On Wed, Jul 14, 2021 at 02:51 AM CEST, Cong Wang wrote:
> On Tue, Jul 13, 2021 at 12:44 AM Jakub Sitnicki <jakub@xxxxxxxxxxxxxx> wrote:
>>
>> Proc socket stats use sk_prot->inuse_idx value to record inuse sock stats.
>> We currently do not set this correctly from sockmap side. The result is
>> reading sock stats '/proc/net/sockstat' gives incorrect values. The
>> socket counter is incremented correctly, but because we don't set the
>> counter correctly when we replace sk_prot we may omit the decrement.
>>
>> To get the correct inuse_idx value move the core_initcall that initializes
>> the udp proto handlers to late_initcall. This way it is initialized after
>> UDP has the chance to assign the inuse_idx value from the register protocol
>> handler.
>
> Interesting. What about IPv6 module? Based on my understanding, it should
> always be loaded before we can trigger udp_bpf_check_v6_needs_rebuild().
> If so, your patch is complete.
That's my understanding as well. The lazy update_proto call chain is:
sock_map_update_common
sock_map_link
sock_map_init_proto
psock->psock_update_sk_prot
udp_bpf_update_proto
udp_bpf_check_v6_needs_rebuild
If that happens we are being passed an AF_INET6 socket. Socket has been
created so IPv6 module must have been loaded.
>>
>> Fixes: 5e21bb4e8125 ("bpf, test: fix NULL pointer dereference on invalid expected_attach_type")
>
> Should be commit edc6741cc66059532ba621928e3f1b02a53a2f39
> (bpf: Add sockmap hooks for UDP sockets), right?
Thanks. Fixed in v2.
