Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
MIME-Version: 1.0
From:   Pintu Agarwal <pintu.ping@gmail.com>
Date:   Wed, 23 Mar 2022 21:15:16 +0530
Message-ID: <CAOuPNLhB8pojv4wqv6d4FvhoBbNvUxD19ciK-6E8HtbbrnjF9g@mail.gmail.com>
Subject: kernel space: finding cpu usage of individual process/thread
To:     open list <linux-kernel@vger.kernel.org>,
        Kernelnewbies <kernelnewbies@kernelnewbies.org>,
        Johannes Weiner <hannes@cmpxchg.org>, mingo@redhat.com,
        peterz@infradead.org, juri.lelli@redhat.com
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

Hi,

For one of my requirements I need to find cpu usage or load_avg of an
individual process or thread using "for_each_process_thread(process,
thread)" where process/thread is a pointer to "task_struct".

But, currently I am not able to figure out which is the right field
from task_struct to get this value, or how to manipulate it.

I see some fields there but I'm not sure if these values are correct.
a) sched_entity -> sched_avg -> load/load_avg/load_sum/util_avg/etc.
b) stime + utime : What time I get from here ?
c) prev_cputime
d) vtime
e) acct_timexpd
f) ???

It will be good if I can get some references in kernel space.

When I use (a) above (load_avg) I get this value:
pid=2 , comm=kthread, load=51
pid=96, comm=cat , load=1024
etc...

When I use (stime + utime) I get:
pid=2 , comm=kthread, load=3978848
pid=96, comm=cat , load=11786176

So, not sure which one is correct ?


Thanks,
Pintu