NetBSD/lib/libnvmm/libnvmm.c
reinoud e0d9d410b7 Implement nvmm_vcpu::stop, a race-free exit from nvmm_vcpu_run() without
signals. This introduces a new kernel and userland NVMM version indicating
this support.

Patch by Kamil Rytarowski <kamil@netbsd.org> and committed on his request.

This is the missing libnvmm part I forgot to include in the origional commit.
2021-04-06 08:40:17 +00:00

574 lines
11 KiB
C

/* $NetBSD: libnvmm.c,v 1.20 2021/04/06 08:40:17 reinoud Exp $ */
/*
* Copyright (c) 2018-2020 Maxime Villard, m00nbsd.net
* All rights reserved.
*
* This code is part of the NVMM hypervisor.
*
* 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 AUTHOR ``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 AUTHOR 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>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <unistd.h>
#include <fcntl.h>
#include <errno.h>
#include <sys/ioctl.h>
#include <sys/mman.h>
#include <sys/queue.h>
#include <machine/vmparam.h>
#include "nvmm.h"
static struct nvmm_capability __capability;
#ifdef __x86_64__
#include "libnvmm_x86.c"
#endif
typedef struct __area {
LIST_ENTRY(__area) list;
gpaddr_t gpa;
uintptr_t hva;
size_t size;
nvmm_prot_t prot;
} area_t;
typedef LIST_HEAD(, __area) area_list_t;
static int nvmm_fd = -1;
/* -------------------------------------------------------------------------- */
static bool
__area_isvalid(struct nvmm_machine *mach, uintptr_t hva, gpaddr_t gpa,
size_t size)
{
area_list_t *areas = mach->areas;
area_t *ent;
LIST_FOREACH(ent, areas, list) {
/* Collision on GPA */
if (gpa >= ent->gpa && gpa < ent->gpa + ent->size) {
return false;
}
if (gpa + size > ent->gpa &&
gpa + size <= ent->gpa + ent->size) {
return false;
}
if (gpa <= ent->gpa && gpa + size >= ent->gpa + ent->size) {
return false;
}
}
return true;
}
static int
__area_add(struct nvmm_machine *mach, uintptr_t hva, gpaddr_t gpa, size_t size,
int prot)
{
area_list_t *areas = mach->areas;
nvmm_prot_t nprot;
area_t *area;
nprot = 0;
if (prot & PROT_READ)
nprot |= NVMM_PROT_READ;
if (prot & PROT_WRITE)
nprot |= NVMM_PROT_WRITE;
if (prot & PROT_EXEC)
nprot |= NVMM_PROT_EXEC;
if (!__area_isvalid(mach, hva, gpa, size)) {
errno = EINVAL;
return -1;
}
area = malloc(sizeof(*area));
if (area == NULL)
return -1;
area->gpa = gpa;
area->hva = hva;
area->size = size;
area->prot = nprot;
LIST_INSERT_HEAD(areas, area, list);
return 0;
}
static int
__area_delete(struct nvmm_machine *mach, uintptr_t hva, gpaddr_t gpa,
size_t size)
{
area_list_t *areas = mach->areas;
area_t *ent, *nxt;
LIST_FOREACH_SAFE(ent, areas, list, nxt) {
if (hva == ent->hva && gpa == ent->gpa && size == ent->size) {
LIST_REMOVE(ent, list);
free(ent);
return 0;
}
}
return -1;
}
static void
__area_remove_all(struct nvmm_machine *mach)
{
area_list_t *areas = mach->areas;
area_t *ent;
while ((ent = LIST_FIRST(areas)) != NULL) {
LIST_REMOVE(ent, list);
free(ent);
}
free(areas);
}
/* -------------------------------------------------------------------------- */
int
nvmm_init(void)
{
if (nvmm_fd != -1)
return 0;
nvmm_fd = open("/dev/nvmm", O_RDONLY | O_CLOEXEC);
if (nvmm_fd == -1)
return -1;
if (nvmm_capability(&__capability) == -1) {
close(nvmm_fd);
nvmm_fd = -1;
return -1;
}
if (__capability.version != NVMM_KERN_VERSION) {
close(nvmm_fd);
nvmm_fd = -1;
errno = EPROGMISMATCH;
return -1;
}
return 0;
}
int
nvmm_root_init(void)
{
if (nvmm_fd != -1)
return 0;
nvmm_fd = open("/dev/nvmm", O_WRONLY | O_CLOEXEC);
if (nvmm_fd == -1)
return -1;
if (nvmm_capability(&__capability) == -1) {
close(nvmm_fd);
nvmm_fd = -1;
return -1;
}
if (__capability.version != NVMM_KERN_VERSION) {
close(nvmm_fd);
nvmm_fd = -1;
errno = EPROGMISMATCH;
return -1;
}
return 0;
}
int
nvmm_capability(struct nvmm_capability *cap)
{
struct nvmm_ioc_capability args;
int ret;
ret = ioctl(nvmm_fd, NVMM_IOC_CAPABILITY, &args);
if (ret == -1)
return -1;
memcpy(cap, &args.cap, sizeof(args.cap));
return 0;
}
int
nvmm_machine_create(struct nvmm_machine *mach)
{
struct nvmm_ioc_machine_create args;
struct nvmm_comm_page **pages;
area_list_t *areas;
int ret;
areas = calloc(1, sizeof(*areas));
if (areas == NULL)
return -1;
pages = calloc(__capability.max_vcpus, sizeof(*pages));
if (pages == NULL) {
free(areas);
return -1;
}
ret = ioctl(nvmm_fd, NVMM_IOC_MACHINE_CREATE, &args);
if (ret == -1) {
free(areas);
return -1;
}
LIST_INIT(areas);
memset(mach, 0, sizeof(*mach));
mach->machid = args.machid;
mach->pages = pages;
mach->areas = areas;
return 0;
}
int
nvmm_machine_destroy(struct nvmm_machine *mach)
{
struct nvmm_ioc_machine_destroy args;
int ret;
args.machid = mach->machid;
ret = ioctl(nvmm_fd, NVMM_IOC_MACHINE_DESTROY, &args);
if (ret == -1)
return -1;
__area_remove_all(mach);
free(mach->pages);
return 0;
}
int
nvmm_machine_configure(struct nvmm_machine *mach, uint64_t op, void *conf)
{
struct nvmm_ioc_machine_configure args;
int ret;
args.machid = mach->machid;
args.op = op;
args.conf = conf;
ret = ioctl(nvmm_fd, NVMM_IOC_MACHINE_CONFIGURE, &args);
if (ret == -1)
return -1;
return 0;
}
int
nvmm_vcpu_create(struct nvmm_machine *mach, nvmm_cpuid_t cpuid,
struct nvmm_vcpu *vcpu)
{
struct nvmm_ioc_vcpu_create args;
struct nvmm_comm_page *comm;
int ret;
args.machid = mach->machid;
args.cpuid = cpuid;
ret = ioctl(nvmm_fd, NVMM_IOC_VCPU_CREATE, &args);
if (ret == -1)
return -1;
comm = mmap(NULL, PAGE_SIZE, PROT_READ|PROT_WRITE, MAP_SHARED|MAP_FILE,
nvmm_fd, NVMM_COMM_OFF(mach->machid, cpuid));
if (comm == MAP_FAILED)
return -1;
mach->pages[cpuid] = comm;
vcpu->cpuid = cpuid;
vcpu->state = &comm->state;
vcpu->event = &comm->event;
vcpu->stop = &comm->stop;
vcpu->exit = malloc(sizeof(*vcpu->exit));
return 0;
}
int
nvmm_vcpu_destroy(struct nvmm_machine *mach, struct nvmm_vcpu *vcpu)
{
struct nvmm_ioc_vcpu_destroy args;
struct nvmm_comm_page *comm;
int ret;
args.machid = mach->machid;
args.cpuid = vcpu->cpuid;
ret = ioctl(nvmm_fd, NVMM_IOC_VCPU_DESTROY, &args);
if (ret == -1)
return -1;
comm = mach->pages[vcpu->cpuid];
munmap(comm, PAGE_SIZE);
free(vcpu->exit);
return 0;
}
int
nvmm_vcpu_configure(struct nvmm_machine *mach, struct nvmm_vcpu *vcpu,
uint64_t op, void *conf)
{
struct nvmm_ioc_vcpu_configure args;
int ret;
switch (op) {
case NVMM_VCPU_CONF_CALLBACKS:
memcpy(&vcpu->cbs, conf, sizeof(vcpu->cbs));
return 0;
}
args.machid = mach->machid;
args.cpuid = vcpu->cpuid;
args.op = op;
args.conf = conf;
ret = ioctl(nvmm_fd, NVMM_IOC_VCPU_CONFIGURE, &args);
if (ret == -1)
return -1;
return 0;
}
int
nvmm_vcpu_setstate(struct nvmm_machine *mach, struct nvmm_vcpu *vcpu,
uint64_t flags)
{
struct nvmm_comm_page *comm;
comm = mach->pages[vcpu->cpuid];
comm->state_commit |= flags;
comm->state_cached |= flags;
return 0;
}
int
nvmm_vcpu_getstate(struct nvmm_machine *mach, struct nvmm_vcpu *vcpu,
uint64_t flags)
{
struct nvmm_ioc_vcpu_getstate args;
struct nvmm_comm_page *comm;
int ret;
comm = mach->pages[vcpu->cpuid];
if (__predict_true((flags & ~comm->state_cached) == 0)) {
return 0;
}
comm->state_wanted = flags & ~comm->state_cached;
args.machid = mach->machid;
args.cpuid = vcpu->cpuid;
ret = ioctl(nvmm_fd, NVMM_IOC_VCPU_GETSTATE, &args);
if (ret == -1)
return -1;
return 0;
}
int
nvmm_vcpu_inject(struct nvmm_machine *mach, struct nvmm_vcpu *vcpu)
{
struct nvmm_comm_page *comm;
comm = mach->pages[vcpu->cpuid];
comm->event_commit = true;
return 0;
}
int
nvmm_vcpu_run(struct nvmm_machine *mach, struct nvmm_vcpu *vcpu)
{
struct nvmm_ioc_vcpu_run args;
int ret;
args.machid = mach->machid;
args.cpuid = vcpu->cpuid;
memset(&args.exit, 0, sizeof(args.exit));
ret = ioctl(nvmm_fd, NVMM_IOC_VCPU_RUN, &args);
if (ret == -1)
return -1;
/* No comm support yet, just copy. */
memcpy(vcpu->exit, &args.exit, sizeof(args.exit));
return 0;
}
int
nvmm_gpa_map(struct nvmm_machine *mach, uintptr_t hva, gpaddr_t gpa,
size_t size, int prot)
{
struct nvmm_ioc_gpa_map args;
int ret;
ret = __area_add(mach, hva, gpa, size, prot);
if (ret == -1)
return -1;
args.machid = mach->machid;
args.hva = hva;
args.gpa = gpa;
args.size = size;
args.prot = prot;
ret = ioctl(nvmm_fd, NVMM_IOC_GPA_MAP, &args);
if (ret == -1) {
/* Can't recover. */
abort();
}
return 0;
}
int
nvmm_gpa_unmap(struct nvmm_machine *mach, uintptr_t hva, gpaddr_t gpa,
size_t size)
{
struct nvmm_ioc_gpa_unmap args;
int ret;
ret = __area_delete(mach, hva, gpa, size);
if (ret == -1)
return -1;
args.machid = mach->machid;
args.gpa = gpa;
args.size = size;
ret = ioctl(nvmm_fd, NVMM_IOC_GPA_UNMAP, &args);
if (ret == -1) {
/* Can't recover. */
abort();
}
return 0;
}
int
nvmm_hva_map(struct nvmm_machine *mach, uintptr_t hva, size_t size)
{
struct nvmm_ioc_hva_map args;
int ret;
args.machid = mach->machid;
args.hva = hva;
args.size = size;
ret = ioctl(nvmm_fd, NVMM_IOC_HVA_MAP, &args);
if (ret == -1)
return -1;
return 0;
}
int
nvmm_hva_unmap(struct nvmm_machine *mach, uintptr_t hva, size_t size)
{
struct nvmm_ioc_hva_unmap args;
int ret;
args.machid = mach->machid;
args.hva = hva;
args.size = size;
ret = ioctl(nvmm_fd, NVMM_IOC_HVA_UNMAP, &args);
if (ret == -1)
return -1;
return 0;
}
/*
* nvmm_gva_to_gpa(): architecture-specific.
*/
int
nvmm_gpa_to_hva(struct nvmm_machine *mach, gpaddr_t gpa, uintptr_t *hva,
nvmm_prot_t *prot)
{
area_list_t *areas = mach->areas;
area_t *ent;
LIST_FOREACH(ent, areas, list) {
if (gpa >= ent->gpa && gpa < ent->gpa + ent->size) {
*hva = ent->hva + (gpa - ent->gpa);
*prot = ent->prot;
return 0;
}
}
errno = ENOENT;
return -1;
}
/*
* nvmm_assist_io(): architecture-specific.
*/
/*
* nvmm_assist_mem(): architecture-specific.
*/
int
nvmm_ctl(int op, void *data, size_t size)
{
struct nvmm_ioc_ctl args;
int ret;
args.op = op;
args.data = data;
args.size = size;
ret = ioctl(nvmm_fd, NVMM_IOC_CTL, &args);
if (ret == -1)
return -1;
return 0;
}
int
nvmm_vcpu_stop(struct nvmm_vcpu *vcpu)
{
*vcpu->stop = 1;
return 0;
}