From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jakub Kicinski Subject: Re: [PATCH bpf-next 1/2] bpftool: introduce cgroup tree command Date: Fri, 6 Jul 2018 14:03:04 -0700 Message-ID: <20180706140304.6de698a9@cakuba.netronome.com> References: <20180706010521.23097-1-guro@fb.com> <20180705190116.7737d66d@cakuba.netronome.com> <20180706182542.GA27167@castle.DHCP.thefacebook.com> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Cc: , , , Quentin Monnet , Daniel Borkmann , Alexei Starovoitov , David Ahern To: Roman Gushchin Return-path: In-Reply-To: <20180706182542.GA27167@castle.DHCP.thefacebook.com> Sender: linux-kernel-owner@vger.kernel.org List-Id: netdev.vger.kernel.org On Fri, 6 Jul 2018 11:25:45 -0700, Roman Gushchin wrote: > > Looks very useful! Minor nits/questions below. I think the reverse > > mapping could also be interesting - similar to how -f flag shows where > > program is pinned, we could add a flag which in > > > > # bpftool prog show/list > > > > adds info about cgroups where the program is attached? Obviously as a > > future extension. > > Well, it would be convenient, but it's not always possible. > A program can be attached to a dying cgroup (a cgroup which was deleted > by a user, but still has some associated resources, e.g. pagecache). Ack, the bpffs and cgroupfs searches are best effort by definition. Thanks for addressing the other comments!