.h>
|
|
#include <sys/sysctl.h>
|
|
#include <sys/kauth.h>
|
|
#include <miscfs/specfs/specdev.h>
|
|
|
|
#include <compat/linux/common/linux_types.h>
|
|
#include <compat/linux/common/linux_signal.h>
|
|
#include <compat/linux/common/linux_util.h>
|
|
#include <compat/linux/common/linux_ioctl.h>
|
|
#include <compat/linux/common/linux_hdio.h>
|
|
#include <compat/linux/common/linux_exec.h>
|
|
#include <compat/linux/common/linux_machdep.h>
|
|
|
|
#include <compat/linux/linux_syscallargs.h>
|
|
|
|
#include <sys/cpu.h>
|
|
#include <machine/psl.h>
|
|
#include <machine/reg.h>
|
|
#include <machine/regnum.h>
|
|
#include <machine/vmparam.h>
|
|
#include <machine/locore.h>
|
|
|
|
#include <mips/cache.h>
|
|
|
|
union linux_ksigframe {
|
|
struct linux_sigframe sf;
|
|
#if !defined(__mips_o32)
|
|
struct linux_sigframe32 sf32;
|
|
#endif
|
|
};
|
|
|
|
/*
|
|
* To see whether wscons is configured (for virtual console ioctl calls).
|
|
*/
|
|
#if defined(_KERNEL_OPT)
|
|
#include "wsdisplay.h"
|
|
#endif
|
|
#if (NWSDISPLAY > 0)
|
|
#include <dev/wscons/wsconsio.h>
|
|
#include <dev/wscons/wsdisplay_usl_io.h>
|
|
#endif
|
|
|
|
/*
|
|
* Set set up registers on exec.
|
|
*/
|
|
void
|
|
linux_setregs(struct lwp *l, struct exec_package *pack, vaddr_t stack)
|
|
{
|
|
setregs(l, pack, stack);
|
|
return;
|
|
}
|
|
|
|
#if !defined(__mips_o32)
|
|
static void
|
|
linux_setup_sigcontext32(struct linux_sigcontext32 *sc,
|
|
const struct trapframe *tf)
|
|
{
|
|
for (u_int i = 0; i < 32; i++) {
|
|
sc->lsc_regs[i] = tf->tf_regs[i];
|
|
}
|
|
sc->lsc_mdhi = tf->tf_regs[_R_MULHI];
|
|
sc->lsc_mdlo = tf->tf_regs[_R_MULLO];
|
|
sc->lsc_pc = tf->tf_regs[_R_PC];
|
|
}
|
|
#endif
|
|
|
|
static void
|
|
linux_setup_sigcontext(struct linux_sigcontext *sc,
|
|
const struct trapframe *tf)
|
|
{
|
|
for (u_int i = 0; i < 32; i++) {
|
|
sc->lsc_regs[i] = tf->tf_regs[i];
|
|
}
|
|
sc->lsc_mdhi = tf->tf_regs[_R_MULHI];
|
|
sc->lsc_mdlo = tf->tf_regs[_R_MULLO];
|
|
sc->lsc_pc = tf->tf_regs[_R_PC];
|
|
}
|
|
|
|
/*
|
|
* Send an interrupt to process.
|
|
*
|
|
* Adapted from sys/arch/mips/mips/mips_machdep.c
|
|
*
|
|
* XXX Does not work well yet with RT signals
|
|
*
|
|
*/
|
|
|
|
void
|
|
linux_sendsig(const ksiginfo_t *ksi, const sigset_t *mask)
|
|
{
|
|
const int sig = ksi->ksi_signo;
|
|
struct lwp * const l = curlwp;
|
|
struct proc * const p = l->l_proc;
|
|
struct trapframe * const tf = l->l_md.md_utf;
|
|
#ifdef __mips_o32
|
|
const int abi = _MIPS_BSD_API_O32;
|
|
#else
|
|
const int abi = p->p_md.md_abi;
|
|
#endif
|
|
union linux_ksigframe ksf, *sf;
|
|
bool onstack;
|
|
int error;
|
|
sig_t catcher = SIGACTION(p, sig).sa_handler;
|
|
|
|
#ifdef DEBUG_LINUX
|
|
printf("linux_sendsig()\n");
|
|
#endif /* DEBUG_LINUX */
|
|
|
|
/*
|
|
* Do we need to jump onto the signal stack?
|
|
*/
|
|
onstack = (l->l_sigstk.ss_flags & (SS_DISABLE | SS_ONSTACK)) == 0
|
|
&& (SIGACTION(p, sig).sa_flags & SA_ONSTACK) != 0;
|
|
|
|
/*
|
|
* Signal stack is broken (see at the end of linux_sigreturn), so we do
|
|
* not use it yet. XXX fix this.
|
|
*/
|
|
onstack = false;
|
|
|
|
/*
|
|
* Allocate space for the signal handler context.
|
|
*/
|
|
sf = (void *)(onstack
|
|
? (uintptr_t) l->l_sigstk.ss_sp + l->l_sigstk.ss_size
|
|
: (uintptr_t) tf->tf_regs[_R_SP]);
|
|
|
|
/*
|
|
* Build stack frame for signal trampoline.
|
|
*/
|
|
memset(&ksf, 0, sizeof ksf);
|
|
|
|
/*
|
|
* This is the signal trampoline used by Linux, we don't use it,
|
|
* but we set it up in case an application expects it to be there
|
|
*/
|
|
ksf.sf.lsf_code[0] = 0x24020000; /* li v0, __NR_sigreturn */
|
|
ksf.sf.lsf_code[1] = 0x0000000c; /* syscall */
|
|
|
|
switch (abi) {
|
|
default:
|
|
native_to_linux_sigset(&ksf.sf.lsf_mask, mask);
|
|
linux_setup_sigcontext(&ksf.sf.lsf_sc, tf);
|
|
break;
|
|
#if !defined(__mips_o32)
|
|
case _MIPS_BSD_API_O32:
|
|
native_to_linux_sigset(&ksf.sf32.lsf_mask, mask);
|
|
linux_setup_sigcontext32(&ksf.sf32.lsf_sc, tf);
|
|
break;
|
|
#endif
|
|
}
|
|
sendsig_reset(l, sig);
|
|
|
|
/*
|
|
* Save signal stack. XXX broken
|
|
*/
|
|
/* kregs.sc_onstack = l->l_sigstk.ss_flags & SS_ONSTACK; */
|
|
|
|
/*
|
|
* Install the sigframe onto the stack
|
|
*/
|
|
sf -= sizeof(*sf);
|
|
mutex_exit(p->p_lock);
|
|
error = copyout(&ksf, sf, sizeof(ksf));
|
|
mutex_enter(p->p_lock);
|
|
|
|
if (error != 0) {
|
|
/*
|
|
* Process has trashed its stack; give it an illegal
|
|
* instruction to halt it in its tracks.
|
|
*/
|
|
#ifdef DEBUG_LINUX
|
|
printf("linux_sendsig: stack trashed\n");
|
|
#endif /* DEBUG_LINUX */
|
|
sigexit(l, SIGILL);
|
|
/* NOTREACHED */
|
|
}
|
|
|
|
/* Set up the registers to return to sigcode. */
|
|
tf->tf_regs[_R_A0] = native_to_linux_signo[sig];
|
|
tf->tf_regs[_R_A1] = 0;
|
|
tf->tf_regs[_R_A2] = (intptr_t)&sf->sf.lsf_sc;
|
|
|
|
#ifdef DEBUG_LINUX
|
|
printf("sigcontext is at %p\n", &sf->sf.lsf_sc);
|
|
#endif /* DEBUG_LINUX */
|
|
|
|
tf->tf_regs[_R_SP] = (intptr_t)sf;
|
|
/* Signal trampoline code is at base of user stack. */
|
|
tf->tf_regs[_R_RA] = (intptr_t)p->p_sigctx.ps_sigcode;
|
|
tf->tf_regs[_R_T9] = (intptr_t)catcher;
|
|
tf->tf_regs[_R_PC] = (intptr_t)catcher;
|
|
|
|
/* Remember that we're now on the signal stack. */
|
|
if (onstack)
|
|
l->l_sigstk.ss_flags |= SS_ONSTACK;
|
|
|
|
return;
|
|
}
|
|
|
|
static void
|
|
linux_putaway_sigcontext(struct trapframe *tf,
|
|
const struct linux_sigcontext *sc)
|
|
{
|
|
for (u_int i = 0; i < 32; i++) {
|
|
tf->tf_regs[i] = sc->lsc_regs[i];
|
|
}
|
|
tf->tf_regs[_R_MULLO] = sc->lsc_mdlo;
|
|
tf->tf_regs[_R_MULHI] = sc->lsc_mdhi;
|
|
tf->tf_regs[_R_PC] = sc->lsc_pc;
|
|
}
|
|
|
|
#ifndef __mips_o32
|
|
static void
|
|
linux_putaway_sigcontext32(struct trapframe *tf,
|
|
const struct linux_sigcontext32 *sc)
|
|
{
|
|
for (u_int i = 0; i < 32; i++) {
|
|
tf->tf_regs[i] = sc->lsc_regs[i];
|
|
}
|
|
tf->tf_regs[_R_MULLO] = sc->lsc_mdlo;
|
|
tf->tf_regs[_R_MULHI] = sc->lsc_mdhi;
|
|
tf->tf_regs[_R_PC] = sc->lsc_pc;
|
|
}
|
|
#endif
|
|
|
|
/*
|
|
* System call to cleanup state after a signal
|
|
* has been taken. Reset signal mask and
|
|
* stack state from context left by sendsig (above).
|
|
*/
|
|
int
|
|
linux_sys_sigreturn(struct lwp *l, const struct linux_sys_sigreturn_args *uap, register_t *retval)
|
|
{
|
|
/* {
|
|
syscallarg(struct linux_sigframe *) sf;
|
|
} */
|
|
struct proc *p = l->l_proc;
|
|
union linux_ksigframe ksf, *sf;
|
|
#ifdef __mips_o32
|
|
const int abi = _MIPS_BSD_API_O32;
|
|
#else
|
|
const int abi = p->p_md.md_abi;
|
|
#endif
|
|
linux_sigset_t *lmask;
|
|
sigset_t mask;
|
|
int error;
|
|
|
|
#ifdef DEBUG_LINUX
|
|
printf("linux_sys_sigreturn()\n");
|
|
#endif /* DEBUG_LINUX */
|
|
|
|
/*
|
|
* The trampoline code hands us the context.
|
|
* It is unsafe to keep track of it ourselves, in the event that a
|
|
* program jumps out of a signal handler.
|
|
*/
|
|
sf = (void *)SCARG(uap, sf);
|
|
|
|
if ((error = copyin(sf, &ksf, sizeof(ksf))) != 0)
|
|
return (error);
|
|
|
|
/* Restore the register context. */
|
|
switch (abi) {
|
|
default:
|
|
lmask = &ksf.sf.lsf_mask;
|
|
linux_putaway_sigcontext(l->l_md.md_utf, &ksf.sf.lsf_sc);
|
|
break;
|
|
#if !defined(__mips_o32)
|
|
case _MIPS_BSD_API_O32:
|
|
lmask = &ksf.sf32.lsf_mask;
|
|
linux_putaway_sigcontext32(l->l_md.md_utf, &ksf.sf32.lsf_sc);
|
|
break;
|
|
#endif
|
|
}
|
|
|
|
mutex_enter(p->p_lock);
|
|
|
|
/* Restore signal stack. */
|
|
l->l_sigstk.ss_flags &= ~SS_ONSTACK;
|
|
|
|
/* Restore signal mask. */
|
|
linux_to_native_sigset(&mask, lmask);
|
|
(void)sigprocmask1(l, SIG_SETMASK, &mask, 0);
|
|
|
|
mutex_exit(p->p_lock);
|
|
|
|
return (EJUSTRETURN);
|
|
}
|
|
|
|
|
|
int
|
|
linux_sys_rt_sigreturn(struct lwp *l, const struct linux_sys_rt_sigreturn_args *v, register_t *retval)
|
|
{
|
|
return (ENOSYS);
|
|
}
|
|
|
|
|
|
/*
|
|
* major device numbers remapping
|
|
*/
|
|
dev_t
|
|
linux_fakedev(dev_t dev, int raw)
|
|
{
|
|
/* XXX write me */
|
|
return dev;
|
|
}
|
|
|
|
/*
|
|
* We come here in a last attempt to satisfy a Linux ioctl() call
|
|
*/
|
|
int
|
|
linux_machdepioctl(struct lwp *l, const struct linux_sys_ioctl_args *uap, register_t *retval)
|
|
{
|
|
return 0;
|
|
}
|
|
|
|
/*
|
|
* See above. If a root process tries to set access to an I/O port,
|
|
* just let it have the whole range.
|
|
*/
|
|
int
|
|
linux_sys_ioperm(struct lwp *l, const struct linux_sys_ioperm_args *uap, register_t *retval)
|
|
{
|
|
/*
|
|
* This syscall is not implemented in Linux/Mips: we should not be here
|
|
*/
|
|
#ifdef DEBUG_LINUX
|
|
printf("linux_sys_ioperm: should not be here.\n");
|
|
#endif /* DEBUG_LINUX */
|
|
return 0;
|
|
}
|
|
|
|
/*
|
|
* wrapper linux_sys_new_uname() -> linux_sys_uname()
|
|
*/
|
|
int
|
|
linux_sys_new_uname(struct lwp *l, const struct linux_sys_new_uname_args *uap, register_t *retval)
|
|
{
|
|
/*
|
|
* Use this if you want to try Linux emulation with a glibc-2.2
|
|
* or higher. Note that signals will not work
|
|
*/
|
|
#if 0
|
|
struct linux_sys_uname_args /* {
|
|
syscallarg(struct linux_utsname *) up;
|
|
} */ *uap = v;
|
|
struct linux_utsname luts;
|
|
|
|
strlcpy(luts.l_sysname, linux_sysname, sizeof(luts.l_sysname));
|
|
strlcpy(luts.l_nodename, hostname, sizeof(luts.l_nodename));
|
|
strlcpy(luts.l_release, "2.4.0", sizeof(luts.l_release));
|
|
strlcpy(luts.l_version, linux_version, sizeof(luts.l_version));
|
|
strlcpy(luts.l_machine, machine, sizeof(luts.l_machine));
|
|
strlcpy(luts.l_domainname, domainname, sizeof(luts.l_domainname));
|
|
|
|
return copyout(&luts, SCARG(uap, up), sizeof(luts));
|
|
#else
|
|
return linux_sys_uname(l, (const void *)uap, retval);
|
|
#endif
|
|
}
|
|
|
|
/*
|
|
* In Linux, cacheflush is currently implemented
|
|
* as a whole cache flush (arguments are ignored)
|
|
* we emulate this broken beahior.
|
|
*/
|
|
int
|
|
linux_sys_cacheflush(struct lwp *l, const struct linux_sys_cacheflush_args *uap, register_t *retval)
|
|
{
|
|
mips_icache_sync_all();
|
|
mips_dcache_wbinv_all();
|
|
return 0;
|
|
}
|
|
|
|
/*
|
|
* This system call is depecated in Linux, but
|
|
* some binaries and some libraries use it.
|
|
*/
|
|
int
|
|
linux_sys_sysmips(struct lwp *l, const struct linux_sys_sysmips_args *uap, register_t *retval)
|
|
{
|
|
/* {
|
|
syscallarg(long) cmd;
|
|
syscallarg(long) arg1;
|
|
syscallarg(long) arg2;
|
|
syscallarg(long) arg3;
|
|
} */
|
|
int error;
|
|
|
|
switch (SCARG(uap, cmd)) {
|
|
case LINUX_SETNAME: {
|
|
char nodename [LINUX___NEW_UTS_LEN + 1];
|
|
int name[2];
|
|
size_t len;
|
|
|
|
if ((error = copyinstr((char *)SCARG(uap, arg1), nodename,
|
|
LINUX___NEW_UTS_LEN, &len)) != 0)
|
|
return error;
|
|
|
|
name[0] = CTL_KERN;
|
|
name[1] = KERN_HOSTNAME;
|
|
return (old_sysctl(&name[0], 2, 0, 0, nodename, len, NULL));
|
|
|
|
break;
|
|
}
|
|
case LINUX_MIPS_ATOMIC_SET: {
|
|
void *addr;
|
|
int s;
|
|
u_int8_t value = 0;
|
|
|
|
addr = (void *)SCARG(uap, arg1);
|
|
|
|
s = splhigh();
|
|
/*
|
|
* No error testing here. This is bad, but Linux does
|
|
* it like this. The source aknowledge "This is broken"
|
|
* in a comment...
|
|
*/
|
|
(void) copyin(addr, &value, 1);
|
|
*retval = value;
|
|
value = (u_int8_t) SCARG(uap, arg2);
|
|
error = copyout(&value, addr, 1);
|
|
splx(s);
|
|
|
|
return 0;
|
|
break;
|
|
}
|
|
case LINUX_MIPS_FIXADE: /* XXX not implemented */
|
|
break;
|
|
case LINUX_FLUSH_CACHE:
|
|
mips_icache_sync_all();
|
|
mips_dcache_wbinv_all();
|
|
break;
|
|
case LINUX_MIPS_RDNVRAM:
|
|
return EIO;
|
|
break;
|
|
default:
|
|
return EINVAL;
|
|
break;
|
|
}
|
|
#ifdef DEBUG_LINUX
|
|
printf("linux_sys_sysmips(): unimplemented command %d\n",
|
|
SCARG(uap,cmd));
|
|
#endif /* DEBUG_LINUX */
|
|
return 0;
|
|
}
|
|
|
|
int
|
|
linux_usertrap(struct lwp *l, vaddr_t trapaddr, void *arg)
|
|
{
|
|
return 0;
|
|
}
|
|
|
|
int
|
|
linux_sys_set_thread_area(struct lwp *l, const struct linux_sys_set_thread_area_args *uap, register_t *retval)
|
|
{
|
|
/* {
|
|
syscallarg(void *) tls;
|
|
} */
|
|
|
|
return lwp_setprivate(l, SCARG(uap, tls));
|
|
}
|