From: Fengguang Wu <wfg@mail.ustc.edu.cn>
To: Ray Lee <ray-lk@madrabbit.org>
Cc: Andrew Morton <akpm@osdl.org>, Matt Mackall <mpm@selenic.com>,
John Berthels <jjberthels@gmail.com>,
linux-kernel@vger.kernel.org
Subject: Re: [PATCH 0/4] process memory footprint info in proc/<pid>/[s|p]maps v2
Date: Mon, 20 Aug 2007 15:55:15 +0800 [thread overview]
Message-ID: <387596515.30798@ustc.edu.cn> (raw)
Message-ID: <20070820075515.GA10476@mail.ustc.edu.cn> (raw)
In-Reply-To: <2c0942db0708192015u2759ccfby320549cd6393dff3@mail.gmail.com>
On Sun, Aug 19, 2007 at 08:15:01PM -0700, Ray Lee wrote:
> On 8/19/07, Fengguang Wu <wfg@mail.ustc.edu.cn> wrote:
> > Inspired by Matt Mackall's pagemap patches and ideas, I worked up these
> > textual interfaces that achieve the same goals. The patches run OK
> > under different sized reads.
> [...]
> > 2b7d6e8f3000-2b7d6ea40000 r-xp 00000000 08:01 1564031 /lib/libc-2.6.so
> > 0 2 YRAU___ 81
> > 2 1 YRAU___ 80
> > 3 1 YRAU___ 81
> > 4 1 YRAU___ 72
> > 5 1 YRAU___ 77
> > 6 1 YRAU___ 79
> > 7 1 YRAU___ 73
> > 8 1 YRAU___ 79
> > 9 1 YRAU___ 78
> > a 1 YRAU___ 77
> > b 1 YRAU___ 72
> > c 1 YRAU___ 75
> > d 1 YRAU___ 81
> > e 1 YRAU___ 72
> > f 1 YRAU___ 78
> > 10 1 YRAU___ 81
> > 11 1 YRAU___ 78
> > 12 1 YRAU___ 80
> > 13 1 YRAU___ 78
> > 14 2 YRAU___ 81
> > 16 1 YRAU___ 49
> > 17 6 YRAU___ 41
> > 1d 1 YRAU___ 80
> > 2a 1 YRAU___ 44
> > 2b 1 YRAU___ 69
> > 2c 1 YRAU___ 28
> > 31 1 YRAU___ 79
> > 32 1 YRAU___ 49
> > 33 1 YRAU___ 73
> > 34 1 YRAU___ 57
> > 35 1 YRAU___ 67
> > 42 1 YRAU___ 77
> > 43 3 YRAU___ 78
> > 46 1 YRAU___ 77
> > 47 1 YRAU___ 70
> > 48 1 YRAU___ 15
> > 4d 1 YRAU___ 78
> > 5f 1 YRAU___ 78
> > 60 1 YRAU___ 75
> > 61 1 YRAU___ 72
> > 62 1 YRAU___ 67
> > 63 1 YRAU___ 59
> > 68 1 YRAU___ 63
> > 69 1 YRAU___ 61
> > 6a 1 YRAU___ 63
> > 6b 1 YRAU___ 75
> > 6c 1 YRAU___ 76
> > 6d 1 YRAU___ 81
> > 6e 1 YRAU___ 77
> > 6f 1 YRAU___ 79
> > 70 1 YRAU___ 44
> > 71 1 YRAU___ 79
> > 72 1 YRAU___ 78
> > 73 1 YRAU___ 79
> > 74 1 YRAU___ 46
> > 75 1 YRAU___ 78
> > 76 1 YRAU___ 60
> > 78 1 YRAU___ 79
> > 79 1 YRAU___ 81
> > 7a 1 YRAU___ 80
> > 7e 1 YRAU___ 41
> > 8a 1 YRAU___ 36
> > 8b 1 YRAU___ 72
> > 8c 1 YRAU___ 28
> > 8d 1 YRAU___ 21
> > 91 2 YRAU___ 19
> > 99 1 YRAU___ 62
> > 9a 1 YRAU___ 76
> > 9b 1 YRAU___ 72
> > c4 1 YRAU___ 80
> > c5 1 YRAU___ 82
> > c6 1 YRAU___ 81
> > cb 1 YRAU___ 42
> > cc 1 YRAU___ 79
> > cd 1 YRAU___ 26
> > cf 1 YRAU___ 12
> > d0 1 YRAU___ 77
> > d1 1 YRAU___ 38
> > d2 1 YRAU___ 45
> > d3 1 YRAU___ 66
> > d4 1 YRAU___ 71
> > e0 1 YRAU___ 52
> > 106 1 YRAU___ 33
> > 107 1 YRAU___ 14
> > 108 1 YRAU___ 63
> > 109 1 YRAU___ 57
> > 10b 1 YRAU___ 70
> > 10c 1 YRAU___ 63
> > 118 1 YRAU___ 65
> > 119 1 YRAU___ 78
> > 11a 1 YRAU___ 69
> > 11b 1 YRAU___ 68
> > 11e 2 YRAU___ 62
> > 120 1 YRAU___ 74
> > 121 1 YRAU___ 42
> > 125 1 YRAU___ 17
>
> Eh, I'd think that pivoting the data set would be a much more natural
> (& therefore shorter) representation.
>
> YRAU___ 0,2:81 2:80 3:81 4:72 5:77 6:79 7:73 8:79 9:78
> ...versus...
> > 0 2 YRAU___ 81
> > 2 1 YRAU___ 80
> > 3 1 YRAU___ 81
> > 4 1 YRAU___ 72
> > 5 1 YRAU___ 77
> > 6 1 YRAU___ 79
> > 7 1 YRAU___ 73
> > 8 1 YRAU___ 79
> > 9 1 YRAU___ 78
>
> So, flags followed by a list of offset[,length]:usage. If the flags
> change, start a new line. If you get a line that's too long, start a
> new line.
>
> No?
It's an interesting suggestion: the compression ratio could be doubled
on average. Thank you!
However I still appreciate the 'beauty' of tabbed output ;-)
I'm working on an interface with similar output(see filecache below). It is
less likely to be improved by the new format - and I want the two interfaces to
be consistent.
As for the format, I'm not quite sure of two things:
- should hex values be used for offset/len?
Hex values are less user friendly but more consistent with the format of
maps. For example, the 001bc000/1bc, 007d1000/7d1 below:
007bc000-007d1000 rw-p 001bc000 08:01 199284 /usr/bin/Xorg
^^^^^
1bc 1 YR_U___ 1
^^^
1bd 5 Y_A_P__ 1
1c3 3 Y_A_P__ 1
1c6 5 YR_U___ 1
1cb 1 Y_A_P__ 1
1cc 1 YR_U___ 1
1cd 4 Y_A_P__ 1
007d1000-0108a000 rw-p 007d1000 00:00 0 [heap]
^^^^^
7d1 4 Y_A_P__ 1
^^^
7d6 3 Y_A_P__ 1
7da 5 Y_A_P__ 1
7e0 41 Y_A_P__ 1
- should the page size be printed as a file header(offset/len are in page
units)? e.g.
# page size 4096
...
But it is also available via the getpagesize() call.
I'd appreciate it a lot if you or other people could give advices on them.
Thank you,
Fengguang
---
filecache: show cached files and their cached pages
===================================================
bash-3.2# cat /proc/filecache
# filecache 1.0
# ino cached cached% refcnt state dev file
0 568 0 0 -- 00:02(bdev) (03:00)
176359 1164 95 2 d- 03:00(hda) /lib/libc-2.3.6.so
176355 92 100 2 d- 03:00(hda) /lib/ld-2.3.6.so
96199 12 100 0 d- 03:00(hda) /bin/cat
32286 4 100 0 -- 03:00(hda) /etc/mtab
96194 40 100 0 d- 03:00(hda) /bin/mount
529061 4 100 0 -- 03:00(hda) /usr/share/terminfo/l/linux
48099 4 100 0 -- 03:00(hda) /.bash_history
176633 40 100 1 -- 03:00(hda) /lib/libnss_files-2.3.6.so
176635 36 100 1 -- 03:00(hda) /lib/libnss_nis-2.3.6.so
176630 76 100 1 -- 03:00(hda) /lib/libnsl-2.3.6.so
176631 36 100 1 -- 03:00(hda) /lib/libnss_compat-2.3.6.so
176362 12 100 1 -- 03:00(hda) /lib/libdl-2.3.6.so
176690 264 100 1 -- 03:00(hda) /lib/libncurses.so.5.5
96202 244 100 0 -- 03:00(hda) /bin/bash
73 568 100 1 -- 00:01(rootfs) /dev/root
bash-3.2# echo /lib/libc-2.3.6.so > /proc/filecache
bash-3.2# cat /proc/filecache
# file /lib/libc-2.3.6.so
# flags R:referenced A:active U:uptodate D:dirty W:writeback M:mmap
# idx len state refcnt
0 16 RAU__M 3
16 3 __U___ 1
19 1 R_U__M 2
1a 5 __U___ 1
1f 1 RAU__M 3
20 1 __U___ 1
21 1 R_U__M 2
22 2 RAU__M 3
24 1 R_U__M 2
25 1 __U___ 1
26 1 RAU__M 3
27 2 __U___ 1
29 2 RAU__M 2
2b 2 RAU__M 3
2d 2 R_U__M 2
2f 9 __U___ 1
38 1 R_U__M 2
39 1 __U___ 1
3a 5 RAU__M 2
3f 1 R_U__M 2
40 4 __U___ 1
44 1 RAU__M 2
45 3 __U___ 1
48 6 R_U__M 2
4e 4 __U___ 1
52 1 R_U__M 2
53 2 RAU__M 2
55 1 R_U__M 2
56 2 RAU__M 2
58 1 R_U__M 2
59 2 __U___ 1
5b 1 RAU__M 2
5c 1 R_U__M 2
5d 5 RAU__M 2
62 1 RAU__M 3
63 2 RAU__M 2
65 1 R_U___ 1
66 3 RAU__M 2
69 1 __U___ 1
6a 3 RAU__M 3
6d 1 R_U__M 2
6e 3 __U___ 1
71 1 R_U__M 2
72 c __U___ 1
7e 1 R_U__M 2
7f 9 __U___ 1
88 2 R_U__M 2
8a 1 __U___ 1
8b 1 R_U__M 2
8c 2 RAU__M 2
8e 1 R_U__M 2
8f 9 __U___ 1
9f f __U___ 1
ae 2 RAU__M 2
b0 9 __U___ 1
b9 2 RAU__M 3
bb 1 RAU__M 2
bc 4 __U___ 1
c0 2 RAU__M 3
c2 1 R_U__M 2
c3 2 __U___ 1
c5 1 RAU__M 2
c6 2 R_U__M 2
c8 1 RAU__M 3
c9 1 RAU__M 2
ca c __U___ 1
d6 1 RAU__M 3
d7 3 __U___ 1
da 3 R_U__M 2
dd f __U___ 1
f3 5 __U___ 1
f8 4 R_U__M 2
fc 1 __U___ 1
fd 2 R_U__M 2
ff 1 RAU__M 3
100 1 R_U__M 2
101 2 __U___ 1
103 1 RAU__M 3
104 5 __U___ 1
109 1 RAU__M 3
10a a __U___ 1
114 1 RAU__M 2
115 1 RAU__M 3
116 1 __U___ 1
117 1 RAU__M 2
118 2 R_U__M 2
11a 2 __U___ 1
11c 1 RAU__M 3
11d 1 RAU__M 2
11e 1 RAU__M 3
11f 2 R_U__M 2
121 1 __U___ 1
122 1 R_U__M 2
123 4 __U___ 1
127 4 RAU___ 1
12b 6 __U___ 1
next prev parent reply other threads:[~2007-08-20 7:55 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-08-19 7:54 [PATCH 0/4] process memory footprint info in proc/<pid>/[s|p]maps v2 Fengguang Wu
2007-08-19 7:54 ` Fengguang Wu
2007-08-20 3:15 ` Ray Lee
2007-08-20 7:55 ` Fengguang Wu [this message]
2007-08-20 7:55 ` Fengguang Wu
2007-08-19 7:54 ` [PATCH 1/4] maps: PSS(proportional set size) accounting in smaps Fengguang Wu
2007-08-19 7:54 ` Fengguang Wu
2007-08-19 7:54 ` [PATCH 2/4] maps: address based vma walking Fengguang Wu
2007-08-19 7:54 ` Fengguang Wu
2007-08-19 7:54 ` [PATCH 3/4] maps: introduce generic_maps_open() Fengguang Wu
2007-08-19 7:54 ` Fengguang Wu
2007-08-19 7:54 ` [PATCH 4/4] maps: /proc/<pid>/pmaps interface - memory maps in granularity of pages Fengguang Wu
2007-08-19 7:54 ` Fengguang Wu
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=387596515.30798@ustc.edu.cn \
--to=wfg@mail.ustc.edu.cn \
--cc=akpm@osdl.org \
--cc=jjberthels@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mpm@selenic.com \
--cc=ray-lk@madrabbit.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.