All of lore.kernel.org
 help / color / mirror / Atom feed
From: Andre Schmidt <andre@osku.de>
To: Greg KH <greg@kroah.com>
Cc: kernelnewbies@kernelnewbies.org
Subject: Re: Simple system stats.
Date: Sat, 2 Nov 2019 14:56:36 +0100	[thread overview]
Message-ID: <20191102145636.60fa2891@ALDOS> (raw)
In-Reply-To: <20191102131308.GA391073@kroah.com>

On Sat, 2 Nov 2019 14:13:08 +0100
Greg KH <greg@kroah.com> wrote:

> On Sat, Nov 02, 2019 at 02:05:19PM +0100, Andre Schmidt wrote:
> > On Sat, 2 Nov 2019 13:07:00 +0100
> > Greg KH <greg@kroah.com> wrote:
> > 
> > > On Sat, Nov 02, 2019 at 12:51:05PM +0100, Andre Schmidt wrote:
> > > > Hello kernel space,
> > > > 
> > > > user space webmonkey here, tinkering on my home desktop setup and spelunking way too many rabbit holes... damn you source code! ;P
> > > > 
> > > > The other day i got frustrated with re-implementing parsing and calculation of /proc/stat and /proc/meminfo for my visual user interface experiments (in various languages), so i "created"* a kernel module that helps me with that (https://github.com/oskude/proc_topstat).
> > > > 
> > > > But i could not get swap info in the kernel module, cause - i guess - it's not exported?
> > > > So i wonder, would/could upstream EXPORT_SYMBOL(si_swapinfo)?
> > > > If yes/maybe, where/how should i propose such change?
> > >
> > > Symbols and functions are only exported if an in-kernel-tree module
> > > needs it.  Sorry.
> > >
> > > What's wrong with parsing the existing exports of this value as-is?
> > 
> > do you mean i can get swap usage info in a kernel module without si_swapinfo?
> 
> Not that I know of, no.
> 
> > (as seen here https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/tree/fs/proc/meminfo.c#n44)
> > 
> > or do you mean what is wrong with /proc/meminfo?
> 
> Yes, what is wrong with that?

I guess there's nothing "wrong" with /proc/meminfo (or /proc/stat), i just want to display their - for me/user relevant - data every 250ms and parsing them felt so sub-optimal... So i just removed what i didn't need and pre-calculated what i could, in the kernel module. So i would not need to worry about it again (except, yeah, when upstream changes).

Also, as i dabble in various programming languages (in userspace), i would need to re-implement the parsing and calculation (that matters to me/user) of /proc/stat and /proc/meminfo every time...
(but i guess that is userspace work? translate kernelspace data to userspace data? does "user" even mean "human"?;P)

But whatever, i now like this more:

$> cat /proc/topstat 
cpu 2085949 109734
cpu 2086513 108671
cpu 2086203 110817
cpu 2086167 127600
mem 8026148 2993832 1033648

cpu <total> <used>
- line number implies cpu number.
- <total> jiffies elapsed.
- <used> amount of <total> in use by programs, that cannot be used by other programs.

mem <total> <used> <cached>
- <total> kilobytes of memory.
- <used> amount of <total> in use by programs, that can not be used by other programs.
- <cached> amount of <used> in use for caches, that can be freed by user.

For now, i can live without swap ui, i avoid swap anyway :D

Cheers
Andre Schmidt

_______________________________________________
Kernelnewbies mailing list
Kernelnewbies@kernelnewbies.org
https://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies

      reply	other threads:[~2019-11-02 13:57 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-11-02 11:51 Simple system stats Andre Schmidt
2019-11-02 12:07 ` Greg KH
2019-11-02 13:05   ` Andre Schmidt
2019-11-02 13:13     ` Greg KH
2019-11-02 13:56       ` Andre Schmidt [this message]

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=20191102145636.60fa2891@ALDOS \
    --to=andre@osku.de \
    --cc=greg@kroah.com \
    --cc=kernelnewbies@kernelnewbies.org \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.