ext3: Unify log messages in ext3
authorAlexey Fisher <bug-track@fisher-privat.net>
Sun, 22 Nov 2009 19:38:55 +0000 (20:38 +0100)
committerJan Kara <jack@suse.cz>
Thu, 10 Dec 2009 14:02:53 +0000 (15:02 +0100)
Make messages produced by ext3 more unified. It should be
easy to parse.

dmesg before patch:
[ 4893.684892] reservations ON
[ 4893.684896] xip option not supported
[ 4893.684964] EXT3-fs warning: maximal mount count reached, running
e2fsck is recommended

dmesg after patch:
[  873.300792] EXT3-fs (loop0): using internal journaln
[  873.300796] EXT3-fs (loop0): mounted filesystem with writeback data mode
[  924.163657] EXT3-fs (loop0): error: can't find ext3 filesystem on dev loop0.
[  723.755642] EXT3-fs (loop0): error: bad blocksize 8192
[  357.874687] EXT3-fs (loop0): error: no journal found. mounting ext3 over ext2?
[  873.300764] EXT3-fs (loop0): warning: maximal mount count reached, running e2fsck is recommended
[  924.163657] EXT3-fs (loop0): error: can't find ext3 filesystem on dev loop0.

Signed-off-by: Alexey Fisher <bug-track@fisher-privat.net>
Signed-off-by: Jan Kara <jack@suse.cz>
fs/ext3/super.c
include/linux/ext3_fs.h

index c1d128d..46433b3 100644 (file)
@@ -135,12 +135,24 @@ void ext3_journal_abort_handle(const char *caller, const char *err_fn,
        if (is_handle_aborted(handle))
                return;
 
-       printk(KERN_ERR "%s: aborting transaction: %s in %s\n",
-              caller, errstr, err_fn);
+       printk(KERN_ERR "EXT3-fs: %s: aborting transaction: %s in %s\n",
+               caller, errstr, err_fn);
 
        journal_abort_handle(handle);
 }
 
+void ext3_msg(struct super_block *sb, const char *prefix,
+               const char *fmt, ...)
+{
+       va_list args;
+
+       va_start(args, fmt);
+       printk("%sEXT3-fs (%s): ", prefix, sb->s_id);
+       vprintk(fmt, args);
+       printk("\n");
+       va_end(args);
+}
+
 /* Deal with the reporting of failure conditions on a filesystem such as
  * inconsistencies detected or read IO failures.
  *
@@ -174,12 +186,13 @@ static void ext3_handle_error(struct super_block *sb)
                        journal_abort(journal, -EIO);
        }
        if (test_opt (sb, ERRORS_RO)) {
-               printk (KERN_CRIT "Remounting filesystem read-only\n");
+               ext3_msg(sb, KERN_CRIT,
+                       "error: remounting filesystem read-only");
                sb->s_flags |= MS_RDONLY;
        }
        ext3_commit_super(sb, es, 1);
        if (test_opt(sb, ERRORS_PANIC))
-               panic("EXT3-fs (device %s): panic forced after error\n",
+               panic("EXT3-fs (%s): panic forced after error\n",
                        sb->s_id);
 }
 
@@ -247,8 +260,7 @@ void __ext3_std_error (struct super_block * sb, const char * function,
                return;
 
        errstr = ext3_decode_error(sb, errno, nbuf);
-       printk (KERN_CRIT "EXT3-fs error (device %s) in %s: %s\n",
-               sb->s_id, function, errstr);
+       ext3_msg(sb, KERN_CRIT, "error in %s: %s", function, errstr);
 
        ext3_handle_error(sb);
 }
@@ -268,21 +280,20 @@ void ext3_abort (struct super_block * sb, const char * function,
 {
        va_list args;
 
-       printk (KERN_CRIT "ext3_abort called.\n");
-
        va_start(args, fmt);
-       printk(KERN_CRIT "EXT3-fs error (device %s): %s: ",sb->s_id, function);
+       printk(KERN_CRIT "EXT3-fs (%s): error: %s: ", sb->s_id, function);
        vprintk(fmt, args);
        printk("\n");
        va_end(args);
 
        if (test_opt(sb, ERRORS_PANIC))
-               panic("EXT3-fs panic from previous error\n");
+               panic("EXT3-fs: panic from previous error\n");
 
        if (sb->s_flags & MS_RDONLY)
                return;
 
-       printk(KERN_CRIT "Remounting filesystem read-only\n");
+       ext3_msg(sb, KERN_CRIT,
+               "error: remounting filesystem read-only");
        EXT3_SB(sb)->s_mount_state |= EXT3_ERROR_FS;
        sb->s_flags |= MS_RDONLY;
        EXT3_SB(sb)->s_mount_opt |= EXT3_MOUNT_ABORT;
@@ -296,7 +307,7 @@ void ext3_warning (struct super_block * sb, const char * function,
        va_list args;
 
        va_start(args, fmt);
-       printk(KERN_WARNING "EXT3-fs warning (device %s): %s: ",
+       printk(KERN_WARNING "EXT3-fs (%s): warning: %s: ",
               sb->s_id, function);
        vprintk(fmt, args);
        printk("\n");
@@ -310,10 +321,10 @@ void ext3_update_dynamic_rev(struct super_block *sb)
        if (le32_to_cpu(es->s_rev_level) > EXT3_GOOD_OLD_REV)
                return;
 
-       ext3_warning(sb, __func__,
-                    "updating to rev %d because of new feature flag, "
-                    "running e2fsck is recommended",
-                    EXT3_DYNAMIC_REV);
+       ext3_msg(sb, KERN_WARNING,
+               "warning: updating to rev %d because of "
+               "new feature flag, running e2fsck is recommended",
+               EXT3_DYNAMIC_REV);
 
        es->s_first_ino = cpu_to_le32(EXT3_GOOD_OLD_FIRST_INO);
        es->s_inode_size = cpu_to_le16(EXT3_GOOD_OLD_INODE_SIZE);
@@ -331,7 +342,7 @@ void ext3_update_dynamic_rev(struct super_block *sb)
 /*
  * Open the external journal device
  */
-static struct block_device *ext3_blkdev_get(dev_t dev)
+static struct block_device *ext3_blkdev_get(dev_t dev, struct super_block *sb)
 {
        struct block_device *bdev;
        char b[BDEVNAME_SIZE];
@@ -342,8 +353,9 @@ static struct block_device *ext3_blkdev_get(dev_t dev)
        return bdev;
 
 fail:
-       printk(KERN_ERR "EXT3: failed to open journal device %s: %ld\n",
-                       __bdevname(dev, b), PTR_ERR(bdev));
+       ext3_msg(sb, "error: failed to open journal device %s: %ld",
+               __bdevname(dev, b), PTR_ERR(bdev));
+
        return NULL;
 }
 
@@ -378,13 +390,13 @@ static void dump_orphan_list(struct super_block *sb, struct ext3_sb_info *sbi)
 {
        struct list_head *l;
 
-       printk(KERN_ERR "sb orphan head is %d\n",
+       ext3_msg(sb, KERN_ERR, "error: sb orphan head is %d",
               le32_to_cpu(sbi->s_es->s_last_orphan));
 
-       printk(KERN_ERR "sb_info orphan list:\n");
+       ext3_msg(sb, KERN_ERR, "sb_info orphan list:");
        list_for_each(l, &sbi->s_orphan) {
                struct inode *inode = orphan_list_entry(l);
-               printk(KERN_ERR "  "
+               ext3_msg(sb, KERN_ERR, "  "
                       "inode %s:%lu at %p: mode %o, nlink %d, next %d\n",
                       inode->i_sb->s_id, inode->i_ino, inode,
                       inode->i_mode, inode->i_nlink,
@@ -849,7 +861,7 @@ static const match_table_t tokens = {
        {Opt_err, NULL},
 };
 
-static ext3_fsblk_t get_sb_block(void **data)
+static ext3_fsblk_t get_sb_block(void **data, struct super_block *sb)
 {
        ext3_fsblk_t    sb_block;
        char            *options = (char *) *data;
@@ -860,7 +872,7 @@ static ext3_fsblk_t get_sb_block(void **data)
        /*todo: use simple_strtoll with >32bit ext3 */
        sb_block = simple_strtoul(options, &options, 0);
        if (*options && *options != ',') {
-               printk("EXT3-fs: Invalid sb specification: %s\n",
+               ext3_msg(sb, "error: invalid sb specification: %s",
                       (char *) *data);
                return 1;
        }
@@ -960,7 +972,8 @@ static int parse_options (char *options, struct super_block *sb,
 #else
                case Opt_user_xattr:
                case Opt_nouser_xattr:
-                       printk("EXT3 (no)user_xattr options not supported\n");
+                       ext3_msg(sb, KERN_INFO,
+                               "(no)user_xattr options not supported");
                        break;
 #endif
 #ifdef CONFIG_EXT3_FS_POSIX_ACL
@@ -973,7 +986,8 @@ static int parse_options (char *options, struct super_block *sb,
 #else
                case Opt_acl:
                case Opt_noacl:
-                       printk("EXT3 (no)acl options not supported\n");
+                       ext3_msg(sb, KERN_INFO,
+                               "(no)acl options not supported");
                        break;
 #endif
                case Opt_reservation:
@@ -989,16 +1003,16 @@ static int parse_options (char *options, struct super_block *sb,
                           user to specify an existing inode to be the
                           journal file. */
                        if (is_remount) {
-                               printk(KERN_ERR "EXT3-fs: cannot specify "
-                                      "journal on remount\n");
+                               ext3_msg(sb, KERN_ERR, "error: cannot specify "
+                                       "journal on remount");
                                return 0;
                        }
                        set_opt (sbi->s_mount_opt, UPDATE_JOURNAL);
                        break;
                case Opt_journal_inum:
                        if (is_remount) {
-                               printk(KERN_ERR "EXT3-fs: cannot specify "
-                                      "journal on remount\n");
+                               ext3_msg(sb, KERN_ERR, "error: cannot specify "
+                                      "journal on remount");
                                return 0;
                        }
                        if (match_int(&args[0], &option))
@@ -1007,8 +1021,8 @@ static int parse_options (char *options, struct super_block *sb,
                        break;
                case Opt_journal_dev:
                        if (is_remount) {
-                               printk(KERN_ERR "EXT3-fs: cannot specify "
-                                      "journal on remount\n");
+                               ext3_msg(sb, KERN_ERR, "error: cannot specify "
+                                      "journal on remount");
                                return 0;
                        }
                        if (match_int(&args[0], &option))
@@ -1040,12 +1054,11 @@ static int parse_options (char *options, struct super_block *sb,
                                if ((sbi->s_mount_opt & EXT3_MOUNT_DATA_FLAGS)
                                                == data_opt)
                                        break;
-                               printk(KERN_ERR
-                                       "EXT3-fs (device %s): Cannot change "
+                               ext3_msg(sb, KERN_ERR,
+                                       "error: cannot change "
                                        "data mode on remount. The filesystem "
                                        "is mounted in data=%s mode and you "
-                                       "try to remount it in data=%s mode.\n",
-                                       sb->s_id,
+                                       "try to remount it in data=%s mode.",
                                        data_mode_string(sbi->s_mount_opt &
                                                        EXT3_MOUNT_DATA_FLAGS),
                                        data_mode_string(data_opt));
@@ -1070,31 +1083,31 @@ static int parse_options (char *options, struct super_block *sb,
 set_qf_name:
                        if (sb_any_quota_loaded(sb) &&
                            !sbi->s_qf_names[qtype]) {
-                               printk(KERN_ERR
-                                       "EXT3-fs: Cannot change journaled "
-                                       "quota options when quota turned on.\n");
+                               ext3_msg(sb, KERN_ERR,
+                                       "error: cannot change journaled "
+                                       "quota options when quota turned on.");
                                return 0;
                        }
                        qname = match_strdup(&args[0]);
                        if (!qname) {
-                               printk(KERN_ERR
-                                       "EXT3-fs: not enough memory for "
-                                       "storing quotafile name.\n");
+                               ext3_msg(sb, KERN_ERR,
+                                       "error: not enough memory for "
+                                       "storing quotafile name.");
                                return 0;
                        }
                        if (sbi->s_qf_names[qtype] &&
                            strcmp(sbi->s_qf_names[qtype], qname)) {
-                               printk(KERN_ERR
-                                       "EXT3-fs: %s quota file already "
-                                       "specified.\n", QTYPE2NAME(qtype));
+                               ext3_msg(sb, KERN_ERR,
+                                       "error: %s quota file already "
+                                       "specified.", QTYPE2NAME(qtype));
                                kfree(qname);
                                return 0;
                        }
                        sbi->s_qf_names[qtype] = qname;
                        if (strchr(sbi->s_qf_names[qtype], '/')) {
-                               printk(KERN_ERR
-                                       "EXT3-fs: quotafile must be on "
-                                       "filesystem root.\n");
+                               ext3_msg(sb, KERN_ERR,
+                                       "error: quotafile must be on "
+                                       "filesystem root.");
                                kfree(sbi->s_qf_names[qtype]);
                                sbi->s_qf_names[qtype] = NULL;
                                return 0;
@@ -1109,9 +1122,9 @@ set_qf_name:
 clear_qf_name:
                        if (sb_any_quota_loaded(sb) &&
                            sbi->s_qf_names[qtype]) {
-                               printk(KERN_ERR "EXT3-fs: Cannot change "
+                               ext3_msg(sb, KERN_ERR, "error: cannot change "
                                        "journaled quota options when "
-                                       "quota turned on.\n");
+                                       "quota turned on.");
                                return 0;
                        }
                        /*
@@ -1128,9 +1141,9 @@ clear_qf_name:
 set_qf_format:
                        if (sb_any_quota_loaded(sb) &&
                            sbi->s_jquota_fmt != qfmt) {
-                               printk(KERN_ERR "EXT3-fs: Cannot change "
+                               ext3_msg(sb, KERN_ERR, "error: cannot change "
                                        "journaled quota options when "
-                                       "quota turned on.\n");
+                                       "quota turned on.");
                                return 0;
                        }
                        sbi->s_jquota_fmt = qfmt;
@@ -1146,8 +1159,8 @@ set_qf_format:
                        break;
                case Opt_noquota:
                        if (sb_any_quota_loaded(sb)) {
-                               printk(KERN_ERR "EXT3-fs: Cannot change quota "
-                                       "options when quota turned on.\n");
+                               ext3_msg(sb, KERN_ERR, "error: cannot change "
+                                       "quota options when quota turned on.");
                                return 0;
                        }
                        clear_opt(sbi->s_mount_opt, QUOTA);
@@ -1158,8 +1171,8 @@ set_qf_format:
                case Opt_quota:
                case Opt_usrquota:
                case Opt_grpquota:
-                       printk(KERN_ERR
-                               "EXT3-fs: quota options not supported.\n");
+                       ext3_msg(sb, KERN_ERR,
+                               "error: quota options not supported.");
                        break;
                case Opt_usrjquota:
                case Opt_grpjquota:
@@ -1167,9 +1180,9 @@ set_qf_format:
                case Opt_offgrpjquota:
                case Opt_jqfmt_vfsold:
                case Opt_jqfmt_vfsv0:
-                       printk(KERN_ERR
-                               "EXT3-fs: journaled quota options not "
-                               "supported.\n");
+                       ext3_msg(sb, KERN_ERR,
+                               "error: journaled quota options not "
+                               "supported.");
                        break;
                case Opt_noquota:
                        break;
@@ -1189,8 +1202,9 @@ set_qf_format:
                        break;
                case Opt_resize:
                        if (!is_remount) {
-                               printk("EXT3-fs: resize option only available "
-                                       "for remount\n");
+                               ext3_msg(sb, KERN_ERR,
+                                       "error: resize option only available "
+                                       "for remount");
                                return 0;
                        }
                        if (match_int(&args[0], &option) != 0)
@@ -1204,9 +1218,9 @@ set_qf_format:
                        clear_opt(sbi->s_mount_opt, NOBH);
                        break;
                default:
-                       printk (KERN_ERR
-                               "EXT3-fs: Unrecognized mount option \"%s\" "
-                               "or missing value\n", p);
+                       ext3_msg(sb, KERN_ERR,
+                               "error: unrecognized mount option \"%s\" "
+                               "or missing value", p);
                        return 0;
                }
        }
@@ -1224,21 +1238,21 @@ set_qf_format:
                                (sbi->s_mount_opt & EXT3_MOUNT_GRPQUOTA)) ||
                    (sbi->s_qf_names[GRPQUOTA] &&
                                (sbi->s_mount_opt & EXT3_MOUNT_USRQUOTA))) {
-                       printk(KERN_ERR "EXT3-fs: old and new quota "
-                                       "format mixing.\n");
+                       ext3_msg(sb, KERN_ERR, "error: old and new quota "
+                                       "format mixing.");
                        return 0;
                }
 
                if (!sbi->s_jquota_fmt) {
-                       printk(KERN_ERR "EXT3-fs: journaled quota format "
-                                       "not specified.\n");
+                       ext3_msg(sb, KERN_ERR, "error: journaled quota format "
+                                       "not specified.");
                        return 0;
                }
        } else {
                if (sbi->s_jquota_fmt) {
-                       printk(KERN_ERR "EXT3-fs: journaled quota format "
+                       ext3_msg(sb, KERN_ERR, "error: journaled quota format "
                                        "specified with no journaling "
-                                       "enabled.\n");
+                                       "enabled.");
                        return 0;
                }
        }
@@ -1253,31 +1267,33 @@ static int ext3_setup_super(struct super_block *sb, struct ext3_super_block *es,
        int res = 0;
 
        if (le32_to_cpu(es->s_rev_level) > EXT3_MAX_SUPP_REV) {
-               printk (KERN_ERR "EXT3-fs warning: revision level too high, "
-                       "forcing read-only mode\n");
+               ext3_msg(sb, KERN_ERR,
+                       "error: revision level too high, "
+                       "forcing read-only mode");
                res = MS_RDONLY;
        }
        if (read_only)
                return res;
        if (!(sbi->s_mount_state & EXT3_VALID_FS))
-               printk (KERN_WARNING "EXT3-fs warning: mounting unchecked fs, "
-                       "running e2fsck is recommended\n");
+               ext3_msg(sb, KERN_WARNING,
+                       "warning: mounting unchecked fs, "
+                       "running e2fsck is recommended");
        else if ((sbi->s_mount_state & EXT3_ERROR_FS))
-               printk (KERN_WARNING
-                       "EXT3-fs warning: mounting fs with errors, "
-                       "running e2fsck is recommended\n");
+               ext3_msg(sb, KERN_WARNING,
+                       "warning: mounting fs with errors, "
+                       "running e2fsck is recommended");
        else if ((__s16) le16_to_cpu(es->s_max_mnt_count) >= 0 &&
                 le16_to_cpu(es->s_mnt_count) >=
                 (unsigned short) (__s16) le16_to_cpu(es->s_max_mnt_count))
-               printk (KERN_WARNING
-                       "EXT3-fs warning: maximal mount count reached, "
-                       "running e2fsck is recommended\n");
+               ext3_msg(sb, KERN_WARNING,
+                       "warning: maximal mount count reached, "
+                       "running e2fsck is recommended");
        else if (le32_to_cpu(es->s_checkinterval) &&
                (le32_to_cpu(es->s_lastcheck) +
                        le32_to_cpu(es->s_checkinterval) <= get_seconds()))
-               printk (KERN_WARNING
-                       "EXT3-fs warning: checktime reached, "
-                       "running e2fsck is recommended\n");
+               ext3_msg(sb, KERN_WARNING,
+                       "warning: checktime reached, "
+                       "running e2fsck is recommended");
 #if 0
                /* @@@ We _will_ want to clear the valid bit if we find
                    inconsistencies, to force a fsck at reboot.  But for
@@ -1294,22 +1310,20 @@ static int ext3_setup_super(struct super_block *sb, struct ext3_super_block *es,
 
        ext3_commit_super(sb, es, 1);
        if (test_opt(sb, DEBUG))
-               printk(KERN_INFO "[EXT3 FS bs=%lu, gc=%lu, "
-                               "bpg=%lu, ipg=%lu, mo=%04lx]\n",
+               ext3_msg(sb, KERN_INFO, "[bs=%lu, gc=%lu, "
+                               "bpg=%lu, ipg=%lu, mo=%04lx]",
                        sb->s_blocksize,
                        sbi->s_groups_count,
                        EXT3_BLOCKS_PER_GROUP(sb),
                        EXT3_INODES_PER_GROUP(sb),
                        sbi->s_mount_opt);
 
-       printk(KERN_INFO "EXT3 FS on %s, ", sb->s_id);
        if (EXT3_SB(sb)->s_journal->j_inode == NULL) {
                char b[BDEVNAME_SIZE];
-
-               printk("external journal on %s\n",
+               ext3_msg(sb, KERN_INFO, "using external journal on %s",
                        bdevname(EXT3_SB(sb)->s_journal->j_dev, b));
        } else {
-               printk("internal journal\n");
+               ext3_msg(sb, KERN_INFO, "using internal journal");
        }
        return res;
 }
@@ -1403,8 +1417,8 @@ static void ext3_orphan_cleanup (struct super_block * sb,
        }
 
        if (bdev_read_only(sb->s_bdev)) {
-               printk(KERN_ERR "EXT3-fs: write access "
-                       "unavailable, skipping orphan cleanup.\n");
+               ext3_msg(sb, KERN_ERR, "error: write access "
+                       "unavailable, skipping orphan cleanup.");
                return;
        }
 
@@ -1418,8 +1432,7 @@ static void ext3_orphan_cleanup (struct super_block * sb,
        }
 
        if (s_flags & MS_RDONLY) {
-               printk(KERN_INFO "EXT3-fs: %s: orphan cleanup on readonly fs\n",
-                      sb->s_id);
+               ext3_msg(sb, KERN_INFO, "orphan cleanup on readonly fs");
                sb->s_flags &= ~MS_RDONLY;
        }
 #ifdef CONFIG_QUOTA
@@ -1430,9 +1443,9 @@ static void ext3_orphan_cleanup (struct super_block * sb,
                if (EXT3_SB(sb)->s_qf_names[i]) {
                        int ret = ext3_quota_on_mount(sb, i);
                        if (ret < 0)
-                               printk(KERN_ERR
-                                       "EXT3-fs: Cannot turn on journaled "
-                                       "quota: error %d\n", ret);
+                               ext3_msg(sb, KERN_ERR,
+                                       "error: cannot turn on journaled "
+                                       "quota: %d", ret);
                }
        }
 #endif
@@ -1470,11 +1483,11 @@ static void ext3_orphan_cleanup (struct super_block * sb,
 #define PLURAL(x) (x), ((x)==1) ? "" : "s"
 
        if (nr_orphans)
-               printk(KERN_INFO "EXT3-fs: %s: %d orphan inode%s deleted\n",
-                      sb->s_id, PLURAL(nr_orphans));
+               ext3_msg(sb, KERN_INFO, "%d orphan inode%s deleted",
+                      PLURAL(nr_orphans));
        if (nr_truncates)
-               printk(KERN_INFO "EXT3-fs: %s: %d truncate%s cleaned up\n",
-                      sb->s_id, PLURAL(nr_truncates));
+               ext3_msg(sb, KERN_INFO, "%d truncate%s cleaned up",
+                      PLURAL(nr_truncates));
 #ifdef CONFIG_QUOTA
        /* Turn quotas off */
        for (i = 0; i < MAXQUOTAS; i++) {
@@ -1558,7 +1571,7 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
        struct ext3_super_block *es = NULL;
        struct ext3_sb_info *sbi;
        ext3_fsblk_t block;
-       ext3_fsblk_t sb_block = get_sb_block(&data);
+       ext3_fsblk_t sb_block = get_sb_block(&data, sb);
        ext3_fsblk_t logic_sb_block;
        unsigned long offset = 0;
        unsigned int journal_inum = 0;
@@ -1594,7 +1607,7 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
 
        blocksize = sb_min_blocksize(sb, EXT3_MIN_BLOCK_SIZE);
        if (!blocksize) {
-               printk(KERN_ERR "EXT3-fs: unable to set blocksize\n");
+               ext3_msg(sb, KERN_ERR, "error: unable to set blocksize");
                goto out_fail;
        }
 
@@ -1610,7 +1623,7 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
        }
 
        if (!(bh = sb_bread(sb, logic_sb_block))) {
-               printk (KERN_ERR "EXT3-fs: unable to read superblock\n");
+               ext3_msg(sb, KERN_ERR, "error: unable to read superblock");
                goto out_fail;
        }
        /*
@@ -1669,9 +1682,9 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
            (EXT3_HAS_COMPAT_FEATURE(sb, ~0U) ||
             EXT3_HAS_RO_COMPAT_FEATURE(sb, ~0U) ||
             EXT3_HAS_INCOMPAT_FEATURE(sb, ~0U)))
-               printk(KERN_WARNING
-                      "EXT3-fs warning: feature flags set on rev 0 fs, "
-                      "running e2fsck is recommended\n");
+               ext3_msg(sb, KERN_WARNING,
+                       "warning: feature flags set on rev 0 fs, "
+                       "running e2fsck is recommended");
        /*
         * Check feature flags regardless of the revision level, since we
         * previously didn't change the revision level when setting the flags,
@@ -1679,25 +1692,25 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
         */
        features = EXT3_HAS_INCOMPAT_FEATURE(sb, ~EXT3_FEATURE_INCOMPAT_SUPP);
        if (features) {
-               printk(KERN_ERR "EXT3-fs: %s: couldn't mount because of "
-                      "unsupported optional features (%x).\n",
-                      sb->s_id, le32_to_cpu(features));
+               ext3_msg(sb, KERN_ERR,
+                       "error: couldn't mount because of unsupported "
+                       "optional features (%x)", le32_to_cpu(features));
                goto failed_mount;
        }
        features = EXT3_HAS_RO_COMPAT_FEATURE(sb, ~EXT3_FEATURE_RO_COMPAT_SUPP);
        if (!(sb->s_flags & MS_RDONLY) && features) {
-               printk(KERN_ERR "EXT3-fs: %s: couldn't mount RDWR because of "
-                      "unsupported optional features (%x).\n",
-                      sb->s_id, le32_to_cpu(features));
+               ext3_msg(sb, KERN_ERR,
+                       "error: couldn't mount RDWR because of unsupported "
+                       "optional features (%x)", le32_to_cpu(features));
                goto failed_mount;
        }
        blocksize = BLOCK_SIZE << le32_to_cpu(es->s_log_block_size);
 
        if (blocksize < EXT3_MIN_BLOCK_SIZE ||
            blocksize > EXT3_MAX_BLOCK_SIZE) {
-               printk(KERN_ERR
-                      "EXT3-fs: Unsupported filesystem blocksize %d on %s.\n",
-                      blocksize, sb->s_id);
+               ext3_msg(sb, KERN_ERR,
+                       "error: couldn't mount because of unsupported "
+                       "filesystem blocksize %d", blocksize);
                goto failed_mount;
        }
 
@@ -1708,30 +1721,31 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
                 * than the hardware sectorsize for the machine.
                 */
                if (blocksize < hblock) {
-                       printk(KERN_ERR "EXT3-fs: blocksize %d too small for "
-                              "device blocksize %d.\n", blocksize, hblock);
+                       ext3_msg(sb, KERN_ERR,
+                               "error: fsblocksize %d too small for "
+                               "hardware sectorsize %d", blocksize, hblock);
                        goto failed_mount;
                }
 
                brelse (bh);
                if (!sb_set_blocksize(sb, blocksize)) {
-                       printk(KERN_ERR "EXT3-fs: bad blocksize %d.\n",
-                               blocksize);
+                       ext3_msg(sb, KERN_ERR,
+                               "error: bad blocksize %d", blocksize);
                        goto out_fail;
                }
                logic_sb_block = (sb_block * EXT3_MIN_BLOCK_SIZE) / blocksize;
                offset = (sb_block * EXT3_MIN_BLOCK_SIZE) % blocksize;
                bh = sb_bread(sb, logic_sb_block);
                if (!bh) {
-                       printk(KERN_ERR
-                              "EXT3-fs: Can't read superblock on 2nd try.\n");
+                       ext3_msg(sb, KERN_ERR,
+                              "error: can't read superblock on 2nd try");
                        goto failed_mount;
                }
                es = (struct ext3_super_block *)(((char *)bh->b_data) + offset);
                sbi->s_es = es;
                if (es->s_magic != cpu_to_le16(EXT3_SUPER_MAGIC)) {
-                       printk (KERN_ERR
-                               "EXT3-fs: Magic mismatch, very weird !\n");
+                       ext3_msg(sb, KERN_ERR,
+                               "error: magic mismatch");
                        goto failed_mount;
                }
        }
@@ -1747,8 +1761,8 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
                if ((sbi->s_inode_size < EXT3_GOOD_OLD_INODE_SIZE) ||
                    (!is_power_of_2(sbi->s_inode_size)) ||
                    (sbi->s_inode_size > blocksize)) {
-                       printk (KERN_ERR
-                               "EXT3-fs: unsupported inode size: %d\n",
+                       ext3_msg(sb, KERN_ERR,
+                               "error: unsupported inode size: %d",
                                sbi->s_inode_size);
                        goto failed_mount;
                }
@@ -1756,8 +1770,8 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
        sbi->s_frag_size = EXT3_MIN_FRAG_SIZE <<
                                   le32_to_cpu(es->s_log_frag_size);
        if (blocksize != sbi->s_frag_size) {
-               printk(KERN_ERR
-                      "EXT3-fs: fragsize %lu != blocksize %u (unsupported)\n",
+               ext3_msg(sb, KERN_ERR,
+                      "error: fragsize %lu != blocksize %u (unsupported)",
                       sbi->s_frag_size, blocksize);
                goto failed_mount;
        }
@@ -1793,31 +1807,31 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
        }
 
        if (sbi->s_blocks_per_group > blocksize * 8) {
-               printk (KERN_ERR
-                       "EXT3-fs: #blocks per group too big: %lu\n",
+               ext3_msg(sb, KERN_ERR,
+                       "#blocks per group too big: %lu",
                        sbi->s_blocks_per_group);
                goto failed_mount;
        }
        if (sbi->s_frags_per_group > blocksize * 8) {
-               printk (KERN_ERR
-                       "EXT3-fs: #fragments per group too big: %lu\n",
+               ext3_msg(sb, KERN_ERR,
+                       "error: #fragments per group too big: %lu",
                        sbi->s_frags_per_group);
                goto failed_mount;
        }
        if (sbi->s_inodes_per_group > blocksize * 8) {
-               printk (KERN_ERR
-                       "EXT3-fs: #inodes per group too big: %lu\n",
+               ext3_msg(sb, KERN_ERR,
+                       "error: #inodes per group too big: %lu",
                        sbi->s_inodes_per_group);
                goto failed_mount;
        }
 
        if (le32_to_cpu(es->s_blocks_count) >
                    (sector_t)(~0ULL) >> (sb->s_blocksize_bits - 9)) {
-               printk(KERN_ERR "EXT3-fs: filesystem on %s:"
-                       " too large to mount safely\n", sb->s_id);
+               ext3_msg(sb, KERN_ERR,
+                       "error: filesystem is too large to mount safely");
                if (sizeof(sector_t) < 8)
-                       printk(KERN_WARNING "EXT3-fs: CONFIG_LBDAF not "
-                                       "enabled\n");
+                       ext3_msg(sb, KERN_ERR,
+                               "error: CONFIG_LBDAF not enabled");
                goto failed_mount;
        }
 
@@ -1831,7 +1845,8 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
        sbi->s_group_desc = kmalloc(db_count * sizeof (struct buffer_head *),
                                    GFP_KERNEL);
        if (sbi->s_group_desc == NULL) {
-               printk (KERN_ERR "EXT3-fs: not enough memory\n");
+               ext3_msg(sb, KERN_ERR,
+                       "error: not enough memory");
                goto failed_mount;
        }
 
@@ -1841,14 +1856,15 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
                block = descriptor_loc(sb, logic_sb_block, i);
                sbi->s_group_desc[i] = sb_bread(sb, block);
                if (!sbi->s_group_desc[i]) {
-                       printk (KERN_ERR "EXT3-fs: "
-                               "can't read group descriptor %d\n", i);
+                       ext3_msg(sb, KERN_ERR,
+                               "error: can't read group descriptor %d", i);
                        db_count = i;
                        goto failed_mount2;
                }
        }
        if (!ext3_check_descriptors (sb)) {
-               printk(KERN_ERR "EXT3-fs: group descriptors corrupted!\n");
+               ext3_msg(sb, KERN_ERR,
+                       "error: group descriptors corrupted");
                goto failed_mount2;
        }
        sbi->s_gdb_count = db_count;
@@ -1866,7 +1882,7 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
                                ext3_count_dirs(sb));
        }
        if (err) {
-               printk(KERN_ERR "EXT3-fs: insufficient memory\n");
+               ext3_msg(sb, KERN_ERR, "error: insufficient memory");
                goto failed_mount3;
        }
 
@@ -1914,9 +1930,9 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
                        goto failed_mount3;
        } else {
                if (!silent)
-                       printk (KERN_ERR
-                               "ext3: No journal on filesystem on %s\n",
-                               sb->s_id);
+                       ext3_msg(sb, KERN_ERR,
+                               "error: no journal found. "
+                               "mounting ext3 over ext2?");
                goto failed_mount3;
        }
 
@@ -1938,8 +1954,9 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
        case EXT3_MOUNT_WRITEBACK_DATA:
                if (!journal_check_available_features
                    (sbi->s_journal, 0, 0, JFS_FEATURE_INCOMPAT_REVOKE)) {
-                       printk(KERN_ERR "EXT3-fs: Journal does not support "
-                              "requested data journaling mode\n");
+                       ext3_msg(sb, KERN_ERR,
+                               "error: journal does not support "
+                               "requested data journaling mode");
                        goto failed_mount4;
                }
        default:
@@ -1948,8 +1965,9 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
 
        if (test_opt(sb, NOBH)) {
                if (!(test_opt(sb, DATA_FLAGS) == EXT3_MOUNT_WRITEBACK_DATA)) {
-                       printk(KERN_WARNING "EXT3-fs: Ignoring nobh option - "
-                               "its supported only with writeback mode\n");
+                       ext3_msg(sb, KERN_WARNING,
+                               "warning: ignoring nobh option - "
+                               "it is supported only with writeback mode");
                        clear_opt(sbi->s_mount_opt, NOBH);
                }
        }
@@ -1960,18 +1978,18 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
 
        root = ext3_iget(sb, EXT3_ROOT_INO);
        if (IS_ERR(root)) {
-               printk(KERN_ERR "EXT3-fs: get root inode failed\n");
+               ext3_msg(sb, KERN_ERR, "error: get root inode failed");
                ret = PTR_ERR(root);
                goto failed_mount4;
        }
        if (!S_ISDIR(root->i_mode) || !root->i_blocks || !root->i_size) {
                iput(root);
-               printk(KERN_ERR "EXT3-fs: corrupt root inode, run e2fsck\n");
+               ext3_msg(sb, KERN_ERR, "error: corrupt root inode, run e2fsck");
                goto failed_mount4;
        }
        sb->s_root = d_alloc_root(root);
        if (!sb->s_root) {
-               printk(KERN_ERR "EXT3-fs: get root dentry failed\n");
+               ext3_msg(sb, KERN_ERR, "error: get root dentry failed");
                iput(root);
                ret = -ENOMEM;
                goto failed_mount4;
@@ -1990,9 +2008,9 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
        ext3_orphan_cleanup(sb, es);
        EXT3_SB(sb)->s_mount_state &= ~EXT3_ORPHAN_FS;
        if (needs_recovery)
-               printk (KERN_INFO "EXT3-fs: recovery complete.\n");
+               ext3_msg(sb, KERN_INFO, "recovery complete");
        ext3_mark_recovery_complete(sb, es);
-       printk (KERN_INFO "EXT3-fs: mounted filesystem with %s data mode.\n",
+       ext3_msg(sb, KERN_INFO, "mounted filesystem with %s data mode",
                test_opt(sb,DATA_FLAGS) == EXT3_MOUNT_JOURNAL_DATA ? "journal":
                test_opt(sb,DATA_FLAGS) == EXT3_MOUNT_ORDERED_DATA ? "ordered":
                "writeback");
@@ -2002,7 +2020,8 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
 
 cantfind_ext3:
        if (!silent)
-               printk(KERN_ERR "VFS: Can't find ext3 filesystem on dev %s.\n",
+               ext3_msg(sb, KERN_INFO,
+                       "error: can't find ext3 filesystem on dev %s.",
                       sb->s_id);
        goto failed_mount;
 
@@ -2070,27 +2089,27 @@ static journal_t *ext3_get_journal(struct super_block *sb,
 
        journal_inode = ext3_iget(sb, journal_inum);
        if (IS_ERR(journal_inode)) {
-               printk(KERN_ERR "EXT3-fs: no journal found.\n");
+               ext3_msg(sb, KERN_ERR, "error: no journal found");
                return NULL;
        }
        if (!journal_inode->i_nlink) {
                make_bad_inode(journal_inode);
                iput(journal_inode);
-               printk(KERN_ERR "EXT3-fs: journal inode is deleted.\n");
+               ext3_msg(sb, KERN_ERR, "error: journal inode is deleted");
                return NULL;
        }
 
        jbd_debug(2, "Journal inode found at %p: %Ld bytes\n",
                  journal_inode, journal_inode->i_size);
        if (!S_ISREG(journal_inode->i_mode)) {
-               printk(KERN_ERR "EXT3-fs: invalid journal inode.\n");
+               ext3_msg(sb, KERN_ERR, "error: invalid journal inode");
                iput(journal_inode);
                return NULL;
        }
 
        journal = journal_init_inode(journal_inode);
        if (!journal) {
-               printk(KERN_ERR "EXT3-fs: Could not load journal inode\n");
+               ext3_msg(sb, KERN_ERR, "error: could not load journal inode");
                iput(journal_inode);
                return NULL;
        }
@@ -2112,13 +2131,13 @@ static journal_t *ext3_get_dev_journal(struct super_block *sb,
        struct ext3_super_block * es;
        struct block_device *bdev;
 
-       bdev = ext3_blkdev_get(j_dev);
+       bdev = ext3_blkdev_get(j_dev, sb);
        if (bdev == NULL)
                return NULL;
 
        if (bd_claim(bdev, sb)) {
-               printk(KERN_ERR
-                       "EXT3: failed to claim external journal device.\n");
+               ext3_msg(sb, KERN_ERR,
+                       "error: failed to claim external journal device");
                blkdev_put(bdev, FMODE_READ|FMODE_WRITE);
                return NULL;
        }
@@ -2126,8 +2145,8 @@ static journal_t *ext3_get_dev_journal(struct super_block *sb,
        blocksize = sb->s_blocksize;
        hblock = bdev_logical_block_size(bdev);
        if (blocksize < hblock) {
-               printk(KERN_ERR
-                       "EXT3-fs: blocksize too small for journal device.\n");
+               ext3_msg(sb, KERN_ERR,
+                       "error: blocksize too small for journal device");
                goto out_bdev;
        }
 
@@ -2135,8 +2154,8 @@ static journal_t *ext3_get_dev_journal(struct super_block *sb,
        offset = EXT3_MIN_BLOCK_SIZE % blocksize;
        set_blocksize(bdev, blocksize);
        if (!(bh = __bread(bdev, sb_block, blocksize))) {
-               printk(KERN_ERR "EXT3-fs: couldn't read superblock of "
-                      "external journal\n");
+               ext3_msg(sb, KERN_ERR, "error: couldn't read superblock of "
+                       "external journal");
                goto out_bdev;
        }
 
@@ -2144,14 +2163,14 @@ static journal_t *ext3_get_dev_journal(struct super_block *sb,
        if ((le16_to_cpu(es->s_magic) != EXT3_SUPER_MAGIC) ||
            !(le32_to_cpu(es->s_feature_incompat) &
              EXT3_FEATURE_INCOMPAT_JOURNAL_DEV)) {
-               printk(KERN_ERR "EXT3-fs: external journal has "
-                                       "bad superblock\n");
+               ext3_msg(sb, KERN_ERR, "error: external journal has "
+                       "bad superblock");
                brelse(bh);
                goto out_bdev;
        }
 
        if (memcmp(EXT3_SB(sb)->s_es->s_journal_uuid, es->s_uuid, 16)) {
-               printk(KERN_ERR "EXT3-fs: journal UUID does not match\n");
+               ext3_msg(sb, KERN_ERR, "error: journal UUID does not match");
                brelse(bh);
                goto out_bdev;
        }
@@ -2163,19 +2182,21 @@ static journal_t *ext3_get_dev_journal(struct super_block *sb,
        journal = journal_init_dev(bdev, sb->s_bdev,
                                        start, len, blocksize);
        if (!journal) {
-               printk(KERN_ERR "EXT3-fs: failed to create device journal\n");
+               ext3_msg(sb, KERN_ERR,
+                       "error: failed to create device journal");
                goto out_bdev;
        }
        journal->j_private = sb;
        ll_rw_block(READ, 1, &journal->j_sb_buffer);
        wait_on_buffer(journal->j_sb_buffer);
        if (!buffer_uptodate(journal->j_sb_buffer)) {
-               printk(KERN_ERR "EXT3-fs: I/O error on journal device\n");
+               ext3_msg(sb, KERN_ERR, "I/O error on journal device");
                goto out_journal;
        }
        if (be32_to_cpu(journal->j_superblock->s_nr_users) != 1) {
-               printk(KERN_ERR "EXT3-fs: External journal has more than one "
-                                       "user (unsupported) - %d\n",
+               ext3_msg(sb, KERN_ERR,
+                       "error: external journal has more than one "
+                       "user (unsupported) - %d",
                        be32_to_cpu(journal->j_superblock->s_nr_users));
                goto out_journal;
        }
@@ -2201,8 +2222,8 @@ static int ext3_load_journal(struct super_block *sb,
 
        if (journal_devnum &&
            journal_devnum != le32_to_cpu(es->s_journal_dev)) {
-               printk(KERN_INFO "EXT3-fs: external journal device major/minor "
-                       "numbers have changed\n");
+               ext3_msg(sb, KERN_INFO, "external journal device major/minor "
+                       "numbers have changed");
                journal_dev = new_decode_dev(journal_devnum);
        } else
                journal_dev = new_decode_dev(le32_to_cpu(es->s_journal_dev));
@@ -2217,21 +2238,21 @@ static int ext3_load_journal(struct super_block *sb,
 
        if (EXT3_HAS_INCOMPAT_FEATURE(sb, EXT3_FEATURE_INCOMPAT_RECOVER)) {
                if (sb->s_flags & MS_RDONLY) {
-                       printk(KERN_INFO "EXT3-fs: INFO: recovery "
-                                       "required on readonly filesystem.\n");
+                       ext3_msg(sb, KERN_INFO,
+                               "recovery required on readonly filesystem");
                        if (really_read_only) {
-                               printk(KERN_ERR "EXT3-fs: write access "
-                                       "unavailable, cannot proceed.\n");
+                               ext3_msg(sb, KERN_ERR, "error: write access "
+                                       "unavailable, cannot proceed");
                                return -EROFS;
                        }
-                       printk (KERN_INFO "EXT3-fs: write access will "
-                                       "be enabled during recovery.\n");
+                       ext3_msg(sb, KERN_INFO,
+                               "write access will be enabled during recovery");
                }
        }
 
        if (journal_inum && journal_dev) {
-               printk(KERN_ERR "EXT3-fs: filesystem has both journal "
-                      "and inode journals!\n");
+               ext3_msg(sb, KERN_ERR, "error: filesystem has both journal "
+                      "and inode journals");
                return -EINVAL;
        }
 
@@ -2246,7 +2267,7 @@ static int ext3_load_journal(struct super_block *sb,
        if (!really_read_only && test_opt(sb, UPDATE_JOURNAL)) {
                err = journal_update_format(journal);
                if (err)  {
-                       printk(KERN_ERR "EXT3-fs: error updating journal.\n");
+                       ext3_msg(sb, KERN_ERR, "error updating journal");
                        journal_destroy(journal);
                        return err;
                }
@@ -2258,7 +2279,7 @@ static int ext3_load_journal(struct super_block *sb,
                err = journal_load(journal);
 
        if (err) {
-               printk(KERN_ERR "EXT3-fs: error loading journal.\n");
+               ext3_msg(sb, KERN_ERR, "error loading journal");
                journal_destroy(journal);
                return err;
        }
@@ -2277,16 +2298,17 @@ static int ext3_load_journal(struct super_block *sb,
        return 0;
 }
 
-static int ext3_create_journal(struct super_block * sb,
-                              struct ext3_super_block * es,
+static int ext3_create_journal(struct super_block *sb,
+                              struct ext3_super_block *es,
                               unsigned int journal_inum)
 {
        journal_t *journal;
        int err;
 
        if (sb->s_flags & MS_RDONLY) {
-               printk(KERN_ERR "EXT3-fs: readonly filesystem when trying to "
-                               "create journal.\n");
+               ext3_msg(sb, KERN_ERR,
+                       "error: readonly filesystem when trying to "
+                       "create journal");
                return -EROFS;
        }
 
@@ -2294,12 +2316,12 @@ static int ext3_create_journal(struct super_block * sb,
        if (!journal)
                return -EINVAL;
 
-       printk(KERN_INFO "EXT3-fs: creating new journal on inode %u\n",
+       ext3_msg(sb, KERN_INFO, "creating new journal on inode %u",
               journal_inum);
 
        err = journal_create(journal);
        if (err) {
-               printk(KERN_ERR "EXT3-fs: error creating journal.\n");
+               ext3_msg(sb, KERN_ERR, "error creating journal");
                journal_destroy(journal);
                return -EIO;
        }
@@ -2380,8 +2402,8 @@ out:
  * has recorded an error from a previous lifetime, move that error to the
  * main filesystem now.
  */
-static void ext3_clear_journal_err(struct super_block * sb,
-                                  struct ext3_super_block * es)
+static void ext3_clear_journal_err(struct super_block *sb,
+                                  struct ext3_super_block *es)
 {
        journal_t *journal;
        int j_errno;
@@ -2572,10 +2594,10 @@ static int ext3_remount (struct super_block * sb, int * flags, char * data)
                        __le32 ret;
                        if ((ret = EXT3_HAS_RO_COMPAT_FEATURE(sb,
                                        ~EXT3_FEATURE_RO_COMPAT_SUPP))) {
-                               printk(KERN_WARNING "EXT3-fs: %s: couldn't "
-                                      "remount RDWR because of unsupported "
-                                      "optional features (%x).\n",
-                                      sb->s_id, le32_to_cpu(ret));
+                               ext3_msg(sb, KERN_WARNING,
+                                       "warning: couldn't remount RDWR "
+                                       "because of unsupported optional "
+                                       "features (%x)", le32_to_cpu(ret));
                                err = -EROFS;
                                goto restore_opts;
                        }
@@ -2586,11 +2608,10 @@ static int ext3_remount (struct super_block * sb, int * flags, char * data)
                         * require a full umount/remount for now.
                         */
                        if (es->s_last_orphan) {
-                               printk(KERN_WARNING "EXT3-fs: %s: couldn't "
+                               ext3_msg(sb, KERN_WARNING, "warning: couldn't "
                                       "remount RDWR because of unprocessed "
                                       "orphan inode list.  Please "
-                                      "umount/remount instead.\n",
-                                      sb->s_id);
+                                      "umount/remount instead.");
                                err = -EINVAL;
                                goto restore_opts;
                        }
@@ -2839,9 +2860,9 @@ static int ext3_quota_on(struct super_block *sb, int type, int format_id,
        if (EXT3_SB(sb)->s_qf_names[type]) {
                /* Quotafile not of fs root? */
                if (path.dentry->d_parent != sb->s_root)
-                       printk(KERN_WARNING
-                               "EXT3-fs: Quota file not on filesystem root. "
-                               "Journaled quota will not work.\n");
+                       ext3_msg(sb, KERN_WARNING,
+                               "warning: Quota file not on filesystem root. "
+                               "Journaled quota will not work.");
        }
 
        /*
@@ -2923,8 +2944,9 @@ static ssize_t ext3_quota_write(struct super_block *sb, int type,
        handle_t *handle = journal_current_handle();
 
        if (!handle) {
-               printk(KERN_WARNING "EXT3-fs: Quota write (off=%Lu, len=%Lu)"
-                       " cancelled because transaction is not started.\n",
+               ext3_msg(sb, KERN_WARNING,
+                       "warning: quota write (off=%llu, len=%llu)"
+                       " cancelled because transaction is not started.",
                        (unsigned long long)off, (unsigned long long)len);
                return -EIO;
        }
index 7499b36..6b04903 100644 (file)
@@ -918,6 +918,8 @@ extern void ext3_abort (struct super_block *, const char *, const char *, ...)
        __attribute__ ((format (printf, 3, 4)));
 extern void ext3_warning (struct super_block *, const char *, const char *, ...)
        __attribute__ ((format (printf, 3, 4)));
+extern void ext3_msg(struct super_block *, const char *, const char *, ...)
+       __attribute__ ((format (printf, 3, 4)));
 extern void ext3_update_dynamic_rev (struct super_block *sb);
 
 #define ext3_std_error(sb, errno)                              \