lkml.org 
[lkml]   [2003]   [Jul]   [28]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
SubjectRe: Need a meaningful memory measure out of /proc/meminfo
Date
On 28 July 2003 11:51, Denis Vlasenko wrote:
> Ok, so I use Used-Buffers-Cached as a memory measure.
> It remains almost unchanged across oom.
>
> Although I am worried that I can lie to user because above test
> shows that some 10M of buffers+cached are not really freeable.
> Also I run this box swapless, what to do with swap numbers?

FYI: this is how oom looks like under nanometer (with swap reporting added):
# nanometer d500 c m s ieth0 ilo
cpu [..........] mem 85M/247M swp 428k eth0 4.2k 3.0k lo 0 0
cpu [..........] mem 85M/247M swp 428k eth0 2.0k 1.7k lo 0 0
cpu [..........] mem 85M/247M swp 428k eth0 5.3k 5.6k lo 0 0
cpu [SSSSU.....] mem 119M/247M swp 428k eth0 3.1k 2.8k lo 0 0
cpu [SSSSSSSSSS] mem 157M/247M swp 4.0M eth0 5.0k 3.8k lo 0 0
cpu [SSSSSU....] mem 171M/247M swp 5.0M eth0 8.0k 4.0k lo 140 140
cpu [SSSSSSSSS.] mem 205M/247M swp 10M eth0 5.1k 3.3k lo 0 0
cpu [SSSS......] mem 205M/247M swp 20M eth0 76k 59k lo 0 0
cpu [SSSS......] mem 211M/247M swp 25M eth0 47k 5.7k lo 0 0
cpu [SSSSSS....] mem 221M/247M swp 32M eth0 391k 9.5k lo 0 0
cpu [SSS.......] mem 220M/247M swp 39M eth0 31k 4.2k lo 0 0
cpu [SSSS......] mem 220M/247M swp 51M eth0 7.9k 840 lo 0 0
cpu [SSSS......] mem 222M/247M swp 58M eth0 92k 7.8k lo 0 0
cpu [SS........] mem 222M/247M swp 60M eth0 85k 5.6k lo 0 0
cpu [SSSS......] mem 225M/247M swp 65M eth0 76k 4.9k lo 140 140
cpu [SSSS......] mem 227M/247M swp 72M eth0 54k 4.5k lo 0 0
cpu [SSSSSS....] mem 229M/247M swp 74M eth0 4.5k 3.2k lo 0 0
cpu [SSSSSSSSSS] mem 232M/247M swp 74M eth0 45k 35k lo 0 0
cpu [SSSSSSSSSS] mem 231M/247M swp 74M eth0 107k 29k lo 0 0
cpu [SSSSSSSSSS] mem 232M/247M swp 74M eth0 171k 6.2k lo 0 0
cpu [SSSSSSSSSS] mem 232M/247M swp 74M eth0 2.3k 1.8k lo 0 0
cpu [SSSSSSSSSS] mem 232M/247M swp 74M eth0 5.7k 5.4k lo 0 0
cpu [SSSSSSSSSS] mem 232M/247M swp 74M eth0 40k 2.9k lo 0 0
cpu [SSSSSSSSSS] mem 236M/247M swp 74M eth0 342k 9.6k lo 0 0
cpu [SSSSSSSSSS] mem 239M/247M swp 74M eth0 3.0k 2.5k lo 0 0
cpu [SSSSSSSSSS] mem 240M/247M swp 74M eth0 6.2k 4.4k lo 0 0
cpu [SSSSSSSSSS] mem 240M/247M swp 74M eth0 313k 26k lo 0 0
cpu [SSSSSSSSSS] mem 240M/247M swp 74M eth0 124k 35k lo 0 0
cpu [SSSSSSSSSS] mem 241M/247M swp 74M eth0 21k 16k lo 0 0
cpu [SSSSSS....] mem 28M/247M swp 60M eth0 380k 15k lo 0 0
cpu [..........] mem 27M/247M swp 61M eth0 10k 5.7k lo 0 0
cpu [..........] mem 27M/247M swp 61M eth0 2.8k 2.1k lo 0 0

# cat /proc/swaps
Filename Type Size Used Priority
/mnt/auto/vfat.hda3/PAGEFILE file 76792 54400 -1

At oom, mem reporting was showing ~6 meg as still free...
I want to do better.
--
vda
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2005-03-22 13:47    [W:0.022 / U:4.628 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site