/*
 * QEMU Guest Agent Linux-specific command implementations
 *
 * Copyright IBM Corp. 2011
 *
 * Authors:
 *  Michael Roth      <mdroth@linux.vnet.ibm.com>
 *  Michal Privoznik  <mprivozn@redhat.com>
 *
 * This work is licensed under the terms of the GNU GPL, version 2 or later.
 * See the COPYING file in the top-level directory.
 */

#include "qemu/osdep.h"
#include "qapi/error.h"
#include "qga-qapi-commands.h"
#include "commands-common.h"
#include "cutils.h"
#include <mntent.h>
#include <sys/ioctl.h>

#if defined(CONFIG_FSFREEZE) || defined(CONFIG_FSTRIM)
static int dev_major_minor(const char *devpath,
                           unsigned int *devmajor, unsigned int *devminor)
{
    struct stat st;

    *devmajor = 0;
    *devminor = 0;

    if (stat(devpath, &st) < 0) {
        slog("failed to stat device file '%s': %s", devpath, strerror(errno));
        return -1;
    }
    if (S_ISDIR(st.st_mode)) {
        /* It is bind mount */
        return -2;
    }
    if (S_ISBLK(st.st_mode)) {
        *devmajor = major(st.st_rdev);
        *devminor = minor(st.st_rdev);
        return 0;
    }
    return -1;
}

static bool build_fs_mount_list_from_mtab(FsMountList *mounts, Error **errp)
{
    struct mntent *ment;
    FsMount *mount;
    char const *mtab = "/proc/self/mounts";
    FILE *fp;
    unsigned int devmajor, devminor;

    fp = setmntent(mtab, "r");
    if (!fp) {
        error_setg(errp, "failed to open mtab file: '%s'", mtab);
        return false;
    }

    while ((ment = getmntent(fp))) {
        /*
         * An entry which device name doesn't start with a '/' is
         * either a dummy file system or a network file system.
         * Add special handling for smbfs and cifs as is done by
         * coreutils as well.
         */
        if ((ment->mnt_fsname[0] != '/') ||
            (strcmp(ment->mnt_type, "smbfs") == 0) ||
            (strcmp(ment->mnt_type, "cifs") == 0)) {
            continue;
        }
        if (dev_major_minor(ment->mnt_fsname, &devmajor, &devminor) == -2) {
            /* Skip bind mounts */
            continue;
        }

        mount = g_new0(FsMount, 1);
        mount->dirname = g_strdup(ment->mnt_dir);
        mount->devtype = g_strdup(ment->mnt_type);
        mount->devmajor = devmajor;
        mount->devminor = devminor;

        QTAILQ_INSERT_TAIL(mounts, mount, next);
    }

    endmntent(fp);
    return true;
}

static void decode_mntname(char *name, int len)
{
    int i, j = 0;
    for (i = 0; i <= len; i++) {
        if (name[i] != '\\') {
            name[j++] = name[i];
        } else if (name[i + 1] == '\\') {
            name[j++] = '\\';
            i++;
        } else if (name[i + 1] >= '0' && name[i + 1] <= '3' &&
                   name[i + 2] >= '0' && name[i + 2] <= '7' &&
                   name[i + 3] >= '0' && name[i + 3] <= '7') {
            name[j++] = (name[i + 1] - '0') * 64 +
                        (name[i + 2] - '0') * 8 +
                        (name[i + 3] - '0');
            i += 3;
        } else {
            name[j++] = name[i];
        }
    }
}

/*
 * Walk the mount table and build a list of local file systems
 */
bool build_fs_mount_list(FsMountList *mounts, Error **errp)
{
    FsMount *mount;
    char const *mountinfo = "/proc/self/mountinfo";
    FILE *fp;
    char *line = NULL, *dash;
    size_t n;
    char check;
    unsigned int devmajor, devminor;
    int ret, dir_s, dir_e, type_s, type_e, dev_s, dev_e;

    fp = fopen(mountinfo, "r");
    if (!fp) {
        return build_fs_mount_list_from_mtab(mounts, errp);
    }

    while (getline(&line, &n, fp) != -1) {
        ret = sscanf(line, "%*u %*u %u:%u %*s %n%*s%n%c",
                     &devmajor, &devminor, &dir_s, &dir_e, &check);
        if (ret < 3) {
            continue;
        }
        dash = strstr(line + dir_e, " - ");
        if (!dash) {
            continue;
        }
        ret = sscanf(dash, " - %n%*s%n %n%*s%n%c",
                     &type_s, &type_e, &dev_s, &dev_e, &check);
        if (ret < 1) {
            continue;
        }
        line[dir_e] = 0;
        dash[type_e] = 0;
        dash[dev_e] = 0;
        decode_mntname(line + dir_s, dir_e - dir_s);
        decode_mntname(dash + dev_s, dev_e - dev_s);
        if (devmajor == 0) {
            /* btrfs reports major number = 0 */
            if (strcmp("btrfs", dash + type_s) != 0 ||
                dev_major_minor(dash + dev_s, &devmajor, &devminor) < 0) {
                continue;
            }
        }

        mount = g_new0(FsMount, 1);
        mount->dirname = g_strdup(line + dir_s);
        mount->devtype = g_strdup(dash + type_s);
        mount->devmajor = devmajor;
        mount->devminor = devminor;

        QTAILQ_INSERT_TAIL(mounts, mount, next);
    }
    free(line);

    fclose(fp);
    return true;
}
#endif /* CONFIG_FSFREEZE || CONFIG_FSTRIM */

#ifdef CONFIG_FSFREEZE
/*
 * Walk list of mounted file systems in the guest, and freeze the ones which
 * are real local file systems.
 */
int64_t qmp_guest_fsfreeze_do_freeze_list(bool has_mountpoints,
                                          strList *mountpoints,
                                          FsMountList mounts,
                                          Error **errp)
{
    struct FsMount *mount;
    strList *list;
    int fd, ret, i = 0;

    QTAILQ_FOREACH_REVERSE(mount, &mounts, next) {
        /* To issue fsfreeze in the reverse order of mounts, check if the
         * mount is listed in the list here */
        if (has_mountpoints) {
            for (list = mountpoints; list; list = list->next) {
                if (strcmp(list->value, mount->dirname) == 0) {
                    break;
                }
            }
            if (!list) {
                continue;
            }
        }

        fd = qga_open_cloexec(mount->dirname, O_RDONLY, 0);
        if (fd == -1) {
            error_setg_errno(errp, errno, "failed to open %s", mount->dirname);
            return -1;
        }

        /* we try to cull filesystems we know won't work in advance, but other
         * filesystems may not implement fsfreeze for less obvious reasons.
         * these will report EOPNOTSUPP. we simply ignore these when tallying
         * the number of frozen filesystems.
         * if a filesystem is mounted more than once (aka bind mount) a
         * consecutive attempt to freeze an already frozen filesystem will
         * return EBUSY.
         *
         * any other error means a failure to freeze a filesystem we
         * expect to be freezable, so return an error in those cases
         * and return system to thawed state.
         */
        ret = ioctl(fd, FIFREEZE);
        if (ret == -1) {
            if (errno != EOPNOTSUPP && errno != EBUSY) {
                error_setg_errno(errp, errno, "failed to freeze %s",
                                 mount->dirname);
                close(fd);
                return -1;
            }
        } else {
            i++;
        }
        close(fd);
    }
    return i;
}

int qmp_guest_fsfreeze_do_thaw(Error **errp)
{
    int ret;
    FsMountList mounts;
    FsMount *mount;
    int fd, i = 0, logged;
    Error *local_err = NULL;

    QTAILQ_INIT(&mounts);
    if (!build_fs_mount_list(&mounts, &local_err)) {
        error_propagate(errp, local_err);
        return -1;
    }

    QTAILQ_FOREACH(mount, &mounts, next) {
        logged = false;
        fd = qga_open_cloexec(mount->dirname, O_RDONLY, 0);
        if (fd == -1) {
            continue;
        }
        /* we have no way of knowing whether a filesystem was actually unfrozen
         * as a result of a successful call to FITHAW, only that if an error
         * was returned the filesystem was *not* unfrozen by that particular
         * call.
         *
         * since multiple preceding FIFREEZEs require multiple calls to FITHAW
         * to unfreeze, continuing issuing FITHAW until an error is returned,
         * in which case either the filesystem is in an unfreezable state, or,
         * more likely, it was thawed previously (and remains so afterward).
         *
         * also, since the most recent successful call is the one that did
         * the actual unfreeze, we can use this to provide an accurate count
         * of the number of filesystems unfrozen by guest-fsfreeze-thaw, which
         * may * be useful for determining whether a filesystem was unfrozen
         * during the freeze/thaw phase by a process other than qemu-ga.
         */
        do {
            ret = ioctl(fd, FITHAW);
            if (ret == 0 && !logged) {
                i++;
                logged = true;
            }
        } while (ret == 0);
        close(fd);
    }

    free_fs_mount_list(&mounts);

    return i;
}
#endif /* CONFIG_FSFREEZE */


#define LINUX_SYS_STATE_FILE "/sys/power/state"
#define SUSPEND_SUPPORTED 0
#define SUSPEND_NOT_SUPPORTED 1

typedef enum {
    SUSPEND_MODE_DISK = 0,
    SUSPEND_MODE_RAM = 1,
    SUSPEND_MODE_HYBRID = 2,
} SuspendMode;

/*
 * Executes a command in a child process using g_spawn_sync,
 * returning an int >= 0 representing the exit status of the
 * process.
 *
 * If the program wasn't found in path, returns -1.
 *
 * If a problem happened when creating the child process,
 * returns -1 and errp is set.
 */
static int run_process_child(const char *command[], Error **errp)
{
    int exit_status, spawn_flag;
    GError *g_err = NULL;
    bool success;

    spawn_flag = G_SPAWN_SEARCH_PATH | G_SPAWN_STDOUT_TO_DEV_NULL |
                 G_SPAWN_STDERR_TO_DEV_NULL;

    success =  g_spawn_sync(NULL, (char **)command, NULL, spawn_flag,
                            NULL, NULL, NULL, NULL,
                            &exit_status, &g_err);

    if (success) {
        return WEXITSTATUS(exit_status);
    }

    if (g_err && (g_err->code != G_SPAWN_ERROR_NOENT)) {
        error_setg(errp, "failed to create child process, error '%s'",
                   g_err->message);
    }

    g_error_free(g_err);
    return -1;
}

static bool systemd_supports_mode(SuspendMode mode, Error **errp)
{
    const char *systemctl_args[3] = {"systemd-hibernate", "systemd-suspend",
                                     "systemd-hybrid-sleep"};
    const char *cmd[4] = {"systemctl", "status", systemctl_args[mode], NULL};
    int status;

    status = run_process_child(cmd, errp);

    /*
     * systemctl status uses LSB return codes so we can expect
     * status > 0 and be ok. To assert if the guest has support
     * for the selected suspend mode, status should be < 4. 4 is
     * the code for unknown service status, the return value when
     * the service does not exist. A common value is status = 3
     * (program is not running).
     */
    if (status > 0 && status < 4) {
        return true;
    }

    return false;
}

static void systemd_suspend(SuspendMode mode, Error **errp)
{
    Error *local_err = NULL;
    const char *systemctl_args[3] = {"hibernate", "suspend", "hybrid-sleep"};
    const char *cmd[3] = {"systemctl", systemctl_args[mode], NULL};
    int status;

    status = run_process_child(cmd, &local_err);

    if (status == 0) {
        return;
    }

    if ((status == -1) && !local_err) {
        error_setg(errp, "the helper program 'systemctl %s' was not found",
                   systemctl_args[mode]);
        return;
    }

    if (local_err) {
        error_propagate(errp, local_err);
    } else {
        error_setg(errp, "the helper program 'systemctl %s' returned an "
                   "unexpected exit status code (%d)",
                   systemctl_args[mode], status);
    }
}

static bool pmutils_supports_mode(SuspendMode mode, Error **errp)
{
    Error *local_err = NULL;
    const char *pmutils_args[3] = {"--hibernate", "--suspend",
                                   "--suspend-hybrid"};
    const char *cmd[3] = {"pm-is-supported", pmutils_args[mode], NULL};
    int status;

    status = run_process_child(cmd, &local_err);

    if (status == SUSPEND_SUPPORTED) {
        return true;
    }

    if ((status == -1) && !local_err) {
        return false;
    }

    if (local_err) {
        error_propagate(errp, local_err);
    } else {
        error_setg(errp,
                   "the helper program '%s' returned an unexpected exit"
                   " status code (%d)", "pm-is-supported", status);
    }

    return false;
}

static void pmutils_suspend(SuspendMode mode, Error **errp)
{
    Error *local_err = NULL;
    const char *pmutils_binaries[3] = {"pm-hibernate", "pm-suspend",
                                       "pm-suspend-hybrid"};
    const char *cmd[2] = {pmutils_binaries[mode], NULL};
    int status;

    status = run_process_child(cmd, &local_err);

    if (status == 0) {
        return;
    }

    if ((status == -1) && !local_err) {
        error_setg(errp, "the helper program '%s' was not found",
                   pmutils_binaries[mode]);
        return;
    }

    if (local_err) {
        error_propagate(errp, local_err);
    } else {
        error_setg(errp,
                   "the helper program '%s' returned an unexpected exit"
                   " status code (%d)", pmutils_binaries[mode], status);
    }
}

static bool linux_sys_state_supports_mode(SuspendMode mode, Error **errp)
{
    const char *sysfile_strs[3] = {"disk", "mem", NULL};
    const char *sysfile_str = sysfile_strs[mode];
    char buf[32]; /* hopefully big enough */
    int fd;
    ssize_t ret;

    if (!sysfile_str) {
        error_setg(errp, "unknown guest suspend mode");
        return false;
    }

    fd = open(LINUX_SYS_STATE_FILE, O_RDONLY);
    if (fd < 0) {
        return false;
    }

    ret = read(fd, buf, sizeof(buf) - 1);
    close(fd);
    if (ret <= 0) {
        return false;
    }
    buf[ret] = '\0';

    if (strstr(buf, sysfile_str)) {
        return true;
    }
    return false;
}

static void linux_sys_state_suspend(SuspendMode mode, Error **errp)
{
    g_autoptr(GError) local_gerr = NULL;
    const char *sysfile_strs[3] = {"disk", "mem", NULL};
    const char *sysfile_str = sysfile_strs[mode];

    if (!sysfile_str) {
        error_setg(errp, "unknown guest suspend mode");
        return;
    }

    if (!g_file_set_contents(LINUX_SYS_STATE_FILE, sysfile_str,
                             -1, &local_gerr)) {
        error_setg(errp, "suspend: cannot write to '%s': %s",
                   LINUX_SYS_STATE_FILE, local_gerr->message);
        return;
    }
}

static void guest_suspend(SuspendMode mode, Error **errp)
{
    Error *local_err = NULL;
    bool mode_supported = false;

    if (systemd_supports_mode(mode, &local_err)) {
        mode_supported = true;
        systemd_suspend(mode, &local_err);

        if (!local_err) {
            return;
        }
    }

    error_free(local_err);
    local_err = NULL;

    if (pmutils_supports_mode(mode, &local_err)) {
        mode_supported = true;
        pmutils_suspend(mode, &local_err);

        if (!local_err) {
            return;
        }
    }

    error_free(local_err);
    local_err = NULL;

    if (linux_sys_state_supports_mode(mode, &local_err)) {
        mode_supported = true;
        linux_sys_state_suspend(mode, &local_err);
    }

    if (!mode_supported) {
        error_free(local_err);
        error_setg(errp,
                   "the requested suspend mode is not supported by the guest");
    } else {
        error_propagate(errp, local_err);
    }
}

void qmp_guest_suspend_disk(Error **errp)
{
    guest_suspend(SUSPEND_MODE_DISK, errp);
}

void qmp_guest_suspend_ram(Error **errp)
{
    guest_suspend(SUSPEND_MODE_RAM, errp);
}

void qmp_guest_suspend_hybrid(Error **errp)
{
    guest_suspend(SUSPEND_MODE_HYBRID, errp);
}

/* Transfer online/offline status between @vcpu and the guest system.
 *
 * On input either @errp or *@errp must be NULL.
 *
 * In system-to-@vcpu direction, the following @vcpu fields are accessed:
 * - R: vcpu->logical_id
 * - W: vcpu->online
 * - W: vcpu->can_offline
 *
 * In @vcpu-to-system direction, the following @vcpu fields are accessed:
 * - R: vcpu->logical_id
 * - R: vcpu->online
 *
 * Written members remain unmodified on error.
 */
static void transfer_vcpu(GuestLogicalProcessor *vcpu, bool sys2vcpu,
                          char *dirpath, Error **errp)
{
    int fd;
    int res;
    int dirfd;
    static const char fn[] = "online";

    dirfd = open(dirpath, O_RDONLY | O_DIRECTORY);
    if (dirfd == -1) {
        error_setg_errno(errp, errno, "open(\"%s\")", dirpath);
        return;
    }

    fd = openat(dirfd, fn, sys2vcpu ? O_RDONLY : O_RDWR);
    if (fd == -1) {
        if (errno != ENOENT) {
            error_setg_errno(errp, errno, "open(\"%s/%s\")", dirpath, fn);
        } else if (sys2vcpu) {
            vcpu->online = true;
            vcpu->can_offline = false;
        } else if (!vcpu->online) {
            error_setg(errp, "logical processor #%" PRId64 " can't be "
                       "offlined", vcpu->logical_id);
        } /* otherwise pretend successful re-onlining */
    } else {
        unsigned char status;

        res = pread(fd, &status, 1, 0);
        if (res == -1) {
            error_setg_errno(errp, errno, "pread(\"%s/%s\")", dirpath, fn);
        } else if (res == 0) {
            error_setg(errp, "pread(\"%s/%s\"): unexpected EOF", dirpath,
                       fn);
        } else if (sys2vcpu) {
            vcpu->online = (status != '0');
            vcpu->can_offline = true;
        } else if (vcpu->online != (status != '0')) {
            status = '0' + vcpu->online;
            if (pwrite(fd, &status, 1, 0) == -1) {
                error_setg_errno(errp, errno, "pwrite(\"%s/%s\")", dirpath,
                                 fn);
            }
        } /* otherwise pretend successful re-(on|off)-lining */

        res = close(fd);
        g_assert(res == 0);
    }

    res = close(dirfd);
    g_assert(res == 0);
}

GuestLogicalProcessorList *qmp_guest_get_vcpus(Error **errp)
{
    GuestLogicalProcessorList *head, **tail;
    const char *cpu_dir = "/sys/devices/system/cpu";
    const gchar *line;
    g_autoptr(GDir) cpu_gdir = NULL;
    Error *local_err = NULL;

    head = NULL;
    tail = &head;
    cpu_gdir = g_dir_open(cpu_dir, 0, NULL);

    if (cpu_gdir == NULL) {
        error_setg_errno(errp, errno, "failed to list entries: %s", cpu_dir);
        return NULL;
    }

    while (local_err == NULL && (line = g_dir_read_name(cpu_gdir)) != NULL) {
        GuestLogicalProcessor *vcpu;
        int64_t id;
        if (sscanf(line, "cpu%" PRId64, &id)) {
            g_autofree char *path = g_strdup_printf("/sys/devices/system/cpu/"
                                                    "cpu%" PRId64 "/", id);
            vcpu = g_malloc0(sizeof *vcpu);
            vcpu->logical_id = id;
            vcpu->has_can_offline = true; /* lolspeak ftw */
            transfer_vcpu(vcpu, true, path, &local_err);
            QAPI_LIST_APPEND(tail, vcpu);
        }
    }

    if (local_err == NULL) {
        /* there's no guest with zero VCPUs */
        g_assert(head != NULL);
        return head;
    }

    qapi_free_GuestLogicalProcessorList(head);
    error_propagate(errp, local_err);
    return NULL;
}

int64_t qmp_guest_set_vcpus(GuestLogicalProcessorList *vcpus, Error **errp)
{
    int64_t processed;
    Error *local_err = NULL;

    processed = 0;
    while (vcpus != NULL) {
        char *path = g_strdup_printf("/sys/devices/system/cpu/cpu%" PRId64 "/",
                                     vcpus->value->logical_id);

        transfer_vcpu(vcpus->value, false, path, &local_err);
        g_free(path);
        if (local_err != NULL) {
            break;
        }
        ++processed;
        vcpus = vcpus->next;
    }

    if (local_err != NULL) {
        if (processed == 0) {
            error_propagate(errp, local_err);
        } else {
            error_free(local_err);
        }
    }

    return processed;
}