X-Git-Url: http://git.ipfire.org/?p=thirdparty%2Fmdadm.git;a=blobdiff_plain;f=util.c;h=21a63c9a638ca9e6bbb6d5a1b2ea4b5e6d3efaa5;hp=818f8392177e4ff584067d8e9ffec2c30e98cb31;hb=99148c19bd9149bb938309ffb6b4dcde20b67934;hpb=41b06495ba3c6da9bddef7ae89c2c633c4c21c5c

diff --git a/util.c b/util.c
index 818f8392..21a63c9a 100644
--- a/util.c
+++ b/util.c
@@ -89,7 +89,6 @@ int dlm_funs_ready(void)
 	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;
@@ -200,16 +199,58 @@ int cluster_release_dlmlock(int lockid)
 out:
 	return ret;
 }
-#else
-int cluster_get_dlmlock(int *lockid)
+
+int md_array_active(int fd)
 {
-	return -1;
+	struct mdinfo *sra;
+	struct mdu_array_info_s array;
+	int ret;
+
+	sra = sysfs_read(fd, NULL, GET_ARRAY_STATE);
+	if (sra) {
+		if (sra->array_state != ARRAY_CLEAR &&
+		    sra->array_state != ARRAY_INACTIVE &&
+		    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, GET_ARRAY_INFO, &array);
+	}
+
+	return !ret;
 }
-int cluster_release_dlmlock(int lockid)
+
+/*
+ * Get array info from the kernel. Longer term we want to deprecate the
+ * ioctl and get it from sysfs.
+ */
+int md_get_array_info(int fd, struct mdu_array_info_s *array)
 {
-	return -1;
+	return ioctl(fd, GET_ARRAY_INFO, array);
+}
+
+/*
+ * Set array info
+ */
+int md_set_array_info(int fd, struct mdu_array_info_s *array)
+{
+	return ioctl(fd, SET_ARRAY_INFO, array);
+}
+
+/*
+ * Get disk info from the kernel.
+ */
+int md_get_disk_info(int fd, struct mdu_disk_info_s *disk)
+{
+	return ioctl(fd, GET_DISK_INFO, disk);
 }
-#endif
 
 /*
  * Parse a 128 bit uuid in 4 integers
@@ -248,35 +289,6 @@ int parse_uuid(char *str, int uuid[4])
 	return 0;
 }
 
-/*
- * 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
- */
-
-int md_get_version(int fd)
-{
-	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;
-}
-
 int get_linux_version()
 {
 	struct utsname name;
@@ -295,7 +307,6 @@ int get_linux_version()
 	return (a*1000000)+(b*1000)+c;
 }
 
-#ifndef MDASSEMBLE
 int mdadm_version(char *version)
 {
 	int a, b, c;
@@ -401,7 +412,6 @@ long parse_num(char *num)
 	else
 		return rv;
 }
-#endif
 
 int parse_cluster_confirm_arg(char *input, char **devname, int *slot)
 {
@@ -532,38 +542,6 @@ int enough(int level, int raid_disks, int layout, int clean, char *avail)
 	}
 }
 
-int enough_fd(int fd)
-{
-	struct mdu_array_info_s array;
-	struct mdu_disk_info_s disk;
-	int i, rv;
-	char *avail;
-
-	if (ioctl(fd, GET_ARRAY_INFO, &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 (ioctl(fd, GET_DISK_INFO, &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)
@@ -644,7 +622,6 @@ char *fname_from_uuid(struct supertype *st, struct mdinfo *info, char *buf, char
 	return __fname_from_uuid(info->uuid, (st->ss == &super1) ? 1 : st->ss->swapuuid, buf, sep);
 }
 
-#ifndef MDASSEMBLE
 int check_ext2(int fd, char *name)
 {
 	/*
@@ -748,7 +725,6 @@ int ask(char *mesg)
 	pr_err("assuming 'no'\n");
 	return 0;
 }
-#endif /* MDASSEMBLE */
 
 int is_standard(char *dev, int *nump)
 {
@@ -808,10 +784,9 @@ unsigned long calc_csum(void *super, int bytes)
 	return csum;
 }
 
-#ifndef MDASSEMBLE
 char *human_size(long long bytes)
 {
-	static char buf[30];
+	static char buf[47];
 
 	/* We convert bytes to either centi-M{ega,ibi}bytes or
 	 * centi-G{igi,ibi}bytes, with appropriate rounding,
@@ -828,14 +803,12 @@ char *human_size(long long bytes)
 		long cMiB = (bytes * 200LL / (1LL<<20) + 1) / 2;
 		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);
+			cMiB/100, cMiB % 100, cMB/100, cMB % 100);
 	} else {
 		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);
+			cGiB/100, cGiB % 100, cGB/100, cGB % 100);
 	}
 	return buf;
 }
@@ -862,22 +835,22 @@ char *human_size_brief(long long bytes, int prefix)
 		if (bytes < 2*1024LL*1024LL*1024LL) {
 			long cMiB = (bytes * 200LL / (1LL<<20) +1) /2;
 			snprintf(buf, sizeof(buf), "%ld.%02ldMiB",
-				cMiB/100 , cMiB % 100);
+				 cMiB/100, cMiB % 100);
 		} else {
 			long cGiB = (bytes * 200LL / (1LL<<30) +1) /2;
 			snprintf(buf, sizeof(buf), "%ld.%02ldGiB",
-					cGiB/100 , cGiB % 100);
+				 cGiB/100, cGiB % 100);
 		}
 	}
 	else if (prefix == JEDEC) {
 		if (bytes < 2*1024LL*1024LL*1024LL) {
 			long cMB  = (bytes / ( 1000000LL / 200LL ) +1) /2;
 			snprintf(buf, sizeof(buf), "%ld.%02ldMB",
-					cMB/100, cMB % 100);
+				 cMB/100, cMB % 100);
 		} else {
 			long cGB  = (bytes / (1000000000LL/200LL ) +1) /2;
 			snprintf(buf, sizeof(buf), "%ld.%02ldGB",
-					cGB/100 , cGB % 100);
+				 cGB/100, cGB % 100);
 		}
 	}
 	else
@@ -902,7 +875,6 @@ void print_r10_layout(int layout)
 	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)
@@ -969,7 +941,6 @@ dev_t devnm2devid(char *devnm)
 	return 0;
 }
 
-#if !defined(MDASSEMBLE) || defined(MDASSEMBLE) && defined(MDASSEMBLE_AUTO)
 char *get_md_name(char *devnm)
 {
 	/* find /dev/md%d or /dev/md/%d or make a device /dev/.tmp.md%d */
@@ -1023,7 +994,6 @@ void put_md_name(char *name)
 	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)
 {
@@ -1093,7 +1063,7 @@ int open_dev_excl(char *devnm)
 	long delay = 1000;
 
 	sprintf(buf, "%d:%d", major(devid), minor(devid));
-	for (i = 0 ; i < 25 ; i++) {
+	for (i = 0; i < 25; i++) {
 		int fd = dev_open(buf, flags|O_EXCL);
 		if (fd >= 0)
 			return fd;
@@ -1134,7 +1104,7 @@ void wait_for(char *dev, int fd)
 	    (stb_want.st_mode & S_IFMT) != S_IFBLK)
 		return;
 
-	for (i = 0 ; i < 25 ; i++) {
+	for (i = 0; i < 25; i++) {
 		struct stat stb;
 		if (stat(dev, &stb) == 0 &&
 		    (stb.st_mode & S_IFMT) == S_IFBLK &&
@@ -1153,9 +1123,8 @@ 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)
 {
@@ -1177,7 +1146,7 @@ struct supertype *super_by_fd(int fd, char **subarrayp)
 		minor = sra->array.minor_version;
 		verstr = sra->text_version;
 	} else {
-		if (ioctl(fd, GET_ARRAY_INFO, &array))
+		if (md_get_array_info(fd, &array))
 			array.major_version = array.minor_version = 0;
 		vers = array.major_version;
 		minor = array.minor_version;
@@ -1205,7 +1174,7 @@ struct supertype *super_by_fd(int fd, char **subarrayp)
 			verstr = "-no-metadata-";
 	}
 
-	for (i = 0; st == NULL && superlist[i] ; i++)
+	for (i = 0; st == NULL && superlist[i]; i++)
 		st = superlist[i]->match_metadata_desc(verstr);
 
 	sysfs_free(sra);
@@ -1220,7 +1189,6 @@ struct supertype *super_by_fd(int fd, char **subarrayp)
 
 	return st;
 }
-#endif /* !defined(MDASSEMBLE) || defined(MDASSEMBLE) && defined(MDASSEMBLE_AUTO) */
 
 int dev_size_from_id(dev_t id, unsigned long long *size)
 {
@@ -1270,7 +1238,7 @@ struct supertype *guess_super_type(int fd, enum guess_types guess_type)
 	st = xcalloc(1, sizeof(*st));
 	st->container_devnm[0] = 0;
 
-	for (i = 0 ; superlist[i]; i++) {
+	for (i = 0; superlist[i]; i++) {
 		int rv;
 		ss = superlist[i];
 		if (guess_type == guess_array && ss->add_to_super == NULL)
@@ -1354,9 +1322,14 @@ int get_dev_sector_size(int fd, char *dname, unsigned int *sectsizep)
  */
 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)
@@ -1435,6 +1408,7 @@ static int get_last_partition_end(int fd, unsigned long long *endofpart)
 	struct MBR boot_sect;
 	unsigned long long curr_part_end;
 	unsigned part_nr;
+	unsigned int sector_size;
 	int retval = 0;
 
 	*endofpart = 0;
@@ -1474,6 +1448,9 @@ static int get_last_partition_end(int fd, unsigned long long *endofpart)
 		/* Unknown partition table */
 		retval = -1;
 	}
+	/* calculate number of 512-byte blocks */
+	if (get_dev_sector_size(fd, NULL, &sector_size))
+		*endofpart *= (sector_size / 512);
  abort:
 	return retval;
 }
@@ -1485,9 +1462,8 @@ int check_partitions(int fd, char *dname, unsigned long long freesize,
 	 * Check where the last partition ends
 	 */
 	unsigned long long endofpart;
-	int ret;
 
-	if ((ret = get_last_partition_end(fd, &endofpart)) > 0) {
+	if (get_last_partition_end(fd, &endofpart) > 0) {
 		/* There appears to be a partition table here */
 		if (freesize == 0) {
 			/* partitions will not be visible in new device */
@@ -1752,7 +1728,7 @@ int add_disk(int mdfd, struct supertype *st,
 {
 	/* 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;
@@ -1772,7 +1748,6 @@ int add_disk(int mdfd, struct supertype *st,
 			}
 		}
 	} else
-#endif
 		rv = ioctl(mdfd, ADD_NEW_DISK, &info->disk);
 	return rv;
 }
@@ -1781,39 +1756,63 @@ int remove_disk(int mdfd, struct supertype *st,
 		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");
 	else
-#endif
 		rv = ioctl(mdfd, HOT_REMOVE_DISK, makedev(info->disk.major,
 							  info->disk.minor));
 	return rv;
 }
 
+int hot_remove_disk(int mdfd, unsigned long dev, int force)
+{
+	int cnt = force ? 500 : 5;
+	int ret;
+
+	/* HOT_REMOVE_DISK can fail with EBUSY if there are
+	 * outstanding IO requests to the device.
+	 * In this case, it can be helpful to wait a little while,
+	 * up to 5 seconds if 'force' is set, or 50 msec if not.
+	 */
+	while ((ret = ioctl(mdfd, HOT_REMOVE_DISK, dev)) == -1 &&
+	       errno == EBUSY &&
+	       cnt-- > 0)
+		usleep(10000);
+
+	return ret;
+}
+
+int sys_hot_remove_disk(int statefd, int force)
+{
+	int cnt = force ? 500 : 5;
+	int ret;
+
+	while ((ret = write(statefd, "remove", 6)) == -1 &&
+	       errno == EBUSY &&
+	       cnt-- > 0)
+		usleep(10000);
+	return ret == 6 ? 0 : -1;
+}
+
 int set_array_info(int mdfd, struct supertype *st, struct mdinfo *info)
 {
 	/* Initialise kernel's knowledge of array.
 	 * 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, 9003);
+		
+	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;
 }
 
@@ -1987,7 +1986,6 @@ use_random:
 	memcpy(buf, r, 16);
 }
 
-#ifndef MDASSEMBLE
 int flush_metadata_updates(struct supertype *st)
 {
 	int sfd;
@@ -2029,7 +2027,6 @@ void append_metadata_update(struct supertype *st, void *buf, int len)
 	*st->update_tail = mu;
 	st->update_tail = &mu->next;
 }
-#endif /* MDASSEMBLE */
 
 #ifdef __TINYC__
 /* tinyc doesn't optimize this check in ioctl.h out ... */
@@ -2169,7 +2166,6 @@ void reopen_mddev(int mdfd)
 		dup2(fd, mdfd);
 }
 
-#ifndef MDASSEMBLE
 static struct cmap_hooks *cmap_hooks = NULL;
 static int is_cmap_hooks_ready = 0;
 
@@ -2244,4 +2240,3 @@ void set_hooks(void)
 	set_dlm_hooks();
 	set_cmap_hooks();
 }
-#endif