Btrfs: use rcu to protect device->name
authorJosef Bacik <josef@redhat.com>
Mon, 4 Jun 2012 18:03:51 +0000 (14:03 -0400)
committerChris Mason <chris.mason@oracle.com>
Fri, 15 Jun 2012 01:29:16 +0000 (21:29 -0400)
Al pointed out that we can just toss out the old name on a device and add a
new one arbitrarily, so anybody who uses device->name in printk could
possibly use free'd memory.  Instead of adding locking around all of this he
suggested doing it with RCU, so I've introduced a struct rcu_string that
does just that and have gone through and protected all accesses to
device->name that aren't under the uuid_mutex with rcu_read_lock().  This
protects us and I will use it for dealing with removing the device that we
used to mount the file system in a later patch.  Thanks,

Reviewed-by: David Sterba <dsterba@suse.cz>
Signed-off-by: Josef Bacik <josef@redhat.com>
fs/btrfs/check-integrity.c
fs/btrfs/disk-io.c
fs/btrfs/extent_io.c
fs/btrfs/ioctl.c
fs/btrfs/rcu-string.h [new file with mode: 0644]
fs/btrfs/scrub.c
fs/btrfs/volumes.c
fs/btrfs/volumes.h

index 9cebb1fd6a3cc59919c7c990d3016caee52b5849..da6e9364a5e3caa48b67c5e17f95a21ded5ee9ec 100644 (file)
@@ -93,6 +93,7 @@
 #include "print-tree.h"
 #include "locking.h"
 #include "check-integrity.h"
+#include "rcu-string.h"
 
 #define BTRFSIC_BLOCK_HASHTABLE_SIZE 0x10000
 #define BTRFSIC_BLOCK_LINK_HASHTABLE_SIZE 0x10000
@@ -843,13 +844,14 @@ static int btrfsic_process_superblock_dev_mirror(
                superblock_tmp->never_written = 0;
                superblock_tmp->mirror_num = 1 + superblock_mirror_num;
                if (state->print_mask & BTRFSIC_PRINT_MASK_SUPERBLOCK_WRITE)
-                       printk(KERN_INFO "New initial S-block (bdev %p, %s)"
-                              " @%llu (%s/%llu/%d)\n",
-                              superblock_bdev, device->name,
-                              (unsigned long long)dev_bytenr,
-                              dev_state->name,
-                              (unsigned long long)dev_bytenr,
-                              superblock_mirror_num);
+                       printk_in_rcu(KERN_INFO "New initial S-block (bdev %p, %s)"
+                                    " @%llu (%s/%llu/%d)\n",
+                                    superblock_bdev,
+                                    rcu_str_deref(device->name),
+                                    (unsigned long long)dev_bytenr,
+                                    dev_state->name,
+                                    (unsigned long long)dev_bytenr,
+                                    superblock_mirror_num);
                list_add(&superblock_tmp->all_blocks_node,
                         &state->all_blocks_list);
                btrfsic_block_hashtable_add(superblock_tmp,
index 5a3bf323e2bd435224ec7ce0ec6a91ada2d343e6..1c9664b16cd19840256fbe47a5a3f45a2e78a83a 100644 (file)
@@ -44,6 +44,7 @@
 #include "free-space-cache.h"
 #include "inode-map.h"
 #include "check-integrity.h"
+#include "rcu-string.h"
 
 static struct extent_io_ops btree_extent_io_ops;
 static void end_workqueue_fn(struct btrfs_work *work);
@@ -2575,8 +2576,9 @@ static void btrfs_end_buffer_write_sync(struct buffer_head *bh, int uptodate)
                struct btrfs_device *device = (struct btrfs_device *)
                        bh->b_private;
 
-               printk_ratelimited(KERN_WARNING "lost page write due to "
-                                  "I/O error on %s\n", device->name);
+               printk_ratelimited_in_rcu(KERN_WARNING "lost page write due to "
+                                         "I/O error on %s\n",
+                                         rcu_str_deref(device->name));
                /* note, we dont' set_buffer_write_io_error because we have
                 * our own ways of dealing with the IO errors
                 */
@@ -2749,8 +2751,8 @@ static int write_dev_flush(struct btrfs_device *device, int wait)
                wait_for_completion(&device->flush_wait);
 
                if (bio_flagged(bio, BIO_EOPNOTSUPP)) {
-                       printk("btrfs: disabling barriers on dev %s\n",
-                              device->name);
+                       printk_in_rcu("btrfs: disabling barriers on dev %s\n",
+                                     rcu_str_deref(device->name));
                        device->nobarriers = 1;
                }
                if (!bio_flagged(bio, BIO_UPTODATE)) {
index 2c8f7b2046173954f720125a6e53e96de3c7727e..aaa12c1eb3483a8371df48e5765b986c436451c4 100644 (file)
@@ -20,6 +20,7 @@
 #include "volumes.h"
 #include "check-integrity.h"
 #include "locking.h"
+#include "rcu-string.h"
 
 static struct kmem_cache *extent_state_cache;
 static struct kmem_cache *extent_buffer_cache;
@@ -1917,9 +1918,9 @@ int repair_io_failure(struct btrfs_mapping_tree *map_tree, u64 start,
                return -EIO;
        }
 
-       printk(KERN_INFO "btrfs read error corrected: ino %lu off %llu (dev %s "
-                       "sector %llu)\n", page->mapping->host->i_ino, start,
-                       dev->name, sector);
+       printk_in_rcu(KERN_INFO "btrfs read error corrected: ino %lu off %llu "
+                     "(dev %s sector %llu)\n", page->mapping->host->i_ino,
+                     start, rcu_str_deref(dev->name), sector);
 
        bio_put(bio);
        return 0;
index 24b776c08d99f7bbb621076f68500464b6829435..c5254dde1f0f7053702e879147dbc55812029471 100644 (file)
@@ -52,6 +52,7 @@
 #include "locking.h"
 #include "inode-map.h"
 #include "backref.h"
+#include "rcu-string.h"
 
 /* Mask out flags that are inappropriate for the given type of inode. */
 static inline __u32 btrfs_mask_flags(umode_t mode, __u32 flags)
@@ -1345,8 +1346,9 @@ static noinline int btrfs_ioctl_resize(struct btrfs_root *root,
        do_div(new_size, root->sectorsize);
        new_size *= root->sectorsize;
 
-       printk(KERN_INFO "btrfs: new size for %s is %llu\n",
-               device->name, (unsigned long long)new_size);
+       printk_in_rcu(KERN_INFO "btrfs: new size for %s is %llu\n",
+                     rcu_str_deref(device->name),
+                     (unsigned long long)new_size);
 
        if (new_size > old_size) {
                trans = btrfs_start_transaction(root, 0);
@@ -2264,7 +2266,12 @@ static long btrfs_ioctl_dev_info(struct btrfs_root *root, void __user *arg)
        di_args->total_bytes = dev->total_bytes;
        memcpy(di_args->uuid, dev->uuid, sizeof(di_args->uuid));
        if (dev->name) {
-               strncpy(di_args->path, dev->name, sizeof(di_args->path));
+               struct rcu_string *name;
+
+               rcu_read_lock();
+               name = rcu_dereference(dev->name);
+               strncpy(di_args->path, name->str, sizeof(di_args->path));
+               rcu_read_unlock();
                di_args->path[sizeof(di_args->path) - 1] = 0;
        } else {
                di_args->path[0] = '\0';
diff --git a/fs/btrfs/rcu-string.h b/fs/btrfs/rcu-string.h
new file mode 100644 (file)
index 0000000..9e111e4
--- /dev/null
@@ -0,0 +1,56 @@
+/*
+ * Copyright (C) 2012 Red Hat.  All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public
+ * License v2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public
+ * License along with this program; if not, write to the
+ * Free Software Foundation, Inc., 59 Temple Place - Suite 330,
+ * Boston, MA 021110-1307, USA.
+ */
+
+struct rcu_string {
+       struct rcu_head rcu;
+       char str[0];
+};
+
+static inline struct rcu_string *rcu_string_strdup(const char *src, gfp_t mask)
+{
+       size_t len = strlen(src) + 1;
+       struct rcu_string *ret = kzalloc(sizeof(struct rcu_string) +
+                                        (len * sizeof(char)), mask);
+       if (!ret)
+               return ret;
+       strncpy(ret->str, src, len);
+       return ret;
+}
+
+static inline void rcu_string_free(struct rcu_string *str)
+{
+       if (str)
+               kfree_rcu(str, rcu);
+}
+
+#define printk_in_rcu(fmt, ...) do {   \
+       rcu_read_lock();                \
+       printk(fmt, __VA_ARGS__);       \
+       rcu_read_unlock();              \
+} while (0)
+
+#define printk_ratelimited_in_rcu(fmt, ...) do {       \
+       rcu_read_lock();                                \
+       printk_ratelimited(fmt, __VA_ARGS__);           \
+       rcu_read_unlock();                              \
+} while (0)
+
+#define rcu_str_deref(rcu_str) ({                              \
+       struct rcu_string *__str = rcu_dereference(rcu_str);    \
+       __str->str;                                             \
+})
index a38cfa4f251ec1065410f561188c4adf5868cea3..b223620cd5a6d59aa4b707e2539e59a8244d2dd1 100644 (file)
@@ -26,6 +26,7 @@
 #include "backref.h"
 #include "extent_io.h"
 #include "check-integrity.h"
+#include "rcu-string.h"
 
 /*
  * This is only the first step towards a full-features scrub. It reads all
@@ -320,10 +321,10 @@ static int scrub_print_warning_inode(u64 inum, u64 offset, u64 root, void *ctx)
         * hold all of the paths here
         */
        for (i = 0; i < ipath->fspath->elem_cnt; ++i)
-               printk(KERN_WARNING "btrfs: %s at logical %llu on dev "
+               printk_in_rcu(KERN_WARNING "btrfs: %s at logical %llu on dev "
                        "%s, sector %llu, root %llu, inode %llu, offset %llu, "
                        "length %llu, links %u (path: %s)\n", swarn->errstr,
-                       swarn->logical, swarn->dev->name,
+                       swarn->logical, rcu_str_deref(swarn->dev->name),
                        (unsigned long long)swarn->sector, root, inum, offset,
                        min(isize - offset, (u64)PAGE_SIZE), nlink,
                        (char *)(unsigned long)ipath->fspath->val[i]);
@@ -332,10 +333,10 @@ static int scrub_print_warning_inode(u64 inum, u64 offset, u64 root, void *ctx)
        return 0;
 
 err:
-       printk(KERN_WARNING "btrfs: %s at logical %llu on dev "
+       printk_in_rcu(KERN_WARNING "btrfs: %s at logical %llu on dev "
                "%s, sector %llu, root %llu, inode %llu, offset %llu: path "
                "resolving failed with ret=%d\n", swarn->errstr,
-               swarn->logical, swarn->dev->name,
+               swarn->logical, rcu_str_deref(swarn->dev->name),
                (unsigned long long)swarn->sector, root, inum, offset, ret);
 
        free_ipath(ipath);
@@ -390,10 +391,11 @@ static void scrub_print_warning(const char *errstr, struct scrub_block *sblock)
                do {
                        ret = tree_backref_for_extent(&ptr, eb, ei, item_size,
                                                        &ref_root, &ref_level);
-                       printk(KERN_WARNING
+                       printk_in_rcu(KERN_WARNING
                                "btrfs: %s at logical %llu on dev %s, "
                                "sector %llu: metadata %s (level %d) in tree "
-                               "%llu\n", errstr, swarn.logical, dev->name,
+                               "%llu\n", errstr, swarn.logical,
+                               rcu_str_deref(dev->name),
                                (unsigned long long)swarn.sector,
                                ref_level ? "node" : "leaf",
                                ret < 0 ? -1 : ref_level,
@@ -580,9 +582,11 @@ out:
                spin_lock(&sdev->stat_lock);
                ++sdev->stat.uncorrectable_errors;
                spin_unlock(&sdev->stat_lock);
-               printk_ratelimited(KERN_ERR
+
+               printk_ratelimited_in_rcu(KERN_ERR
                        "btrfs: unable to fixup (nodatasum) error at logical %llu on dev %s\n",
-                       (unsigned long long)fixup->logical, sdev->dev->name);
+                       (unsigned long long)fixup->logical,
+                       rcu_str_deref(sdev->dev->name));
        }
 
        btrfs_free_path(path);
@@ -936,18 +940,20 @@ corrected_error:
                        spin_lock(&sdev->stat_lock);
                        sdev->stat.corrected_errors++;
                        spin_unlock(&sdev->stat_lock);
-                       printk_ratelimited(KERN_ERR
+                       printk_ratelimited_in_rcu(KERN_ERR
                                "btrfs: fixed up error at logical %llu on dev %s\n",
-                               (unsigned long long)logical, sdev->dev->name);
+                               (unsigned long long)logical,
+                               rcu_str_deref(sdev->dev->name));
                }
        } else {
 did_not_correct_error:
                spin_lock(&sdev->stat_lock);
                sdev->stat.uncorrectable_errors++;
                spin_unlock(&sdev->stat_lock);
-               printk_ratelimited(KERN_ERR
+               printk_ratelimited_in_rcu(KERN_ERR
                        "btrfs: unable to fixup (regular) error at logical %llu on dev %s\n",
-                       (unsigned long long)logical, sdev->dev->name);
+                       (unsigned long long)logical,
+                       rcu_str_deref(sdev->dev->name));
        }
 
 out:
index 7782020996feccd4b7103528a4c2989230f79b71..8a3d2594b80726b3a4da08c5924cb51b8e28a0ba 100644 (file)
@@ -35,6 +35,7 @@
 #include "volumes.h"
 #include "async-thread.h"
 #include "check-integrity.h"
+#include "rcu-string.h"
 
 static int init_first_rw_device(struct btrfs_trans_handle *trans,
                                struct btrfs_root *root,
@@ -64,7 +65,7 @@ static void free_fs_devices(struct btrfs_fs_devices *fs_devices)
                device = list_entry(fs_devices->devices.next,
                                    struct btrfs_device, dev_list);
                list_del(&device->dev_list);
-               kfree(device->name);
+               rcu_string_free(device->name);
                kfree(device);
        }
        kfree(fs_devices);
@@ -334,8 +335,8 @@ static noinline int device_list_add(const char *path,
 {
        struct btrfs_device *device;
        struct btrfs_fs_devices *fs_devices;
+       struct rcu_string *name;
        u64 found_transid = btrfs_super_generation(disk_super);
-       char *name;
 
        fs_devices = find_fsid(disk_super->fsid);
        if (!fs_devices) {
@@ -369,11 +370,13 @@ static noinline int device_list_add(const char *path,
                memcpy(device->uuid, disk_super->dev_item.uuid,
                       BTRFS_UUID_SIZE);
                spin_lock_init(&device->io_lock);
-               device->name = kstrdup(path, GFP_NOFS);
-               if (!device->name) {
+
+               name = rcu_string_strdup(path, GFP_NOFS);
+               if (!name) {
                        kfree(device);
                        return -ENOMEM;
                }
+               rcu_assign_pointer(device->name, name);
                INIT_LIST_HEAD(&device->dev_alloc_list);
 
                /* init readahead state */
@@ -390,12 +393,12 @@ static noinline int device_list_add(const char *path,
 
                device->fs_devices = fs_devices;
                fs_devices->num_devices++;
-       } else if (!device->name || strcmp(device->name, path)) {
-               name = kstrdup(path, GFP_NOFS);
+       } else if (!device->name || strcmp(device->name->str, path)) {
+               name = rcu_string_strdup(path, GFP_NOFS);
                if (!name)
                        return -ENOMEM;
-               kfree(device->name);
-               device->name = name;
+               rcu_string_free(device->name);
+               rcu_assign_pointer(device->name, name);
                if (device->missing) {
                        fs_devices->missing_devices--;
                        device->missing = 0;
@@ -430,15 +433,22 @@ static struct btrfs_fs_devices *clone_fs_devices(struct btrfs_fs_devices *orig)
 
        /* We have held the volume lock, it is safe to get the devices. */
        list_for_each_entry(orig_dev, &orig->devices, dev_list) {
+               struct rcu_string *name;
+
                device = kzalloc(sizeof(*device), GFP_NOFS);
                if (!device)
                        goto error;
 
-               device->name = kstrdup(orig_dev->name, GFP_NOFS);
-               if (!device->name) {
+               /*
+                * This is ok to do without rcu read locked because we hold the
+                * uuid mutex so nothing we touch in here is going to disappear.
+                */
+               name = rcu_string_strdup(orig_dev->name->str, GFP_NOFS);
+               if (!name) {
                        kfree(device);
                        goto error;
                }
+               rcu_assign_pointer(device->name, name);
 
                device->devid = orig_dev->devid;
                device->work.func = pending_bios_fn;
@@ -491,7 +501,7 @@ again:
                }
                list_del_init(&device->dev_list);
                fs_devices->num_devices--;
-               kfree(device->name);
+               rcu_string_free(device->name);
                kfree(device);
        }
 
@@ -516,7 +526,7 @@ static void __free_device(struct work_struct *work)
        if (device->bdev)
                blkdev_put(device->bdev, device->mode);
 
-       kfree(device->name);
+       rcu_string_free(device->name);
        kfree(device);
 }
 
@@ -540,6 +550,7 @@ static int __btrfs_close_devices(struct btrfs_fs_devices *fs_devices)
        mutex_lock(&fs_devices->device_list_mutex);
        list_for_each_entry(device, &fs_devices->devices, dev_list) {
                struct btrfs_device *new_device;
+               struct rcu_string *name;
 
                if (device->bdev)
                        fs_devices->open_devices--;
@@ -555,8 +566,11 @@ static int __btrfs_close_devices(struct btrfs_fs_devices *fs_devices)
                new_device = kmalloc(sizeof(*new_device), GFP_NOFS);
                BUG_ON(!new_device); /* -ENOMEM */
                memcpy(new_device, device, sizeof(*new_device));
-               new_device->name = kstrdup(device->name, GFP_NOFS);
-               BUG_ON(device->name && !new_device->name); /* -ENOMEM */
+
+               /* Safe because we are under uuid_mutex */
+               name = rcu_string_strdup(device->name->str, GFP_NOFS);
+               BUG_ON(device->name && !name); /* -ENOMEM */
+               rcu_assign_pointer(new_device->name, name);
                new_device->bdev = NULL;
                new_device->writeable = 0;
                new_device->in_fs_metadata = 0;
@@ -621,9 +635,9 @@ static int __btrfs_open_devices(struct btrfs_fs_devices *fs_devices,
                if (!device->name)
                        continue;
 
-               bdev = blkdev_get_by_path(device->name, flags, holder);
+               bdev = blkdev_get_by_path(device->name->str, flags, holder);
                if (IS_ERR(bdev)) {
-                       printk(KERN_INFO "open %s failed\n", device->name);
+                       printk(KERN_INFO "open %s failed\n", device->name->str);
                        goto error;
                }
                filemap_write_and_wait(bdev->bd_inode->i_mapping);
@@ -1632,6 +1646,7 @@ int btrfs_init_new_device(struct btrfs_root *root, char *device_path)
        struct block_device *bdev;
        struct list_head *devices;
        struct super_block *sb = root->fs_info->sb;
+       struct rcu_string *name;
        u64 total_bytes;
        int seeding_dev = 0;
        int ret = 0;
@@ -1671,23 +1686,24 @@ int btrfs_init_new_device(struct btrfs_root *root, char *device_path)
                goto error;
        }
 
-       device->name = kstrdup(device_path, GFP_NOFS);
-       if (!device->name) {
+       name = rcu_string_strdup(device_path, GFP_NOFS);
+       if (!name) {
                kfree(device);
                ret = -ENOMEM;
                goto error;
        }
+       rcu_assign_pointer(device->name, name);
 
        ret = find_next_devid(root, &device->devid);
        if (ret) {
-               kfree(device->name);
+               rcu_string_free(device->name);
                kfree(device);
                goto error;
        }
 
        trans = btrfs_start_transaction(root, 0);
        if (IS_ERR(trans)) {
-               kfree(device->name);
+               rcu_string_free(device->name);
                kfree(device);
                ret = PTR_ERR(trans);
                goto error;
@@ -1796,7 +1812,7 @@ error_trans:
        unlock_chunks(root);
        btrfs_abort_transaction(trans, root, ret);
        btrfs_end_transaction(trans, root);
-       kfree(device->name);
+       rcu_string_free(device->name);
        kfree(device);
 error:
        blkdev_put(bdev, FMODE_EXCL);
@@ -4204,10 +4220,17 @@ int btrfs_map_bio(struct btrfs_root *root, int rw, struct bio *bio,
                bio->bi_sector = bbio->stripes[dev_nr].physical >> 9;
                dev = bbio->stripes[dev_nr].dev;
                if (dev && dev->bdev && (rw != WRITE || dev->writeable)) {
+#ifdef DEBUG
+                       struct rcu_string *name;
+
+                       rcu_read_lock();
+                       name = rcu_dereference(dev->name);
                        pr_debug("btrfs_map_bio: rw %d, secor=%llu, dev=%lu "
                                 "(%s id %llu), size=%u\n", rw,
                                 (u64)bio->bi_sector, (u_long)dev->bdev->bd_dev,
-                                dev->name, dev->devid, bio->bi_size);
+                                name->str, dev->devid, bio->bi_size);
+                       rcu_read_unlock();
+#endif
                        bio->bi_bdev = dev->bdev;
                        if (async_submit)
                                schedule_bio(root, dev, rw, bio);
@@ -4694,8 +4717,9 @@ int btrfs_init_dev_stats(struct btrfs_fs_info *fs_info)
                key.offset = device->devid;
                ret = btrfs_search_slot(NULL, dev_root, &key, path, 0, 0);
                if (ret) {
-                       printk(KERN_WARNING "btrfs: no dev_stats entry found for device %s (devid %llu) (OK on first mount after mkfs)\n",
-                              device->name, (unsigned long long)device->devid);
+                       printk_in_rcu(KERN_WARNING "btrfs: no dev_stats entry found for device %s (devid %llu) (OK on first mount after mkfs)\n",
+                                     rcu_str_deref(device->name),
+                                     (unsigned long long)device->devid);
                        __btrfs_reset_dev_stats(device);
                        device->dev_stats_valid = 1;
                        btrfs_release_path(path);
@@ -4747,8 +4771,8 @@ static int update_dev_stat_item(struct btrfs_trans_handle *trans,
        BUG_ON(!path);
        ret = btrfs_search_slot(trans, dev_root, &key, path, -1, 1);
        if (ret < 0) {
-               printk(KERN_WARNING "btrfs: error %d while searching for dev_stats item for device %s!\n",
-                      ret, device->name);
+               printk_in_rcu(KERN_WARNING "btrfs: error %d while searching for dev_stats item for device %s!\n",
+                             ret, rcu_str_deref(device->name));
                goto out;
        }
 
@@ -4757,8 +4781,8 @@ static int update_dev_stat_item(struct btrfs_trans_handle *trans,
                /* need to delete old one and insert a new one */
                ret = btrfs_del_item(trans, dev_root, path);
                if (ret != 0) {
-                       printk(KERN_WARNING "btrfs: delete too small dev_stats item for device %s failed %d!\n",
-                              device->name, ret);
+                       printk_in_rcu(KERN_WARNING "btrfs: delete too small dev_stats item for device %s failed %d!\n",
+                                     rcu_str_deref(device->name), ret);
                        goto out;
                }
                ret = 1;
@@ -4770,8 +4794,8 @@ static int update_dev_stat_item(struct btrfs_trans_handle *trans,
                ret = btrfs_insert_empty_item(trans, dev_root, path,
                                              &key, sizeof(*ptr));
                if (ret < 0) {
-                       printk(KERN_WARNING "btrfs: insert dev_stats item for device %s failed %d!\n",
-                              device->name, ret);
+                       printk_in_rcu(KERN_WARNING "btrfs: insert dev_stats item for device %s failed %d!\n",
+                                     rcu_str_deref(device->name), ret);
                        goto out;
                }
        }
@@ -4823,9 +4847,9 @@ void btrfs_dev_stat_print_on_error(struct btrfs_device *dev)
 {
        if (!dev->dev_stats_valid)
                return;
-       printk_ratelimited(KERN_ERR
+       printk_ratelimited_in_rcu(KERN_ERR
                           "btrfs: bdev %s errs: wr %u, rd %u, flush %u, corrupt %u, gen %u\n",
-                          dev->name,
+                          rcu_str_deref(dev->name),
                           btrfs_dev_stat_read(dev, BTRFS_DEV_STAT_WRITE_ERRS),
                           btrfs_dev_stat_read(dev, BTRFS_DEV_STAT_READ_ERRS),
                           btrfs_dev_stat_read(dev, BTRFS_DEV_STAT_FLUSH_ERRS),
@@ -4837,8 +4861,8 @@ void btrfs_dev_stat_print_on_error(struct btrfs_device *dev)
 
 static void btrfs_dev_stat_print_on_load(struct btrfs_device *dev)
 {
-       printk(KERN_INFO "btrfs: bdev %s errs: wr %u, rd %u, flush %u, corrupt %u, gen %u\n",
-              dev->name,
+       printk_in_rcu(KERN_INFO "btrfs: bdev %s errs: wr %u, rd %u, flush %u, corrupt %u, gen %u\n",
+              rcu_str_deref(dev->name),
               btrfs_dev_stat_read(dev, BTRFS_DEV_STAT_WRITE_ERRS),
               btrfs_dev_stat_read(dev, BTRFS_DEV_STAT_READ_ERRS),
               btrfs_dev_stat_read(dev, BTRFS_DEV_STAT_FLUSH_ERRS),
index 3406a88ca83e023429b8af19f2d6aa64d4cac6f8..74366f27a76bbc7272e4b41f2f92bf9520bca813 100644 (file)
@@ -58,7 +58,7 @@ struct btrfs_device {
        /* the mode sent to blkdev_get */
        fmode_t mode;
 
-       char *name;
+       struct rcu_string *name;
 
        /* the internal btrfs device id */
        u64 devid;