KVM: Don't pre-reserve xarray entries when storing empty/NULL attributesgrafted

Skip the xarray reservation loop when clearing all memory attributes, as
storing NULL only erases the entry and never needs to allocate, so no
reservation (and no cleanup of a failed one) is required in that case.

Suggested-by: Sean Christopherson <seanjc@google.com>
Cc: David Ballesteros <davimaba.v@proton.me>
Signed-off-by: Zeng Chi <zengchi@kylinos.cn>
Link: https://patch.msgid.link/20260921102442.1232375-1-zeng_chi911@163.com
[sean: split to separate patch]
Signed-off-by: Sean Christopherson <seanjc@google.com>

1 files changed, 180 insertions, 0 deletions

diff --git a/net/sysctl_net.c b/net/sysctl_net.c
new file mode 100644
index 000000000..e190a639e
--- /dev/null
+++ b/net/sysctl_net.c
@@ -0,0 +1,180 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* -*- linux-c -*-
+ * sysctl_net.c: sysctl interface to net subsystem.
+ *
+ * Begun April 1, 1996, Mike Shaver.
+ * Added /proc/sys/net directories for each protocol family. [MS]
+ *
+ * Revision 1.2  1996/05/08  20:24:40  shaver
+ * Added bits for NET_BRIDGE and the NET_IPV4_ARP stuff and
+ * NET_IPV4_IP_FORWARD.
+ *
+ *
+ */
+
+#include <linux/mm.h>
+#include <linux/export.h>
+#include <linux/sysctl.h>
+#include <linux/nsproxy.h>
+
+#include <net/sock.h>
+
+#ifdef CONFIG_INET
+#include <net/ip.h>
+#endif
+
+#ifdef CONFIG_NET
+#include <linux/if_ether.h>
+#endif
+
+static struct ctl_table_set *
+net_ctl_header_lookup(struct ctl_table_root *root)
+{
+	return &current->nsproxy->net_ns->sysctls;
+}
+
+static int is_seen(struct ctl_table_set *set)
+{
+	return &current->nsproxy->net_ns->sysctls == set;
+}
+
+/* Return standard mode bits for table entry. */
+static int net_ctl_permissions(struct ctl_table_header *head,
+			       const struct ctl_table *table)
+{
+	struct net *net = container_of(head->set, struct net, sysctls);
+
+	/* Allow network administrator to have same access as root. */
+	if (ns_capable_noaudit(net->user_ns, CAP_NET_ADMIN)) {
+		int mode = (table->mode >> 6) & 7;
+		return (mode << 6) | (mode << 3) | mode;
+	}
+
+	return table->mode;
+}
+
+static void net_ctl_set_ownership(struct ctl_table_header *head,
+				  kuid_t *uid, kgid_t *gid)
+{
+	struct net *net = container_of(head->set, struct net, sysctls);
+	kuid_t ns_root_uid;
+	kgid_t ns_root_gid;
+
+	ns_root_uid = make_kuid(net->user_ns, 0);
+	if (uid_valid(ns_root_uid))
+		*uid = ns_root_uid;
+
+	ns_root_gid = make_kgid(net->user_ns, 0);
+	if (gid_valid(ns_root_gid))
+		*gid = ns_root_gid;
+}
+
+static struct ctl_table_root net_sysctl_root = {
+	.lookup = net_ctl_header_lookup,
+	.permissions = net_ctl_permissions,
+	.set_ownership = net_ctl_set_ownership,
+};
+
+static int __net_init sysctl_net_init(struct net *net)
+{
+	setup_sysctl_set(&net->sysctls, &net_sysctl_root, is_seen);
+	return 0;
+}
+
+static void __net_exit sysctl_net_exit(struct net *net)
+{
+	retire_sysctl_set(&net->sysctls);
+}
+
+static struct pernet_operations sysctl_pernet_ops = {
+	.init = sysctl_net_init,
+	.exit = sysctl_net_exit,
+};
+
+static struct ctl_table_header *net_header;
+__init int net_sysctl_init(void)
+{
+	static struct ctl_table empty[1];
+	int ret = -ENOMEM;
+	/* Avoid limitations in the sysctl implementation by
+	 * registering "/proc/sys/net" as an empty directory not in a
+	 * network namespace.
+	 */
+	net_header = register_sysctl_sz("net", empty, 0);
+	if (!net_header)
+		goto out;
+	ret = register_pernet_subsys(&sysctl_pernet_ops);
+	if (ret)
+		goto out1;
+out:
+	return ret;
+out1:
+	unregister_sysctl_table(net_header);
+	net_header = NULL;
+	goto out;
+}
+
+/* Return error when sysctls for non-init netns are unsafe by verifying:
+ * 1) being read-only, or
+ * 2) having a data pointer which points outside of the global kernel/module
+ *    data segment, and rather into the heap where a per-net object was
+ *    allocated.
+ */
+static int ensure_safe_net_sysctl(struct net *net, const char *path,
+				  const struct ctl_table *table,
+				  size_t table_size)
+{
+	const struct ctl_table *ent;
+
+	pr_debug("Registering net sysctl (net %p): %s\n", net, path);
+	ent = table;
+	for (size_t i = 0; i < table_size; ent++, i++) {
+		unsigned long addr;
+		const char *where;
+
+		pr_debug("  procname=%s mode=%o proc_handler=%ps data=%p\n",
+			 ent->procname, ent->mode, ent->proc_handler, ent->data);
+
+		/* If it's not writable inside the netns, then it can't hurt. */
+		if ((ent->mode & 0222) == 0) {
+			pr_debug("    Not writable by anyone\n");
+			continue;
+		}
+
+		/* Where does data point? */
+		addr = (unsigned long)ent->data;
+		if (is_module_address(addr))
+			where = "module";
+		else if (is_kernel_core_data(addr))
+			where = "kernel";
+		else
+			continue;
+
+		/* Warn on netns leak. */
+		WARN(1, "sysctl %s/%s: data points to %s global data: %ps\n",
+		     path, ent->procname, where, ent->data);
+
+		return -EACCES;
+	}
+
+	return 0;
+}
+
+struct ctl_table_header *register_net_sysctl_sz(struct net *net,
+						const char *path,
+						const struct ctl_table *table,
+						size_t table_size)
+{
+	if (!net_eq(net, &init_net))
+		if (ensure_safe_net_sysctl(net, path, table, table_size))
+			return NULL;
+
+	return __register_sysctl_table(&net->sysctls, path, table, table_size);
+}
+EXPORT_SYMBOL_GPL(register_net_sysctl_sz);
+
+void unregister_net_sysctl_table(struct ctl_table_header *header)
+{
+	unregister_sysctl_table(header);
+}
+EXPORT_SYMBOL_GPL(unregister_net_sysctl_table);