Re: [PATCH] net: inet: diag: expose sockets cgroup classid

From	"Levin, Alexander (Sasha Levin)" <alexander.levin@verizon.com>
Newsgroups	linux.kernel
Subject	Re: [PATCH] net: inet: diag: expose sockets cgroup classid
Date	2017-07-27 01:10 +0200
Message-ID	<u7DBg-6H2-13@gated-at.bofh.it> (permalink)
References	<u7yie-3eE-9@gated-at.bofh.it> <u7CFc-65z-9@gated-at.bofh.it>
Organization	linux.* mail to news gateway

Show all headers | View raw

On Wed, Jul 26, 2017 at 03:01:32PM -0700, Cong Wang wrote:
>On Wed, Jul 26, 2017 at 10:22 AM, Levin, Alexander (Sasha Levin)
><alexander.levin@verizon.com> wrote:
>> +       if (ext & (1 << (INET_DIAG_CLASS_ID - 1))) {
>> +               u32 classid = 0;
>> +
>> +#ifdef CONFIG_SOCK_CGROUP_DATA
>> +               classid = sock_cgroup_classid(&sk->sk_cgrp_data);
>> +#endif
>
>
>If CONFIG_SOCK_CGROUP_DATA is not enabled, should we put 0
>or put nothing (that is, skipping this nla_put())?

My logic was that if CONFIG_SOCK_CGROUP_DATA is disabled, then all sockets have the same default classid ==> 0, rather than having to deal with whether that config is enabled or not.

>> +
>> +               nla_put(skb, INET_DIAG_CLASS_ID, sizeof(classid), &classid);
>
>nla_put_u32()

Will fix, thanks!

-- 

Thanks,
Sasha

Back to linux.kernel | Previous | Next — Previous in thread | Find similar | Unroll thread

Thread

[PATCH] net: inet: diag: expose sockets cgroup classid "Levin, Alexander (Sasha Levin)" <alexander.levin@verizon.com> - 2017-07-26 19:30 +0200
  Re: [PATCH] net: inet: diag: expose sockets cgroup classid Cong Wang <xiyou.wangcong@gmail.com> - 2017-07-27 00:10 +0200
    Re: [PATCH] net: inet: diag: expose sockets cgroup classid "Levin, Alexander (Sasha Levin)" <alexander.levin@verizon.com> - 2017-07-27 01:10 +0200

csiph-web