xref: /openbsd-src/sys/arch/amd64/amd64/machdep.c (revision 20bdc91c785fb949f0d36715e4f9c8ac93a86082)

/*	$OpenBSD: machdep.c,v 1.297 2024/09/21 19:06:07 deraadt Exp $	*/
/*	$NetBSD: machdep.c,v 1.3 2003/05/07 22:58:18 fvdl Exp $	*/

/*-
 * Copyright (c) 1996, 1997, 1998, 2000 The NetBSD Foundation, Inc.
 * All rights reserved.
 *
 * This code is derived from software contributed to The NetBSD Foundation
 * by Charles M. Hannum and by Jason R. Thorpe of the Numerical Aerospace
 * Simulation Facility, NASA Ames Research Center.
 *
 * 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.
 */

/*-
 * Copyright (c) 1982, 1987, 1990 The Regents of the University of California.
 * All rights reserved.
 *
 * This code is derived from software contributed to Berkeley by
 * William Jolitz.
 *
 * 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.
 * 3. Neither the name of the University nor the names of its contributors
 *    may be used to endorse or promote products derived from this software
 *    without specific prior written permission.
 *
 * THIS SOFTWARE IS PROVIDED BY THE REGENTS 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 REGENTS 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.
 *
 *	@(#)machdep.c	7.4 (Berkeley) 6/3/91
 */

#include <sys/param.h>
#include <sys/systm.h>
#include <sys/signal.h>
#include <sys/signalvar.h>
#include <sys/proc.h>
#include <sys/user.h>
#include <sys/exec.h>
#include <sys/buf.h>
#include <sys/reboot.h>
#include <sys/conf.h>
#include <sys/msgbuf.h>
#include <sys/mount.h>
#include <sys/extent.h>
#include <sys/core.h>
#include <sys/kcore.h>
#include <sys/syscallargs.h>

#include <dev/cons.h>
#include <stand/boot/bootarg.h>

#include <net/if.h>
#include <uvm/uvm_extern.h>

#include <sys/sysctl.h>

#include <machine/cpu_full.h>
#include <machine/cpufunc.h>
#include <machine/pio.h>
#include <machine/psl.h>
#include <machine/reg.h>
#include <machine/fpu.h>
#include <machine/biosvar.h>
#include <machine/mpbiosvar.h>
#include <machine/kcore.h>
#include <machine/tss.h>

#include <dev/isa/isareg.h>
#include <dev/ic/i8042reg.h>

#ifdef DDB
#include <machine/db_machdep.h>
#include <ddb/db_extern.h>
extern int db_console;
#endif

#include "isa.h"
#include "isadma.h"
#include "ksyms.h"

#include "acpi.h"
#if NACPI > 0
#include <dev/acpi/acpireg.h>
#include <dev/acpi/acpivar.h>
#endif

#include "com.h"
#if NCOM > 0
#include <sys/tty.h>
#include <dev/ic/comvar.h>
#include <dev/ic/comreg.h>
#endif

#include "efi.h"
#if NEFI > 0
#include <dev/efi/efi.h>
#endif

#include "softraid.h"
#if NSOFTRAID > 0
#include <dev/softraidvar.h>
#endif

#ifdef HIBERNATE
#include <machine/hibernate_var.h>
#endif /* HIBERNATE */

#include "ukbd.h"
#include "pckbc.h"
#if NPCKBC > 0 && NUKBD > 0
#include <dev/ic/pckbcvar.h>
#endif

/* #define MACHDEP_DEBUG */

#ifdef MACHDEP_DEBUG
#define DPRINTF(x...)	do { printf(x); } while(0)
#else
#define DPRINTF(x...)
#endif /* MACHDEP_DEBUG */

/* the following is used externally (sysctl_hw) */
char machine[] = MACHINE;

/*
 * switchto vectors
 */
void cpu_idle_cycle_hlt(void);
void (*cpu_idle_cycle_fcn)(void) = &cpu_idle_cycle_hlt;
void (*cpu_suspend_cycle_fcn)(void);

/* the following is used externally for concurrent handlers */
int setperf_prio = 0;

#ifdef CPURESET_DELAY
int	cpureset_delay = CPURESET_DELAY;
#else
int     cpureset_delay = 0;
#endif

char *ssym = 0, *esym = 0;	/* start and end of symbol table */
dev_t bootdev = 0;		/* device we booted from */
int biosbasemem = 0;		/* base memory reported by BIOS */
u_int bootapiver = 0;		/* /boot API version */

int	physmem;
extern int	boothowto;

paddr_t	dumpmem_paddr;
vaddr_t	dumpmem_vaddr;
psize_t	dumpmem_sz;

vaddr_t kern_end;

vaddr_t	msgbuf_vaddr;
paddr_t msgbuf_paddr;

vaddr_t	idt_vaddr;
paddr_t	idt_paddr;

vaddr_t lo32_vaddr;
paddr_t lo32_paddr;
paddr_t tramp_pdirpa;

int kbd_reset;
int lid_action = 1;
int pwr_action = 1;
int forceukbd;

/*
 * safepri is a safe priority for sleep to set for a spin-wait
 * during autoconfiguration or after a panic.
 */
int	safepri = 0;

struct vm_map *exec_map = NULL;
struct vm_map *phys_map = NULL;

/* UVM constraint ranges. */
struct uvm_constraint_range  isa_constraint = { 0x0, 0x00ffffffUL };
struct uvm_constraint_range  dma_constraint = { 0x0, 0xffffffffUL };
struct uvm_constraint_range *uvm_md_constraints[] = {
    &isa_constraint,
    &dma_constraint,
    NULL,
};

paddr_t avail_start;
paddr_t avail_end;

void (*delay_func)(int) = i8254_delay;
void (*initclock_func)(void) = i8254_initclocks;
void (*startclock_func)(void) = i8254_start_both_clocks;

/*
 * Format of boot information passed to us by 32-bit /boot
 */
typedef struct _boot_args32 {
	int	ba_type;
	int	ba_size;
	int	ba_nextX;	/* a ptr in 32-bit world, but not here */
	char	ba_arg[1];
} bootarg32_t;

#define BOOTARGC_MAX	NBPG	/* one page */

bios_bootmac_t *bios_bootmac;

/* locore copies the arguments from /boot to here for us */
char bootinfo[BOOTARGC_MAX];
int bootinfo_size = BOOTARGC_MAX;

void getbootinfo(char *, int);

/* Data passed to us by /boot, filled in by getbootinfo() */
bios_diskinfo_t	*bios_diskinfo;
bios_memmap_t	*bios_memmap;
u_int32_t	bios_cksumlen;
bios_efiinfo_t	*bios_efiinfo;
bios_ucode_t	*bios_ucode;

#if NEFI > 0
EFI_MEMORY_DESCRIPTOR *mmap;
#endif

/*
 * Size of memory segments, before any memory is stolen.
 */
phys_ram_seg_t mem_clusters[VM_PHYSSEG_MAX];
int	mem_cluster_cnt;

int	cpu_dump(void);
int	cpu_dumpsize(void);
u_long	cpu_dump_mempagecnt(void);
void	dumpsys(void);
void	cpu_init_extents(void);
void	map_tramps(void);
void	init_x86_64(paddr_t);
void	(*cpuresetfn)(void);
void	enter_shared_special_pages(void);

#ifdef APERTURE
int allowaperture = 0;
#endif

/*
 * Machine-dependent startup code
 */
void
cpu_startup(void)
{
	vaddr_t minaddr, maxaddr;

	msgbuf_vaddr = PMAP_DIRECT_MAP(msgbuf_paddr);
	initmsgbuf((caddr_t)msgbuf_vaddr, round_page(MSGBUFSIZE));

	printf("%s", version);
	startclocks();
	rtcinit();

	printf("real mem = %lu (%luMB)\n", ptoa((psize_t)physmem),
	    ptoa((psize_t)physmem)/1024/1024);

	/*
	 * Allocate a submap for exec arguments.  This map effectively
	 * limits the number of processes exec'ing at any time.
	 */
	minaddr = vm_map_min(kernel_map);
	exec_map = uvm_km_suballoc(kernel_map, &minaddr, &maxaddr,
				   16*NCARGS, VM_MAP_PAGEABLE, FALSE, NULL);

	/*
	 * Allocate a submap for physio
	 */
	minaddr = vm_map_min(kernel_map);
	phys_map = uvm_km_suballoc(kernel_map, &minaddr, &maxaddr,
				   VM_PHYS_SIZE, 0, FALSE, NULL);

	printf("avail mem = %lu (%luMB)\n", ptoa((psize_t)uvmexp.free),
	    ptoa((psize_t)uvmexp.free)/1024/1024);

	bufinit();

	if (boothowto & RB_CONFIG) {
#ifdef BOOT_CONFIG
		user_config();
#else
		printf("kernel does not support -c; continuing..\n");
#endif
	}

	/* Safe for i/o port / memory space allocation to use malloc now. */
	x86_bus_space_mallocok();

#ifndef SMALL_KERNEL
	cpu_ucode_setup();
	cpu_ucode_apply(&cpu_info_primary);
#endif
	cpu_tsx_disable(&cpu_info_primary);

	/* enter the IDT and trampoline code in the u-k maps */
	enter_shared_special_pages();

	/* initialize CPU0's TSS and GDT and put them in the u-k maps */
	cpu_enter_pages(&cpu_info_full_primary);
}

/*
 * enter_shared_special_pages
 *
 * Requests mapping of various special pages required in the Intel Meltdown
 * case (to be entered into the U-K page table):
 *
 *  1 IDT page
 *  Various number of pages covering the U-K ".kutext" section. This section
 *   contains code needed during trampoline operation
 *  Various number of pages covering the U-K ".kudata" section. This section
 *   contains data accessed by the trampoline, before switching to U+K
 *   (for example, various shared global variables used by IPIs, etc)
 *
 * The linker script places the required symbols in the sections above.
 *
 * On CPUs not affected by Meltdown, the calls to pmap_enter_special below
 * become no-ops.
 */
void
enter_shared_special_pages(void)
{
	extern char __kutext_start[], __kutext_end[], __kernel_kutext_phys[];
	extern char __text_page_start[], __text_page_end[];
	extern char __kernel_kutext_page_phys[];
	extern char __kudata_start[], __kudata_end[], __kernel_kudata_phys[];
	vaddr_t va;
	paddr_t pa;

	/* idt */
	pmap_enter_special(idt_vaddr, idt_paddr, PROT_READ);
	DPRINTF("%s: entered idt page va 0x%llx pa 0x%llx\n", __func__,
	    (uint64_t)idt_vaddr, (uint64_t)idt_paddr);

	/* .kutext section */
	va = (vaddr_t)__kutext_start;
	pa = (paddr_t)__kernel_kutext_phys;
	while (va < (vaddr_t)__kutext_end) {
		pmap_enter_special(va, pa, PROT_READ | PROT_EXEC);
		DPRINTF("%s: entered kutext page va 0x%llx pa 0x%llx\n",
		    __func__, (uint64_t)va, (uint64_t)pa);
		va += PAGE_SIZE;
		pa += PAGE_SIZE;
	}

	/* .kutext.page section */
	va = (vaddr_t)__text_page_start;
	pa = (paddr_t)__kernel_kutext_page_phys;
	while (va < (vaddr_t)__text_page_end) {
		pmap_enter_special(va, pa, PROT_READ | PROT_EXEC);
		DPRINTF("%s: entered kutext.page va 0x%llx pa 0x%llx\n",
		    __func__, (uint64_t)va, (uint64_t)pa);
		va += PAGE_SIZE;
		pa += PAGE_SIZE;
	}

	/* .kudata section */
	va = (vaddr_t)__kudata_start;
	pa = (paddr_t)__kernel_kudata_phys;
	while (va < (vaddr_t)__kudata_end) {
		pmap_enter_special(va, pa, PROT_READ | PROT_WRITE);
		DPRINTF("%s: entered kudata page va 0x%llx pa 0x%llx\n",
		    __func__, (uint64_t)va, (uint64_t)pa);
		va += PAGE_SIZE;
		pa += PAGE_SIZE;
	}
}

/*
 * Set up proc0's PCB and the cpu's TSS.
 */
void
x86_64_proc0_tss_ldt_init(void)
{
	struct pcb *pcb;

	cpu_info_primary.ci_curpcb = pcb = &proc0.p_addr->u_pcb;
	pcb->pcb_fsbase = 0;
	pcb->pcb_kstack = (u_int64_t)proc0.p_addr + USPACE - 16;
	proc0.p_md.md_regs = (struct trapframe *)pcb->pcb_kstack - 1;

	ltr(GSYSSEL(GPROC0_SEL, SEL_KPL));
	lldt(0);
}

bios_diskinfo_t *
bios_getdiskinfo(dev_t dev)
{
	bios_diskinfo_t *pdi;

	if (bios_diskinfo == NULL)
		return NULL;

	for (pdi = bios_diskinfo; pdi->bios_number != -1; pdi++) {
		if ((dev & B_MAGICMASK) == B_DEVMAGIC) { /* search by bootdev */
			if (pdi->bsd_dev == dev)
				break;
		} else {
			if (pdi->bios_number == dev)
				break;
		}
	}

	if (pdi->bios_number == -1)
		return NULL;
	else
		return pdi;
}

int
bios_sysctl(int *name, u_int namelen, void *oldp, size_t *oldlenp, void *newp,
    size_t newlen, struct proc *p)
{
	bios_diskinfo_t *pdi;
	int biosdev;

	/* all sysctl names at this level except diskinfo are terminal */
	if (namelen != 1 && name[0] != BIOS_DISKINFO)
		return (ENOTDIR);	       /* overloaded */

	if (!(bootapiver & BAPIV_VECTOR))
		return EOPNOTSUPP;

	switch (name[0]) {
	case BIOS_DEV:
		if ((pdi = bios_getdiskinfo(bootdev)) == NULL)
			return ENXIO;
		biosdev = pdi->bios_number;
		return sysctl_rdint(oldp, oldlenp, newp, biosdev);
	case BIOS_DISKINFO:
		if (namelen != 2)
			return ENOTDIR;
		if ((pdi = bios_getdiskinfo(name[1])) == NULL)
			return ENXIO;
		return sysctl_rdstruct(oldp, oldlenp, newp, pdi, sizeof(*pdi));
	case BIOS_CKSUMLEN:
		return sysctl_rdint(oldp, oldlenp, newp, bios_cksumlen);
	default:
		return EOPNOTSUPP;
	}
	/* NOTREACHED */
}

extern int tsc_is_invariant;
extern int amd64_has_xcrypt;
extern int need_retpoline;

const struct sysctl_bounded_args cpuctl_vars[] = {
	{ CPU_LIDACTION, &lid_action, -1, 2 },
	{ CPU_PWRACTION, &pwr_action, 0, 2 },
	{ CPU_CPUID, &cpu_id, SYSCTL_INT_READONLY },
	{ CPU_CPUFEATURE, &cpu_feature, SYSCTL_INT_READONLY },
	{ CPU_XCRYPT, &amd64_has_xcrypt, SYSCTL_INT_READONLY },
	{ CPU_INVARIANTTSC, &tsc_is_invariant, SYSCTL_INT_READONLY },
	{ CPU_RETPOLINE, &need_retpoline, SYSCTL_INT_READONLY },
};

/*
 * machine dependent system variables.
 */
int
cpu_sysctl(int *name, u_int namelen, void *oldp, size_t *oldlenp, void *newp,
    size_t newlen, struct proc *p)
{
	extern uint64_t tsc_frequency;
	dev_t consdev;
	dev_t dev;

	switch (name[0]) {
	case CPU_CONSDEV:
		if (namelen != 1)
			return (ENOTDIR);		/* overloaded */
		if (cn_tab != NULL)
			consdev = cn_tab->cn_dev;
		else
			consdev = NODEV;
		return (sysctl_rdstruct(oldp, oldlenp, newp, &consdev,
		    sizeof consdev));
	case CPU_CHR2BLK:
		if (namelen != 2)
			return (ENOTDIR);		/* overloaded */
		dev = chrtoblk((dev_t)name[1]);
		return sysctl_rdstruct(oldp, oldlenp, newp, &dev, sizeof(dev));
	case CPU_BIOS:
		return bios_sysctl(name + 1, namelen - 1, oldp, oldlenp,
		    newp, newlen, p);
	case CPU_CPUVENDOR:
		return (sysctl_rdstring(oldp, oldlenp, newp, cpu_vendor));
	case CPU_KBDRESET:
		return (sysctl_securelevel_int(oldp, oldlenp, newp, newlen,
		    &kbd_reset));
	case CPU_ALLOWAPERTURE:
		if (namelen != 1)
			return (ENOTDIR);		/* overloaded */
#ifdef APERTURE
		if (securelevel > 0)
			return (sysctl_int_lower(oldp, oldlenp, newp, newlen,
			    &allowaperture));
		else
			return (sysctl_int(oldp, oldlenp, newp, newlen,
			    &allowaperture));
#else
		return (sysctl_rdint(oldp, oldlenp, newp, 0));
#endif
#if NPCKBC > 0 && NUKBD > 0
	case CPU_FORCEUKBD:
		{
		int error;

		if (forceukbd)
			return (sysctl_rdint(oldp, oldlenp, newp, forceukbd));

		error = sysctl_int(oldp, oldlenp, newp, newlen, &forceukbd);
		if (forceukbd)
			pckbc_release_console();
		return (error);
		}
#endif
	case CPU_TSCFREQ:
		return (sysctl_rdquad(oldp, oldlenp, newp, tsc_frequency));
	default:
		return (sysctl_bounded_arr(cpuctl_vars, nitems(cpuctl_vars),
		    name, namelen, oldp, oldlenp, newp, newlen));
	}
	/* NOTREACHED */
}

static inline void
maybe_enable_user_cet(struct proc *p)
{
#ifndef SMALL_KERNEL
	/* Enable indirect-branch tracking if present and not disabled */
	if ((xsave_mask & XFEATURE_CET_U) &&
	    (p->p_p->ps_flags & PS_NOBTCFI) == 0) {
		uint64_t msr = rdmsr(MSR_U_CET);
		wrmsr(MSR_U_CET, msr | MSR_CET_ENDBR_EN | MSR_CET_NO_TRACK_EN);
	}
#endif
}

static inline void
initialize_thread_xstate(struct proc *p)
{
	if (cpu_use_xsaves) {
		xrstors(fpu_cleandata, xsave_mask);
		maybe_enable_user_cet(p);
	} else {
		/* Reset FPU state in PCB */
		memcpy(&p->p_addr->u_pcb.pcb_savefpu, fpu_cleandata,
		    fpu_save_len);

		if (curcpu()->ci_pflags & CPUPF_USERXSTATE) {
			/* state in CPU is obsolete; reset it */
			fpureset();
		}
	}

	/* The reset state _is_ the userspace state for this thread now */
	curcpu()->ci_pflags |= CPUPF_USERXSTATE;
}

/*
 * Copy out the FPU state, massaging it to be usable from userspace
 * and acceptable to xrstor_user()
 */
static inline int
copyoutfpu(struct savefpu *sfp, char *sp, size_t len)
{
	uint64_t bvs[2];

	if (copyout(sfp, sp, len))
		return 1;
	if (len > offsetof(struct savefpu, fp_xstate.xstate_bv)) {
		sp  += offsetof(struct savefpu, fp_xstate.xstate_bv);
		len -= offsetof(struct savefpu, fp_xstate.xstate_bv);
		bvs[0] = sfp->fp_xstate.xstate_bv & XFEATURE_XCR0_MASK;
		bvs[1] = sfp->fp_xstate.xstate_xcomp_bv &
		    (XFEATURE_XCR0_MASK | XFEATURE_COMPRESSED);
		if (copyout(bvs, sp, min(len, sizeof bvs)))
			return 1;
	}
	return 0;
}

/*
 * Send an interrupt to process.
 *
 * Stack is set up to allow sigcode to call routine, followed by
 * syscall to sigreturn routine below.  After sigreturn resets the
 * signal mask, the stack, and the frame pointer, it returns to the
 * user specified pc.
 */
int
sendsig(sig_t catcher, int sig, sigset_t mask, const siginfo_t *ksip,
    int info, int onstack)
{
	struct proc *p = curproc;
	struct trapframe *tf = p->p_md.md_regs;
	struct sigcontext ksc;
	struct savefpu *sfp = &p->p_addr->u_pcb.pcb_savefpu;
	register_t sp, scp, sip;
	u_long sss;

	memset(&ksc, 0, sizeof ksc);
	ksc.sc_rdi = tf->tf_rdi;
	ksc.sc_rsi = tf->tf_rsi;
	ksc.sc_rdx = tf->tf_rdx;
	ksc.sc_rcx = tf->tf_rcx;
	ksc.sc_r8  = tf->tf_r8;
	ksc.sc_r9  = tf->tf_r9;
	ksc.sc_r10 = tf->tf_r10;
	ksc.sc_r11 = tf->tf_r11;
	ksc.sc_r12 = tf->tf_r12;
	ksc.sc_r13 = tf->tf_r13;
	ksc.sc_r14 = tf->tf_r14;
	ksc.sc_r15 = tf->tf_r15;
	ksc.sc_rbx = tf->tf_rbx;
	ksc.sc_rax = tf->tf_rax;
	ksc.sc_rbp = tf->tf_rbp;
	ksc.sc_rip = tf->tf_rip;
	ksc.sc_cs  = tf->tf_cs;
	ksc.sc_rflags = tf->tf_rflags;
	ksc.sc_rsp = tf->tf_rsp;
	ksc.sc_ss  = tf->tf_ss;
	ksc.sc_mask = mask;

	/* Allocate space for the signal handler context. */
	if ((p->p_sigstk.ss_flags & SS_DISABLE) == 0 &&
	    !sigonstack(tf->tf_rsp) && onstack)
		sp = trunc_page((vaddr_t)p->p_sigstk.ss_sp + p->p_sigstk.ss_size);
	else
		sp = tf->tf_rsp - 128;

	sp -= fpu_save_len;
	if (cpu_use_xsaves)
		sp &= ~63ULL;	/* just in case */
	else
		sp &= ~15ULL;	/* just in case */

	/* Save FPU state to PCB if necessary, then copy it out */
	if (curcpu()->ci_pflags & CPUPF_USERXSTATE)
		fpusave(&p->p_addr->u_pcb.pcb_savefpu);
	if (copyoutfpu(sfp, (void *)sp, fpu_save_len))
		return 1;

	initialize_thread_xstate(p);

	ksc.sc_fpstate = (struct fxsave64 *)sp;
	sss = (sizeof(ksc) + 15) & ~15;
	sip = 0;
	if (info) {
		sip = sp - ((sizeof(*ksip) + 15) & ~15);
		sss += (sizeof(*ksip) + 15) & ~15;

		if (copyout(ksip, (void *)sip, sizeof(*ksip)))
			return 1;
	}
	scp = sp - sss;

	ksc.sc_cookie = (long)scp ^ p->p_p->ps_sigcookie;
	if (copyout(&ksc, (void *)scp, sizeof(ksc)))
		return 1;

	/*
	 * Build context to run handler in.
	 */
	tf->tf_rax = (u_int64_t)catcher;
	tf->tf_rdi = sig;
	tf->tf_rsi = sip;
	tf->tf_rdx = scp;

	tf->tf_rip = (u_int64_t)p->p_p->ps_sigcode;
	tf->tf_cs = GSEL(GUCODE_SEL, SEL_UPL);
	tf->tf_rflags &= ~(PSL_T|PSL_D|PSL_VM|PSL_AC);
	tf->tf_rsp = scp;
	tf->tf_ss = GSEL(GUDATA_SEL, SEL_UPL);

	return 0;
}

/*
 * System call to cleanup state after a signal
 * has been taken.  Reset signal mask and
 * stack state from context left by sendsig (above).
 * Return to previous pc and psl as specified by
 * context left by sendsig. Check carefully to
 * make sure that the user has not modified the
 * psl to gain improper privileges or to cause
 * a machine fault.
 */
int
sys_sigreturn(struct proc *p, void *v, register_t *retval)
{
	struct sys_sigreturn_args /* {
		syscallarg(struct sigcontext *) sigcntxp;
	} */ *uap = v;
	struct sigcontext ksc, *scp = SCARG(uap, sigcntxp);
	struct trapframe *tf = p->p_md.md_regs;
	struct savefpu *sfp = &p->p_addr->u_pcb.pcb_savefpu;
	int error;

	if (PROC_PC(p) != p->p_p->ps_sigcoderet) {
		sigexit(p, SIGILL);
		return (EPERM);
	}

	if ((error = copyin((caddr_t)scp, &ksc, sizeof ksc)))
		return (error);

	if (ksc.sc_cookie != ((long)scp ^ p->p_p->ps_sigcookie)) {
		sigexit(p, SIGILL);
		return (EFAULT);
	}

	/* Prevent reuse of the sigcontext cookie */
	ksc.sc_cookie = 0;
	(void)copyout(&ksc.sc_cookie, (caddr_t)scp +
	    offsetof(struct sigcontext, sc_cookie), sizeof (ksc.sc_cookie));

	if (((ksc.sc_rflags ^ tf->tf_rflags) & PSL_USERSTATIC) != 0 ||
	    !USERMODE(ksc.sc_cs, ksc.sc_eflags))
		return (EINVAL);

	/* Current FPU state is obsolete; toss it and force a reload */
	if (curcpu()->ci_pflags & CPUPF_USERXSTATE) {
		curcpu()->ci_pflags &= ~CPUPF_USERXSTATE;
		fpureset();
	}

	/* Copy in the FPU state to restore */
	if (__predict_true(ksc.sc_fpstate != NULL)) {
		if ((error = copyin(ksc.sc_fpstate, sfp, fpu_save_len)))
			return error;
		if (xrstor_user(sfp, xsave_mask)) {
			memcpy(sfp, fpu_cleandata, fpu_save_len);
			return EINVAL;
		}
		maybe_enable_user_cet(p);
		curcpu()->ci_pflags |= CPUPF_USERXSTATE;
	} else {
		/* shouldn't happen, but handle it */
		initialize_thread_xstate(p);
	}

	tf->tf_rdi = ksc.sc_rdi;
	tf->tf_rsi = ksc.sc_rsi;
	tf->tf_rdx = ksc.sc_rdx;
	tf->tf_rcx = ksc.sc_rcx;
	tf->tf_r8  = ksc.sc_r8;
	tf->tf_r9  = ksc.sc_r9;
	tf->tf_r10 = ksc.sc_r10;
	tf->tf_r11 = ksc.sc_r11;
	tf->tf_r12 = ksc.sc_r12;
	tf->tf_r13 = ksc.sc_r13;
	tf->tf_r14 = ksc.sc_r14;
	tf->tf_r15 = ksc.sc_r15;
	tf->tf_rbx = ksc.sc_rbx;
	tf->tf_rax = ksc.sc_rax;
	tf->tf_rbp = ksc.sc_rbp;
	tf->tf_rip = ksc.sc_rip;
	tf->tf_cs  = ksc.sc_cs;
	tf->tf_rflags = ksc.sc_rflags;
	tf->tf_rsp = ksc.sc_rsp;
	tf->tf_ss  = ksc.sc_ss;

	/* Restore signal mask. */
	p->p_sigmask = ksc.sc_mask & ~sigcantmask;

	/*
	 * sigreturn() needs to return to userspace via the 'iretq'
	 * method, so that if the process was interrupted (by tick,
	 * an IPI, whatever) as opposed to already being in the kernel
	 * when a signal was being delivered, the process will be
	 * completely restored, including the userland %rcx and %r11
	 * registers which the 'sysretq' instruction cannot restore.
	 * Also need to make sure we can handle faulting on xrstor.
	 */
	p->p_md.md_flags |= MDP_IRET;

	return (EJUSTRETURN);
}

#ifdef MULTIPROCESSOR
/* force a CPU into the kernel, whether or not it's idle */
void
cpu_kick(struct cpu_info *ci)
{
	/* only need to kick other CPUs */
	if (ci != curcpu()) {
		if (cpu_mwait_size > 0) {
			/*
			 * If not idling, then send an IPI, else
			 * just clear the "keep idling" bit.
			 */
			if ((ci->ci_mwait & MWAIT_IN_IDLE) == 0)
				x86_send_ipi(ci, X86_IPI_NOP);
			else
				atomic_clearbits_int(&ci->ci_mwait,
				    MWAIT_KEEP_IDLING);
		} else {
			/* no mwait, so need an IPI */
			x86_send_ipi(ci, X86_IPI_NOP);
		}
	}
}
#endif

/*
 * Notify the current process (p) that it has a signal pending,
 * process as soon as possible.
 */
void
signotify(struct proc *p)
{
	aston(p);
	cpu_kick(p->p_cpu);
}

#ifdef MULTIPROCESSOR
void
cpu_unidle(struct cpu_info *ci)
{
	if (cpu_mwait_size > 0 && (ci->ci_mwait & MWAIT_ONLY)) {
		/*
		 * Just clear the "keep idling" bit; if it wasn't
		 * idling then we didn't need to do anything anyway.
		 */
		atomic_clearbits_int(&ci->ci_mwait, MWAIT_KEEP_IDLING);
		return;
	}

	if (ci != curcpu())
		x86_send_ipi(ci, X86_IPI_NOP);
}
#endif

int	waittime = -1;
struct pcb dumppcb;

__dead void
boot(int howto)
{
#if NACPI > 0
	if ((howto & RB_POWERDOWN) != 0 && acpi_softc)
		acpi_softc->sc_state = ACPI_STATE_S5;
#endif

	if ((howto & RB_POWERDOWN) != 0)
		lid_action = 0;

	if ((howto & RB_RESET) != 0)
		goto doreset;

	if (cold) {
		if ((howto & RB_USERREQ) == 0)
			howto |= RB_HALT;
		goto haltsys;
	}

	boothowto = howto;
	if ((howto & RB_NOSYNC) == 0 && waittime < 0) {
		waittime = 0;
		vfs_shutdown(curproc);

		if ((howto & RB_TIMEBAD) == 0) {
			resettodr();
		} else {
			printf("WARNING: not updating battery clock\n");
		}
	}
	if_downall();

	uvm_shutdown();
	splhigh();
	cold = 1;

	if ((howto & RB_DUMP) != 0)
		dumpsys();

haltsys:
	config_suspend_all(DVACT_POWERDOWN);

#ifdef MULTIPROCESSOR
	x86_broadcast_ipi(X86_IPI_HALT);
#endif

	if ((howto & RB_HALT) != 0) {
#if NACPI > 0 && !defined(SMALL_KERNEL)
		extern int acpi_enabled;

		if (acpi_enabled) {
			delay(500000);
			if ((howto & RB_POWERDOWN) != 0)
				acpi_powerdown();
		}
#endif
		printf("\n");
		printf("The operating system has halted.\n");
		printf("Please press any key to reboot.\n\n");
		cnpollc(1);	/* for proper keyboard command handling */
		cngetc();
		cnpollc(0);
	}

doreset:
	printf("rebooting...\n");
	if (cpureset_delay > 0)
		delay(cpureset_delay * 1000);
	cpu_reset();
	for (;;)
		continue;
	/* NOTREACHED */
}

/*
 * These variables are needed by /sbin/savecore
 */
u_long	dumpmag = 0x8fca0101;	/* magic number */
int 	dumpsize = 0;		/* pages */
long	dumplo = 0; 		/* blocks */

/*
 * cpu_dump: dump the machine-dependent kernel core dump headers.
 */
int
cpu_dump(void)
{
	int (*dump)(dev_t, daddr_t, caddr_t, size_t);
	char buf[dbtob(1)];
	kcore_seg_t *segp;
	cpu_kcore_hdr_t *cpuhdrp;
	phys_ram_seg_t *memsegp;
	caddr_t va;
	int i;

	dump = bdevsw[major(dumpdev)].d_dump;

	memset(buf, 0, sizeof buf);
	segp = (kcore_seg_t *)buf;
	cpuhdrp = (cpu_kcore_hdr_t *)&buf[ALIGN(sizeof(*segp))];
	memsegp = (phys_ram_seg_t *)&buf[ALIGN(sizeof(*segp)) +
	    ALIGN(sizeof(*cpuhdrp))];

	/*
	 * Generate a segment header.
	 */
	CORE_SETMAGIC(*segp, KCORE_MAGIC, MID_MACHINE, CORE_CPU);
	segp->c_size = dbtob(1) - ALIGN(sizeof(*segp));

	/*
	 * Add the machine-dependent header info.
	 */
	cpuhdrp->ptdpaddr = proc0.p_addr->u_pcb.pcb_cr3;
	cpuhdrp->nmemsegs = mem_cluster_cnt;

	/*
	 * Fill in the memory segment descriptors.
	 */
	for (i = 0; i < mem_cluster_cnt; i++) {
		memsegp[i].start = mem_clusters[i].start;
		memsegp[i].size = mem_clusters[i].size & ~PAGE_MASK;
	}

	/*
	 * If we have dump memory then assume the kernel stack is in high
	 * memory and bounce
	 */
	if (dumpmem_vaddr != 0) {
		memcpy((char *)dumpmem_vaddr, buf, sizeof(buf));
		va = (caddr_t)dumpmem_vaddr;
	} else {
		va = (caddr_t)buf;
	}
	return (dump(dumpdev, dumplo, va, dbtob(1)));
}

/*
 * This is called by main to set dumplo and dumpsize.
 * Dumps always skip the first PAGE_SIZE of disk space
 * in case there might be a disk label stored there.
 * If there is extra space, put dump at the end to
 * reduce the chance that swapping trashes it.
 */
void
dumpconf(void)
{
	int nblks, dumpblks;	/* size of dump area */

	if (dumpdev == NODEV ||
	    (nblks = (bdevsw[major(dumpdev)].d_psize)(dumpdev)) == 0)
		return;
	if (nblks <= ctod(1))
		return;

	dumpblks = cpu_dumpsize();
	if (dumpblks < 0)
		return;
	dumpblks += ctod(cpu_dump_mempagecnt());

	/* If dump won't fit (incl. room for possible label), punt. */
	if (dumpblks > (nblks - ctod(1)))
		return;

	/* Put dump at end of partition */
	dumplo = nblks - dumpblks;

	/* dumpsize is in page units, and doesn't include headers. */
	dumpsize = cpu_dump_mempagecnt();
}

/*
 * Doadump comes here after turning off memory management and
 * getting on the dump stack, either when called above, or by
 * the auto-restart code.
 */
#define BYTES_PER_DUMP  MAXPHYS /* must be a multiple of pagesize */

void
dumpsys(void)
{
	u_long totalbytesleft, bytes, i, n, memseg;
	u_long maddr;
	daddr_t blkno;
	void *va;
	int (*dump)(dev_t, daddr_t, caddr_t, size_t);
	int error;

	/* Save registers. */
	savectx(&dumppcb);

	if (dumpdev == NODEV)
		return;

	/*
	 * For dumps during autoconfiguration,
	 * if dump device has already configured...
	 */
	if (dumpsize == 0)
		dumpconf();
	if (dumplo <= 0 || dumpsize == 0) {
		printf("\ndump to dev %u,%u not possible\n", major(dumpdev),
		    minor(dumpdev));
		return;
	}
	printf("\ndumping to dev %u,%u offset %ld\n", major(dumpdev),
	    minor(dumpdev), dumplo);

	error = (*bdevsw[major(dumpdev)].d_psize)(dumpdev);
	printf("dump ");
	if (error == -1) {
		printf("area unavailable\n");
		return;
	}

	if ((error = cpu_dump()) != 0)
		goto err;

	totalbytesleft = ptoa(cpu_dump_mempagecnt());
	blkno = dumplo + cpu_dumpsize();
	dump = bdevsw[major(dumpdev)].d_dump;
	error = 0;

	for (memseg = 0; memseg < mem_cluster_cnt; memseg++) {
		maddr = mem_clusters[memseg].start;
		bytes = mem_clusters[memseg].size;

		for (i = 0; i < bytes; i += n, totalbytesleft -= n) {
			/* Print out how many MBs we have left to go. */
			if ((totalbytesleft % (1024*1024)) < BYTES_PER_DUMP)
				printf("%ld ", totalbytesleft / (1024 * 1024));

			/* Limit size for next transfer. */
			n = bytes - i;
			if (n > BYTES_PER_DUMP)
				n = BYTES_PER_DUMP;
			if (maddr > 0xffffffff) {
				va = (void *)dumpmem_vaddr;
				if (n > dumpmem_sz)
					n = dumpmem_sz;
				memcpy(va, (void *)PMAP_DIRECT_MAP(maddr), n);
			} else {
				va = (void *)PMAP_DIRECT_MAP(maddr);
			}

			error = (*dump)(dumpdev, blkno, va, n);
			if (error)
				goto err;
			maddr += n;
			blkno += btodb(n);		/* XXX? */

#if 0	/* XXX this doesn't work.  grr. */
			/* operator aborting dump? */
			if (sget() != NULL) {
				error = EINTR;
				break;
			}
#endif
		}
	}

 err:
	switch (error) {

	case ENXIO:
		printf("device bad\n");
		break;

	case EFAULT:
		printf("device not ready\n");
		break;

	case EINVAL:
		printf("area improper\n");
		break;

	case EIO:
		printf("i/o error\n");
		break;

	case EINTR:
		printf("aborted from console\n");
		break;

	case 0:
		printf("succeeded\n");
		break;

	default:
		printf("error %d\n", error);
		break;
	}
	printf("\n\n");
	delay(5000000);		/* 5 seconds */
}

/*
 * Force the userspace FS.base to be reloaded from the PCB on return from
 * the kernel, and reset the segment registers (%ds, %es, %fs, and %gs)
 * to their expected userspace value.
 */
void
reset_segs(void)
{
	/*
	 * This operates like the cpu_switchto() sequence: if we
	 * haven't reset %[defg]s already, do so now.
	*/
	if (curcpu()->ci_pflags & CPUPF_USERSEGS) {
		curcpu()->ci_pflags &= ~CPUPF_USERSEGS;
		__asm volatile(
		    "movw %%ax,%%ds\n\t"
		    "movw %%ax,%%es\n\t"
		    "movw %%ax,%%fs\n\t"
		    "cli\n\t"		/* block intr when on user GS.base */
		    "swapgs\n\t"	/* swap from kernel to user GS.base */
		    "movw %%ax,%%gs\n\t"/* set %gs to UDATA and GS.base to 0 */
		    "swapgs\n\t"	/* back to kernel GS.base */
		    "sti" : : "a"(GSEL(GUDATA_SEL, SEL_UPL)));
	}
}

/*
 * Clear registers on exec
 */
void
setregs(struct proc *p, struct exec_package *pack, u_long stack,
    struct ps_strings *arginfo)
{
	struct trapframe *tf;

	initialize_thread_xstate(p);

	/* To reset all registers we have to return via iretq */
	p->p_md.md_flags |= MDP_IRET;

	reset_segs();
	p->p_addr->u_pcb.pcb_fsbase = 0;

	tf = p->p_md.md_regs;
	memset(tf, 0, sizeof *tf);
	tf->tf_rip = pack->ep_entry;
	tf->tf_cs = GSEL(GUCODE_SEL, SEL_UPL);
	tf->tf_rflags = PSL_USERSET;
	tf->tf_rsp = stack;
	tf->tf_ss = GSEL(GUDATA_SEL, SEL_UPL);
}

/*
 * Initialize segments and descriptor tables
 */

struct gate_descriptor *idt;
char idt_allocmap[NIDT];
struct user *proc0paddr = NULL;

void
setgate(struct gate_descriptor *gd, void *func, int ist, int type, int dpl,
    int sel)
{
	gd->gd_looffset = (u_int64_t)func & 0xffff;
	gd->gd_selector = sel;
	gd->gd_ist = ist;
	gd->gd_type = type;
	gd->gd_dpl = dpl;
	gd->gd_p = 1;
	gd->gd_hioffset = (u_int64_t)func >> 16;
	gd->gd_zero = 0;
	gd->gd_xx1 = 0;
	gd->gd_xx2 = 0;
	gd->gd_xx3 = 0;
}

void
unsetgate(struct gate_descriptor *gd)
{
	memset(gd, 0, sizeof (*gd));
}

void
setregion(struct region_descriptor *rd, void *base, u_int16_t limit)
{
	rd->rd_limit = limit;
	rd->rd_base = (u_int64_t)base;
}

/*
 * Note that the base and limit fields are ignored in long mode.
 */
void
set_mem_segment(struct mem_segment_descriptor *sd, void *base, size_t limit,
    int type, int dpl, int gran, int def32, int is64)
{
	sd->sd_lolimit = (unsigned)limit;
	sd->sd_lobase = (unsigned long)base;
	sd->sd_type = type;
	sd->sd_dpl = dpl;
	sd->sd_p = 1;
	sd->sd_hilimit = (unsigned)limit >> 16;
	sd->sd_avl = 0;
	sd->sd_long = is64;
	sd->sd_def32 = def32;
	sd->sd_gran = gran;
	sd->sd_hibase = (unsigned long)base >> 24;
}

void
set_sys_segment(struct sys_segment_descriptor *sd, void *base, size_t limit,
    int type, int dpl, int gran)
{
	memset(sd, 0, sizeof *sd);
	sd->sd_lolimit = (unsigned)limit;
	sd->sd_lobase = (u_int64_t)base;
	sd->sd_type = type;
	sd->sd_dpl = dpl;
	sd->sd_p = 1;
	sd->sd_hilimit = (unsigned)limit >> 16;
	sd->sd_gran = gran;
	sd->sd_hibase = (u_int64_t)base >> 24;
}

void
cpu_init_idt(void)
{
	struct region_descriptor region;

	setregion(&region, idt, NIDT * sizeof(idt[0]) - 1);
	lidt(&region);
}

void
cpu_init_extents(void)
{
	extern struct extent *iomem_ex;
	static int already_done;
	int i;

	/* We get called for each CPU, only first should do this */
	if (already_done)
		return;

	/*
	 * Allocate the physical addresses used by RAM from the iomem
	 * extent map.
	 */
	for (i = 0; i < mem_cluster_cnt; i++) {
		if (extent_alloc_region(iomem_ex, mem_clusters[i].start,
		    mem_clusters[i].size, EX_NOWAIT)) {
			/* XXX What should we do? */
			printf("WARNING: CAN'T ALLOCATE RAM (%llx-%llx)"
			    " FROM IOMEM EXTENT MAP!\n", mem_clusters[i].start,
			    mem_clusters[i].start + mem_clusters[i].size - 1);
		}
	}

	already_done = 1;
}

void
map_tramps(void)
{
#if defined(MULTIPROCESSOR) || \
    (NACPI > 0 && !defined(SMALL_KERNEL))
	struct pmap *kmp = pmap_kernel();
	extern paddr_t tramp_pdirpa;
#ifdef MULTIPROCESSOR
	extern u_char cpu_spinup_trampoline[];
	extern u_char cpu_spinup_trampoline_end[];
	extern u_char mp_tramp_data_start[];
	extern u_char mp_tramp_data_end[];
	extern u_int32_t mp_pdirpa;
#endif

	/*
	 * The initial PML4 pointer must be below 4G, so if the
	 * current one isn't, use a "bounce buffer" and save it
	 * for tramps to use.
	 */
	if (kmp->pm_pdirpa > 0xffffffff) {
		pmap_kenter_pa(lo32_vaddr, lo32_paddr, PROT_READ | PROT_WRITE);
		memcpy((void *)lo32_vaddr, kmp->pm_pdir, PAGE_SIZE);
		tramp_pdirpa = lo32_paddr;
		pmap_kremove(lo32_vaddr, PAGE_SIZE);
	} else
		tramp_pdirpa = kmp->pm_pdirpa;


#ifdef MULTIPROCESSOR
	/* Map MP tramp code and data pages RW for copy */
	pmap_kenter_pa(MP_TRAMPOLINE, MP_TRAMPOLINE,
	    PROT_READ | PROT_WRITE);

	pmap_kenter_pa(MP_TRAMP_DATA, MP_TRAMP_DATA,
	    PROT_READ | PROT_WRITE);

	memset((caddr_t)MP_TRAMPOLINE, 0xcc, PAGE_SIZE);
	memset((caddr_t)MP_TRAMP_DATA, 0xcc, PAGE_SIZE);

	memcpy((caddr_t)MP_TRAMPOLINE,
	    cpu_spinup_trampoline,
	    cpu_spinup_trampoline_end-cpu_spinup_trampoline);

	memcpy((caddr_t)MP_TRAMP_DATA,
		mp_tramp_data_start,
		mp_tramp_data_end - mp_tramp_data_start);

	/*
	 * We need to patch this after we copy the tramp data,
	 * the symbol points into the copied tramp data page.
	 */
	mp_pdirpa = tramp_pdirpa;

	/* Unmap, will be remapped in cpu_start_secondary */
	pmap_kremove(MP_TRAMPOLINE, PAGE_SIZE);
	pmap_kremove(MP_TRAMP_DATA, PAGE_SIZE);
#endif /* MULTIPROCESSOR */
#endif
}

void
cpu_set_vendor(struct cpu_info *ci, int level, const char *vendor)
{
	ci->ci_cpuid_level = level;
	cpuid_level = MIN(cpuid_level, level);

	/* map the vendor string to an integer */
	if (strcmp(vendor, "AuthenticAMD") == 0)
		ci->ci_vendor = CPUV_AMD;
	else if (strcmp(vendor, "GenuineIntel") == 0)
		ci->ci_vendor = CPUV_INTEL;
	else if (strcmp(vendor, "CentaurHauls") == 0)
		ci->ci_vendor = CPUV_VIA;
	else
		ci->ci_vendor = CPUV_UNKNOWN;
}

#define	IDTVEC(name)	__CONCAT(X, name)
typedef void (vector)(void);
extern vector *IDTVEC(exceptions)[];

paddr_t early_pte_pages;

void
init_x86_64(paddr_t first_avail)
{
	struct region_descriptor region;
	bios_memmap_t *bmp;
	int x, ist;
	uint64_t max_dm_size = ((uint64_t)512 * NUM_L4_SLOT_DIRECT) << 30;

	/*
	 * locore0 mapped 3 pages for use before the pmap is initialized
	 * starting at first_avail. These pages are currently used by
	 * efifb to create early-use VAs for the framebuffer before efifb
	 * is attached.
	 */
	early_pte_pages = first_avail;
	first_avail += 3 * NBPG;

	cpu_set_vendor(&cpu_info_primary, cpuid_level, cpu_vendor);
	cpu_init_msrs(&cpu_info_primary);

	proc0.p_addr = proc0paddr;
	cpu_info_primary.ci_curpcb = &proc0.p_addr->u_pcb;

	x86_bus_space_init();

	i8254_startclock();

	/*
	 * Initialize PAGE_SIZE-dependent variables.
	 */
	uvm_setpagesize();

	/*
	 * Boot arguments are in a single page specified by /boot.
	 *
	 * We require the "new" vector form, as well as memory ranges
	 * to be given in bytes rather than KB.
	 *
	 * locore copies the data into bootinfo[] for us.
	 */
	if ((bootapiver & (BAPIV_VECTOR | BAPIV_BMEMMAP)) ==
	    (BAPIV_VECTOR | BAPIV_BMEMMAP)) {
		if (bootinfo_size >= sizeof(bootinfo))
			panic("boot args too big");

		getbootinfo(bootinfo, bootinfo_size);
	} else
		panic("invalid /boot");

	cninit();

/*
 * Memory on the AMD64 port is described by three different things.
 *
 * 1. biosbasemem - This is outdated, and should really only be used to
 *    sanitize the other values. This is what we get back from the BIOS
 *    using the legacy routines, describing memory below 640KB.
 *
 * 2. bios_memmap[] - This is the memory map as the bios has returned
 *    it to us.  It includes memory the kernel occupies, etc.
 *
 * 3. mem_cluster[] - This is the massaged free memory segments after
 *    taking into account the contents of bios_memmap, biosbasemem,
 *    and locore/machdep/pmap kernel allocations of physical
 *    pages.
 *
 * The other thing is that the physical page *RANGE* is described by
 * three more variables:
 *
 * avail_start - This is a physical address of the start of available
 *               pages, until IOM_BEGIN.  This is basically the start
 *               of the UVM managed range of memory, with some holes...
 *
 * avail_end - This is the end of physical pages.  All physical pages
 *             that UVM manages are between avail_start and avail_end.
 *             There are holes...
 *
 * first_avail - This is the first available physical page after the
 *               kernel, page tables, etc.
 *
 * We skip the first few pages for trampolines, hibernate, and to avoid
 * buggy SMI implementations that could corrupt the first 64KB.
 */
	avail_start = 16*PAGE_SIZE;

#ifdef MULTIPROCESSOR
	if (avail_start < MP_TRAMPOLINE + PAGE_SIZE)
		avail_start = MP_TRAMPOLINE + PAGE_SIZE;
	if (avail_start < MP_TRAMP_DATA + PAGE_SIZE)
		avail_start = MP_TRAMP_DATA + PAGE_SIZE;
#endif

#if (NACPI > 0 && !defined(SMALL_KERNEL))
	if (avail_start < ACPI_TRAMPOLINE + PAGE_SIZE)
		avail_start = ACPI_TRAMPOLINE + PAGE_SIZE;
	if (avail_start < ACPI_TRAMP_DATA + PAGE_SIZE)
		avail_start = ACPI_TRAMP_DATA + PAGE_SIZE;
#endif

#ifdef HIBERNATE
	if (avail_start < HIBERNATE_HIBALLOC_PAGE + PAGE_SIZE)
		avail_start = HIBERNATE_HIBALLOC_PAGE + PAGE_SIZE;
#endif /* HIBERNATE */

	/*
	 * We need to go through the BIOS memory map given, and
	 * fill out mem_clusters and mem_cluster_cnt stuff, taking
	 * into account all the points listed above.
	 */
	avail_end = mem_cluster_cnt = 0;
	for (bmp = bios_memmap; bmp->type != BIOS_MAP_END; bmp++) {
		paddr_t s1, s2, e1, e2;

		/* Ignore non-free memory */
		if (bmp->type != BIOS_MAP_FREE)
			continue;
		if (bmp->size < PAGE_SIZE)
			continue;

		/* Init our segment(s), round/trunc to pages */
		s1 = round_page(bmp->addr);
		e1 = trunc_page(bmp->addr + bmp->size);
		s2 = e2 = 0;

		/*
		 * XXX Some buggy ACPI BIOSes use memory that they
		 * declare as free.  Current worst offender is
		 * Supermicro 5019D-FTN4.  Typically the affected memory
		 * areas are small blocks between areas reserved for
		 * ACPI and other BIOS goo.  So skip areas smaller
		 * than 32 MB above the 16 MB boundary (to avoid
		 * affecting legacy stuff).
		 */
		if (s1 > 16*1024*1024 && (e1 - s1) < 32*1024*1024)
			continue;

		/* Check and adjust our segment(s) */
		/* Nuke low pages */
		if (s1 < avail_start) {
			s1 = avail_start;
			if (s1 > e1)
				continue;
		}

		/*
		 * The direct map is limited to 512GB * NUM_L4_SLOT_DIRECT of
		 * memory, so discard anything above that.
		 */
		if (e1 >= max_dm_size) {
			e1 = max_dm_size;
			if (s1 > e1)
				continue;
		}

		/* Crop stuff into "640K hole" */
		if (s1 < IOM_BEGIN && e1 > IOM_BEGIN)
			e1 = IOM_BEGIN;
		if (s1 < biosbasemem && e1 > biosbasemem)
			e1 = biosbasemem;

		/* Split any segments straddling the 16MB boundary */
		if (s1 < 16*1024*1024 && e1 > 16*1024*1024) {
			e2 = e1;
			s2 = e1 = 16*1024*1024;
		}

		/* Store segment(s) */
		if (e1 - s1 >= PAGE_SIZE) {
			mem_clusters[mem_cluster_cnt].start = s1;
			mem_clusters[mem_cluster_cnt].size = e1 - s1;
			mem_cluster_cnt++;
		}
		if (e2 - s2 >= PAGE_SIZE) {
			mem_clusters[mem_cluster_cnt].start = s2;
			mem_clusters[mem_cluster_cnt].size = e2 - s2;
			mem_cluster_cnt++;
		}
		if (avail_end < e1) avail_end = e1;
		if (avail_end < e2) avail_end = e2;
	}

	/*
	 * Call pmap initialization to make new kernel address space.
	 * We must do this before loading pages into the VM system.
	 */
	first_avail = pmap_bootstrap(first_avail, trunc_page(avail_end));

#if NEFI > 0
	/* Relocate the EFI memory map. */
	if (bios_efiinfo && bios_efiinfo->mmap_start) {
		mmap = (EFI_MEMORY_DESCRIPTOR *)PMAP_DIRECT_MAP(first_avail);
		memcpy(mmap, (void *)PMAP_DIRECT_MAP(bios_efiinfo->mmap_start),
		    bios_efiinfo->mmap_size);
		first_avail += round_page(bios_efiinfo->mmap_size);
	}
#endif

	/* Allocate these out of the 640KB base memory */
	if (avail_start != PAGE_SIZE)
		avail_start = pmap_prealloc_lowmem_ptps(avail_start);

	cpu_init_extents();

	/* Make sure the end of the space used by the kernel is rounded. */
	first_avail = round_page(first_avail);
	kern_end = KERNBASE + first_avail;

	/*
	 * Now, load the memory clusters (which have already been
	 * flensed) into the VM system.
	 */
	for (x = 0; x < mem_cluster_cnt; x++) {
		paddr_t seg_start = mem_clusters[x].start;
		paddr_t seg_end = seg_start + mem_clusters[x].size;

		if (seg_start < first_avail) seg_start = first_avail;
		if (seg_start > seg_end) continue;
		if (seg_end - seg_start < PAGE_SIZE) continue;

		physmem += atop(mem_clusters[x].size);

#if DEBUG_MEMLOAD
		printf("loading 0x%lx-0x%lx (0x%lx-0x%lx)\n",
		    seg_start, seg_end, atop(seg_start), atop(seg_end));
#endif
		uvm_page_physload(atop(seg_start), atop(seg_end),
		    atop(seg_start), atop(seg_end), 0);
	}

	/*
         * Now, load the memory between the end of I/O memory "hole"
         * and the kernel.
	 */
	{
		paddr_t seg_start = round_page(IOM_END);
		paddr_t seg_end = trunc_page(KERNTEXTOFF - KERNBASE);

		if (seg_start < seg_end) {
#if DEBUG_MEMLOAD
			printf("loading 0x%lx-0x%lx\n", seg_start, seg_end);
#endif
			uvm_page_physload(atop(seg_start), atop(seg_end),
			    atop(seg_start), atop(seg_end), 0);
		}
	}

#if DEBUG_MEMLOAD
	printf("avail_start = 0x%lx\n", avail_start);
	printf("avail_end = 0x%lx\n", avail_end);
	printf("first_avail = 0x%lx\n", first_avail);
#endif

	/*
	 * Steal memory for the message buffer (at end of core).
	 */
	{
		struct vm_physseg *vps = NULL;
		psize_t sz = round_page(MSGBUFSIZE);
		psize_t reqsz = sz;

		for (x = 0; x < vm_nphysseg; x++) {
			vps = &vm_physmem[x];
			if (ptoa(vps->avail_end) == avail_end)
				break;
		}
		if (x == vm_nphysseg)
			panic("init_x86_64: can't find end of memory");

		/* Shrink so it'll fit in the last segment. */
		if ((vps->avail_end - vps->avail_start) < atop(sz))
			sz = ptoa(vps->avail_end - vps->avail_start);

		vps->avail_end -= atop(sz);
		vps->end -= atop(sz);
		msgbuf_paddr = ptoa(vps->avail_end);

		/* Remove the last segment if it now has no pages. */
		if (vps->start == vps->end) {
			for (vm_nphysseg--; x < vm_nphysseg; x++)
				vm_physmem[x] = vm_physmem[x + 1];
		}

		/* Now find where the new avail_end is. */
		for (avail_end = 0, x = 0; x < vm_nphysseg; x++)
			if (vm_physmem[x].avail_end > avail_end)
				avail_end = vm_physmem[x].avail_end;
		avail_end = ptoa(avail_end);

		/* Warn if the message buffer had to be shrunk. */
		if (sz != reqsz)
			printf("WARNING: %ld bytes not available for msgbuf "
			    "in last cluster (%ld used)\n", reqsz, sz);
	}

	/*
	 * Steal some memory for a dump bouncebuffer if we have memory over
	 * the 32-bit barrier.
	 */
	if (avail_end > 0xffffffff) {
		struct vm_physseg *vps = NULL;
		psize_t sz = round_page(MAX(BYTES_PER_DUMP, dbtob(1)));

		/* XXX assumes segments are ordered */
		for (x = 0; x < vm_nphysseg; x++) {
			vps = &vm_physmem[x];
			/* Find something between 16meg and 4gig */
			if (ptoa(vps->avail_end) <= 0xffffffff &&
			    ptoa(vps->avail_start) >= 0xffffff)
				break;
		}
		if (x == vm_nphysseg)
			panic("init_x86_64: no memory between "
			    "0xffffff-0xffffffff");

		/* Shrink so it'll fit in the segment. */
		if ((vps->avail_end - vps->avail_start) < atop(sz))
			sz = ptoa(vps->avail_end - vps->avail_start);

		vps->avail_end -= atop(sz);
		vps->end -= atop(sz);
		dumpmem_paddr = ptoa(vps->avail_end);
		dumpmem_vaddr = PMAP_DIRECT_MAP(dumpmem_paddr);
		dumpmem_sz = sz;

		/* Remove the last segment if it now has no pages. */
		if (vps->start == vps->end) {
			for (vm_nphysseg--; x < vm_nphysseg; x++)
				vm_physmem[x] = vm_physmem[x + 1];
		}
	}

	pmap_growkernel(VM_MIN_KERNEL_ADDRESS + 32 * 1024 * 1024);

	pmap_kenter_pa(idt_vaddr, idt_paddr, PROT_READ | PROT_WRITE);

	idt = (struct gate_descriptor *)idt_vaddr;
	cpu_info_primary.ci_tss = &cpu_info_full_primary.cif_tss;
	cpu_info_primary.ci_gdt = &cpu_info_full_primary.cif_gdt;

	/* make gdt gates and memory segments */
	set_mem_segment(GDT_ADDR_MEM(cpu_info_primary.ci_gdt, GCODE_SEL), 0,
	    0xfffff, SDT_MEMERA, SEL_KPL, 1, 0, 1);

	set_mem_segment(GDT_ADDR_MEM(cpu_info_primary.ci_gdt, GDATA_SEL), 0,
	    0xfffff, SDT_MEMRWA, SEL_KPL, 1, 0, 1);

	set_mem_segment(GDT_ADDR_MEM(cpu_info_primary.ci_gdt, GUDATA_SEL), 0,
	    atop(VM_MAXUSER_ADDRESS) - 1, SDT_MEMRWA, SEL_UPL, 1, 0, 1);

	set_mem_segment(GDT_ADDR_MEM(cpu_info_primary.ci_gdt, GUCODE_SEL), 0,
	    atop(VM_MAXUSER_ADDRESS) - 1, SDT_MEMERA, SEL_UPL, 1, 0, 1);

	set_sys_segment(GDT_ADDR_SYS(cpu_info_primary.ci_gdt, GPROC0_SEL),
	    cpu_info_primary.ci_tss, sizeof (struct x86_64_tss)-1,
	    SDT_SYS386TSS, SEL_KPL, 0);

	/* exceptions */
	for (x = 0; x < 32; x++) {
		/* trap2 == NMI, trap8 == double fault */
		ist = (x == 2) ? 2 : (x == 8) ? 1 : 0;
		setgate(&idt[x], IDTVEC(exceptions)[x], ist, SDT_SYS386IGT,
		    (x == 3) ? SEL_UPL : SEL_KPL,
		    GSEL(GCODE_SEL, SEL_KPL));
		idt_allocmap[x] = 1;
	}

	setregion(&region, cpu_info_primary.ci_gdt, GDT_SIZE - 1);
	lgdt(&region);

	cpu_init_idt();

	intr_default_setup();

	fpuinit(&cpu_info_primary);

	softintr_init();
	splraise(IPL_IPI);
	intr_enable();

#ifdef DDB
	db_machine_init();
	ddb_init();
	if (boothowto & RB_KDB)
		db_enter();
#endif
}

void
cpu_reset(void)
{
	intr_disable();

	if (cpuresetfn)
		(*cpuresetfn)();

	/*
	 * The keyboard controller has 4 random output pins, one of which is
	 * connected to the RESET pin on the CPU in many PCs.  We tell the
	 * keyboard controller to pulse this line a couple of times.
	 */
	outb(IO_KBD + KBCMDP, KBC_PULSE0);
	delay(100000);
	outb(IO_KBD + KBCMDP, KBC_PULSE0);
	delay(100000);

	/*
	 * Try to cause a triple fault and watchdog reset by making the IDT
	 * invalid and causing a fault.
	 */
	memset((caddr_t)idt, 0, NIDT * sizeof(idt[0]));
	__asm volatile("divl %0,%1" : : "q" (0), "a" (0));

	for (;;)
		continue;
	/* NOTREACHED */
}

/*
 * cpu_dumpsize: calculate size of machine-dependent kernel core dump headers.
 */
int
cpu_dumpsize(void)
{
	int size;

	size = ALIGN(sizeof(kcore_seg_t)) +
	    ALIGN(mem_cluster_cnt * sizeof(phys_ram_seg_t));
	if (roundup(size, dbtob(1)) != dbtob(1))
		return (-1);

	return (1);
}

/*
 * cpu_dump_mempagecnt: calculate the size of RAM (in pages) to be dumped.
 */
u_long
cpu_dump_mempagecnt(void)
{
	u_long i, n;

	n = 0;
	for (i = 0; i < mem_cluster_cnt; i++)
		n += atop(mem_clusters[i].size);
	return (n);
}

/*
 * Figure out which portions of memory are used by the kernel/system.
 */
int
amd64_pa_used(paddr_t addr)
{
	struct vm_page	*pg;

	/* Kernel manages these */
	if ((pg = PHYS_TO_VM_PAGE(addr)) && (pg->pg_flags & PG_DEV) == 0)
		return 1;

	/* Kernel is loaded here */
	if (addr > IOM_END && addr < (kern_end - KERNBASE))
		return 1;

	/* Low memory used for various bootstrap things */
	if (addr < avail_start)
		return 1;

	/*
	 * The only regions I can think of that are left are the things
	 * we steal away from UVM.  The message buffer?
	 * XXX - ignore these for now.
	 */

	return 0;
}

void
cpu_initclocks(void)
{
	(*initclock_func)();
}

void
cpu_startclock(void)
{
	(*startclock_func)();
}

void
need_resched(struct cpu_info *ci)
{
	ci->ci_want_resched = 1;

	/* There's a risk we'll be called before the idle threads start */
	if (ci->ci_curproc) {
		aston(ci->ci_curproc);
		cpu_kick(ci);
	}
}

/*
 * Allocate an IDT vector slot within the given range.
 * XXX needs locking to avoid MP allocation races.
 */

int
idt_vec_alloc(int low, int high)
{
	int vec;

	for (vec = low; vec <= high; vec++) {
		if (idt_allocmap[vec] == 0) {
			idt_allocmap[vec] = 1;
			return vec;
		}
	}
	return 0;
}

int
idt_vec_alloc_range(int low, int high, int num)
{
	int i, vec;

	KASSERT(powerof2(num));
	low = (low + num - 1) & ~(num - 1);
	high = ((high + 1) & ~(num - 1)) - 1;

	for (vec = low; vec <= high; vec += num) {
		for (i = 0; i < num; i++) {
			if (idt_allocmap[vec + i] != 0)
				break;
		}
		if (i == num) {
			for (i = 0; i < num; i++)
				idt_allocmap[vec + i] = 1;
			return vec;
		}
	}
	return 0;
}

void
idt_vec_set(int vec, void (*function)(void))
{
	/*
	 * Vector should be allocated, so no locking needed.
	 */
	KASSERT(idt_allocmap[vec] == 1);
	setgate(&idt[vec], function, 0, SDT_SYS386IGT, SEL_KPL,
	    GSEL(GCODE_SEL, SEL_KPL));
}

void
idt_vec_free(int vec)
{
	unsetgate(&idt[vec]);
	idt_allocmap[vec] = 0;
}

#ifdef DIAGNOSTIC
void
splassert_check(int wantipl, const char *func)
{
	int cpl = curcpu()->ci_ilevel;
	int floor = curcpu()->ci_handled_intr_level;

	if (cpl < wantipl) {
		splassert_fail(wantipl, cpl, func);
	}
	if (floor > wantipl) {
		splassert_fail(wantipl, floor, func);
	}

}
#endif

int
copyin32(const uint32_t *uaddr, uint32_t *kaddr)
{
	if ((vaddr_t)uaddr & 0x3)
		return EFAULT;

	/* copyin(9) is atomic */
	return copyin(uaddr, kaddr, sizeof(uint32_t));
}

void
getbootinfo(char *bootinfo, int bootinfo_size)
{
	bootarg32_t *q;
	bios_ddb_t *bios_ddb;
	bios_bootduid_t *bios_bootduid;
	bios_bootsr_t *bios_bootsr;
#undef BOOTINFO_DEBUG
#ifdef BOOTINFO_DEBUG
	printf("bootargv:");
#endif

	for (q = (bootarg32_t *)bootinfo;
	    (q->ba_type != BOOTARG_END) &&
	    ((((char *)q) - bootinfo) < bootinfo_size);
	    q = (bootarg32_t *)(((char *)q) + q->ba_size)) {

		switch (q->ba_type) {
		case BOOTARG_MEMMAP:
			bios_memmap = (bios_memmap_t *)q->ba_arg;
#ifdef BOOTINFO_DEBUG
			printf(" memmap %p", bios_memmap);
#endif
			break;
		case BOOTARG_DISKINFO:
			bios_diskinfo = (bios_diskinfo_t *)q->ba_arg;
#ifdef BOOTINFO_DEBUG
			printf(" diskinfo %p", bios_diskinfo);
#endif
			break;
		case BOOTARG_APMINFO:
			/* generated by i386 boot loader */
			break;
		case BOOTARG_CKSUMLEN:
			bios_cksumlen = *(u_int32_t *)q->ba_arg;
#ifdef BOOTINFO_DEBUG
			printf(" cksumlen %d", bios_cksumlen);
#endif
			break;
		case BOOTARG_PCIINFO:
			/* generated by i386 boot loader */
			break;
		case BOOTARG_CONSDEV: {
#if NCOM > 0
			bios_consdev_t *cdp = (bios_consdev_t*)q->ba_arg;
			static const int ports[] =
			    { 0x3f8, 0x2f8, 0x3e8, 0x2e8 };
			int unit = minor(cdp->consdev);
			uint64_t consaddr = cdp->consaddr;
			if (consaddr == -1 && unit >= 0 && unit < nitems(ports))
				consaddr = ports[unit];
			if (major(cdp->consdev) == 8 && consaddr != -1) {
				comconsunit = unit;
				comconsaddr = consaddr;
				comconsrate = cdp->conspeed;
				comconsfreq = cdp->consfreq;
				comcons_reg_width = cdp->reg_width;
				comcons_reg_shift = cdp->reg_shift;
				if (cdp->flags & BCD_MMIO)
					comconsiot = X86_BUS_SPACE_MEM;
				else
					comconsiot = X86_BUS_SPACE_IO;
			}
#endif
#ifdef BOOTINFO_DEBUG
			printf(" console 0x%x:%d", cdp->consdev, cdp->conspeed);
#endif
			break;
		}
		case BOOTARG_BOOTMAC:
			bios_bootmac = (bios_bootmac_t *)q->ba_arg;
			break;

		case BOOTARG_DDB:
			bios_ddb = (bios_ddb_t *)q->ba_arg;
#ifdef DDB
			db_console = bios_ddb->db_console;
#endif
			break;

		case BOOTARG_BOOTDUID:
			bios_bootduid = (bios_bootduid_t *)q->ba_arg;
			memcpy(bootduid, bios_bootduid, sizeof(bootduid));
			break;

		case BOOTARG_BOOTSR:
			bios_bootsr = (bios_bootsr_t *)q->ba_arg;
#if NSOFTRAID > 0
			memcpy(&sr_bootuuid, &bios_bootsr->uuid,
			    sizeof(sr_bootuuid));
			memcpy(&sr_bootkey, &bios_bootsr->maskkey,
			    sizeof(sr_bootkey));
#endif
			explicit_bzero(bios_bootsr, sizeof(bios_bootsr_t));
			break;

		case BOOTARG_EFIINFO:
			bios_efiinfo = (bios_efiinfo_t *)q->ba_arg;
			break;

		case BOOTARG_UCODE:
			bios_ucode = (bios_ucode_t *)q->ba_arg;
			break;

		default:
#ifdef BOOTINFO_DEBUG
			printf(" unsupported arg (%d) %p", q->ba_type,
			    q->ba_arg);
#endif
			break;
		}
	}
#ifdef BOOTINFO_DEBUG
	printf("\n");
#endif
}

int
check_context(const struct reg *regs, struct trapframe *tf)
{
	uint16_t sel;

	if (((regs->r_rflags ^ tf->tf_rflags) & PSL_USERSTATIC) != 0)
		return EINVAL;

	sel = regs->r_ss & 0xffff;
	if (!VALID_USER_DSEL(sel))
		return EINVAL;

	sel = regs->r_cs & 0xffff;
	if (!VALID_USER_CSEL(sel))
		return EINVAL;

	if (regs->r_rip >= VM_MAXUSER_ADDRESS)
		return EINVAL;

	return 0;
}

int amd64_delay_quality;

void
delay_init(void(*fn)(int), int fn_quality)
{
	if (fn_quality > amd64_delay_quality) {
		delay_func = fn;
		amd64_delay_quality = fn_quality;
	}
}

void
delay_fini(void (*fn)(int))
{
	if (fn == delay_func) {
		delay_func = i8254_delay;
		amd64_delay_quality = 0;
	}
}