On Fri, Jul 21, 2023 at 02:22:46PM -0600, Daniel Xu wrote:
> This commit adds support for enabling IP defrag using pre-existing
> netfilter defrag support. Basically all the flag does is bump a refcnt
> while the link the active. Checks are also added to ensure the prog
> requesting defrag support is run _after_ netfilter defrag hooks.
>
> We also take care to avoid any issues w.r.t. module unloading -- while
> defrag is active on a link, the module is prevented from unloading.
>
> Signed-off-by: Daniel Xu <dxu@xxxxxxxxx>
> ---
> include/uapi/linux/bpf.h | 5 ++
> net/netfilter/nf_bpf_link.c | 123 +++++++++++++++++++++++++++++----
> tools/include/uapi/linux/bpf.h | 5 ++
> 3 files changed, 118 insertions(+), 15 deletions(-)
>
> diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
> index 739c15906a65..12a5480314a2 100644
> --- a/include/uapi/linux/bpf.h
> +++ b/include/uapi/linux/bpf.h
> @@ -1187,6 +1187,11 @@ enum bpf_perf_event_type {
> */
> #define BPF_F_KPROBE_MULTI_RETURN (1U << 0)
>
> +/* link_create.netfilter.flags used in LINK_CREATE command for
> + * BPF_PROG_TYPE_NETFILTER to enable IP packet defragmentation.
> + */
> +#define BPF_F_NETFILTER_IP_DEFRAG (1U << 0)
> +
> /* When BPF ldimm64's insn[0].src_reg != 0 then this can have
> * the following extensions:
> *
> diff --git a/net/netfilter/nf_bpf_link.c b/net/netfilter/nf_bpf_link.c
> index c36da56d756f..8fe594bbc7e2 100644
> --- a/net/netfilter/nf_bpf_link.c
> +++ b/net/netfilter/nf_bpf_link.c
> @@ -1,6 +1,8 @@
> // SPDX-License-Identifier: GPL-2.0
> #include <linux/bpf.h>
> #include <linux/filter.h>
> +#include <linux/kmod.h>
> +#include <linux/module.h>
> #include <linux/netfilter.h>
>
> #include <net/netfilter/nf_bpf_link.h>
> @@ -23,8 +25,88 @@ struct bpf_nf_link {
> struct nf_hook_ops hook_ops;
> struct net *net;
> u32 dead;
> + const struct nf_defrag_hook *defrag_hook;
> };
>
> +static const struct nf_defrag_hook *
> +get_proto_defrag_hook(struct bpf_nf_link *link,
> + const struct nf_defrag_hook __rcu *global_hook,
> + const char *mod)
> +{
> + const struct nf_defrag_hook *hook;
> + int err;
> +
> + /* RCU protects us from races against module unloading */
> + rcu_read_lock();
> + hook = rcu_dereference(global_hook);
> + if (!hook) {
> + rcu_read_unlock();
> + err = request_module(mod);
> + if (err)
> + return ERR_PTR(err < 0 ? err : -EINVAL);
> +
> + rcu_read_lock();
> + hook = rcu_dereference(global_hook);
> + }
> +
> + if (hook && try_module_get(hook->owner)) {
> + /* Once we have a refcnt on the module, we no longer need RCU */
> + hook = rcu_pointer_handoff(hook);
> + } else {
> + WARN_ONCE(!hook, "%s has bad registration", mod);
> + hook = ERR_PTR(-ENOENT);
> + }
> + rcu_read_unlock();
> +
> + if (!IS_ERR(hook)) {
> + err = hook->enable(link->net);
> + if (err) {
> + module_put(hook->owner);
> + hook = ERR_PTR(err);
> + }
> + }
> +
> + return hook;
The rcu + module_get logic looks correct to me, but you've dropped all Florian's acks.
What's going on?
We need explicit acks to merge this through bpf-next.
