public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Ingo Molnar <mingo@kernel.org>
To: David Ahern <dsahern@gmail.com>
Cc: linux-kernel@vger.kernel.org, acme@ghostprotocols.net,
	fweisbec@gmail.com, peterz@infradead.org, jolsa@redhat.com,
	namhyung@kernel.org, xiaoguangrong@linux.vnet.ibm.com
Subject: Re: [PATCH 10/13] perf kvm: add live mode
Date: Thu, 9 May 2013 10:51:37 +0200	[thread overview]
Message-ID: <20130509085137.GA8813@gmail.com> (raw)
In-Reply-To: <1368073909-8825-11-git-send-email-dsahern@gmail.com>


* David Ahern <dsahern@gmail.com> wrote:

> perf kvm stat currently requires back to back record and report
> commands to see stats. e.g,.
> 
>   perf kvm stat record -p $pid -- sleep 1
>   perf kvm stat report
> 
> This is inconvenvient for on box monitoring of a VM. This patch
> introduces a 'live' mode that in effect combines the record plus
> report into one command. e.g., to monitor a single VM:
> 
>   perf kvm stat live -p $pid
> 
> or all VMs:
> 
>   perf kvm stat live
> 
> Same stats options for the record+report path work with the live mode.
> Display rate defaults to 1 second and can be changed using the -d option.

Nice!

1)

Regarding the basic syntax - is the 'perf stat kvm' command already used 
up? If not then it might make sense to alias 'perf kvm stat live' to 'perf 
kvm stat', making it really easy to use:

  perf kvm stat
  perf kvm stat -p <pid>

etc. I'd expect most 'perf kvm stat' users to utilize this syntax - so it 
makes sense to make 'live' the default and main command.

2)

Another suggestion: it would be nice to allow a "--repeat <n>" option as 
well, for a limited number of iterations - similar to what the 'top' tool 
allows:

  top -b -d <delay> -n <iterations>

--repeat 0 means infinite repeats, --repeat 1 means a single step.

3)

It might make sense to create 'perf kvm stat top' which is a small 
variation to 'perf kvm stat live': it outputs a perf stat output per 
second, but also outputs a clear-screen sequence before the output.

To keep it simple you could do this with plain console text output, not 
libslang, the C code to generate a clear screen sequence can be generated 
via this shell oneliner:

  echo 'prinf("'$(clear)'");' > clearscreen.c

This way people could see constantly updating top-alike KVM statistics.

Thanks,

	Ingo

  parent reply	other threads:[~2013-05-09  8:51 UTC|newest]

Thread overview: 30+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-05-09  4:31 [PATCH 00/13] perf: kvm live mode David Ahern
2013-05-09  4:31 ` [PATCH 01/13] perf evlist: restore methods removed in earlier cleanup David Ahern
2013-05-09  4:31 ` [PATCH 02/13] perf evlist: move tracepoint processing code to evlist.c David Ahern
2013-05-09  4:31 ` [PATCH 03/13] perf evlist: add initialzation function for tracepoints David Ahern
2013-05-09  8:40   ` Namhyung Kim
2013-05-09  4:31 ` [PATCH 04/13] perf session: export a few functions for event processing David Ahern
2013-05-10 12:14   ` Jiri Olsa
2013-05-10 14:07     ` David Ahern
2013-05-09  4:31 ` [PATCH 05/13] perf top: move CONSOLE_CLEAR to header file David Ahern
2013-05-09  4:31 ` [PATCH 06/13] perf kvm: handle realloc failures David Ahern
2013-05-09  4:31 ` [PATCH 07/13] perf kvm: split out tracepoints from record args David Ahern
2013-05-10 12:39   ` Jiri Olsa
2013-05-26  0:10     ` David Ahern
2013-05-09  4:31 ` [PATCH 08/13] perf stats: fix divide by 0 in variance David Ahern
2013-05-09  4:31 ` [PATCH 09/13] perf stats: add max and min stats David Ahern
2013-05-09  4:31 ` [PATCH 10/13] perf kvm: add live mode David Ahern
2013-05-09  8:45   ` Namhyung Kim
2013-05-09 13:56     ` David Ahern
2013-05-09  8:51   ` Ingo Molnar [this message]
2013-05-09 13:54     ` David Ahern
2013-05-10  0:28       ` Namhyung Kim
2013-05-10  0:34         ` David Ahern
2013-05-10  8:08       ` Ingo Molnar
2013-05-09 12:32   ` Runzhen
2013-05-09 13:29     ` David Ahern
2013-05-09  4:31 ` [PATCH 11/13] perf kvm: add min and max stats to display David Ahern
2013-05-09  4:31 ` [PATCH 12/13] perf kvm: option to print events that exceed a threshold David Ahern
2013-05-09  4:31 ` [PATCH 13/13] perf kvm: debug for missing vmexit/vmentry event David Ahern
2013-05-09 12:53 ` [PATCH 00/13] perf: kvm live mode Xiao Guangrong
2013-05-13  9:40   ` Gleb Natapov

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20130509085137.GA8813@gmail.com \
    --to=mingo@kernel.org \
    --cc=acme@ghostprotocols.net \
    --cc=dsahern@gmail.com \
    --cc=fweisbec@gmail.com \
    --cc=jolsa@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=namhyung@kernel.org \
    --cc=peterz@infradead.org \
    --cc=xiaoguangrong@linux.vnet.ibm.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox