Files
linux/net/handshake/netlink.c
Chuck Lever ea5fe6a73c net/handshake: Drain pending requests at net namespace exit
The arguments to list_splice_init() in handshake_net_exit() are
reversed. The call moves the local empty "requests" list onto
hn->hn_requests, leaving the local list empty, so the subsequent
drain loop runs zero iterations. Pending handshake requests that
had not yet been accepted are not torn down when the net namespace
is destroyed; each one keeps a reference on a socket file and on
the handshake_req allocation.

Pass the source and destination in the documented order
(list_splice_init(list, head) moves list onto head) so the pending
list is transferred to the local scratch list and drained through
handshake_complete().

Fixing the splice direction exposes a list-corruption race. After
the splice each req->hr_list still has non-empty link pointers,
threading the stack-local scratch list rather than hn_requests.
A concurrent handshake_req_cancel() -- for example, from sunrpc's
TLS timeout on a kernel socket whose netns reference was not
taken -- finds the request through the rhashtable, calls
remove_pending(), and sees !list_empty(&req->hr_list).
__remove_pending_locked() then list_del_init()s an entry off the
scratch list while the drain iterates, corrupting it. The same
call arriving after the drain loop has run list_del() on an
entry hits LIST_POISON instead.

Have remove_pending() check HANDSHAKE_F_NET_DRAINING under
hn_lock and report not-found when drain is in progress. The
drain has already taken ownership; handshake_complete()'s existing
test_and_set on HANDSHAKE_F_REQ_COMPLETED still arbitrates
between drain and cancel for who calls the consumer's hp_done. Use
list_del_init() rather than list_del() in the drain so req->hr_list
does not carry LIST_POISON after drain releases the entry.

The DRAINING guard in remove_pending() makes cancel return false,
but cancel still falls through to test_and_set_bit on
HANDSHAKE_F_REQ_COMPLETED and drops the request's hr_file reference.
Without another pin, if that is the last reference, sk_destruct frees
the request while it is still linked on the drain loop's local list.
Pin each request's hr_file under hn_lock before releasing the list,
and drop that drain pin after the loop finishes with the request.

Fixes: 3b3009ea8a ("net/handshake: Create a NETLINK service for handling handshake requests")
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Reviewed-by: Hannes Reinecke <hare@kernel.org>
Link: https://patch.msgid.link/20260525-handshake-file-pin-v3-8-66c616906ead@oracle.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2026-05-28 13:35:32 +02:00

288 lines
7.0 KiB
C

// SPDX-License-Identifier: GPL-2.0-only
/*
* Generic netlink handshake service
*
* Author: Chuck Lever <chuck.lever@oracle.com>
*
* Copyright (c) 2023, Oracle and/or its affiliates.
*/
#include <linux/types.h>
#include <linux/socket.h>
#include <linux/kernel.h>
#include <linux/module.h>
#include <linux/skbuff.h>
#include <linux/mm.h>
#include <net/sock.h>
#include <net/genetlink.h>
#include <net/netns/generic.h>
#include <kunit/visibility.h>
#include <uapi/linux/handshake.h>
#include "handshake.h"
#include "genl.h"
#include <trace/events/handshake.h>
/**
* handshake_genl_notify - Notify handlers that a request is waiting
* @net: target network namespace
* @proto: handshake protocol
* @flags: memory allocation control flags
*
* Returns zero on success or a negative errno if notification failed.
*/
int handshake_genl_notify(struct net *net, const struct handshake_proto *proto,
gfp_t flags)
{
struct sk_buff *msg;
void *hdr;
/* Disable notifications during unit testing */
if (!test_bit(HANDSHAKE_F_PROTO_NOTIFY, &proto->hp_flags))
return 0;
if (!genl_has_listeners(&handshake_nl_family, net,
proto->hp_handler_class))
return -ESRCH;
msg = genlmsg_new(GENLMSG_DEFAULT_SIZE, flags);
if (!msg)
return -ENOMEM;
hdr = genlmsg_put(msg, 0, 0, &handshake_nl_family, 0,
HANDSHAKE_CMD_READY);
if (!hdr)
goto out_free;
if (nla_put_u32(msg, HANDSHAKE_A_ACCEPT_HANDLER_CLASS,
proto->hp_handler_class) < 0) {
genlmsg_cancel(msg, hdr);
goto out_free;
}
genlmsg_end(msg, hdr);
return genlmsg_multicast_netns(&handshake_nl_family, net, msg,
0, proto->hp_handler_class, flags);
out_free:
nlmsg_free(msg);
return -EMSGSIZE;
}
/**
* handshake_genl_put - Create a generic netlink message header
* @msg: buffer in which to create the header
* @info: generic netlink message context
*
* Returns a ready-to-use header, or NULL.
*/
struct nlmsghdr *handshake_genl_put(struct sk_buff *msg,
struct genl_info *info)
{
return genlmsg_put(msg, info->snd_portid, info->snd_seq,
&handshake_nl_family, 0, info->genlhdr->cmd);
}
EXPORT_SYMBOL(handshake_genl_put);
int handshake_nl_accept_doit(struct sk_buff *skb, struct genl_info *info)
{
struct net *net = sock_net(skb->sk);
struct handshake_net *hn = handshake_pernet(net);
struct handshake_req *req = NULL;
int class, err;
err = -EOPNOTSUPP;
if (!hn)
goto out_status;
err = -EINVAL;
if (GENL_REQ_ATTR_CHECK(info, HANDSHAKE_A_ACCEPT_HANDLER_CLASS))
goto out_status;
class = nla_get_u32(info->attrs[HANDSHAKE_A_ACCEPT_HANDLER_CLASS]);
err = -EAGAIN;
req = handshake_req_next(hn, class);
if (req) {
FD_PREPARE(fdf, O_CLOEXEC, req->hr_file);
if (fdf.err) {
fput(req->hr_file); /* drop ref from handshake_req_next() */
err = fdf.err;
goto out_complete;
}
err = req->hr_proto->hp_accept(req, info, fd_prepare_fd(fdf));
if (err)
goto out_complete; /* Automatic cleanup handles fput */
trace_handshake_cmd_accept(net, req, req->hr_sk, fd_prepare_fd(fdf));
fd_publish(fdf);
return 0;
}
out_complete:
if (req)
handshake_complete(req, -EIO, NULL);
out_status:
trace_handshake_cmd_accept_err(net, req, NULL, err);
return err;
}
int handshake_nl_done_doit(struct sk_buff *skb, struct genl_info *info)
{
struct net *net = sock_net(skb->sk);
struct handshake_req *req;
struct socket *sock;
int fd, status, err;
if (GENL_REQ_ATTR_CHECK(info, HANDSHAKE_A_DONE_SOCKFD))
return -EINVAL;
fd = nla_get_s32(info->attrs[HANDSHAKE_A_DONE_SOCKFD]);
sock = sockfd_lookup(fd, &err);
if (!sock)
return err;
req = handshake_req_hash_lookup(sock->sk);
if (!req) {
err = -EBUSY;
trace_handshake_cmd_done_err(net, req, sock->sk, err);
sockfd_put(sock);
return err;
}
trace_handshake_cmd_done(net, req, sock->sk, fd);
status = -EIO;
if (info->attrs[HANDSHAKE_A_DONE_STATUS])
status = -(int)nla_get_u32(info->attrs[HANDSHAKE_A_DONE_STATUS]);
handshake_complete(req, status, info);
sockfd_put(sock);
return 0;
}
static unsigned int handshake_net_id;
static int __net_init handshake_net_init(struct net *net)
{
struct handshake_net *hn = net_generic(net, handshake_net_id);
unsigned long tmp;
struct sysinfo si;
/*
* Arbitrary limit to prevent handshakes that do not make
* progress from clogging up the system. The cap scales up
* with the amount of physical memory on the system.
*/
si_meminfo(&si);
tmp = si.totalram / (25 * si.mem_unit);
hn->hn_pending_max = clamp(tmp, 3UL, 50UL);
spin_lock_init(&hn->hn_lock);
hn->hn_pending = 0;
hn->hn_flags = 0;
INIT_LIST_HEAD(&hn->hn_requests);
return 0;
}
static void __net_exit handshake_net_exit(struct net *net)
{
struct handshake_net *hn = net_generic(net, handshake_net_id);
struct handshake_req *req;
LIST_HEAD(requests);
/*
* Drain the net's pending list. Requests that have been
* accepted and are in progress will be destroyed when
* the socket is closed.
*/
spin_lock_bh(&hn->hn_lock);
set_bit(HANDSHAKE_F_NET_DRAINING, &hn->hn_flags);
list_splice_init(&hn->hn_requests, &requests);
list_for_each_entry(req, &requests, hr_list)
get_file(req->hr_file);
spin_unlock_bh(&hn->hn_lock);
while (!list_empty(&requests)) {
struct file *file;
req = list_first_entry(&requests, struct handshake_req, hr_list);
file = req->hr_file;
list_del_init(&req->hr_list);
handshake_complete(req, -ETIMEDOUT, NULL);
fput(file);
}
}
static struct pernet_operations handshake_genl_net_ops = {
.init = handshake_net_init,
.exit = handshake_net_exit,
.id = &handshake_net_id,
.size = sizeof(struct handshake_net),
};
/**
* handshake_pernet - Get the handshake private per-net structure
* @net: network namespace
*
* Returns a pointer to the net's private per-net structure for the
* handshake module, or NULL if handshake_init() failed.
*/
struct handshake_net *handshake_pernet(struct net *net)
{
return handshake_net_id ?
net_generic(net, handshake_net_id) : NULL;
}
EXPORT_SYMBOL_IF_KUNIT(handshake_pernet);
static int __init handshake_init(void)
{
int ret;
ret = handshake_req_hash_init();
if (ret) {
pr_warn("handshake: hash initialization failed (%d)\n", ret);
return ret;
}
ret = genl_register_family(&handshake_nl_family);
if (ret) {
pr_warn("handshake: netlink registration failed (%d)\n", ret);
handshake_req_hash_destroy();
return ret;
}
/*
* ORDER: register_pernet_subsys must be done last.
*
* If initialization does not make it past pernet_subsys
* registration, then handshake_net_id will remain 0. That
* shunts the handshake consumer API to return ENOTSUPP
* to prevent it from dereferencing something that hasn't
* been allocated.
*/
ret = register_pernet_subsys(&handshake_genl_net_ops);
if (ret) {
pr_warn("handshake: pernet registration failed (%d)\n", ret);
genl_unregister_family(&handshake_nl_family);
handshake_req_hash_destroy();
}
return ret;
}
static void __exit handshake_exit(void)
{
unregister_pernet_subsys(&handshake_genl_net_ops);
handshake_net_id = 0;
handshake_req_hash_destroy();
genl_unregister_family(&handshake_nl_family);
}
module_init(handshake_init);
module_exit(handshake_exit);