nilfs2: use a more common logging style
authorJoe Perches <joe@perches.com>
Wed, 12 Aug 2020 01:35:49 +0000 (18:35 -0700)
committerLinus Torvalds <torvalds@linux-foundation.org>
Wed, 12 Aug 2020 17:58:01 +0000 (10:58 -0700)
Add macros for nilfs_<level>(sb, fmt, ...) and convert the uses of
'nilfs_msg(sb, KERN_<LEVEL>, ...)' to 'nilfs_<level>(sb, ...)' so nilfs2
uses a logging style more like the typical kernel logging style.

Miscellanea:

o Realign arguments for these uses

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ryusuke Konishi <konishi.ryusuke@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Link: http://lkml.kernel.org/r/1595860111-3920-4-git-send-email-konishi.ryusuke@gmail.com
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
20 files changed:
fs/nilfs2/alloc.c
fs/nilfs2/btree.c
fs/nilfs2/cpfile.c
fs/nilfs2/dat.c
fs/nilfs2/direct.c
fs/nilfs2/gcinode.c
fs/nilfs2/ifile.c
fs/nilfs2/inode.c
fs/nilfs2/ioctl.c
fs/nilfs2/mdt.c
fs/nilfs2/namei.c
fs/nilfs2/nilfs.h
fs/nilfs2/page.c
fs/nilfs2/recovery.c
fs/nilfs2/segbuf.c
fs/nilfs2/segment.c
fs/nilfs2/sufile.c
fs/nilfs2/super.c
fs/nilfs2/sysfs.c
fs/nilfs2/the_nilfs.c

index 235b959..adf3bb0 100644 (file)
@@ -613,10 +613,10 @@ void nilfs_palloc_commit_free_entry(struct inode *inode,
        lock = nilfs_mdt_bgl_lock(inode, group);
 
        if (!nilfs_clear_bit_atomic(lock, group_offset, bitmap))
-               nilfs_msg(inode->i_sb, KERN_WARNING,
-                         "%s (ino=%lu): entry number %llu already freed",
-                         __func__, inode->i_ino,
-                         (unsigned long long)req->pr_entry_nr);
+               nilfs_warn(inode->i_sb,
+                          "%s (ino=%lu): entry number %llu already freed",
+                          __func__, inode->i_ino,
+                          (unsigned long long)req->pr_entry_nr);
        else
                nilfs_palloc_group_desc_add_entries(desc, lock, 1);
 
@@ -654,10 +654,10 @@ void nilfs_palloc_abort_alloc_entry(struct inode *inode,
        lock = nilfs_mdt_bgl_lock(inode, group);
 
        if (!nilfs_clear_bit_atomic(lock, group_offset, bitmap))
-               nilfs_msg(inode->i_sb, KERN_WARNING,
-                         "%s (ino=%lu): entry number %llu already freed",
-                         __func__, inode->i_ino,
-                         (unsigned long long)req->pr_entry_nr);
+               nilfs_warn(inode->i_sb,
+                          "%s (ino=%lu): entry number %llu already freed",
+                          __func__, inode->i_ino,
+                          (unsigned long long)req->pr_entry_nr);
        else
                nilfs_palloc_group_desc_add_entries(desc, lock, 1);
 
@@ -763,10 +763,10 @@ int nilfs_palloc_freev(struct inode *inode, __u64 *entry_nrs, size_t nitems)
                do {
                        if (!nilfs_clear_bit_atomic(lock, group_offset,
                                                    bitmap)) {
-                               nilfs_msg(inode->i_sb, KERN_WARNING,
-                                         "%s (ino=%lu): entry number %llu already freed",
-                                         __func__, inode->i_ino,
-                                         (unsigned long long)entry_nrs[j]);
+                               nilfs_warn(inode->i_sb,
+                                          "%s (ino=%lu): entry number %llu already freed",
+                                          __func__, inode->i_ino,
+                                          (unsigned long long)entry_nrs[j]);
                        } else {
                                n++;
                        }
@@ -808,10 +808,10 @@ int nilfs_palloc_freev(struct inode *inode, __u64 *entry_nrs, size_t nitems)
                        ret = nilfs_palloc_delete_entry_block(inode,
                                                              last_nrs[k]);
                        if (ret && ret != -ENOENT)
-                               nilfs_msg(inode->i_sb, KERN_WARNING,
-                                         "error %d deleting block that object (entry=%llu, ino=%lu) belongs to",
-                                         ret, (unsigned long long)last_nrs[k],
-                                         inode->i_ino);
+                               nilfs_warn(inode->i_sb,
+                                          "error %d deleting block that object (entry=%llu, ino=%lu) belongs to",
+                                          ret, (unsigned long long)last_nrs[k],
+                                          inode->i_ino);
                }
 
                desc_kaddr = kmap_atomic(desc_bh->b_page);
@@ -826,9 +826,9 @@ int nilfs_palloc_freev(struct inode *inode, __u64 *entry_nrs, size_t nitems)
                if (nfree == nilfs_palloc_entries_per_group(inode)) {
                        ret = nilfs_palloc_delete_bitmap_block(inode, group);
                        if (ret && ret != -ENOENT)
-                               nilfs_msg(inode->i_sb, KERN_WARNING,
-                                         "error %d deleting bitmap block of group=%lu, ino=%lu",
-                                         ret, group, inode->i_ino);
+                               nilfs_warn(inode->i_sb,
+                                          "error %d deleting bitmap block of group=%lu, ino=%lu",
+                                          ret, group, inode->i_ino);
                }
        }
        return 0;
index 23e043e..f42ab57 100644 (file)
@@ -351,10 +351,10 @@ static int nilfs_btree_node_broken(const struct nilfs_btree_node *node,
                     (flags & NILFS_BTREE_NODE_ROOT) ||
                     nchildren < 0 ||
                     nchildren > NILFS_BTREE_NODE_NCHILDREN_MAX(size))) {
-               nilfs_msg(inode->i_sb, KERN_CRIT,
-                         "bad btree node (ino=%lu, blocknr=%llu): level = %d, flags = 0x%x, nchildren = %d",
-                         inode->i_ino, (unsigned long long)blocknr, level,
-                         flags, nchildren);
+               nilfs_crit(inode->i_sb,
+                          "bad btree node (ino=%lu, blocknr=%llu): level = %d, flags = 0x%x, nchildren = %d",
+                          inode->i_ino, (unsigned long long)blocknr, level,
+                          flags, nchildren);
                ret = 1;
        }
        return ret;
@@ -381,9 +381,9 @@ static int nilfs_btree_root_broken(const struct nilfs_btree_node *node,
                     level >= NILFS_BTREE_LEVEL_MAX ||
                     nchildren < 0 ||
                     nchildren > NILFS_BTREE_ROOT_NCHILDREN_MAX)) {
-               nilfs_msg(inode->i_sb, KERN_CRIT,
-                         "bad btree root (ino=%lu): level = %d, flags = 0x%x, nchildren = %d",
-                         inode->i_ino, level, flags, nchildren);
+               nilfs_crit(inode->i_sb,
+                          "bad btree root (ino=%lu): level = %d, flags = 0x%x, nchildren = %d",
+                          inode->i_ino, level, flags, nchildren);
                ret = 1;
        }
        return ret;
@@ -450,10 +450,10 @@ static int nilfs_btree_bad_node(const struct nilfs_bmap *btree,
 {
        if (unlikely(nilfs_btree_node_get_level(node) != level)) {
                dump_stack();
-               nilfs_msg(btree->b_inode->i_sb, KERN_CRIT,
-                         "btree level mismatch (ino=%lu): %d != %d",
-                         btree->b_inode->i_ino,
-                         nilfs_btree_node_get_level(node), level);
+               nilfs_crit(btree->b_inode->i_sb,
+                          "btree level mismatch (ino=%lu): %d != %d",
+                          btree->b_inode->i_ino,
+                          nilfs_btree_node_get_level(node), level);
                return 1;
        }
        return 0;
@@ -508,7 +508,7 @@ static int __nilfs_btree_get_block(const struct nilfs_bmap *btree, __u64 ptr,
 
  out_no_wait:
        if (!buffer_uptodate(bh)) {
-               nilfs_msg(btree->b_inode->i_sb, KERN_ERR,
+               nilfs_err(btree->b_inode->i_sb,
                          "I/O error reading b-tree node block (ino=%lu, blocknr=%llu)",
                          btree->b_inode->i_ino, (unsigned long long)ptr);
                brelse(bh);
@@ -2074,10 +2074,10 @@ static int nilfs_btree_propagate(struct nilfs_bmap *btree,
        ret = nilfs_btree_do_lookup(btree, path, key, NULL, level + 1, 0);
        if (ret < 0) {
                if (unlikely(ret == -ENOENT))
-                       nilfs_msg(btree->b_inode->i_sb, KERN_CRIT,
-                                 "writing node/leaf block does not appear in b-tree (ino=%lu) at key=%llu, level=%d",
-                                 btree->b_inode->i_ino,
-                                 (unsigned long long)key, level);
+                       nilfs_crit(btree->b_inode->i_sb,
+                                  "writing node/leaf block does not appear in b-tree (ino=%lu) at key=%llu, level=%d",
+                                  btree->b_inode->i_ino,
+                                  (unsigned long long)key, level);
                goto out;
        }
 
@@ -2114,11 +2114,11 @@ static void nilfs_btree_add_dirty_buffer(struct nilfs_bmap *btree,
        if (level < NILFS_BTREE_LEVEL_NODE_MIN ||
            level >= NILFS_BTREE_LEVEL_MAX) {
                dump_stack();
-               nilfs_msg(btree->b_inode->i_sb, KERN_WARNING,
-                         "invalid btree level: %d (key=%llu, ino=%lu, blocknr=%llu)",
-                         level, (unsigned long long)key,
-                         btree->b_inode->i_ino,
-                         (unsigned long long)bh->b_blocknr);
+               nilfs_warn(btree->b_inode->i_sb,
+                          "invalid btree level: %d (key=%llu, ino=%lu, blocknr=%llu)",
+                          level, (unsigned long long)key,
+                          btree->b_inode->i_ino,
+                          (unsigned long long)bh->b_blocknr);
                return;
        }
 
index 8d41311..86d4d85 100644 (file)
@@ -322,7 +322,7 @@ int nilfs_cpfile_delete_checkpoints(struct inode *cpfile,
        int ret, ncps, nicps, nss, count, i;
 
        if (unlikely(start == 0 || start > end)) {
-               nilfs_msg(cpfile->i_sb, KERN_ERR,
+               nilfs_err(cpfile->i_sb,
                          "cannot delete checkpoints: invalid range [%llu, %llu)",
                          (unsigned long long)start, (unsigned long long)end);
                return -EINVAL;
@@ -376,7 +376,7 @@ int nilfs_cpfile_delete_checkpoints(struct inode *cpfile,
                                                                   cpfile, cno);
                                        if (ret == 0)
                                                continue;
-                                       nilfs_msg(cpfile->i_sb, KERN_ERR,
+                                       nilfs_err(cpfile->i_sb,
                                                  "error %d deleting checkpoint block",
                                                  ret);
                                        break;
@@ -981,12 +981,10 @@ int nilfs_cpfile_read(struct super_block *sb, size_t cpsize,
        int err;
 
        if (cpsize > sb->s_blocksize) {
-               nilfs_msg(sb, KERN_ERR,
-                         "too large checkpoint size: %zu bytes", cpsize);
+               nilfs_err(sb, "too large checkpoint size: %zu bytes", cpsize);
                return -EINVAL;
        } else if (cpsize < NILFS_MIN_CHECKPOINT_SIZE) {
-               nilfs_msg(sb, KERN_ERR,
-                         "too small checkpoint size: %zu bytes", cpsize);
+               nilfs_err(sb, "too small checkpoint size: %zu bytes", cpsize);
                return -EINVAL;
        }
 
index 6f40666..8bccdf1 100644 (file)
@@ -340,11 +340,11 @@ int nilfs_dat_move(struct inode *dat, __u64 vblocknr, sector_t blocknr)
        kaddr = kmap_atomic(entry_bh->b_page);
        entry = nilfs_palloc_block_get_entry(dat, vblocknr, entry_bh, kaddr);
        if (unlikely(entry->de_blocknr == cpu_to_le64(0))) {
-               nilfs_msg(dat->i_sb, KERN_CRIT,
-                         "%s: invalid vblocknr = %llu, [%llu, %llu)",
-                         __func__, (unsigned long long)vblocknr,
-                         (unsigned long long)le64_to_cpu(entry->de_start),
-                         (unsigned long long)le64_to_cpu(entry->de_end));
+               nilfs_crit(dat->i_sb,
+                          "%s: invalid vblocknr = %llu, [%llu, %llu)",
+                          __func__, (unsigned long long)vblocknr,
+                          (unsigned long long)le64_to_cpu(entry->de_start),
+                          (unsigned long long)le64_to_cpu(entry->de_end));
                kunmap_atomic(kaddr);
                brelse(entry_bh);
                return -EINVAL;
@@ -471,11 +471,11 @@ int nilfs_dat_read(struct super_block *sb, size_t entry_size,
        int err;
 
        if (entry_size > sb->s_blocksize) {
-               nilfs_msg(sb, KERN_ERR, "too large DAT entry size: %zu bytes",
+               nilfs_err(sb, "too large DAT entry size: %zu bytes",
                          entry_size);
                return -EINVAL;
        } else if (entry_size < NILFS_MIN_DAT_ENTRY_SIZE) {
-               nilfs_msg(sb, KERN_ERR, "too small DAT entry size: %zu bytes",
+               nilfs_err(sb, "too small DAT entry size: %zu bytes",
                          entry_size);
                return -EINVAL;
        }
index 533e24e..f353101 100644 (file)
@@ -328,16 +328,18 @@ static int nilfs_direct_assign(struct nilfs_bmap *bmap,
 
        key = nilfs_bmap_data_get_key(bmap, *bh);
        if (unlikely(key > NILFS_DIRECT_KEY_MAX)) {
-               nilfs_msg(bmap->b_inode->i_sb, KERN_CRIT,
-                         "%s (ino=%lu): invalid key: %llu", __func__,
-                         bmap->b_inode->i_ino, (unsigned long long)key);
+               nilfs_crit(bmap->b_inode->i_sb,
+                          "%s (ino=%lu): invalid key: %llu",
+                          __func__,
+                          bmap->b_inode->i_ino, (unsigned long long)key);
                return -EINVAL;
        }
        ptr = nilfs_direct_get_ptr(bmap, key);
        if (unlikely(ptr == NILFS_BMAP_INVALID_PTR)) {
-               nilfs_msg(bmap->b_inode->i_sb, KERN_CRIT,
-                         "%s (ino=%lu): invalid pointer: %llu", __func__,
-                         bmap->b_inode->i_ino, (unsigned long long)ptr);
+               nilfs_crit(bmap->b_inode->i_sb,
+                          "%s (ino=%lu): invalid pointer: %llu",
+                          __func__,
+                          bmap->b_inode->i_ino, (unsigned long long)ptr);
                return -EINVAL;
        }
 
index aa3c328..4483204 100644 (file)
@@ -142,7 +142,7 @@ int nilfs_gccache_wait_and_mark_dirty(struct buffer_head *bh)
        if (!buffer_uptodate(bh)) {
                struct inode *inode = bh->b_page->mapping->host;
 
-               nilfs_msg(inode->i_sb, KERN_ERR,
+               nilfs_err(inode->i_sb,
                          "I/O error reading %s block for GC (ino=%lu, vblocknr=%llu)",
                          buffer_nilfs_node(bh) ? "node" : "data",
                          inode->i_ino, (unsigned long long)bh->b_blocknr);
index 4140d23..02727ed 100644 (file)
@@ -142,8 +142,8 @@ int nilfs_ifile_get_inode_block(struct inode *ifile, ino_t ino,
 
        err = nilfs_palloc_get_entry_block(ifile, ino, 0, out_bh);
        if (unlikely(err))
-               nilfs_msg(sb, KERN_WARNING, "error %d reading inode: ino=%lu",
-                         err, (unsigned long)ino);
+               nilfs_warn(sb, "error %d reading inode: ino=%lu",
+                          err, (unsigned long)ino);
        return err;
 }
 
index 3318dd1..745d371 100644 (file)
@@ -104,10 +104,10 @@ int nilfs_get_block(struct inode *inode, sector_t blkoff,
                                 * However, the page having this block must
                                 * be locked in this case.
                                 */
-                               nilfs_msg(inode->i_sb, KERN_WARNING,
-                                         "%s (ino=%lu): a race condition while inserting a data block at offset=%llu",
-                                         __func__, inode->i_ino,
-                                         (unsigned long long)blkoff);
+                               nilfs_warn(inode->i_sb,
+                                          "%s (ino=%lu): a race condition while inserting a data block at offset=%llu",
+                                          __func__, inode->i_ino,
+                                          (unsigned long long)blkoff);
                                err = 0;
                        }
                        nilfs_transaction_abort(inode->i_sb);
@@ -707,9 +707,8 @@ repeat:
                goto repeat;
 
 failed:
-       nilfs_msg(ii->vfs_inode.i_sb, KERN_WARNING,
-                 "error %d truncating bmap (ino=%lu)", ret,
-                 ii->vfs_inode.i_ino);
+       nilfs_warn(ii->vfs_inode.i_sb, "error %d truncating bmap (ino=%lu)",
+                  ret, ii->vfs_inode.i_ino);
 }
 
 void nilfs_truncate(struct inode *inode)
@@ -920,9 +919,9 @@ int nilfs_set_file_dirty(struct inode *inode, unsigned int nr_dirty)
                         * This will happen when somebody is freeing
                         * this inode.
                         */
-                       nilfs_msg(inode->i_sb, KERN_WARNING,
-                                 "cannot set file dirty (ino=%lu): the file is being freed",
-                                 inode->i_ino);
+                       nilfs_warn(inode->i_sb,
+                                  "cannot set file dirty (ino=%lu): the file is being freed",
+                                  inode->i_ino);
                        spin_unlock(&nilfs->ns_inode_lock);
                        return -EINVAL; /*
                                         * NILFS_I_DIRTY may remain for
@@ -943,9 +942,9 @@ int __nilfs_mark_inode_dirty(struct inode *inode, int flags)
 
        err = nilfs_load_inode_block(inode, &ibh);
        if (unlikely(err)) {
-               nilfs_msg(inode->i_sb, KERN_WARNING,
-                         "cannot mark inode dirty (ino=%lu): error %d loading inode block",
-                         inode->i_ino, err);
+               nilfs_warn(inode->i_sb,
+                          "cannot mark inode dirty (ino=%lu): error %d loading inode block",
+                          inode->i_ino, err);
                return err;
        }
        nilfs_update_inode(inode, ibh, flags);
@@ -971,8 +970,8 @@ void nilfs_dirty_inode(struct inode *inode, int flags)
        struct nilfs_mdt_info *mdi = NILFS_MDT(inode);
 
        if (is_bad_inode(inode)) {
-               nilfs_msg(inode->i_sb, KERN_WARNING,
-                         "tried to mark bad_inode dirty. ignored.");
+               nilfs_warn(inode->i_sb,
+                          "tried to mark bad_inode dirty. ignored.");
                dump_stack();
                return;
        }
index 4ba73db..07d26f6 100644 (file)
@@ -569,25 +569,25 @@ static int nilfs_ioctl_move_inode_block(struct inode *inode,
 
        if (unlikely(ret < 0)) {
                if (ret == -ENOENT)
-                       nilfs_msg(inode->i_sb, KERN_CRIT,
-                                 "%s: invalid virtual block address (%s): ino=%llu, cno=%llu, offset=%llu, blocknr=%llu, vblocknr=%llu",
-                                 __func__, vdesc->vd_flags ? "node" : "data",
-                                 (unsigned long long)vdesc->vd_ino,
-                                 (unsigned long long)vdesc->vd_cno,
-                                 (unsigned long long)vdesc->vd_offset,
-                                 (unsigned long long)vdesc->vd_blocknr,
-                                 (unsigned long long)vdesc->vd_vblocknr);
+                       nilfs_crit(inode->i_sb,
+                                  "%s: invalid virtual block address (%s): ino=%llu, cno=%llu, offset=%llu, blocknr=%llu, vblocknr=%llu",
+                                  __func__, vdesc->vd_flags ? "node" : "data",
+                                  (unsigned long long)vdesc->vd_ino,
+                                  (unsigned long long)vdesc->vd_cno,
+                                  (unsigned long long)vdesc->vd_offset,
+                                  (unsigned long long)vdesc->vd_blocknr,
+                                  (unsigned long long)vdesc->vd_vblocknr);
                return ret;
        }
        if (unlikely(!list_empty(&bh->b_assoc_buffers))) {
-               nilfs_msg(inode->i_sb, KERN_CRIT,
-                         "%s: conflicting %s buffer: ino=%llu, cno=%llu, offset=%llu, blocknr=%llu, vblocknr=%llu",
-                         __func__, vdesc->vd_flags ? "node" : "data",
-                         (unsigned long long)vdesc->vd_ino,
-                         (unsigned long long)vdesc->vd_cno,
-                         (unsigned long long)vdesc->vd_offset,
-                         (unsigned long long)vdesc->vd_blocknr,
-                         (unsigned long long)vdesc->vd_vblocknr);
+               nilfs_crit(inode->i_sb,
+                          "%s: conflicting %s buffer: ino=%llu, cno=%llu, offset=%llu, blocknr=%llu, vblocknr=%llu",
+                          __func__, vdesc->vd_flags ? "node" : "data",
+                          (unsigned long long)vdesc->vd_ino,
+                          (unsigned long long)vdesc->vd_cno,
+                          (unsigned long long)vdesc->vd_offset,
+                          (unsigned long long)vdesc->vd_blocknr,
+                          (unsigned long long)vdesc->vd_vblocknr);
                brelse(bh);
                return -EEXIST;
        }
@@ -837,8 +837,7 @@ int nilfs_ioctl_prepare_clean_segments(struct the_nilfs *nilfs,
        return 0;
 
  failed:
-       nilfs_msg(nilfs->ns_sb, KERN_ERR, "error %d preparing GC: %s", ret,
-                 msg);
+       nilfs_err(nilfs->ns_sb, "error %d preparing GC: %s", ret, msg);
        return ret;
 }
 
@@ -947,7 +946,7 @@ static int nilfs_ioctl_clean_segments(struct inode *inode, struct file *filp,
 
        ret = nilfs_ioctl_move_blocks(inode->i_sb, &argv[0], kbufs[0]);
        if (ret < 0) {
-               nilfs_msg(inode->i_sb, KERN_ERR,
+               nilfs_err(inode->i_sb,
                          "error %d preparing GC: cannot read source blocks",
                          ret);
        } else {
index 700870a..c0361ce 100644 (file)
@@ -199,7 +199,7 @@ static int nilfs_mdt_read_block(struct inode *inode, unsigned long block,
  out_no_wait:
        err = -EIO;
        if (!buffer_uptodate(first_bh)) {
-               nilfs_msg(inode->i_sb, KERN_ERR,
+               nilfs_err(inode->i_sb,
                          "I/O error reading meta-data file (ino=%lu, block-offset=%lu)",
                          inode->i_ino, block);
                goto failed_bh;
index 9fe6d4a..a6ec796 100644 (file)
@@ -272,9 +272,9 @@ static int nilfs_do_unlink(struct inode *dir, struct dentry *dentry)
                goto out;
 
        if (!inode->i_nlink) {
-               nilfs_msg(inode->i_sb, KERN_WARNING,
-                         "deleting nonexistent file (ino=%lu), %d",
-                         inode->i_ino, inode->i_nlink);
+               nilfs_warn(inode->i_sb,
+                          "deleting nonexistent file (ino=%lu), %d",
+                          inode->i_ino, inode->i_nlink);
                set_nlink(inode, 1);
        }
        err = nilfs_delete_entry(de, page);
index 979a410..f8450ee 100644 (file)
@@ -317,6 +317,15 @@ void __nilfs_error(struct super_block *sb, const char *function,
 
 #endif /* CONFIG_PRINTK */
 
+#define nilfs_crit(sb, fmt, ...)                                       \
+       nilfs_msg(sb, KERN_CRIT, fmt, ##__VA_ARGS__)
+#define nilfs_err(sb, fmt, ...)                                                \
+       nilfs_msg(sb, KERN_ERR, fmt, ##__VA_ARGS__)
+#define nilfs_warn(sb, fmt, ...)                                       \
+       nilfs_msg(sb, KERN_WARNING, fmt, ##__VA_ARGS__)
+#define nilfs_info(sb, fmt, ...)                                       \
+       nilfs_msg(sb, KERN_INFO, fmt, ##__VA_ARGS__)
+
 extern struct nilfs_super_block *
 nilfs_read_super_block(struct super_block *, u64, int, struct buffer_head **);
 extern int nilfs_store_magic_and_option(struct super_block *,
index d7fc8d3..b175f13 100644 (file)
@@ -391,9 +391,8 @@ void nilfs_clear_dirty_page(struct page *page, bool silent)
        BUG_ON(!PageLocked(page));
 
        if (!silent)
-               nilfs_msg(sb, KERN_WARNING,
-                         "discard dirty page: offset=%lld, ino=%lu",
-                         page_offset(page), inode->i_ino);
+               nilfs_warn(sb, "discard dirty page: offset=%lld, ino=%lu",
+                          page_offset(page), inode->i_ino);
 
        ClearPageUptodate(page);
        ClearPageMappedToDisk(page);
@@ -409,9 +408,9 @@ void nilfs_clear_dirty_page(struct page *page, bool silent)
                do {
                        lock_buffer(bh);
                        if (!silent)
-                               nilfs_msg(sb, KERN_WARNING,
-                                         "discard dirty block: blocknr=%llu, size=%zu",
-                                         (u64)bh->b_blocknr, bh->b_size);
+                               nilfs_warn(sb,
+                                          "discard dirty block: blocknr=%llu, size=%zu",
+                                          (u64)bh->b_blocknr, bh->b_size);
 
                        set_mask_bits(&bh->b_state, clear_bits, 0);
                        unlock_buffer(bh);
index 140b663..0b453ef 100644 (file)
@@ -51,7 +51,7 @@ static int nilfs_warn_segment_error(struct super_block *sb, int err)
 
        switch (err) {
        case NILFS_SEG_FAIL_IO:
-               nilfs_msg(sb, KERN_ERR, "I/O error reading segment");
+               nilfs_err(sb, "I/O error reading segment");
                return -EIO;
        case NILFS_SEG_FAIL_MAGIC:
                msg = "Magic number mismatch";
@@ -72,10 +72,10 @@ static int nilfs_warn_segment_error(struct super_block *sb, int err)
                msg = "No super root in the last segment";
                break;
        default:
-               nilfs_msg(sb, KERN_ERR, "unrecognized segment error %d", err);
+               nilfs_err(sb, "unrecognized segment error %d", err);
                return -EINVAL;
        }
-       nilfs_msg(sb, KERN_WARNING, "invalid segment: %s", msg);
+       nilfs_warn(sb, "invalid segment: %s", msg);
        return -EINVAL;
 }
 
@@ -543,10 +543,10 @@ static int nilfs_recover_dsync_blocks(struct the_nilfs *nilfs,
                put_page(page);
 
  failed_inode:
-               nilfs_msg(sb, KERN_WARNING,
-                         "error %d recovering data block (ino=%lu, block-offset=%llu)",
-                         err, (unsigned long)rb->ino,
-                         (unsigned long long)rb->blkoff);
+               nilfs_warn(sb,
+                          "error %d recovering data block (ino=%lu, block-offset=%llu)",
+                          err, (unsigned long)rb->ino,
+                          (unsigned long long)rb->blkoff);
                if (!err2)
                        err2 = err;
  next:
@@ -669,8 +669,7 @@ static int nilfs_do_roll_forward(struct the_nilfs *nilfs,
        }
 
        if (nsalvaged_blocks) {
-               nilfs_msg(sb, KERN_INFO, "salvaged %lu blocks",
-                         nsalvaged_blocks);
+               nilfs_info(sb, "salvaged %lu blocks", nsalvaged_blocks);
                ri->ri_need_recovery = NILFS_RECOVERY_ROLLFORWARD_DONE;
        }
  out:
@@ -681,7 +680,7 @@ static int nilfs_do_roll_forward(struct the_nilfs *nilfs,
  confused:
        err = -EINVAL;
  failed:
-       nilfs_msg(sb, KERN_ERR,
+       nilfs_err(sb,
                  "error %d roll-forwarding partial segment at blocknr = %llu",
                  err, (unsigned long long)pseg_start);
        goto out;
@@ -703,8 +702,8 @@ static void nilfs_finish_roll_forward(struct the_nilfs *nilfs,
        set_buffer_dirty(bh);
        err = sync_dirty_buffer(bh);
        if (unlikely(err))
-               nilfs_msg(nilfs->ns_sb, KERN_WARNING,
-                         "buffer sync write failed during post-cleaning of recovery.");
+               nilfs_warn(nilfs->ns_sb,
+                          "buffer sync write failed during post-cleaning of recovery.");
        brelse(bh);
 }
 
@@ -739,8 +738,7 @@ int nilfs_salvage_orphan_logs(struct the_nilfs *nilfs,
 
        err = nilfs_attach_checkpoint(sb, ri->ri_cno, true, &root);
        if (unlikely(err)) {
-               nilfs_msg(sb, KERN_ERR,
-                         "error %d loading the latest checkpoint", err);
+               nilfs_err(sb, "error %d loading the latest checkpoint", err);
                return err;
        }
 
@@ -751,8 +749,7 @@ int nilfs_salvage_orphan_logs(struct the_nilfs *nilfs,
        if (ri->ri_need_recovery == NILFS_RECOVERY_ROLLFORWARD_DONE) {
                err = nilfs_prepare_segment_for_recovery(nilfs, sb, ri);
                if (unlikely(err)) {
-                       nilfs_msg(sb, KERN_ERR,
-                                 "error %d preparing segment for recovery",
+                       nilfs_err(sb, "error %d preparing segment for recovery",
                                  err);
                        goto failed;
                }
@@ -766,8 +763,7 @@ int nilfs_salvage_orphan_logs(struct the_nilfs *nilfs,
                nilfs_detach_log_writer(sb);
 
                if (unlikely(err)) {
-                       nilfs_msg(sb, KERN_ERR,
-                                 "error %d writing segment for recovery",
+                       nilfs_err(sb, "error %d writing segment for recovery",
                                  err);
                        goto failed;
                }
index 20c479b..1a8729e 100644 (file)
@@ -505,7 +505,7 @@ static int nilfs_segbuf_wait(struct nilfs_segment_buffer *segbuf)
        } while (--segbuf->sb_nbio > 0);
 
        if (unlikely(atomic_read(&segbuf->sb_err) > 0)) {
-               nilfs_msg(segbuf->sb_super, KERN_ERR,
+               nilfs_err(segbuf->sb_super,
                          "I/O error writing log (start-blocknr=%llu, block-count=%lu) in segment %llu",
                          (unsigned long long)segbuf->sb_pseg_start,
                          segbuf->sb_sum.nblocks,
index 91b58c8..a651e82 100644 (file)
@@ -158,7 +158,7 @@ static int nilfs_prepare_segment_lock(struct super_block *sb,
                 * it is saved and will be restored on
                 * nilfs_transaction_commit().
                 */
-               nilfs_msg(sb, KERN_WARNING, "journal info from a different FS");
+               nilfs_warn(sb, "journal info from a different FS");
                save = current->journal_info;
        }
        if (!ti) {
@@ -1940,9 +1940,9 @@ static int nilfs_segctor_collect_dirty_files(struct nilfs_sc_info *sci,
                        err = nilfs_ifile_get_inode_block(
                                ifile, ii->vfs_inode.i_ino, &ibh);
                        if (unlikely(err)) {
-                               nilfs_msg(sci->sc_super, KERN_WARNING,
-                                         "log writer: error %d getting inode block (ino=%lu)",
-                                         err, ii->vfs_inode.i_ino);
+                               nilfs_warn(sci->sc_super,
+                                          "log writer: error %d getting inode block (ino=%lu)",
+                                          err, ii->vfs_inode.i_ino);
                                return err;
                        }
                        spin_lock(&nilfs->ns_inode_lock);
@@ -2449,7 +2449,7 @@ int nilfs_clean_segments(struct super_block *sb, struct nilfs_argv *argv,
                if (likely(!err))
                        break;
 
-               nilfs_msg(sb, KERN_WARNING, "error %d cleaning segments", err);
+               nilfs_warn(sb, "error %d cleaning segments", err);
                set_current_state(TASK_INTERRUPTIBLE);
                schedule_timeout(sci->sc_interval);
        }
@@ -2457,9 +2457,9 @@ int nilfs_clean_segments(struct super_block *sb, struct nilfs_argv *argv,
                int ret = nilfs_discard_segments(nilfs, sci->sc_freesegs,
                                                 sci->sc_nfreesegs);
                if (ret) {
-                       nilfs_msg(sb, KERN_WARNING,
-                                 "error %d on discard request, turning discards off for the device",
-                                 ret);
+                       nilfs_warn(sb,
+                                  "error %d on discard request, turning discards off for the device",
+                                  ret);
                        nilfs_clear_opt(nilfs, DISCARD);
                }
        }
@@ -2540,9 +2540,9 @@ static int nilfs_segctor_thread(void *arg)
        /* start sync. */
        sci->sc_task = current;
        wake_up(&sci->sc_wait_task); /* for nilfs_segctor_start_thread() */
-       nilfs_msg(sci->sc_super, KERN_INFO,
-                 "segctord starting. Construction interval = %lu seconds, CP frequency < %lu seconds",
-                 sci->sc_interval / HZ, sci->sc_mjcp_freq / HZ);
+       nilfs_info(sci->sc_super,
+                  "segctord starting. Construction interval = %lu seconds, CP frequency < %lu seconds",
+                  sci->sc_interval / HZ, sci->sc_mjcp_freq / HZ);
 
        spin_lock(&sci->sc_state_lock);
  loop:
@@ -2616,8 +2616,8 @@ static int nilfs_segctor_start_thread(struct nilfs_sc_info *sci)
        if (IS_ERR(t)) {
                int err = PTR_ERR(t);
 
-               nilfs_msg(sci->sc_super, KERN_ERR,
-                         "error %d creating segctord thread", err);
+               nilfs_err(sci->sc_super, "error %d creating segctord thread",
+                         err);
                return err;
        }
        wait_event(sci->sc_wait_task, sci->sc_task != NULL);
@@ -2727,14 +2727,14 @@ static void nilfs_segctor_destroy(struct nilfs_sc_info *sci)
                nilfs_segctor_write_out(sci);
 
        if (!list_empty(&sci->sc_dirty_files)) {
-               nilfs_msg(sci->sc_super, KERN_WARNING,
-                         "disposed unprocessed dirty file(s) when stopping log writer");
+               nilfs_warn(sci->sc_super,
+                          "disposed unprocessed dirty file(s) when stopping log writer");
                nilfs_dispose_list(nilfs, &sci->sc_dirty_files, 1);
        }
 
        if (!list_empty(&sci->sc_iput_queue)) {
-               nilfs_msg(sci->sc_super, KERN_WARNING,
-                         "disposed unprocessed inode(s) in iput queue when stopping log writer");
+               nilfs_warn(sci->sc_super,
+                          "disposed unprocessed inode(s) in iput queue when stopping log writer");
                nilfs_dispose_list(nilfs, &sci->sc_iput_queue, 1);
        }
 
@@ -2812,8 +2812,8 @@ void nilfs_detach_log_writer(struct super_block *sb)
        spin_lock(&nilfs->ns_inode_lock);
        if (!list_empty(&nilfs->ns_dirty_files)) {
                list_splice_init(&nilfs->ns_dirty_files, &garbage_list);
-               nilfs_msg(sb, KERN_WARNING,
-                         "disposed unprocessed dirty file(s) when detaching log writer");
+               nilfs_warn(sb,
+                          "disposed unprocessed dirty file(s) when detaching log writer");
        }
        spin_unlock(&nilfs->ns_inode_lock);
        up_write(&nilfs->ns_segctor_sem);
index bf3f8f0..42ff67c 100644 (file)
@@ -171,9 +171,9 @@ int nilfs_sufile_updatev(struct inode *sufile, __u64 *segnumv, size_t nsegs,
        down_write(&NILFS_MDT(sufile)->mi_sem);
        for (seg = segnumv; seg < segnumv + nsegs; seg++) {
                if (unlikely(*seg >= nilfs_sufile_get_nsegments(sufile))) {
-                       nilfs_msg(sufile->i_sb, KERN_WARNING,
-                                 "%s: invalid segment number: %llu",
-                                 __func__, (unsigned long long)*seg);
+                       nilfs_warn(sufile->i_sb,
+                                  "%s: invalid segment number: %llu",
+                                  __func__, (unsigned long long)*seg);
                        nerr++;
                }
        }
@@ -230,9 +230,8 @@ int nilfs_sufile_update(struct inode *sufile, __u64 segnum, int create,
        int ret;
 
        if (unlikely(segnum >= nilfs_sufile_get_nsegments(sufile))) {
-               nilfs_msg(sufile->i_sb, KERN_WARNING,
-                         "%s: invalid segment number: %llu",
-                         __func__, (unsigned long long)segnum);
+               nilfs_warn(sufile->i_sb, "%s: invalid segment number: %llu",
+                          __func__, (unsigned long long)segnum);
                return -EINVAL;
        }
        down_write(&NILFS_MDT(sufile)->mi_sem);
@@ -410,9 +409,8 @@ void nilfs_sufile_do_cancel_free(struct inode *sufile, __u64 segnum,
        kaddr = kmap_atomic(su_bh->b_page);
        su = nilfs_sufile_block_get_segment_usage(sufile, segnum, su_bh, kaddr);
        if (unlikely(!nilfs_segment_usage_clean(su))) {
-               nilfs_msg(sufile->i_sb, KERN_WARNING,
-                         "%s: segment %llu must be clean", __func__,
-                         (unsigned long long)segnum);
+               nilfs_warn(sufile->i_sb, "%s: segment %llu must be clean",
+                          __func__, (unsigned long long)segnum);
                kunmap_atomic(kaddr);
                return;
        }
@@ -468,9 +466,8 @@ void nilfs_sufile_do_free(struct inode *sufile, __u64 segnum,
        kaddr = kmap_atomic(su_bh->b_page);
        su = nilfs_sufile_block_get_segment_usage(sufile, segnum, su_bh, kaddr);
        if (nilfs_segment_usage_clean(su)) {
-               nilfs_msg(sufile->i_sb, KERN_WARNING,
-                         "%s: segment %llu is already clean",
-                         __func__, (unsigned long long)segnum);
+               nilfs_warn(sufile->i_sb, "%s: segment %llu is already clean",
+                          __func__, (unsigned long long)segnum);
                kunmap_atomic(kaddr);
                return;
        }
@@ -1168,12 +1165,12 @@ int nilfs_sufile_read(struct super_block *sb, size_t susize,
        int err;
 
        if (susize > sb->s_blocksize) {
-               nilfs_msg(sb, KERN_ERR,
-                         "too large segment usage size: %zu bytes", susize);
+               nilfs_err(sb, "too large segment usage size: %zu bytes",
+                         susize);
                return -EINVAL;
        } else if (susize < NILFS_MIN_SEGMENT_USAGE_SIZE) {
-               nilfs_msg(sb, KERN_ERR,
-                         "too small segment usage size: %zu bytes", susize);
+               nilfs_err(sb, "too small segment usage size: %zu bytes",
+                         susize);
                return -EINVAL;
        }
 
index fef4821..2eee5fb 100644 (file)
@@ -112,7 +112,7 @@ static void nilfs_set_error(struct super_block *sb)
  *
  * This implements the body of nilfs_error() macro.  Normally,
  * nilfs_error() should be used.  As for sustainable errors such as a
- * single-shot I/O error, nilfs_msg() should be used instead.
+ * single-shot I/O error, nilfs_err() should be used instead.
  *
  * Callers should not add a trailing newline since this will do it.
  */
@@ -184,8 +184,7 @@ static int nilfs_sync_super(struct super_block *sb, int flag)
        }
 
        if (unlikely(err)) {
-               nilfs_msg(sb, KERN_ERR, "unable to write superblock: err=%d",
-                         err);
+               nilfs_err(sb, "unable to write superblock: err=%d", err);
                if (err == -EIO && nilfs->ns_sbh[1]) {
                        /*
                         * sbp[0] points to newer log than sbp[1],
@@ -255,7 +254,7 @@ struct nilfs_super_block **nilfs_prepare_super(struct super_block *sb,
                    sbp[1]->s_magic == cpu_to_le16(NILFS_SUPER_MAGIC)) {
                        memcpy(sbp[0], sbp[1], nilfs->ns_sbsize);
                } else {
-                       nilfs_msg(sb, KERN_CRIT, "superblock broke");
+                       nilfs_crit(sb, "superblock broke");
                        return NULL;
                }
        } else if (sbp[1] &&
@@ -365,9 +364,9 @@ static int nilfs_move_2nd_super(struct super_block *sb, loff_t sb2off)
        offset = sb2off & (nilfs->ns_blocksize - 1);
        nsbh = sb_getblk(sb, newblocknr);
        if (!nsbh) {
-               nilfs_msg(sb, KERN_WARNING,
-                         "unable to move secondary superblock to block %llu",
-                         (unsigned long long)newblocknr);
+               nilfs_warn(sb,
+                          "unable to move secondary superblock to block %llu",
+                          (unsigned long long)newblocknr);
                ret = -EIO;
                goto out;
        }
@@ -530,7 +529,7 @@ int nilfs_attach_checkpoint(struct super_block *sb, __u64 cno, int curr_mnt,
        up_read(&nilfs->ns_segctor_sem);
        if (unlikely(err)) {
                if (err == -ENOENT || err == -EINVAL) {
-                       nilfs_msg(sb, KERN_ERR,
+                       nilfs_err(sb,
                                  "Invalid checkpoint (checkpoint number=%llu)",
                                  (unsigned long long)cno);
                        err = -EINVAL;
@@ -628,8 +627,7 @@ static int nilfs_statfs(struct dentry *dentry, struct kstatfs *buf)
        err = nilfs_ifile_count_free_inodes(root->ifile,
                                            &nmaxinodes, &nfreeinodes);
        if (unlikely(err)) {
-               nilfs_msg(sb, KERN_WARNING,
-                         "failed to count free inodes: err=%d", err);
+               nilfs_warn(sb, "failed to count free inodes: err=%d", err);
                if (err == -ERANGE) {
                        /*
                         * If nilfs_palloc_count_max_entries() returns
@@ -761,7 +759,7 @@ static int parse_options(char *options, struct super_block *sb, int is_remount)
                        break;
                case Opt_snapshot:
                        if (is_remount) {
-                               nilfs_msg(sb, KERN_ERR,
+                               nilfs_err(sb,
                                          "\"%s\" option is invalid for remount",
                                          p);
                                return 0;
@@ -777,8 +775,7 @@ static int parse_options(char *options, struct super_block *sb, int is_remount)
                        nilfs_clear_opt(nilfs, DISCARD);
                        break;
                default:
-                       nilfs_msg(sb, KERN_ERR,
-                                 "unrecognized mount option \"%s\"", p);
+                       nilfs_err(sb, "unrecognized mount option \"%s\"", p);
                        return 0;
                }
        }
@@ -814,10 +811,10 @@ static int nilfs_setup_super(struct super_block *sb, int is_mount)
        mnt_count = le16_to_cpu(sbp[0]->s_mnt_count);
 
        if (nilfs->ns_mount_state & NILFS_ERROR_FS) {
-               nilfs_msg(sb, KERN_WARNING, "mounting fs with errors");
+               nilfs_warn(sb, "mounting fs with errors");
 #if 0
        } else if (max_mnt_count >= 0 && mnt_count >= max_mnt_count) {
-               nilfs_msg(sb, KERN_WARNING, "maximal mount count reached");
+               nilfs_warn(sb, "maximal mount count reached");
 #endif
        }
        if (!max_mnt_count)
@@ -880,7 +877,7 @@ int nilfs_check_feature_compatibility(struct super_block *sb,
        features = le64_to_cpu(sbp->s_feature_incompat) &
                ~NILFS_FEATURE_INCOMPAT_SUPP;
        if (features) {
-               nilfs_msg(sb, KERN_ERR,
+               nilfs_err(sb,
                          "couldn't mount because of unsupported optional features (%llx)",
                          (unsigned long long)features);
                return -EINVAL;
@@ -888,7 +885,7 @@ int nilfs_check_feature_compatibility(struct super_block *sb,
        features = le64_to_cpu(sbp->s_feature_compat_ro) &
                ~NILFS_FEATURE_COMPAT_RO_SUPP;
        if (!sb_rdonly(sb) && features) {
-               nilfs_msg(sb, KERN_ERR,
+               nilfs_err(sb,
                          "couldn't mount RDWR because of unsupported optional features (%llx)",
                          (unsigned long long)features);
                return -EINVAL;
@@ -907,12 +904,12 @@ static int nilfs_get_root_dentry(struct super_block *sb,
        inode = nilfs_iget(sb, root, NILFS_ROOT_INO);
        if (IS_ERR(inode)) {
                ret = PTR_ERR(inode);
-               nilfs_msg(sb, KERN_ERR, "error %d getting root inode", ret);
+               nilfs_err(sb, "error %d getting root inode", ret);
                goto out;
        }
        if (!S_ISDIR(inode->i_mode) || !inode->i_blocks || !inode->i_size) {
                iput(inode);
-               nilfs_msg(sb, KERN_ERR, "corrupt root inode");
+               nilfs_err(sb, "corrupt root inode");
                ret = -EINVAL;
                goto out;
        }
@@ -940,7 +937,7 @@ static int nilfs_get_root_dentry(struct super_block *sb,
        return ret;
 
  failed_dentry:
-       nilfs_msg(sb, KERN_ERR, "error %d getting root dentry", ret);
+       nilfs_err(sb, "error %d getting root dentry", ret);
        goto out;
 }
 
@@ -960,7 +957,7 @@ static int nilfs_attach_snapshot(struct super_block *s, __u64 cno,
                ret = (ret == -ENOENT) ? -EINVAL : ret;
                goto out;
        } else if (!ret) {
-               nilfs_msg(s, KERN_ERR,
+               nilfs_err(s,
                          "The specified checkpoint is not a snapshot (checkpoint number=%llu)",
                          (unsigned long long)cno);
                ret = -EINVAL;
@@ -969,7 +966,7 @@ static int nilfs_attach_snapshot(struct super_block *s, __u64 cno,
 
        ret = nilfs_attach_checkpoint(s, cno, false, &root);
        if (ret) {
-               nilfs_msg(s, KERN_ERR,
+               nilfs_err(s,
                          "error %d while loading snapshot (checkpoint number=%llu)",
                          ret, (unsigned long long)cno);
                goto out;
@@ -1066,7 +1063,7 @@ nilfs_fill_super(struct super_block *sb, void *data, int silent)
        cno = nilfs_last_cno(nilfs);
        err = nilfs_attach_checkpoint(sb, cno, true, &fsroot);
        if (err) {
-               nilfs_msg(sb, KERN_ERR,
+               nilfs_err(sb,
                          "error %d while loading last checkpoint (checkpoint number=%llu)",
                          err, (unsigned long long)cno);
                goto failed_unload;
@@ -1128,8 +1125,8 @@ static int nilfs_remount(struct super_block *sb, int *flags, char *data)
        err = -EINVAL;
 
        if (!nilfs_valid_fs(nilfs)) {
-               nilfs_msg(sb, KERN_WARNING,
-                         "couldn't remount because the filesystem is in an incomplete recovery state");
+               nilfs_warn(sb,
+                          "couldn't remount because the filesystem is in an incomplete recovery state");
                goto restore_opts;
        }
 
@@ -1161,9 +1158,9 @@ static int nilfs_remount(struct super_block *sb, int *flags, char *data)
                        ~NILFS_FEATURE_COMPAT_RO_SUPP;
                up_read(&nilfs->ns_sem);
                if (features) {
-                       nilfs_msg(sb, KERN_WARNING,
-                                 "couldn't remount RDWR because of unsupported optional features (%llx)",
-                                 (unsigned long long)features);
+                       nilfs_warn(sb,
+                                  "couldn't remount RDWR because of unsupported optional features (%llx)",
+                                  (unsigned long long)features);
                        err = -EROFS;
                        goto restore_opts;
                }
@@ -1222,7 +1219,7 @@ static int nilfs_parse_snapshot_option(const char *option,
        return 0;
 
 parse_error:
-       nilfs_msg(NULL, KERN_ERR, "invalid option \"%s\": %s", option, msg);
+       nilfs_err(NULL, "invalid option \"%s\": %s", option, msg);
        return 1;
 }
 
@@ -1325,7 +1322,7 @@ nilfs_mount(struct file_system_type *fs_type, int flags,
        } else if (!sd.cno) {
                if (nilfs_tree_is_busy(s->s_root)) {
                        if ((flags ^ s->s_flags) & SB_RDONLY) {
-                               nilfs_msg(s, KERN_ERR,
+                               nilfs_err(s,
                                          "the device already has a %s mount.",
                                          sb_rdonly(s) ? "read-only" : "read/write");
                                err = -EBUSY;
index e60be7b..303d714 100644 (file)
@@ -263,8 +263,8 @@ nilfs_checkpoints_checkpoints_number_show(struct nilfs_checkpoints_attr *attr,
        err = nilfs_cpfile_get_stat(nilfs->ns_cpfile, &cpstat);
        up_read(&nilfs->ns_segctor_sem);
        if (err < 0) {
-               nilfs_msg(nilfs->ns_sb, KERN_ERR,
-                         "unable to get checkpoint stat: err=%d", err);
+               nilfs_err(nilfs->ns_sb, "unable to get checkpoint stat: err=%d",
+                         err);
                return err;
        }
 
@@ -286,8 +286,8 @@ nilfs_checkpoints_snapshots_number_show(struct nilfs_checkpoints_attr *attr,
        err = nilfs_cpfile_get_stat(nilfs->ns_cpfile, &cpstat);
        up_read(&nilfs->ns_segctor_sem);
        if (err < 0) {
-               nilfs_msg(nilfs->ns_sb, KERN_ERR,
-                         "unable to get checkpoint stat: err=%d", err);
+               nilfs_err(nilfs->ns_sb, "unable to get checkpoint stat: err=%d",
+                         err);
                return err;
        }
 
@@ -405,8 +405,8 @@ nilfs_segments_dirty_segments_show(struct nilfs_segments_attr *attr,
        err = nilfs_sufile_get_stat(nilfs->ns_sufile, &sustat);
        up_read(&nilfs->ns_segctor_sem);
        if (err < 0) {
-               nilfs_msg(nilfs->ns_sb, KERN_ERR,
-                         "unable to get segment stat: err=%d", err);
+               nilfs_err(nilfs->ns_sb, "unable to get segment stat: err=%d",
+                         err);
                return err;
        }
 
@@ -779,15 +779,15 @@ nilfs_superblock_sb_update_frequency_store(struct nilfs_superblock_attr *attr,
 
        err = kstrtouint(skip_spaces(buf), 0, &val);
        if (err) {
-               nilfs_msg(nilfs->ns_sb, KERN_ERR,
-                         "unable to convert string: err=%d", err);
+               nilfs_err(nilfs->ns_sb, "unable to convert string: err=%d",
+                         err);
                return err;
        }
 
        if (val < NILFS_SB_FREQ) {
                val = NILFS_SB_FREQ;
-               nilfs_msg(nilfs->ns_sb, KERN_WARNING,
-                         "superblock update frequency cannot be lesser than 10 seconds");
+               nilfs_warn(nilfs->ns_sb,
+                          "superblock update frequency cannot be lesser than 10 seconds");
        }
 
        down_write(&nilfs->ns_sem);
@@ -990,8 +990,7 @@ int nilfs_sysfs_create_device_group(struct super_block *sb)
        nilfs->ns_dev_subgroups = kzalloc(devgrp_size, GFP_KERNEL);
        if (unlikely(!nilfs->ns_dev_subgroups)) {
                err = -ENOMEM;
-               nilfs_msg(sb, KERN_ERR,
-                         "unable to allocate memory for device group");
+               nilfs_err(sb, "unable to allocate memory for device group");
                goto failed_create_device_group;
        }
 
@@ -1101,15 +1100,13 @@ int __init nilfs_sysfs_init(void)
        nilfs_kset = kset_create_and_add(NILFS_ROOT_GROUP_NAME, NULL, fs_kobj);
        if (!nilfs_kset) {
                err = -ENOMEM;
-               nilfs_msg(NULL, KERN_ERR,
-                         "unable to create sysfs entry: err=%d", err);
+               nilfs_err(NULL, "unable to create sysfs entry: err=%d", err);
                goto failed_sysfs_init;
        }
 
        err = sysfs_create_group(&nilfs_kset->kobj, &nilfs_feature_attr_group);
        if (unlikely(err)) {
-               nilfs_msg(NULL, KERN_ERR,
-                         "unable to create feature group: err=%d", err);
+               nilfs_err(NULL, "unable to create feature group: err=%d", err);
                goto cleanup_sysfs_init;
        }
 
index 484785c..221a1cc 100644 (file)
@@ -183,7 +183,7 @@ static int nilfs_store_log_cursor(struct the_nilfs *nilfs,
                nilfs_get_segnum_of_block(nilfs, nilfs->ns_last_pseg);
        nilfs->ns_cno = nilfs->ns_last_cno + 1;
        if (nilfs->ns_segnum >= nilfs->ns_nsegments) {
-               nilfs_msg(nilfs->ns_sb, KERN_ERR,
+               nilfs_err(nilfs->ns_sb,
                          "pointed segment number is out of range: segnum=%llu, nsegments=%lu",
                          (unsigned long long)nilfs->ns_segnum,
                          nilfs->ns_nsegments);
@@ -210,12 +210,12 @@ int load_nilfs(struct the_nilfs *nilfs, struct super_block *sb)
        int err;
 
        if (!valid_fs) {
-               nilfs_msg(sb, KERN_WARNING, "mounting unchecked fs");
+               nilfs_warn(sb, "mounting unchecked fs");
                if (s_flags & SB_RDONLY) {
-                       nilfs_msg(sb, KERN_INFO,
-                                 "recovery required for readonly filesystem");
-                       nilfs_msg(sb, KERN_INFO,
-                                 "write access will be enabled during recovery");
+                       nilfs_info(sb,
+                                  "recovery required for readonly filesystem");
+                       nilfs_info(sb,
+                                  "write access will be enabled during recovery");
                }
        }
 
@@ -230,12 +230,11 @@ int load_nilfs(struct the_nilfs *nilfs, struct super_block *sb)
                        goto scan_error;
 
                if (!nilfs_valid_sb(sbp[1])) {
-                       nilfs_msg(sb, KERN_WARNING,
-                                 "unable to fall back to spare super block");
+                       nilfs_warn(sb,
+                                  "unable to fall back to spare super block");
                        goto scan_error;
                }
-               nilfs_msg(sb, KERN_INFO,
-                         "trying rollback from an earlier position");
+               nilfs_info(sb, "trying rollback from an earlier position");
 
                /*
                 * restore super block with its spare and reconfigure
@@ -248,9 +247,9 @@ int load_nilfs(struct the_nilfs *nilfs, struct super_block *sb)
                /* verify consistency between two super blocks */
                blocksize = BLOCK_SIZE << le32_to_cpu(sbp[0]->s_log_block_size);
                if (blocksize != nilfs->ns_blocksize) {
-                       nilfs_msg(sb, KERN_WARNING,
-                                 "blocksize differs between two super blocks (%d != %d)",
-                                 blocksize, nilfs->ns_blocksize);
+                       nilfs_warn(sb,
+                                  "blocksize differs between two super blocks (%d != %d)",
+                                  blocksize, nilfs->ns_blocksize);
                        goto scan_error;
                }
 
@@ -269,8 +268,7 @@ int load_nilfs(struct the_nilfs *nilfs, struct super_block *sb)
 
        err = nilfs_load_super_root(nilfs, sb, ri.ri_super_root);
        if (unlikely(err)) {
-               nilfs_msg(sb, KERN_ERR, "error %d while loading super root",
-                         err);
+               nilfs_err(sb, "error %d while loading super root", err);
                goto failed;
        }
 
@@ -281,28 +279,28 @@ int load_nilfs(struct the_nilfs *nilfs, struct super_block *sb)
                __u64 features;
 
                if (nilfs_test_opt(nilfs, NORECOVERY)) {
-                       nilfs_msg(sb, KERN_INFO,
-                                 "norecovery option specified, skipping roll-forward recovery");
+                       nilfs_info(sb,
+                                  "norecovery option specified, skipping roll-forward recovery");
                        goto skip_recovery;
                }
                features = le64_to_cpu(nilfs->ns_sbp[0]->s_feature_compat_ro) &
                        ~NILFS_FEATURE_COMPAT_RO_SUPP;
                if (features) {
-                       nilfs_msg(sb, KERN_ERR,
+                       nilfs_err(sb,
                                  "couldn't proceed with recovery because of unsupported optional features (%llx)",
                                  (unsigned long long)features);
                        err = -EROFS;
                        goto failed_unload;
                }
                if (really_read_only) {
-                       nilfs_msg(sb, KERN_ERR,
+                       nilfs_err(sb,
                                  "write access unavailable, cannot proceed");
                        err = -EROFS;
                        goto failed_unload;
                }
                sb->s_flags &= ~SB_RDONLY;
        } else if (nilfs_test_opt(nilfs, NORECOVERY)) {
-               nilfs_msg(sb, KERN_ERR,
+               nilfs_err(sb,
                          "recovery cancelled because norecovery option was specified for a read/write mount");
                err = -EINVAL;
                goto failed_unload;
@@ -318,12 +316,12 @@ int load_nilfs(struct the_nilfs *nilfs, struct super_block *sb)
        up_write(&nilfs->ns_sem);
 
        if (err) {
-               nilfs_msg(sb, KERN_ERR,
+               nilfs_err(sb,
                          "error %d updating super block. recovery unfinished.",
                          err);
                goto failed_unload;
        }
-       nilfs_msg(sb, KERN_INFO, "recovery complete");
+       nilfs_info(sb, "recovery complete");
 
  skip_recovery:
        nilfs_clear_recovery_info(&ri);
@@ -331,7 +329,7 @@ int load_nilfs(struct the_nilfs *nilfs, struct super_block *sb)
        return 0;
 
  scan_error:
-       nilfs_msg(sb, KERN_ERR, "error %d while searching super root", err);
+       nilfs_err(sb, "error %d while searching super root", err);
        goto failed;
 
  failed_unload:
@@ -378,7 +376,7 @@ static int nilfs_store_disk_layout(struct the_nilfs *nilfs,
                                   struct nilfs_super_block *sbp)
 {
        if (le32_to_cpu(sbp->s_rev_level) < NILFS_MIN_SUPP_REV) {
-               nilfs_msg(nilfs->ns_sb, KERN_ERR,
+               nilfs_err(nilfs->ns_sb,
                          "unsupported revision (superblock rev.=%d.%d, current rev.=%d.%d). Please check the version of mkfs.nilfs(2).",
                          le32_to_cpu(sbp->s_rev_level),
                          le16_to_cpu(sbp->s_minor_rev_level),
@@ -391,13 +389,11 @@ static int nilfs_store_disk_layout(struct the_nilfs *nilfs,
 
        nilfs->ns_inode_size = le16_to_cpu(sbp->s_inode_size);
        if (nilfs->ns_inode_size > nilfs->ns_blocksize) {
-               nilfs_msg(nilfs->ns_sb, KERN_ERR,
-                         "too large inode size: %d bytes",
+               nilfs_err(nilfs->ns_sb, "too large inode size: %d bytes",
                          nilfs->ns_inode_size);
                return -EINVAL;
        } else if (nilfs->ns_inode_size < NILFS_MIN_INODE_SIZE) {
-               nilfs_msg(nilfs->ns_sb, KERN_ERR,
-                         "too small inode size: %d bytes",
+               nilfs_err(nilfs->ns_sb, "too small inode size: %d bytes",
                          nilfs->ns_inode_size);
                return -EINVAL;
        }
@@ -406,8 +402,7 @@ static int nilfs_store_disk_layout(struct the_nilfs *nilfs,
 
        nilfs->ns_blocks_per_segment = le32_to_cpu(sbp->s_blocks_per_segment);
        if (nilfs->ns_blocks_per_segment < NILFS_SEG_MIN_BLOCKS) {
-               nilfs_msg(nilfs->ns_sb, KERN_ERR,
-                         "too short segment: %lu blocks",
+               nilfs_err(nilfs->ns_sb, "too short segment: %lu blocks",
                          nilfs->ns_blocks_per_segment);
                return -EINVAL;
        }
@@ -417,7 +412,7 @@ static int nilfs_store_disk_layout(struct the_nilfs *nilfs,
                le32_to_cpu(sbp->s_r_segments_percentage);
        if (nilfs->ns_r_segments_percentage < 1 ||
            nilfs->ns_r_segments_percentage > 99) {
-               nilfs_msg(nilfs->ns_sb, KERN_ERR,
+               nilfs_err(nilfs->ns_sb,
                          "invalid reserved segments percentage: %lu",
                          nilfs->ns_r_segments_percentage);
                return -EINVAL;
@@ -503,16 +498,16 @@ static int nilfs_load_super_block(struct the_nilfs *nilfs,
 
        if (!sbp[0]) {
                if (!sbp[1]) {
-                       nilfs_msg(sb, KERN_ERR, "unable to read superblock");
+                       nilfs_err(sb, "unable to read superblock");
                        return -EIO;
                }
-               nilfs_msg(sb, KERN_WARNING,
-                         "unable to read primary superblock (blocksize = %d)",
-                         blocksize);
+               nilfs_warn(sb,
+                          "unable to read primary superblock (blocksize = %d)",
+                          blocksize);
        } else if (!sbp[1]) {
-               nilfs_msg(sb, KERN_WARNING,
-                         "unable to read secondary superblock (blocksize = %d)",
-                         blocksize);
+               nilfs_warn(sb,
+                          "unable to read secondary superblock (blocksize = %d)",
+                          blocksize);
        }
 
        /*
@@ -534,14 +529,14 @@ static int nilfs_load_super_block(struct the_nilfs *nilfs,
        }
        if (!valid[swp]) {
                nilfs_release_super_block(nilfs);
-               nilfs_msg(sb, KERN_ERR, "couldn't find nilfs on the device");
+               nilfs_err(sb, "couldn't find nilfs on the device");
                return -EINVAL;
        }
 
        if (!valid[!swp])
-               nilfs_msg(sb, KERN_WARNING,
-                         "broken superblock, retrying with spare superblock (blocksize = %d)",
-                         blocksize);
+               nilfs_warn(sb,
+                          "broken superblock, retrying with spare superblock (blocksize = %d)",
+                          blocksize);
        if (swp)
                nilfs_swap_super_block(nilfs);
 
@@ -575,7 +570,7 @@ int init_nilfs(struct the_nilfs *nilfs, struct super_block *sb, char *data)
 
        blocksize = sb_min_blocksize(sb, NILFS_MIN_BLOCK_SIZE);
        if (!blocksize) {
-               nilfs_msg(sb, KERN_ERR, "unable to set blocksize");
+               nilfs_err(sb, "unable to set blocksize");
                err = -EINVAL;
                goto out;
        }
@@ -594,7 +589,7 @@ int init_nilfs(struct the_nilfs *nilfs, struct super_block *sb, char *data)
        blocksize = BLOCK_SIZE << le32_to_cpu(sbp->s_log_block_size);
        if (blocksize < NILFS_MIN_BLOCK_SIZE ||
            blocksize > NILFS_MAX_BLOCK_SIZE) {
-               nilfs_msg(sb, KERN_ERR,
+               nilfs_err(sb,
                          "couldn't mount because of unsupported filesystem blocksize %d",
                          blocksize);
                err = -EINVAL;
@@ -604,7 +599,7 @@ int init_nilfs(struct the_nilfs *nilfs, struct super_block *sb, char *data)
                int hw_blocksize = bdev_logical_block_size(sb->s_bdev);
 
                if (blocksize < hw_blocksize) {
-                       nilfs_msg(sb, KERN_ERR,
+                       nilfs_err(sb,
                                  "blocksize %d too small for device (sector-size = %d)",
                                  blocksize, hw_blocksize);
                        err = -EINVAL;