[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20170118142608.GF6515@twins.programming.kicks-ass.net>
Date: Wed, 18 Jan 2017 15:26:08 +0100
From: Peter Zijlstra <peterz@...radead.org>
To: Hari Bathini <hbathini@...ux.vnet.ibm.com>
Cc: ast@...com, lkml <linux-kernel@...r.kernel.org>, acme@...nel.org,
alexander.shishkin@...ux.intel.com, mingo@...hat.com,
daniel@...earbox.net, rostedt@...dmis.org,
Ananth N Mavinakayanahalli <ananth@...ux.vnet.ibm.com>,
ebiederm@...ssion.com, sargun@...gun.me,
Aravinda Prasad <aravinda@...ux.vnet.ibm.com>,
brendan.d.gregg@...il.com, jolsa@...hat.com
Subject: Re: [PATCH v5 1/3] perf: add PERF_RECORD_NAMESPACES to include
namespaces related info
On Tue, Jan 17, 2017 at 07:47:22PM +0530, Hari Bathini wrote:
> With the advert of container technologies like docker, that depend
> on namespaces for isolation, there is a need for tracing support for
> namespaces. This patch introduces new PERF_RECORD_NAMESPACES event
> for tracing based on namespaces related info.
This fails to mention the whole namespace of namespaces issue Eric
brought up. Where are we on that, and or why isn't it required?
> +/*
> + * The maximum size of the name of each namespace
> + */
> +#define NS_NAME_SIZE 8
> +
> +struct perf_ns_link_info {
> + char name[NS_NAME_SIZE];
Do we really need a name? why?
> + __u64 dev;
> + __u64 ino;
> +};
> +
Powered by blists - more mailing lists