#include "md_p.h"
#include <ctype.h>
-#define REGISTER_DEV _IO (MD_MAJOR, 1)
-#define START_MD _IO (MD_MAJOR, 2)
-#define STOP_MD _IO (MD_MAJOR, 3)
-
int Manage_ro(char *devname, int fd, int readonly)
{
/* switch to readonly or rw
* use RESTART_ARRAY_RW or STOP_ARRAY_RO
*
*/
- mdu_array_info_t array;
-#ifndef MDASSEMBLE
struct mdinfo *mdi;
-#endif
int rv = 0;
- if (md_get_version(fd) < 9000) {
- pr_err("need md driver version 0.90.0 or later\n");
- return 1;
- }
-#ifndef MDASSEMBLE
/* If this is an externally-managed array, we need to modify the
* metadata_version so that mdmon doesn't undo our change.
*/
}
goto out;
}
-#endif
- if (ioctl(fd, GET_ARRAY_INFO, &array)) {
- pr_err("%s does not appear to be active.\n",
- devname);
+
+ if (!md_array_active(fd)) {
+ pr_err("%s does not appear to be active.\n", devname);
rv = 1;
goto out;
}
}
}
out:
-#ifndef MDASSEMBLE
- if (mdi)
- sysfs_free(mdi);
-#endif
+ sysfs_free(mdi);
return rv;
}
-#ifndef MDASSEMBLE
-
static void remove_devices(char *devnm, char *path)
{
/*
*/
char nm[32], *nmp;
- if (md_get_version(fd) < 9000) {
- pr_err("need md driver version 0.90.0 or later\n");
- return 1;
- }
nmp = fd2devnm(fd);
if (!nmp) {
pr_err("Cannot find %s in sysfs!!\n", devname);
if (will_retry && verbose == 0)
verbose = -1;
- if (md_get_version(fd) < 9000) {
- if (ioctl(fd, STOP_MD, 0) == 0)
- return 0;
- pr_err("stopping device %s failed: %s\n",
- devname, strerror(errno));
- return 1;
- }
-
strcpy(devnm, fd2devnm(fd));
/* Get EXCL access first. If this fails, then attempting
* to stop is probably a bad idea.
count = 5;
while (((fd = ((devname[0] == '/')
?open(devname, O_RDONLY|O_EXCL)
- :open_dev_flags(devnm, O_RDONLY|O_EXCL))) < 0
- || strcmp(fd2devnm(fd), devnm) != 0)
- && container[0]
- && mdmon_running(container)
- && count) {
+ :open_dev_flags(devnm, O_RDONLY|O_EXCL))) < 0 ||
+ strcmp(fd2devnm(fd), devnm) != 0) && container[0] &&
+ mdmon_running(container) && count) {
/* Can't open, so something might be wrong. However it
* is a container, so we might be racing with mdmon, so
* retry for a bit.
while (count &&
(err = sysfs_set_str(mdi, NULL,
"array_state",
- "inactive")) < 0
- && errno == EBUSY) {
+ "inactive")) < 0 &&
+ errno == EBUSY) {
usleep(200000);
count--;
}
* so it is reasonable to retry for a while - 5 seconds.
*/
count = 25; err = 0;
- while (count && fd >= 0
- && (err = ioctl(fd, STOP_ARRAY, NULL)) < 0
- && errno == EBUSY) {
+ while (count && fd >= 0 &&
+ (err = ioctl(fd, STOP_ARRAY, NULL)) < 0 && errno == EBUSY) {
usleep(200000);
count --;
}
rv = 1;
goto out;
}
- /* prior to 2.6.28, KOBJ_CHANGE was not sent when an md array
- * was stopped, so We'll do it here just to be sure. Drop any
- * partitions as well...
- */
- if (fd >= 0)
- ioctl(fd, BLKRRPART, 0);
- if (mdi)
- sysfs_uevent(mdi, "change");
+
+ if (get_linux_version() < 2006028) {
+ /* prior to 2.6.28, KOBJ_CHANGE was not sent when an md array
+ * was stopped, so We'll do it here just to be sure. Drop any
+ * partitions as well...
+ */
+ if (fd >= 0)
+ ioctl(fd, BLKRRPART, 0);
+ if (mdi)
+ sysfs_uevent(mdi, "change");
+ }
if (devnm[0] && use_udev()) {
struct map_ent *mp = map_by_devnm(&map, devnm);
map_remove(&map, devnm);
map_unlock(&map);
out:
- if (mdi)
- sysfs_free(mdi);
+ sysfs_free(mdi);
return rv;
}
int remaining_disks;
int i;
- if (ioctl(fd, GET_ARRAY_INFO, &array) != 0)
+ if (md_get_array_info(fd, &array) != 0)
return;
remaining_disks = array.nr_disks;
for (i = 0; i < MAX_DISKS && remaining_disks > 0; i++) {
char buf[40];
disk.number = i;
- if (ioctl(fd, GET_DISK_INFO, &disk) != 0)
+ if (md_get_disk_info(fd, &disk) != 0)
continue;
if (disk.major == 0 && disk.minor == 0)
continue;
int remaining_disks;
int i;
- if (ioctl(fd, GET_ARRAY_INFO, &array) != 0)
+ if (md_get_array_info(fd, &array) != 0)
return;
remaining_disks = array.nr_disks;
char buf[40];
int sfd;
disk.number = i;
- if (ioctl(fd, GET_DISK_INFO, &disk) != 0)
+ if (md_get_disk_info(fd, &disk) != 0)
continue;
if (disk.major == 0 && disk.minor == 0)
continue;
int copies, set;
int i;
- if (ioctl(fd, GET_ARRAY_INFO, &array) != 0)
+ if (md_get_array_info(fd, &array) != 0)
return;
if (array.level != 10)
return;
for (i = 0; i < MAX_DISKS && remaining_disks > 0; i++) {
char buf[40];
disk.number = i;
- if (ioctl(fd, GET_DISK_INFO, &disk) != 0)
+ if (md_get_disk_info(fd, &disk) != 0)
continue;
if (disk.major == 0 && disk.minor == 0)
continue;
get_linux_version() <= 2006018)
goto skip_re_add;
disc.number = mdi.disk.number;
- if (ioctl(fd, GET_DISK_INFO, &disc) != 0
- || disc.major != 0 || disc.minor != 0
- )
+ if (md_get_disk_info(fd, &disc) != 0 ||
+ disc.major != 0 || disc.minor != 0)
goto skip_re_add;
disc.major = major(rdev);
disc.minor = minor(rdev);
else
disc.state |= (1 << MD_DISK_CLUSTER_ADD);
}
- if (dv->writemostly == 1)
+ if (dv->writemostly == FlagSet)
disc.state |= 1 << MD_DISK_WRITEMOSTLY;
- if (dv->writemostly == 2)
+ if (dv->writemostly == FlagClear)
disc.state &= ~(1 << MD_DISK_WRITEMOSTLY);
+ if (dv->failfast == FlagSet)
+ disc.state |= 1 << MD_DISK_FAILFAST;
+ if (dv->failfast == FlagClear)
+ disc.state &= ~(1 << MD_DISK_FAILFAST);
remove_partitions(tfd);
- if (update || dv->writemostly > 0) {
+ if (update || dv->writemostly != FlagDefault ||
+ dv->failfast != FlagDefault) {
int rv = -1;
tfd = dev_open(dv->devname, O_RDWR);
if (tfd < 0) {
return -1;
}
- if (dv->writemostly == 1)
+ if (dv->writemostly == FlagSet)
rv = dev_st->ss->update_super(
dev_st, NULL, "writemostly",
devname, verbose, 0, NULL);
- if (dv->writemostly == 2)
+ if (dv->writemostly == FlagClear)
rv = dev_st->ss->update_super(
dev_st, NULL, "readwrite",
devname, verbose, 0, NULL);
+ if (dv->failfast == FlagSet)
+ rv = dev_st->ss->update_super(
+ dev_st, NULL, "failfast",
+ devname, verbose, 0, NULL);
+ if (dv->failfast == FlagClear)
+ rv = dev_st->ss->update_super(
+ dev_st, NULL, "nofailfast",
+ devname, verbose, 0, NULL);
if (update)
rv = dev_st->ss->update_super(
dev_st, NULL, update,
int raid_slot)
{
unsigned long long ldsize;
- struct supertype *dev_st = NULL;
+ struct supertype *dev_st;
int j;
mdu_disk_info_t disc;
" Adding anyway as --force was given.\n",
dv->devname, devname);
}
- if (!tst->ss->external &&
- array->major_version == 0 &&
- md_get_version(fd)%100 < 2) {
+ if (!tst->ss->external && array->major_version == 0) {
if (ioctl(fd, HOT_ADD_DISK, rdev)==0) {
if (verbose >= 0)
pr_err("hot added %s\n",
char *dev;
int dfd;
disc.number = j;
- if (ioctl(fd, GET_DISK_INFO, &disc))
+ if (md_get_disk_info(fd, &disc))
continue;
if (disc.major==0 && disc.minor==0)
continue;
break;
}
/* FIXME this is a bad test to be using */
- if (!tst->sb && (dv->disposition != 'a'
- && dv->disposition != 'S')) {
+ if (!tst->sb && (dv->disposition != 'a' &&
+ dv->disposition != 'S')) {
/* we are re-adding a device to a
* completely dead array - have to depend
* on kernel to check
* simply re-add it.
*/
- if (array->not_persistent==0) {
+ if (array->not_persistent == 0) {
dev_st = dup_super(tst);
dev_st->ss->load_super(dev_st, tfd, NULL);
- }
- if (dev_st && dev_st->sb && dv->disposition != 'S') {
- int rv = attempt_re_add(fd, tfd, dv,
- dev_st, tst,
- rdev,
- update, devname,
- verbose,
- array);
- dev_st->ss->free_super(dev_st);
- if (rv)
- return rv;
+ if (dev_st->sb && dv->disposition != 'S') {
+ int rv;
+
+ rv = attempt_re_add(fd, tfd, dv, dev_st, tst,
+ rdev, update, devname,
+ verbose, array);
+ dev_st->ss->free_super(dev_st);
+ if (rv)
+ return rv;
+ }
}
if (dv->disposition == 'M') {
if (verbose > 0)
for (d = 0; d < MAX_DISKS && found < array->nr_disks; d++) {
disc.number = d;
- if (ioctl(fd, GET_DISK_INFO, &disc))
+ if (md_get_disk_info(fd, &disc))
continue;
if (disc.major == 0 && disc.minor == 0)
continue;
- found++;
if (!(disc.state & (1<<MD_DISK_SYNC)))
continue;
avail[disc.raid_disk] = 1;
+ found++;
}
array_failed = !enough(array->level, array->raid_disks,
array->layout, 1, avail);
*/
for (j = array->raid_disks; j < tst->max_devs; j++) {
disc.number = j;
- if (ioctl(fd, GET_DISK_INFO, &disc))
+ if (md_get_disk_info(fd, &disc))
break;
if (disc.major==0 && disc.minor==0)
break;
/* only add journal to array that supports journaling */
if (dv->disposition == 'j') {
- struct mdinfo mdi;
struct mdinfo *mdp;
mdp = sysfs_read(fd, NULL, GET_ARRAY_STATE);
-
- if (strncmp(mdp->sysfs_array_state, "readonly", 8) != 0) {
- pr_err("%s is not readonly, cannot add journal.\n", devname);
+ if (!mdp) {
+ pr_err("%s unable to read array state.\n", devname);
return -1;
}
- tst->ss->getinfo_super(tst, &mdi, NULL);
- if (mdi.journal_device_required == 0) {
- pr_err("%s does not support journal device.\n", devname);
+ if (mdp->array_state != ARRAY_READONLY) {
+ sysfs_free(mdp);
+ pr_err("%s is not readonly, cannot add journal.\n", devname);
return -1;
}
+
+ sysfs_free(mdp);
+
disc.raid_disk = 0;
}
int dfd;
if (dv->disposition == 'j')
disc.state |= (1 << MD_DISK_JOURNAL) | (1 << MD_DISK_SYNC);
- if (dv->writemostly == 1)
+ if (dv->writemostly == FlagSet)
disc.state |= 1 << MD_DISK_WRITEMOSTLY;
+ if (dv->failfast == FlagSet)
+ disc.state |= 1 << MD_DISK_FAILFAST;
dfd = dev_open(dv->devname, O_RDWR | O_EXCL|O_DIRECT);
if (tst->ss->add_to_super(tst, &disc, dfd,
dv->devname, INVALID_SECTORS))
for (j = 0; j < tst->max_devs; j++) {
mdu_disk_info_t disc2;
disc2.number = j;
- if (ioctl(fd, GET_DISK_INFO, &disc2))
+ if (md_get_disk_info(fd, &disc2))
continue;
if (disc2.major==0 && disc2.minor==0)
continue;
disc.state |= (1 << MD_DISK_CLUSTER_ADD);
}
- if (dv->writemostly == 1)
+ if (dv->writemostly == FlagSet)
disc.state |= (1 << MD_DISK_WRITEMOSTLY);
+ if (dv->failfast == FlagSet)
+ disc.state |= (1 << MD_DISK_FAILFAST);
if (tst->ss->external) {
/* add a disk
* to an external metadata container */
}
int Manage_remove(struct supertype *tst, int fd, struct mddev_dev *dv,
- int sysfd, unsigned long rdev, int verbose, char *devname)
+ int sysfd, unsigned long rdev, int force, int verbose, char *devname)
{
int lfd = -1;
int err;
*/
if (rdev == 0)
ret = -1;
- else
- ret = sysfs_unique_holder(devnm, rdev);
- if (ret == 0) {
- pr_err("%s is not a member, cannot remove.\n",
- dv->devname);
- close(lfd);
- return -1;
- }
- if (ret >= 2) {
- pr_err("%s is still in use, cannot remove.\n",
- dv->devname);
- close(lfd);
- return -1;
+ else {
+ /*
+ * The drive has already been set to 'faulty', however
+ * monitor might not have had time to process it and the
+ * drive might still have an entry in the 'holders'
+ * directory. Try a few times to avoid a false error
+ */
+ int count = 20;
+
+ do {
+ ret = sysfs_unique_holder(devnm, rdev);
+ if (ret < 2)
+ break;
+ usleep(100 * 1000); /* 100ms */
+ } while (--count > 0);
+
+ if (ret == 0) {
+ pr_err("%s is not a member, cannot remove.\n",
+ dv->devname);
+ close(lfd);
+ return -1;
+ }
+ if (ret >= 2) {
+ pr_err("%s is still in use, cannot remove.\n",
+ dv->devname);
+ close(lfd);
+ return -1;
+ }
}
}
/* FIXME check that it is a current member */
/* device has been removed and we don't know
* the major:minor number
*/
- int n = write(sysfd, "remove", 6);
- if (n != 6)
- err = -1;
- else
- err = 0;
+ err = sys_hot_remove_disk(sysfd, force);
} else {
- err = ioctl(fd, HOT_REMOVE_DISK, rdev);
+ err = hot_remove_disk(fd, rdev, force);
if (err && errno == ENODEV) {
/* Old kernels rejected this if no personality
* is registered */
"state", "remove");
else
err = -1;
- if (sra)
- sysfs_free(sra);
+ sysfs_free(sra);
}
}
if (err) {
int busy = 0;
int raid_slot = -1;
- if (ioctl(fd, GET_ARRAY_INFO, &array)) {
- pr_err("Cannot get array info for %s\n",
- devname);
+ if (sysfs_init(&info, fd, NULL)) {
+ pr_err("sysfs not availabile for %s\n", devname);
goto abort;
}
- sysfs_init(&info, fd, NULL);
+ if (md_get_array_info(fd, &array)) {
+ pr_err("Cannot get array info for %s\n", devname);
+ goto abort;
+ }
/* array.size is only 32 bits and may be truncated.
* So read from sysfs if possible, and record number of sectors
*/
}
for (dv = devlist; dv; dv = dv->next) {
- unsigned long rdev = 0; /* device to add/remove etc */
+ dev_t rdev = 0; /* device to add/remove etc */
int rv;
int mj,mn;
if (strcmp(dv->devname, "failed") == 0 ||
strcmp(dv->devname, "faulty") == 0) {
- if (dv->disposition != 'A'
- && dv->disposition != 'r') {
+ if (dv->disposition != 'A' && dv->disposition != 'r') {
pr_err("%s only meaningful with -r or --re-add, not -%c\n",
dv->devname, dv->disposition);
goto abort;
}
if (strcmp(dv->devname, "missing") == 0) {
- struct mddev_dev *add_devlist = NULL;
+ struct mddev_dev *add_devlist;
struct mddev_dev **dp;
if (dv->disposition == 'c') {
rv = ioctl(fd, CLUSTERED_DISK_NACK, NULL);
}
add_devlist = conf_get_devs();
if (add_devlist == NULL) {
- pr_err("no devices to scan for missing members.");
+ pr_err("no devices to scan for missing members.\n");
continue;
}
for (dp = &add_devlist; *dp; dp = & (*dp)->next)
goto abort;
}
}
- } else if ((dv->disposition == 'r' || dv->disposition == 'f')
- && get_maj_min(dv->devname, &mj, &mn)) {
+ } else if ((dv->disposition == 'r' ||
+ dv->disposition == 'f') &&
+ get_maj_min(dv->devname, &mj, &mn)) {
/* for 'fail' and 'remove', the device might
* not exist.
*/
rdev = makedev(mj, mn);
} else {
- struct stat stb;
tfd = dev_open(dv->devname, O_RDONLY);
- if (tfd >= 0)
- fstat(tfd, &stb);
- else {
+ if (tfd >= 0) {
+ fstat_is_blkdev(tfd, dv->devname, &rdev);
+ close(tfd);
+ } else {
int open_err = errno;
- if (stat(dv->devname, &stb) != 0) {
- pr_err("Cannot find %s: %s\n",
- dv->devname, strerror(errno));
- goto abort;
- }
- if ((stb.st_mode & S_IFMT) != S_IFBLK) {
+ if (!stat_is_blkdev(dv->devname, &rdev)) {
if (dv->disposition == 'M')
/* non-fatal. Also improbable */
continue;
- pr_err("%s is not a block device.\n",
- dv->devname);
goto abort;
}
if (dv->disposition == 'r')
goto abort;
}
}
- rdev = stb.st_rdev;
}
switch(dv->disposition){
default:
if (dv->disposition == 'F')
/* Need to remove first */
- ioctl(fd, HOT_REMOVE_DISK, rdev);
+ hot_remove_disk(fd, rdev, force);
/* Make sure it isn't in use (in 2.6 or later) */
tfd = dev_open(dv->devname, O_RDONLY|O_EXCL);
if (tfd >= 0) {
rv = -1;
} else
rv = Manage_remove(tst, fd, dv, sysfd,
- rdev, verbose,
+ rdev, verbose, force,
devname);
if (sysfd >= 0)
close(sysfd);
devlist.next = NULL;
devlist.used = 0;
- devlist.writemostly = 0;
+ devlist.writemostly = FlagDefault;
+ devlist.failfast = FlagDefault;
devlist.devname = devname;
sprintf(devname, "%d:%d", major(devid), minor(devid));
close(fd2);
return 0;
}
-#endif