From: Christoph Lameter <clameter@sgi.com>
To: Andrew Morton <akpm@linux-foundation.org>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>,
linux-mm <linux-mm@kvack.org>,
mel@skynet.ie, y-goto@jp.fujitsu.com,
Kamezawa Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>,
Eric Whitney <eric.whitney@hp.com>
Subject: Re: [PATCH/RFC] Add node 'states' sysfs class attribute - V2
Date: Mon, 27 Aug 2007 22:53:15 -0700 (PDT) [thread overview]
Message-ID: <Pine.LNX.4.64.0708272235580.9834@schroedinger.engr.sgi.com> (raw)
In-Reply-To: <20070827222912.8b364352.akpm@linux-foundation.org>
On Mon, 27 Aug 2007, Andrew Morton wrote:
> Your claim here is, I believe, that a human user interface should be
> implemented in the kernel because the cost (to you) (short-term) of doing
> that is lower that the cost of implementing a simpler kernel interface and
> a bit of userspace human presentation code. Even though the long-term
> cost to the kernel maintainers is higher, and the resulting output is
> harder for programs to parse.
The long term cost is zero since there is already a kernel function
to process these lists. See bitmap_parselist(). The kernel already allows
output and input of these lists.
> Please type "cat /proc/stat". The world hasn't ended.
Yea that the prime example of a bad use of the proc filesystem. All these
numbers better be split up into individual files.
The cpu affinity is a horror to see on 4096 cpu systems. If you
want to figure out to which cpu the process has restricted itself then you
need to do some quick hex conversions in your mind.
margin:/proc/1 # cat /proc/1/stat
1 (init) S 0 1 1 0 -1 4194560 77 340052 10 1575 0 463 1256 1246 20 0 1 0
67 1802240 47 18446744073709551615 4611686018427387904 4611686018428481976
6953557824659209808 16140902370223191856 11529215046068536865 0 0
1467013372 680207875 11529215050365063520 0 0 0 2 0 0 0
> > Well I keep ending up cat this and that proc entry for debugging and its
> > difficult to do if one sysfs file spews huge amounts of illegible binary
> > data to you.
>
> Nobody ever said "binary". Please try to keep up.
What you get right now from this patch is a series of hex digits and you
have the task of converting that to a series of 0 and 1's in your mind and
then figure out which node it was that had a 1 there.
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
next prev parent reply other threads:[~2007-08-28 5:53 UTC|newest]
Thread overview: 43+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <200708242228.l7OMS5fU017948@imap1.linux-foundation.org>
2007-08-27 15:58 ` [PATCH] 2.6.23-rc3-mm1 - update N_HIGH_MEMORY node state for memory hotadd Lee Schermerhorn
2007-08-27 17:48 ` [PATCH/RFC] Add node 'states' sysfs class attribute Lee Schermerhorn
2007-08-27 19:11 ` Christoph Lameter
2007-08-27 20:08 ` Lee Schermerhorn
2007-08-27 20:15 ` Christoph Lameter
2007-08-27 21:02 ` [PATCH/RFC] Add node 'states' sysfs class attribute - V2 Lee Schermerhorn
2007-08-27 21:04 ` Christoph Lameter
2007-08-28 0:01 ` Andrew Morton
2007-08-28 0:08 ` Christoph Lameter
2007-08-28 1:14 ` Andrew Morton
2007-08-28 1:29 ` Christoph Lameter
2007-08-28 3:18 ` Andrew Morton
2007-08-28 5:15 ` Christoph Lameter
2007-08-28 5:29 ` Andrew Morton
2007-08-28 5:34 ` Andrew Morton
2007-08-28 5:53 ` Christoph Lameter [this message]
2007-08-28 6:12 ` Andrew Morton
2007-08-28 14:05 ` Lee Schermerhorn
2007-08-28 22:02 ` Christoph Lameter
2007-08-28 22:13 ` Nish Aravamudan
2007-08-29 14:43 ` Lee Schermerhorn
2007-08-29 17:39 ` Christoph Lameter
2007-08-29 21:31 ` [PATCH/RFC] Add node states sysfs class attributeS - V3 Lee Schermerhorn
2007-08-29 22:14 ` Christoph Lameter
2007-08-30 13:34 ` Lee Schermerhorn
2007-08-29 22:36 ` Nish Aravamudan
2007-08-30 15:19 ` [PATCH/RFC] Add node states sysfs class attributeS - V4 Lee Schermerhorn
2007-08-30 16:44 ` Nish Aravamudan
2007-08-30 18:20 ` Christoph Lameter
2007-08-30 18:19 ` Christoph Lameter
2007-08-30 18:41 ` Lee Schermerhorn
2007-09-11 13:56 ` [PATCH/RFC] Add node states sysfs class attributeS - V5 Lee Schermerhorn
2007-09-11 20:25 ` Christoph Lameter
2007-09-14 10:50 ` Andrew Morton
2007-09-14 11:35 ` Andy Whitcroft
2007-09-14 14:34 ` Lee Schermerhorn
2007-09-14 14:43 ` Mel Gorman
2007-09-14 15:00 ` Paul Mundt
2007-09-16 12:10 ` Mel Gorman
2007-09-14 16:00 ` Martin J. Bligh
2007-08-28 19:34 ` [PATCH/RFC] Add node 'states' sysfs class attribute - V2 Christoph Lameter
2007-08-28 1:16 ` Yasunori Goto
2007-08-28 1:21 ` Yasunori Goto
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=Pine.LNX.4.64.0708272235580.9834@schroedinger.engr.sgi.com \
--to=clameter@sgi.com \
--cc=Lee.Schermerhorn@hp.com \
--cc=akpm@linux-foundation.org \
--cc=eric.whitney@hp.com \
--cc=kamezawa.hiroyu@jp.fujitsu.com \
--cc=linux-mm@kvack.org \
--cc=mel@skynet.ie \
--cc=y-goto@jp.fujitsu.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