Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
MIME-Version: 1.0
References: <20210512095823.99162-1-yunbo.xufeng@linux.alibaba.com>
 <20210512095823.99162-2-yunbo.xufeng@linux.alibaba.com> <20210512225504.3kt6ij4xqzbtyej5@ast-mbp.dhcp.thefacebook.com>
 <9ae7e503-8f49-a7a4-3e18-0288c7989484@linux.alibaba.com> <CAADnVQK1s4US2LPe4+XsQBEjH8iG8Jdh58n6X1yXAafDAw+M4Q@mail.gmail.com>
In-Reply-To: <CAADnVQK1s4US2LPe4+XsQBEjH8iG8Jdh58n6X1yXAafDAw+M4Q@mail.gmail.com>
From:   KP Singh <kpsingh@kernel.org>
Date:   Fri, 14 May 2021 13:21:41 +0200
Message-ID: <CACYkzJ4xmJBtHBbRfDRnmrCLw=NTf6tYSYXDS6==57RMeBz-4Q@mail.gmail.com>
Subject: Re: [RFC] [PATCH bpf-next 1/1] bpf: Add a BPF helper for getting the
 cgroup path of current task
To:     Alexei Starovoitov <alexei.starovoitov@gmail.com>
Cc:     xufeng zhang <yunbo.xufeng@linux.alibaba.com>,
        Alexei Starovoitov <ast@kernel.org>,
        Daniel Borkmann <daniel@iogearbox.net>,
        bpf <bpf@vger.kernel.org>, LKML <linux-kernel@vger.kernel.org>,
        LSM List <linux-security-module@vger.kernel.org>,
        Florent Revest <revest@chromium.org>,
        Brendan Jackman <jackmanb@chromium.org>,
        Yonghong Song <yhs@fb.com>, Song Liu <songliubraving@fb.com>,
        Martin KaFai Lau <kafai@fb.com>,
        John Fastabend <john.fastabend@gmail.com>,
        Joe Stringer <joe@cilium.io>,
        Quentin Monnet <quentin@isovalent.com>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

> > About your suggestion, the *cgroup_id helpers only works for cgroup v2,
> > however, we're still using cgroup v1 in product=EF=BC=8Cand even for cg=
roup v2,
> > I'm not sure if there is any way for user space to get this cgroup id
> > timely(after container created, but before container start to run)=E3=
=80=82
> >
> > So if there is any effective way works for cgroup v1?
>
> https://github.com/systemd/systemd/blob/main/NEWS#L379

I agree that we should not focus on cgroup v1 if we do add a helper.