#include <sys/un.h>
#include <sys/resource.h>
#include <sys/vfs.h>
+#include <sys/mman.h>
#include <linux/magic.h>
#include <poll.h>
#include <ctype.h>
#include <dirent.h>
-#include <signal.h>
#include <dlfcn.h>
-
+#include <limits.h>
/*
* following taken from linux/blkpg.h because they aren't
return is_dlm_hooks_ready ? 1 : 0;
}
-#ifndef MDASSEMBLE
static struct dlm_hooks *dlm_hooks = NULL;
struct dlm_lock_resource *dlm_lock_res = NULL;
static int ast_called = 0;
static char *cluster_name = NULL;
/* Create the lockspace, take bitmapXXX locks on all the bitmaps. */
-int cluster_get_dlmlock(int *lockid)
+int cluster_get_dlmlock(void)
{
int ret = -1;
char str[64];
int flags = LKF_NOQUEUE;
+ int retry_count = 0;
+
+ if (!dlm_funs_ready()) {
+ pr_err("Something wrong with dlm library\n");
+ return -1;
+ }
ret = get_cluster_name(&cluster_name);
if (ret) {
}
dlm_lock_res = xmalloc(sizeof(struct dlm_lock_resource));
- dlm_lock_res->ls = dlm_hooks->create_lockspace(cluster_name, O_RDWR);
+ dlm_lock_res->ls = dlm_hooks->open_lockspace(cluster_name);
if (!dlm_lock_res->ls) {
- pr_err("%s failed to create lockspace\n", cluster_name);
- return -ENOMEM;
+ dlm_lock_res->ls = dlm_hooks->create_lockspace(cluster_name, O_RDWR);
+ if (!dlm_lock_res->ls) {
+ pr_err("%s failed to create lockspace\n", cluster_name);
+ return -ENOMEM;
+ }
+ } else {
+ pr_err("open existed %s lockspace\n", cluster_name);
}
snprintf(str, 64, "bitmap%s", cluster_name);
- ret = dlm_hooks->ls_lock(dlm_lock_res->ls, LKM_PWMODE, &dlm_lock_res->lksb,
- flags, str, strlen(str), 0, dlm_ast,
- dlm_lock_res, NULL, NULL);
+retry:
+ ret = dlm_hooks->ls_lock(dlm_lock_res->ls, LKM_PWMODE,
+ &dlm_lock_res->lksb, flags, str, strlen(str),
+ 0, dlm_ast, dlm_lock_res, NULL, NULL);
if (ret) {
pr_err("error %d when get PW mode on lock %s\n", errno, str);
+ /* let's try several times if EAGAIN happened */
+ if (dlm_lock_res->lksb.sb_status == EAGAIN && retry_count < 10) {
+ sleep_for(10, 0, true);
+ retry_count++;
+ goto retry;
+ }
dlm_hooks->release_lockspace(cluster_name, dlm_lock_res->ls, 1);
return ret;
}
/* Wait for it to complete */
poll_for_ast(dlm_lock_res->ls);
- *lockid = dlm_lock_res->lksb.sb_lkid;
- return dlm_lock_res->lksb.sb_status;
+ if (dlm_lock_res->lksb.sb_status) {
+ pr_err("failed to lock cluster\n");
+ return -1;
+ }
+ return 1;
}
-int cluster_release_dlmlock(int lockid)
+int cluster_release_dlmlock(void)
{
int ret = -1;
if (!cluster_name)
- return -1;
+ goto out;
+
+ if (!dlm_lock_res->lksb.sb_lkid)
+ goto out;
- ret = dlm_hooks->ls_unlock(dlm_lock_res->ls, lockid, 0,
- &dlm_lock_res->lksb, dlm_lock_res);
+ ret = dlm_hooks->ls_unlock_wait(dlm_lock_res->ls,
+ dlm_lock_res->lksb.sb_lkid, 0,
+ &dlm_lock_res->lksb);
if (ret) {
pr_err("error %d happened when unlock\n", errno);
/* XXX make sure the lock is unlocked eventually */
errno = dlm_lock_res->lksb.sb_status;
if (errno != EUNLOCK) {
- pr_err("error %d happened in ast when unlock lockspace\n", errno);
+ pr_err("error %d happened in ast when unlock lockspace\n",
+ errno);
/* XXX make sure the lockspace is unlocked eventually */
goto out;
}
out:
return ret;
}
-#else
-int cluster_get_dlmlock(int *lockid)
+
+int md_array_valid(int fd)
{
- return -1;
+ struct mdinfo *sra;
+ int ret;
+
+ sra = sysfs_read(fd, NULL, GET_ARRAY_STATE);
+ if (sra) {
+ if (sra->array_state != ARRAY_UNKNOWN_STATE)
+ ret = 0;
+ else
+ ret = -ENODEV;
+
+ free(sra);
+ } else {
+ /*
+ * GET_ARRAY_INFO doesn't provide access to the proper state
+ * information, so fallback to a basic check for raid_disks != 0
+ */
+ ret = ioctl(fd, RAID_VERSION);
+ }
+
+ return !ret;
}
-int cluster_release_dlmlock(int lockid)
+
+int md_array_active(int fd)
{
- return -1;
+ struct mdinfo *sra;
+ struct mdu_array_info_s array;
+ int ret = 0;
+
+ sra = sysfs_read(fd, NULL, GET_ARRAY_STATE);
+ if (sra) {
+ if (!md_array_is_active(sra))
+ ret = -ENODEV;
+
+ free(sra);
+ } else {
+ /*
+ * GET_ARRAY_INFO doesn't provide access to the proper state
+ * information, so fallback to a basic check for raid_disks != 0
+ */
+ ret = md_get_array_info(fd, &array);
+ }
+
+ return !ret;
+}
+
+int md_array_is_active(struct mdinfo *info)
+{
+ return (info->array_state != ARRAY_CLEAR &&
+ info->array_state != ARRAY_INACTIVE &&
+ info->array_state != ARRAY_UNKNOWN_STATE);
}
-#endif
/*
* Get array info from the kernel. Longer term we want to deprecate the
}
/*
- * Get disk info from the kernel.
+ * Set array info
*/
-int md_get_disk_info(int fd, struct mdu_disk_info_s *disk)
+int md_set_array_info(int fd, struct mdu_array_info_s *array)
{
- return ioctl(fd, GET_DISK_INFO, disk);
-}
-
-/*
- * Parse a 128 bit uuid in 4 integers
- * format is 32 hexx nibbles with options :.<space> separator
- * If not exactly 32 hex digits are found, return 0
- * else return 1
- */
-int parse_uuid(char *str, int uuid[4])
-{
- int hit = 0; /* number of Hex digIT */
- int i;
- char c;
- for (i = 0; i < 4; i++)
- uuid[i] = 0;
-
- while ((c = *str++) != 0) {
- int n;
- if (c >= '0' && c <= '9')
- n = c-'0';
- else if (c >= 'a' && c <= 'f')
- n = 10 + c - 'a';
- else if (c >= 'A' && c <= 'F')
- n = 10 + c - 'A';
- else if (strchr(":. -", c))
- continue;
- else return 0;
-
- if (hit<32) {
- uuid[hit/8] <<= 4;
- uuid[hit/8] += n;
- }
- hit++;
- }
- if (hit == 32)
- return 1;
- return 0;
+ return ioctl(fd, SET_ARRAY_INFO, array);
}
/*
- * Get the md version number.
- * We use the RAID_VERSION ioctl if it is supported
- * If not, but we have a block device with major '9', we assume
- * 0.36.0
- *
- * Return version number as 24 but number - assume version parts
- * always < 255
+ * Get disk info from the kernel.
*/
-
-int md_get_version(int fd)
+int md_get_disk_info(int fd, struct mdu_disk_info_s *disk)
{
- struct stat stb;
- mdu_version_t vers;
-
- if (fstat(fd, &stb)<0)
- return -1;
- if ((S_IFMT&stb.st_mode) != S_IFBLK)
- return -1;
-
- if (ioctl(fd, RAID_VERSION, &vers) == 0)
- return (vers.major*10000) + (vers.minor*100) + vers.patchlevel;
- if (errno == EACCES)
- return -1;
- if (major(stb.st_rdev) == MD_MAJOR)
- return (3600);
- return -1;
+ return ioctl(fd, GET_DISK_INFO, disk);
}
int get_linux_version()
return (a*1000000)+(b*1000)+c;
}
-#ifndef MDASSEMBLE
int mdadm_version(char *version)
{
int a, b, c;
unsigned long long parse_size(char *size)
{
/* parse 'size' which should be a number optionally
- * followed by 'K', 'M', or 'G'.
+ * followed by 'K', 'M'. 'G' or 'T'.
* Without a suffix, K is assumed.
* Number returned is in sectors (half-K)
* INVALID_SECTORS returned on error.
c++;
s *= 1024 * 1024 * 2;
break;
+ case 'T':
+ c++;
+ s *= 1024 * 1024 * 1024 * 2LL;
+ break;
case 's': /* sectors */
c++;
break;
return s;
}
+int is_near_layout_10(int layout)
+{
+ int fc, fo;
+
+ fc = (layout >> 8) & 255;
+ fo = layout & (1 << 16);
+ if (fc > 1 || fo > 0)
+ return 0;
+ return 1;
+}
+
int parse_layout_10(char *layout)
{
int copies, rv;
int parse_layout_faulty(char *layout)
{
+ int ln, mode;
+ char *m;
+
+ if (!layout)
+ return -1;
+
/* Parse the layout string for 'faulty' */
- int ln = strcspn(layout, "0123456789");
- char *m = xstrdup(layout);
- int mode;
+ ln = strcspn(layout, "0123456789");
+ m = xstrdup(layout);
m[ln] = 0;
mode = map_name(faultylayout, m);
+ free(m);
+
if (mode == UnSet)
return -1;
return mode | (atoi(layout+ln)<< ModeShift);
}
-long parse_num(char *num)
-{
- /* Either return a valid number, or -1 */
- char *c;
- long rv = strtol(num, &c, 10);
- if (rv < 0 || *c || !num[0])
- return -1;
- else
- return rv;
-}
-#endif
-
int parse_cluster_confirm_arg(char *input, char **devname, int *slot)
{
char *dev;
}
}
-int enough_fd(int fd)
-{
- struct mdu_array_info_s array;
- struct mdu_disk_info_s disk;
- int i, rv;
- char *avail;
-
- if (md_get_array_info(fd, &array) != 0 || array.raid_disks <= 0)
- return 0;
- avail = xcalloc(array.raid_disks, 1);
- for (i = 0; i < MAX_DISKS && array.nr_disks > 0; i++) {
- disk.number = i;
- if (md_get_disk_info(fd, &disk) != 0)
- continue;
- if (disk.major == 0 && disk.minor == 0)
- continue;
- array.nr_disks--;
-
- if (! (disk.state & (1<<MD_DISK_SYNC)))
- continue;
- if (disk.raid_disk < 0 || disk.raid_disk >= array.raid_disks)
- continue;
- avail[disk.raid_disk] = 1;
- }
- /* This is used on an active array, so assume it is clean */
- rv = enough(array.level, array.raid_disks, array.layout,
- 1, avail);
- free(avail);
- return rv;
-}
-
-const int uuid_zero[4] = { 0, 0, 0, 0 };
-
-int same_uuid(int a[4], int b[4], int swapuuid)
-{
- if (swapuuid) {
- /* parse uuids are hostendian.
- * uuid's from some superblocks are big-ending
- * if there is a difference, we need to swap..
- */
- unsigned char *ac = (unsigned char *)a;
- unsigned char *bc = (unsigned char *)b;
- int i;
- for (i = 0; i < 16; i += 4) {
- if (ac[i+0] != bc[i+3] ||
- ac[i+1] != bc[i+2] ||
- ac[i+2] != bc[i+1] ||
- ac[i+3] != bc[i+0])
- return 0;
- }
- return 1;
- } else {
- if (a[0]==b[0] &&
- a[1]==b[1] &&
- a[2]==b[2] &&
- a[3]==b[3])
- return 1;
- return 0;
- }
-}
-
-void copy_uuid(void *a, int b[4], int swapuuid)
-{
- if (swapuuid) {
- /* parse uuids are hostendian.
- * uuid's from some superblocks are big-ending
- * if there is a difference, we need to swap..
- */
- unsigned char *ac = (unsigned char *)a;
- unsigned char *bc = (unsigned char *)b;
- int i;
- for (i = 0; i < 16; i += 4) {
- ac[i+0] = bc[i+3];
- ac[i+1] = bc[i+2];
- ac[i+2] = bc[i+1];
- ac[i+3] = bc[i+0];
- }
- } else
- memcpy(a, b, 16);
-}
-
char *__fname_from_uuid(int id[4], int swap, char *buf, char sep)
{
int i, j;
}
-char *fname_from_uuid(struct supertype *st, struct mdinfo *info, char *buf, char sep)
+/**
+ * fname_from_uuid() - generate uuid string. Should not be used with super1.
+ * @info: info with uuid
+ * @buf: buf to fill.
+ *
+ * This routine should not be used with super1. See detail_fname_from_uuid() for details. It does
+ * not use superswitch swapuuid as it should be 0 but it has to do UUID conversion if host is big
+ * endian- left for backward compatibility.
+ */
+char *fname_from_uuid(struct mdinfo *info, char *buf)
{
- // dirty hack to work around an issue with super1 superblocks...
- // super1 superblocks need swapuuid set in order for assembly to
- // work, but can't have it set if we want this printout to match
- // all the other uuid printouts in super1.c, so we force swapuuid
- // to 1 to make our printout match the rest of super1
- return __fname_from_uuid(info->uuid, (st->ss == &super1) ? 1 : st->ss->swapuuid, buf, sep);
+#if __BYTE_ORDER == BIG_ENDIAN
+ return __fname_from_uuid(info->uuid, true, buf, ':');
+#else
+ return __fname_from_uuid(info->uuid, false, buf, ':');
+#endif
}
-#ifndef MDASSEMBLE
int check_ext2(int fd, char *name)
{
/*
return 1;
}
+int fstat_is_blkdev(int fd, char *devname, dev_t *rdev)
+{
+ struct stat stb;
+
+ if (fstat(fd, &stb) != 0) {
+ pr_err("fstat failed for %s: %s\n", devname, strerror(errno));
+ return 0;
+ }
+ if ((S_IFMT & stb.st_mode) != S_IFBLK) {
+ pr_err("%s is not a block device.\n", devname);
+ return 0;
+ }
+ if (rdev)
+ *rdev = stb.st_rdev;
+ return 1;
+}
+
+int stat_is_blkdev(char *devname, dev_t *rdev)
+{
+ struct stat stb;
+
+ if (stat(devname, &stb) != 0) {
+ pr_err("stat failed for %s: %s\n", devname, strerror(errno));
+ return 0;
+ }
+ if ((S_IFMT & stb.st_mode) != S_IFBLK) {
+ pr_err("%s is not a block device.\n", devname);
+ return 0;
+ }
+ if (rdev)
+ *rdev = stb.st_rdev;
+ return 1;
+}
+
+/**
+ * ask() - prompt user for "yes/no" dialog.
+ * @mesg: message to be printed, without '?' sign.
+ * Returns: 1 if 'Y/y', 0 otherwise.
+ *
+ * The default value is 'N/n', thus the caps on "N" on prompt.
+ */
int ask(char *mesg)
{
- char *add = "";
- int i;
- for (i = 0; i < 5; i++) {
- char buf[100];
- fprintf(stderr, "%s%s", mesg, add);
- fflush(stderr);
- if (fgets(buf, 100, stdin)==NULL)
- return 0;
- if (buf[0]=='y' || buf[0]=='Y')
- return 1;
- if (buf[0]=='n' || buf[0]=='N')
- return 0;
- add = "(y/n) ";
+ char buf[3] = {0};
+
+ fprintf(stderr, "%s [y/N]? ", mesg);
+ fflush(stderr);
+ if (fgets(buf, 3, stdin) == NULL)
+ return 0;
+ if (strlen(buf) == 1) {
+ pr_err("assuming no.\n");
+ return 0;
}
- pr_err("assuming 'no'\n");
+ if (buf[1] != '\n')
+ goto bad_option;
+ if (toupper(buf[0]) == 'Y')
+ return 1;
+ if (toupper(buf[0]) == 'N')
+ return 0;
+bad_option:
+ pr_err("bad option.\n");
return 0;
}
-#endif /* MDASSEMBLE */
int is_standard(char *dev, int *nump)
{
return csum;
}
-#ifndef MDASSEMBLE
char *human_size(long long bytes)
{
static char buf[47];
- /* We convert bytes to either centi-M{ega,ibi}bytes or
- * centi-G{igi,ibi}bytes, with appropriate rounding,
- * and then print 1/100th of those as a decimal.
+ /* We convert bytes to either centi-M{ega,ibi}bytes,
+ * centi-G{igi,ibi}bytes or centi-T{era,ebi}bytes
+ * with appropriate rounding, and then print
+ * 1/100th of those as a decimal.
* We allow upto 2048Megabytes before converting to
- * gigabytes, as that shows more precision and isn't
+ * gigabytes and 2048Gigabytes before converting to
+ * terabytes, as that shows more precision and isn't
* too large a number.
- * Terabytes are not yet handled.
*/
if (bytes < 5000*1024)
long cMB = (bytes / ( 1000000LL / 200LL ) +1) /2;
snprintf(buf, sizeof(buf), " (%ld.%02ld MiB %ld.%02ld MB)",
cMiB/100, cMiB % 100, cMB/100, cMB % 100);
- } else {
+ } else if (bytes < 2*1024LL*1024LL*1024LL*1024LL) {
long cGiB = (bytes * 200LL / (1LL<<30) +1) / 2;
long cGB = (bytes / (1000000000LL/200LL ) +1) /2;
snprintf(buf, sizeof(buf), " (%ld.%02ld GiB %ld.%02ld GB)",
cGiB/100, cGiB % 100, cGB/100, cGB % 100);
+ } else {
+ long cTiB = (bytes * 200LL / (1LL<<40) + 1) / 2;
+ long cTB = (bytes / (1000000000000LL / 200LL) + 1) / 2;
+ snprintf(buf, sizeof(buf), " (%ld.%02ld TiB %ld.%02ld TB)",
+ cTiB/100, cTiB % 100, cTB/100, cTB % 100);
}
return buf;
}
{
static char buf[30];
- /* We convert bytes to either centi-M{ega,ibi}bytes or
- * centi-G{igi,ibi}bytes, with appropriate rounding,
- * and then print 1/100th of those as a decimal.
+ /* We convert bytes to either centi-M{ega,ibi}bytes,
+ * centi-G{igi,ibi}bytes or centi-T{era,ebi}bytes
+ * with appropriate rounding, and then print
+ * 1/100th of those as a decimal.
* We allow upto 2048Megabytes before converting to
- * gigabytes, as that shows more precision and isn't
+ * gigabytes and 2048Gigabytes before converting to
+ * terabytes, as that shows more precision and isn't
* too large a number.
- * Terabytes are not yet handled.
*
* If prefix == IEC, we mean prefixes like kibi,mebi,gibi etc.
* If prefix == JEDEC, we mean prefixes like kilo,mega,giga etc.
long cMiB = (bytes * 200LL / (1LL<<20) +1) /2;
snprintf(buf, sizeof(buf), "%ld.%02ldMiB",
cMiB/100, cMiB % 100);
- } else {
+ } else if (bytes < 2*1024LL*1024LL*1024LL*1024LL) {
long cGiB = (bytes * 200LL / (1LL<<30) +1) /2;
snprintf(buf, sizeof(buf), "%ld.%02ldGiB",
cGiB/100, cGiB % 100);
+ } else {
+ long cTiB = (bytes * 200LL / (1LL<<40) + 1) / 2;
+ snprintf(buf, sizeof(buf), "%ld.%02ldTiB",
+ cTiB/100, cTiB % 100);
}
}
else if (prefix == JEDEC) {
long cMB = (bytes / ( 1000000LL / 200LL ) +1) /2;
snprintf(buf, sizeof(buf), "%ld.%02ldMB",
cMB/100, cMB % 100);
- } else {
+ } else if (bytes < 2*1024LL*1024LL*1024LL*1024LL) {
long cGB = (bytes / (1000000000LL/200LL ) +1) /2;
snprintf(buf, sizeof(buf), "%ld.%02ldGB",
cGB/100, cGB % 100);
+ } else {
+ long cTB = (bytes / (1000000000000LL / 200LL) + 1) / 2;
+ snprintf(buf, sizeof(buf), "%ld.%02ldTB",
+ cTB/100, cTB % 100);
}
}
else
if (near*far == 1)
printf("NO REDUNDANCY");
}
-#endif
unsigned long long calc_array_size(int level, int raid_disks, int layout,
int chunksize, unsigned long long devsize)
/* First look in /sys/block/$DEVNM/dev for %d:%d
* If that fails, try parsing out a number
*/
- char path[100];
+ char path[PATH_MAX];
char *ep;
int fd;
int mjr,mnr;
- sprintf(path, "/sys/block/%s/dev", devnm);
+ snprintf(path, sizeof(path), "/sys/block/%s/dev", devnm);
fd = open(path, O_RDONLY);
if (fd >= 0) {
char buf[20];
return 0;
}
-#if !defined(MDASSEMBLE) || defined(MDASSEMBLE) && defined(MDASSEMBLE_AUTO)
-char *get_md_name(char *devnm)
+/**
+ * is_devname_numbered() - helper for numbered devname verification.
+ * @devname: path or name to check.
+ * @pref: expected devname prefix.
+ * @pref_len: prefix len.
+ */
+static bool is_devname_numbered(const char *devname, const char *pref, const int pref_len)
+{
+ int val;
+
+ assert(devname && pref);
+
+ if (strncmp(devname, pref, pref_len) != 0)
+ return false;
+
+ if (parse_num(&val, devname + pref_len) != 0)
+ return false;
+
+ if (val > 127)
+ return false;
+
+ return true;
+}
+
+/**
+ * is_devname_md_numbered() - check if &devname is numbered MD device (md).
+ * @devname: path or name to check.
+ */
+bool is_devname_md_numbered(const char *devname)
+{
+ return is_devname_numbered(devname, DEV_NUM_PREF, DEV_NUM_PREF_LEN);
+}
+
+/**
+ * is_devname_md_d_numbered() - check if &devname is secondary numbered MD device (md_d).
+ * @devname: path or name to check.
+ */
+bool is_devname_md_d_numbered(const char *devname)
{
- /* find /dev/md%d or /dev/md/%d or make a device /dev/.tmp.md%d */
- /* if dev < 0, want /dev/md/d%d or find mdp in /proc/devices ... */
+ static const char d_dev[] = DEV_NUM_PREF "_d";
- static char devname[50];
+ return is_devname_numbered(devname, d_dev, sizeof(d_dev) - 1);
+}
+
+/**
+ * get_md_name() - Get main dev node of the md device.
+ * @devnm: Md device name or path.
+ *
+ * Function checks if the full name was passed and returns md name
+ * if it is the MD device.
+ *
+ * Return: Main dev node of the md device or NULL if not found.
+ */
+char *get_md_name(char *devnm)
+{
+ static char devname[NAME_MAX];
struct stat stb;
- dev_t rdev = devnm2devid(devnm);
- char *dn;
- if (rdev == 0)
- return 0;
- if (strncmp(devnm, "md_", 3) == 0) {
- snprintf(devname, sizeof(devname), "/dev/md/%s",
- devnm + 3);
- if (stat(devname, &stb) == 0
- && (S_IFMT&stb.st_mode) == S_IFBLK
- && (stb.st_rdev == rdev))
- return devname;
- }
- snprintf(devname, sizeof(devname), "/dev/%s", devnm);
- if (stat(devname, &stb) == 0
- && (S_IFMT&stb.st_mode) == S_IFBLK
- && (stb.st_rdev == rdev))
- return devname;
+ if (strncmp(devnm, "/dev/", 5) == 0)
+ snprintf(devname, sizeof(devname), "%s", devnm);
+ else
+ snprintf(devname, sizeof(devname), "/dev/%s", devnm);
- snprintf(devname, sizeof(devname), "/dev/md/%s", devnm+2);
- if (stat(devname, &stb) == 0
- && (S_IFMT&stb.st_mode) == S_IFBLK
- && (stb.st_rdev == rdev))
+ if (!is_mddev(devname))
+ return NULL;
+ if (stat(devname, &stb) == 0 && (S_IFMT&stb.st_mode) == S_IFBLK)
return devname;
- dn = map_dev(major(rdev), minor(rdev), 0);
- if (dn)
- return dn;
- snprintf(devname, sizeof(devname), "/dev/.tmp.%s", devnm);
- if (mknod(devname, S_IFBLK | 0600, rdev) == -1)
- if (errno != EEXIST)
- return NULL;
-
- if (stat(devname, &stb) == 0
- && (S_IFMT&stb.st_mode) == S_IFBLK
- && (stb.st_rdev == rdev))
- return devname;
- unlink(devname);
return NULL;
}
if (strncmp(name, "/dev/.tmp.md", 12) == 0)
unlink(name);
}
-#endif /* !defined(MDASSEMBLE) || defined(MDASSEMBLE) && defined(MDASSEMBLE_AUTO) */
int get_maj_min(char *dev, int *major, int *minor)
{
*e == 0);
}
+/**
+ * is_bit_set() - get bit value by index.
+ * @val: value.
+ * @index: index of the bit (LSB numbering).
+ *
+ * Return: bit value.
+ */
+bool is_bit_set(int *val, unsigned char index)
+{
+ if ((*val) & (1 << index))
+ return true;
+ return false;
+}
+
int dev_open(char *dev, int flags)
{
/* like 'open', but if 'dev' matches %d:%d, create a temp
}
if (fd < 0) {
/* Try /tmp as /dev appear to be read-only */
- snprintf(devname, sizeof(devname), "/tmp/.tmp.md.%d:%d:%d",
+ snprintf(devname, sizeof(devname),
+ "/tmp/.tmp.md.%d:%d:%d",
(int)getpid(), major, minor);
- if (mknod(devname, S_IFBLK|0600, makedev(major, minor)) == 0) {
+ if (mknod(devname, S_IFBLK|0600,
+ makedev(major, minor)) == 0) {
fd = open(devname, flags);
unlink(devname);
}
int i;
int flags = O_RDWR;
dev_t devid = devnm2devid(devnm);
- long delay = 1000;
+ unsigned int delay = 1; // miliseconds
sprintf(buf, "%d:%d", major(devid), minor(devid));
for (i = 0; i < 25; i++) {
}
if (errno != EBUSY)
return fd;
- usleep(delay);
- if (delay < 200000)
+ sleep_for(0, MSEC_TO_NSEC(delay), true);
+ if (delay < 200)
delay *= 2;
}
return -1;
{
int i;
struct stat stb_want;
- long delay = 1000;
+ unsigned int delay = 1; // miliseconds
if (fstat(fd, &stb_want) != 0 ||
(stb_want.st_mode & S_IFMT) != S_IFBLK)
(stb.st_mode & S_IFMT) == S_IFBLK &&
(stb.st_rdev == stb_want.st_rdev))
return;
- usleep(delay);
- if (delay < 200000)
+ sleep_for(0, MSEC_TO_NSEC(delay), true);
+ if (delay < 200)
delay *= 2;
}
if (i == 25)
- dprintf("timeout waiting for %s\n", dev);
+ pr_err("timeout waiting for %s\n", dev);
}
struct superswitch *superlist[] =
&super0, &super1,
&super_ddf, &super_imsm,
&mbr, &gpt,
- NULL };
-
-#if !defined(MDASSEMBLE) || defined(MDASSEMBLE) && defined(MDASSEMBLE_AUTO)
+ NULL
+};
struct supertype *super_by_fd(int fd, char **subarrayp)
{
int i;
char *subarray = NULL;
char container[32] = "";
+ char *devnm = NULL;
+
+ devnm = fd2devnm(fd);
+ if (!devnm)
+ return NULL;
sra = sysfs_read(fd, NULL, GET_VERSION);
if (subarrayp)
*subarrayp = subarray;
strcpy(st->container_devnm, container);
- strcpy(st->devnm, fd2devnm(fd));
+ strncpy(st->devnm, devnm, MD_NAME_MAX - 1);
} else
free(subarray);
return st;
}
-#endif /* !defined(MDASSEMBLE) || defined(MDASSEMBLE) && defined(MDASSEMBLE_AUTO) */
-
-int dev_size_from_id(dev_t id, unsigned long long *size)
-{
- char buf[20];
- int fd;
-
- sprintf(buf, "%d:%d", major(id), minor(id));
- fd = dev_open(buf, O_RDONLY);
- if (fd < 0)
- return 0;
- if (get_dev_size(fd, NULL, size)) {
- close(fd);
- return 1;
- }
- close(fd);
- return 0;
-}
struct supertype *dup_super(struct supertype *orig)
{
*/
int must_be_container(int fd)
{
+ struct mdinfo *mdi;
unsigned long long size;
- if (md_get_version(fd) < 0)
+
+ mdi = sysfs_read(fd, NULL, GET_VERSION);
+ if (!mdi)
return 0;
+ sysfs_free(mdi);
+
if (get_dev_size(fd, NULL, &size) == 0)
return 1;
if (size == 0)
/* 'fd' is a block device. Find out if it is in use
* by a container, and return an open fd on that container.
*/
- char path[256];
+ char path[288];
char *e;
DIR *dir;
struct dirent *de;
{
/* Add a device to an array, in one of 2 ways. */
int rv;
-#ifndef MDASSEMBLE
+
if (st->ss->external) {
if (info->disk.state & (1<<MD_DISK_SYNC))
info->recovery_start = MaxSector;
}
}
} else
-#endif
rv = ioctl(mdfd, ADD_NEW_DISK, &info->disk);
return rv;
}
struct mdinfo *sra, struct mdinfo *info)
{
int rv;
+
/* Remove the disk given by 'info' from the array */
-#ifndef MDASSEMBLE
if (st->ss->external)
- rv = sysfs_set_str(sra, info, "slot", "none");
+ rv = sysfs_set_str(sra, info, "slot", STR_COMMON_NONE);
else
-#endif
rv = ioctl(mdfd, HOT_REMOVE_DISK, makedev(info->disk.major,
info->disk.minor));
return rv;
while ((ret = ioctl(mdfd, HOT_REMOVE_DISK, dev)) == -1 &&
errno == EBUSY &&
cnt-- > 0)
- usleep(10000);
+ sleep_for(0, MSEC_TO_NSEC(10), true);
return ret;
}
while ((ret = write(statefd, "remove", 6)) == -1 &&
errno == EBUSY &&
cnt-- > 0)
- usleep(10000);
+ sleep_for(0, MSEC_TO_NSEC(10), true);
return ret == 6 ? 0 : -1;
}
* This varies between externally managed arrays
* and older kernels
*/
- int vers = md_get_version(mdfd);
+ mdu_array_info_t inf;
int rv;
-#ifndef MDASSEMBLE
if (st->ss->external)
- rv = sysfs_set_array(info, vers);
- else
-#endif
- if ((vers % 100) >= 1) { /* can use different versions */
- mdu_array_info_t inf;
- memset(&inf, 0, sizeof(inf));
- inf.major_version = info->array.major_version;
- inf.minor_version = info->array.minor_version;
- rv = ioctl(mdfd, SET_ARRAY_INFO, &inf);
- } else
- rv = ioctl(mdfd, SET_ARRAY_INFO, NULL);
+ return sysfs_set_array(info);
+
+ memset(&inf, 0, sizeof(inf));
+ inf.major_version = info->array.major_version;
+ inf.minor_version = info->array.minor_version;
+ rv = md_set_array_info(mdfd, &inf);
+
return rv;
}
int start_mdmon(char *devnm)
{
- int i, skipped;
+ int i;
int len;
pid_t pid;
int status;
+ char *prefix = in_initrd() ? "initrd-" : "";
char pathbuf[1024];
char *paths[4] = {
pathbuf,
if (check_env("MDADM_NO_MDMON"))
return 0;
+ if (continue_via_systemd(devnm, MDMON_SERVICE, prefix))
+ return 0;
+ /* That failed, try running mdmon directly */
len = readlink("/proc/self/exe", pathbuf, sizeof(pathbuf)-1);
if (len > 0) {
char *sl;
} else
pathbuf[0] = '\0';
- /* First try to run systemctl */
- if (!check_env("MDADM_NO_SYSTEMCTL"))
- switch(fork()) {
- case 0:
- /* FIXME yuk. CLOSE_EXEC?? */
- skipped = 0;
- for (i = 3; skipped < 20; i++)
- if (close(i) < 0)
- skipped++;
- else
- skipped = 0;
-
- /* Don't want to see error messages from
- * systemctl. If the service doesn't exist,
- * we start mdmon ourselves.
- */
- close(2);
- open("/dev/null", O_WRONLY);
- snprintf(pathbuf, sizeof(pathbuf), "mdmon@%s.service",
- devnm);
- status = execl("/usr/bin/systemctl", "systemctl",
- "start",
- pathbuf, NULL);
- status = execl("/bin/systemctl", "systemctl", "start",
- pathbuf, NULL);
- exit(1);
- case -1: pr_err("cannot run mdmon. Array remains readonly\n");
- return -1;
- default: /* parent - good */
- pid = wait(&status);
- if (pid >= 0 && status == 0)
- return 0;
- }
-
- /* That failed, try running mdmon directly */
switch(fork()) {
case 0:
- /* FIXME yuk. CLOSE_EXEC?? */
- skipped = 0;
- for (i = 3; skipped < 20; i++)
- if (close(i) < 0)
- skipped++;
- else
- skipped = 0;
-
+ manage_fork_fds(1);
for (i = 0; paths[i]; i++)
if (paths[i][0]) {
execl(paths[i], paths[i],
memcpy(buf, r, 16);
}
-#ifndef MDASSEMBLE
int flush_metadata_updates(struct supertype *st)
{
int sfd;
*st->update_tail = mu;
st->update_tail = &mu->next;
}
-#endif /* MDASSEMBLE */
#ifdef __TINYC__
/* tinyc doesn't optimize this check in ioctl.h out ... */
unsigned int __invalid_size_argument_for_IOC = 0;
#endif
-int experimental(void)
+/**
+ * disk_fd_matches_criteria() - check if device matches spare criteria.
+ * @st: supertype, not NULL.
+ * @disk_fd: file descriptor of the disk.
+ * @sc: criteria to test.
+ *
+ * Return: true if disk matches criteria, false otherwise.
+ */
+bool disk_fd_matches_criteria(struct supertype *st, int disk_fd, struct spare_criteria *sc)
{
- if (check_env("MDADM_EXPERIMENTAL"))
- return 1;
- else {
- pr_err("To use this feature MDADM_EXPERIMENTAL environment variable has to be defined.\n");
- return 0;
- }
+ unsigned int dev_sector_size = 0;
+ unsigned long long dev_size = 0;
+
+ if (!sc->criteria_set)
+ return true;
+
+ if (!get_dev_size(disk_fd, NULL, &dev_size) || dev_size < sc->min_size)
+ return false;
+
+ if (!get_dev_sector_size(disk_fd, NULL, &dev_sector_size) ||
+ sc->sector_size != dev_sector_size)
+ return false;
+
+ if (drive_test_and_add_policies(st, &sc->pols, disk_fd, 0))
+ return false;
+
+ return true;
+}
+
+/**
+ * devid_matches_criteria() - check if device referenced by devid matches spare criteria.
+ * @st: supertype, not NULL.
+ * @devid: devid of the device to check.
+ * @sc: criteria to test.
+ *
+ * Return: true if disk matches criteria, false otherwise.
+ */
+bool devid_matches_criteria(struct supertype *st, dev_t devid, struct spare_criteria *sc)
+{
+ char buf[NAME_MAX];
+ bool ret;
+ int fd;
+
+ if (!sc->criteria_set)
+ return true;
+
+ snprintf(buf, NAME_MAX, "%d:%d", major(devid), minor(devid));
+
+ fd = dev_open(buf, O_RDONLY);
+ if (!is_fd_valid(fd))
+ return false;
+
+ /* Error code inherited */
+ ret = disk_fd_matches_criteria(st, fd, sc);
+
+ close(fd);
+ return ret;
}
/* Pick all spares matching given criteria from a container
* if spare_group given add it to domains of each spare
* metadata allows to test domains using metadata of destination array */
struct mdinfo *container_choose_spares(struct supertype *st,
- unsigned long long min_size,
+ struct spare_criteria *criteria,
struct domainlist *domlist,
char *spare_group,
const char *metadata, int get_one)
dp = &disks->devs;
disks->array.spare_disks = 0;
while (*dp) {
- int found = 0;
+ bool found = false;
+
d = *dp;
if (d->disk.state == 0) {
- /* check if size is acceptable */
- unsigned long long dev_size;
dev_t dev = makedev(d->disk.major,d->disk.minor);
- if (!min_size ||
- (dev_size_from_id(dev, &dev_size) &&
- dev_size >= min_size))
- found = 1;
+ found = devid_matches_criteria(st, dev, criteria);
+
/* check if domain matches */
if (found && domlist) {
struct dev_policy *pol = devid_policy(dev);
pol_add(&pol, pol_domain,
spare_group, NULL);
if (domain_test(domlist, pol, metadata) != 1)
- found = 0;
+ found = false;
+
dev_policy_free(pol);
}
}
{
unsigned int fds = 20 + devices;
struct rlimit lim;
- if (getrlimit(RLIMIT_NOFILE, &lim) != 0
- || lim.rlim_cur >= fds)
+ if (getrlimit(RLIMIT_NOFILE, &lim) != 0 || lim.rlim_cur >= fds)
return;
if (lim.rlim_max < fds)
lim.rlim_max = fds;
setrlimit(RLIMIT_NOFILE, &lim);
}
+/* Close all opened descriptors if needed and redirect
+ * streams to /dev/null.
+ * For debug purposed, leave STDOUT and STDERR untouched
+ * Returns:
+ * 1- if any error occurred
+ * 0- otherwise
+ */
+void manage_fork_fds(int close_all)
+{
+ DIR *dir;
+ struct dirent *dirent;
+
+ close(0);
+ open("/dev/null", O_RDWR);
+
+#ifndef DEBUG
+ dup2(0, 1);
+ dup2(0, 2);
+#endif
+
+ if (close_all == 0)
+ return;
+
+ dir = opendir("/proc/self/fd");
+ if (!dir) {
+ pr_err("Cannot open /proc/self/fd directory.\n");
+ return;
+ }
+ for (dirent = readdir(dir); dirent; dirent = readdir(dir)) {
+ int fd = -1;
+
+ if ((strcmp(dirent->d_name, ".") == 0) ||
+ (strcmp(dirent->d_name, "..")) == 0)
+ continue;
+
+ fd = strtol(dirent->d_name, NULL, 10);
+ if (fd > 2)
+ close(fd);
+ }
+}
+
+/* In a systemd/udev world, it is best to get systemd to
+ * run daemon rather than running in the background.
+ * Returns:
+ * 1- if systemd service has been started
+ * 0- otherwise
+ */
+int continue_via_systemd(char *devnm, char *service_name, char *prefix)
+{
+ int pid, status;
+ char pathbuf[1024];
+
+ /* Simply return that service cannot be started */
+ if (check_env("MDADM_NO_SYSTEMCTL"))
+ return 0;
+ switch (fork()) {
+ case 0:
+ manage_fork_fds(1);
+ snprintf(pathbuf, sizeof(pathbuf),
+ "%s@%s%s.service", service_name, prefix ?: "", devnm);
+ status = execl("/usr/bin/systemctl", "systemctl", "restart",
+ pathbuf, NULL);
+ status = execl("/bin/systemctl", "systemctl", "restart",
+ pathbuf, NULL);
+ exit(1);
+ case -1: /* Just do it ourselves. */
+ break;
+ default: /* parent - good */
+ pid = wait(&status);
+ if (pid >= 0 && status == 0)
+ return 1;
+ }
+ return 0;
+}
+
int in_initrd(void)
{
- /* This is based on similar function in systemd. */
- struct statfs s;
- /* statfs.f_type is signed long on s390x and MIPS, causing all
- sorts of sign extension problems with RAMFS_MAGIC being
- defined as 0x858458f6 */
- return statfs("/", &s) >= 0 &&
- ((unsigned long)s.f_type == TMPFS_MAGIC ||
- ((unsigned long)s.f_type & 0xFFFFFFFFUL) ==
- ((unsigned long)RAMFS_MAGIC & 0xFFFFFFFFUL));
+ return access("/etc/initrd-release", F_OK) >= 0;
}
void reopen_mddev(int mdfd)
dup2(fd, mdfd);
}
-#ifndef MDASSEMBLE
static struct cmap_hooks *cmap_hooks = NULL;
static int is_cmap_hooks_ready = 0;
if (!cmap_hooks->cmap_handle)
return;
- cmap_hooks->initialize = dlsym(cmap_hooks->cmap_handle, "cmap_initialize");
- cmap_hooks->get_string = dlsym(cmap_hooks->cmap_handle, "cmap_get_string");
+ cmap_hooks->initialize =
+ dlsym(cmap_hooks->cmap_handle, "cmap_initialize");
+ cmap_hooks->get_string =
+ dlsym(cmap_hooks->cmap_handle, "cmap_get_string");
cmap_hooks->finalize = dlsym(cmap_hooks->cmap_handle, "cmap_finalize");
if (!cmap_hooks->initialize || !cmap_hooks->get_string ||
if (!dlm_hooks->dlm_handle)
return;
- dlm_hooks->create_lockspace = dlsym(dlm_hooks->dlm_handle, "dlm_create_lockspace");
- dlm_hooks->release_lockspace = dlsym(dlm_hooks->dlm_handle, "dlm_release_lockspace");
+ dlm_hooks->open_lockspace =
+ dlsym(dlm_hooks->dlm_handle, "dlm_open_lockspace");
+ dlm_hooks->create_lockspace =
+ dlsym(dlm_hooks->dlm_handle, "dlm_create_lockspace");
+ dlm_hooks->release_lockspace =
+ dlsym(dlm_hooks->dlm_handle, "dlm_release_lockspace");
dlm_hooks->ls_lock = dlsym(dlm_hooks->dlm_handle, "dlm_ls_lock");
- dlm_hooks->ls_unlock = dlsym(dlm_hooks->dlm_handle, "dlm_ls_unlock");
+ dlm_hooks->ls_unlock_wait =
+ dlsym(dlm_hooks->dlm_handle, "dlm_ls_unlock_wait");
dlm_hooks->ls_get_fd = dlsym(dlm_hooks->dlm_handle, "dlm_ls_get_fd");
dlm_hooks->dispatch = dlsym(dlm_hooks->dlm_handle, "dlm_dispatch");
- if (!dlm_hooks->create_lockspace || !dlm_hooks->ls_lock ||
- !dlm_hooks->ls_unlock || !dlm_hooks->release_lockspace ||
- !dlm_hooks->ls_get_fd || !dlm_hooks->dispatch)
+ if (!dlm_hooks->open_lockspace || !dlm_hooks->create_lockspace ||
+ !dlm_hooks->ls_lock || !dlm_hooks->ls_unlock_wait ||
+ !dlm_hooks->release_lockspace || !dlm_hooks->ls_get_fd ||
+ !dlm_hooks->dispatch)
dlclose(dlm_hooks->dlm_handle);
else
is_dlm_hooks_ready = 1;
set_dlm_hooks();
set_cmap_hooks();
}
-#endif
+
+int zero_disk_range(int fd, unsigned long long sector, size_t count)
+{
+ int ret = 0;
+ int fd_zero;
+ void *addr = NULL;
+ size_t written = 0;
+ size_t len = count * 512;
+ ssize_t n;
+
+ fd_zero = open("/dev/zero", O_RDONLY);
+ if (fd_zero < 0) {
+ pr_err("Cannot open /dev/zero\n");
+ return -1;
+ }
+
+ if (lseek64(fd, sector * 512, SEEK_SET) < 0) {
+ ret = -errno;
+ pr_err("Failed to seek offset for zeroing\n");
+ goto out;
+ }
+
+ addr = mmap(NULL, len, PROT_READ, MAP_PRIVATE, fd_zero, 0);
+
+ if (addr == MAP_FAILED) {
+ ret = -errno;
+ pr_err("Mapping /dev/zero failed\n");
+ goto out;
+ }
+
+ do {
+ n = write(fd, addr + written, len - written);
+ if (n < 0) {
+ if (errno == EINTR)
+ continue;
+ ret = -errno;
+ pr_err("Zeroing disk range failed\n");
+ break;
+ }
+ written += n;
+ } while (written != len);
+
+ munmap(addr, len);
+
+out:
+ close(fd_zero);
+ return ret;
+}
+
+/**
+ * sleep_for() - Sleeps for specified time.
+ * @sec: Seconds to sleep for.
+ * @nsec: Nanoseconds to sleep for, has to be less than one second.
+ * @wake_after_interrupt: If set, wake up if interrupted.
+ *
+ * Function immediately returns if error different than EINTR occurs.
+ */
+void sleep_for(unsigned int sec, long nsec, bool wake_after_interrupt)
+{
+ struct timespec delay = {.tv_sec = sec, .tv_nsec = nsec};
+
+ assert(nsec < MSEC_TO_NSEC(1000));
+
+ do {
+ errno = 0;
+ nanosleep(&delay, &delay);
+ if (errno != 0 && errno != EINTR) {
+ pr_err("Error sleeping for %us %ldns: %s\n", sec, nsec, strerror(errno));
+ return;
+ }
+ } while (!wake_after_interrupt && errno == EINTR);
+}
+
+/* is_directory() - Checks if directory provided by path is indeed a regular directory.
+ * @path: directory path to be checked
+ *
+ * Doesn't accept symlinks.
+ *
+ * Return: true if is a directory, false if not
+ */
+bool is_directory(const char *path)
+{
+ struct stat st;
+
+ if (lstat(path, &st) != 0) {
+ pr_err("%s: %s\n", strerror(errno), path);
+ return false;
+ }
+
+ if (!S_ISDIR(st.st_mode))
+ return false;
+
+ return true;
+}
+
+/*
+ * is_file() - Checks if file provided by path is indeed a regular file.
+ * @path: file path to be checked
+ *
+ * Doesn't accept symlinks.
+ *
+ * Return: true if is a file, false if not
+ */
+bool is_file(const char *path)
+{
+ struct stat st;
+
+ if (lstat(path, &st) != 0) {
+ pr_err("%s: %s\n", strerror(errno), path);
+ return false;
+ }
+
+ if (!S_ISREG(st.st_mode))
+ return false;
+
+ return true;
+}
projects / thirdparty / mdadm.git / blobdiff