On Thu, 2 Jun 2011 18:13:22 +0200 David Sterba <dsterba@xxxxxxx> wrote: > fs_info is now ~9kb, more than fits into one page. This will cause > mount failure when memory is too fragmented. Top space consumers are > super block structures super_copy and super_for_commit, ~2.8kb each. > Allocate them dynamically. fs_info will be ~3.5kb. (measured on x86_64) > > Add a wrapper for freeing fs_info. > > Signed-off-by: David Sterba <dsterba@xxxxxxx> > --- ... > @@ -801,6 +800,15 @@ static struct dentry *btrfs_mount(struct file_system_type *fs_type, int flags, > fs_info->fs_devices = fs_devices; > tree_root->fs_info = fs_info; > > + fs_info->super_copy = kzalloc(sizeof(struct btrfs_super_block), > + GFP_NOFS); > + fs_info->super_for_commit = kzalloc(sizeof(struct btrfs_super_block), > + GFP_NOFS); In light of http://thread.gmane.org/gmane.comp.file-systems.btrfs/11004/focus=11038 what do you think of allocating BTRFS_SUPER_INFO_SIZE instead of sizeof? > + if (!fs_info->super_copy || !fs_info->super_for_commit) { > + error = -ENOMEM; > + goto error_close_devices; > + } > + > bdev = fs_devices->latest_bdev; > s = sget(fs_type, btrfs_test_super, btrfs_set_super, tree_root); > if (IS_ERR(s)) -- Sergei
Attachment:
signature.asc
Description: PGP signature
