sched_ext: Maintain an online cid mask in the scheduler arenagrafted

Schedulers on the default cid mapping treat [0, nr_online_cids) as the
online set and restart on hotplug. Schedulers that install their own mapping
with scx_bpf_cid_override() have no way to learn which cids are online: the
count no longer identifies members and the CPU-form cpumask is unusable from
cid programs. This is an obvious hole in the cid API.

Add scx_bpf_online_cmask(), a kernel-maintained cmask in the scheduler's
arena, allocated alongside the per-CPU scratch masks and populated after the
cid mapping is finalized and before ops.init(), for child schedulers too.
The pointer stays valid through ops.exit() with no reference to take. It is
the arena offset as a void pointer, the same form struct_ops arena arguments
arrive in. The verifier types the void return as a scalar for the program's
arena cast.

The mask follows the SCX hotplug notifications: seeded from cpu_active_mask
and updated before ops.cid_online/offline() runs, so it lags cpu_online_mask
only inside a hotplug transition. Updates walk the scheduler list under the
lock that also serializes unlinking. Reads are live, not atomic snapshots.
Root initialization excludes hotplug.

v2: Reworded the getter kerneldoc (Andrea Righi).

Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>

1 files changed, 174 insertions, 0 deletions

diff --git a/fs/ecryptfs/super.c b/fs/ecryptfs/super.c
new file mode 100644
index 000000000..3bc21d677
--- /dev/null
+++ b/fs/ecryptfs/super.c
@@ -0,0 +1,174 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * eCryptfs: Linux filesystem encryption layer
+ *
+ * Copyright (C) 1997-2003 Erez Zadok
+ * Copyright (C) 2001-2003 Stony Brook University
+ * Copyright (C) 2004-2006 International Business Machines Corp.
+ *   Author(s): Michael A. Halcrow <mahalcro@us.ibm.com>
+ *              Michael C. Thompson <mcthomps@us.ibm.com>
+ */
+
+#include <linux/fs.h>
+#include <linux/mount.h>
+#include <linux/key.h>
+#include <linux/slab.h>
+#include <linux/seq_file.h>
+#include <linux/file.h>
+#include <linux/statfs.h>
+#include <linux/magic.h>
+#include "ecryptfs_kernel.h"
+
+struct kmem_cache *ecryptfs_inode_info_cache;
+
+/**
+ * ecryptfs_alloc_inode - allocate an ecryptfs inode
+ * @sb: Pointer to the ecryptfs super block
+ *
+ * Called to bring an inode into existence.
+ *
+ * Only handle allocation, setting up structures should be done in
+ * ecryptfs_read_inode. This is because the kernel, between now and
+ * then, will 0 out the private data pointer.
+ *
+ * Returns a pointer to a newly allocated inode, NULL otherwise
+ */
+static struct inode *ecryptfs_alloc_inode(struct super_block *sb)
+{
+	struct ecryptfs_inode_info *inode_info;
+	struct inode *inode = NULL;
+
+	inode_info = alloc_inode_sb(sb, ecryptfs_inode_info_cache, GFP_KERNEL);
+	if (unlikely(!inode_info))
+		goto out;
+	ecryptfs_init_crypt_stat(&inode_info->crypt_stat);
+	mutex_init(&inode_info->lower_file_mutex);
+	atomic_set(&inode_info->lower_file_count, 0);
+	inode_info->lower_file = NULL;
+	inode = &inode_info->vfs_inode;
+out:
+	return inode;
+}
+
+static void ecryptfs_free_inode(struct inode *inode)
+{
+	struct ecryptfs_inode_info *inode_info;
+	inode_info = ecryptfs_inode_to_private(inode);
+
+	kmem_cache_free(ecryptfs_inode_info_cache, inode_info);
+}
+
+/**
+ * ecryptfs_destroy_inode
+ * @inode: The ecryptfs inode
+ *
+ * This is used during the final destruction of the inode.  All
+ * allocation of memory related to the inode, including allocated
+ * memory in the crypt_stat struct, will be released here.
+ * There should be no chance that this deallocation will be missed.
+ */
+static void ecryptfs_destroy_inode(struct inode *inode)
+{
+	struct ecryptfs_inode_info *inode_info;
+
+	inode_info = ecryptfs_inode_to_private(inode);
+	BUG_ON(inode_info->lower_file);
+	ecryptfs_destroy_crypt_stat(&inode_info->crypt_stat);
+}
+
+/**
+ * ecryptfs_statfs
+ * @dentry: The ecryptfs dentry
+ * @buf: The struct kstatfs to fill in with stats
+ *
+ * Get the filesystem statistics. Currently, we let this pass right through
+ * to the lower filesystem and take no action ourselves.
+ */
+static int ecryptfs_statfs(struct dentry *dentry, struct kstatfs *buf)
+{
+	struct dentry *lower_dentry = ecryptfs_dentry_to_lower(dentry);
+	int rc;
+
+	if (!lower_dentry->d_sb->s_op->statfs)
+		return -ENOSYS;
+
+	rc = lower_dentry->d_sb->s_op->statfs(lower_dentry, buf);
+	if (rc)
+		return rc;
+
+	buf->f_type = ECRYPTFS_SUPER_MAGIC;
+	rc = ecryptfs_set_f_namelen(&buf->f_namelen, buf->f_namelen,
+	       &ecryptfs_superblock_to_private(dentry->d_sb)->mount_crypt_stat);
+
+	return rc;
+}
+
+/**
+ * ecryptfs_evict_inode
+ * @inode: The ecryptfs inode
+ *
+ * Called by iput() when the inode reference count reached zero
+ * and the inode is not hashed anywhere.  Used to clear anything
+ * that needs to be, before the inode is completely destroyed and put
+ * on the inode free list. We use this to drop out reference to the
+ * lower inode.
+ */
+static void ecryptfs_evict_inode(struct inode *inode)
+{
+	truncate_inode_pages_final(&inode->i_data);
+	clear_inode(inode);
+	iput(ecryptfs_inode_to_lower(inode));
+}
+
+/*
+ * ecryptfs_show_options
+ *
+ * Prints the mount options for a given superblock.
+ * Returns zero; does not fail.
+ */
+static int ecryptfs_show_options(struct seq_file *m, struct dentry *root)
+{
+	struct super_block *sb = root->d_sb;
+	struct ecryptfs_mount_crypt_stat *mount_crypt_stat =
+		&ecryptfs_superblock_to_private(sb)->mount_crypt_stat;
+	struct ecryptfs_global_auth_tok *walker;
+
+	mutex_lock(&mount_crypt_stat->global_auth_tok_list_mutex);
+	list_for_each_entry(walker,
+			    &mount_crypt_stat->global_auth_tok_list,
+			    mount_crypt_stat_list) {
+		if (walker->flags & ECRYPTFS_AUTH_TOK_FNEK)
+			seq_printf(m, ",ecryptfs_fnek_sig=%s", walker->sig);
+		else
+			seq_printf(m, ",ecryptfs_sig=%s", walker->sig);
+	}
+	mutex_unlock(&mount_crypt_stat->global_auth_tok_list_mutex);
+
+	seq_printf(m, ",ecryptfs_cipher=%s",
+		mount_crypt_stat->global_default_cipher_name);
+
+	if (mount_crypt_stat->global_default_cipher_key_size)
+		seq_printf(m, ",ecryptfs_key_bytes=%zd",
+			   mount_crypt_stat->global_default_cipher_key_size);
+	if (mount_crypt_stat->flags & ECRYPTFS_PLAINTEXT_PASSTHROUGH_ENABLED)
+		seq_printf(m, ",ecryptfs_passthrough");
+	if (mount_crypt_stat->flags & ECRYPTFS_XATTR_METADATA_ENABLED)
+		seq_printf(m, ",ecryptfs_xattr_metadata");
+	if (mount_crypt_stat->flags & ECRYPTFS_ENCRYPTED_VIEW_ENABLED)
+		seq_printf(m, ",ecryptfs_encrypted_view");
+	if (mount_crypt_stat->flags & ECRYPTFS_UNLINK_SIGS)
+		seq_printf(m, ",ecryptfs_unlink_sigs");
+	if (mount_crypt_stat->flags & ECRYPTFS_GLOBAL_MOUNT_AUTH_TOK_ONLY)
+		seq_printf(m, ",ecryptfs_mount_auth_tok_only");
+
+	return 0;
+}
+
+const struct super_operations ecryptfs_sops = {
+	.alloc_inode = ecryptfs_alloc_inode,
+	.destroy_inode = ecryptfs_destroy_inode,
+	.free_inode = ecryptfs_free_inode,
+	.statfs = ecryptfs_statfs,
+	.evict_inode = ecryptfs_evict_inode,
+	.show_options = ecryptfs_show_options
+};