btrfs: use GFP_KERNEL in mount and remount
authorDavid Sterba <dsterba@suse.com>
Thu, 22 Jun 2017 00:26:54 +0000 (02:26 +0200)
committerDavid Sterba <dsterba@suse.com>
Wed, 16 Aug 2017 14:12:03 +0000 (16:12 +0200)
We don't need to restrict the allocation flags in btrfs_mount or
_remount. No big filesystem locks are held (possibly s_umount but that
does no count here).

Signed-off-by: David Sterba <dsterba@suse.com>
fs/btrfs/super.c

index e4c268c250f501405254ca8e50cbae054e9e235d..64981fc8e39e8c85b95b66a15928c2acda5a5729 100644 (file)
@@ -426,7 +426,7 @@ int btrfs_parse_options(struct btrfs_fs_info *info, char *options,
         * strsep changes the string, duplicate it because parse_options
         * gets called twice
         */
-       options = kstrdup(options, GFP_NOFS);
+       options = kstrdup(options, GFP_KERNEL);
        if (!options)
                return -ENOMEM;
 
@@ -950,7 +950,7 @@ static char *get_subvol_name_from_objectid(struct btrfs_fs_info *fs_info,
        }
        path->leave_spinning = 1;
 
-       name = kmalloc(PATH_MAX, GFP_NOFS);
+       name = kmalloc(PATH_MAX, GFP_KERNEL);
        if (!name) {
                ret = -ENOMEM;
                goto err;
@@ -1336,10 +1336,11 @@ static char *setup_root_args(char *args)
        char *buf, *dst, *sep;
 
        if (!args)
-               return kstrdup("subvolid=0", GFP_NOFS);
+               return kstrdup("subvolid=0", GFP_KERNEL);
 
        /* The worst case is that we add ",subvolid=0" to the end. */
-       buf = dst = kmalloc(strlen(args) + strlen(",subvolid=0") + 1, GFP_NOFS);
+       buf = dst = kmalloc(strlen(args) + strlen(",subvolid=0") + 1,
+                       GFP_KERNEL);
        if (!buf)
                return NULL;
 
@@ -1568,7 +1569,7 @@ static struct dentry *btrfs_mount(struct file_system_type *fs_type, int flags,
         * it for searching for existing supers, so this lets us do that and
         * then open_ctree will properly initialize everything later.
         */
-       fs_info = kzalloc(sizeof(struct btrfs_fs_info), GFP_NOFS);
+       fs_info = kzalloc(sizeof(struct btrfs_fs_info), GFP_KERNEL);
        if (!fs_info) {
                error = -ENOMEM;
                goto error_sec_opts;
@@ -1576,8 +1577,8 @@ static struct dentry *btrfs_mount(struct file_system_type *fs_type, int flags,
 
        fs_info->fs_devices = fs_devices;
 
-       fs_info->super_copy = kzalloc(BTRFS_SUPER_INFO_SIZE, GFP_NOFS);
-       fs_info->super_for_commit = kzalloc(BTRFS_SUPER_INFO_SIZE, GFP_NOFS);
+       fs_info->super_copy = kzalloc(BTRFS_SUPER_INFO_SIZE, GFP_KERNEL);
+       fs_info->super_for_commit = kzalloc(BTRFS_SUPER_INFO_SIZE, GFP_KERNEL);
        security_init_mnt_opts(&fs_info->security_opts);
        if (!fs_info->super_copy || !fs_info->super_for_commit) {
                error = -ENOMEM;