[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: [libvirt] [PATCH] virsh: fix memtune's help message for swap_hard_limit



于 2011年03月03日 14:10, KAMEZAWA Hiroyuki 写道:
On Thu, 03 Mar 2011 14:02:32 +0800
Osier Yang<jyang redhat com>  wrote:

于 2011年03月03日 13:25, KAMEZAWA Hiroyuki 写道:
I found this when I used virsh memtune...

Happened to see a bug for this yesterday.

Oh ;)

BTW, how to fix http://libvirt.org/formatdomain.html 's memtune description ?

How about be consitent with what cgroup doc says? :-)

http://git.kernel.org/?p=linux/kernel/git/next/linux-next.git;a=blob;f=Documentation/cgroups/memory.txt;h=b6ed61c958560d987e40cc28a9eca7f7d12401a1;hb=HEAD


  memory.memsw.usage_in_bytes     # show current memory+Swap usage
  memory.limit_in_bytes           # set/show limit of memory usage
  memory.memsw.limit_in_bytes     # set/show limit of memory+Swap usage
  memory.failcnt                  # show the number of memory usage hits limits

Maybe nice.



==
 From 541ae04430f376e8168b413a20b35dce49779816 Mon Sep 17 00:00:00 2001
From: KAMEZAWA Hiroyuki<kamezawa hiroyu jp fujitsu com>
Date: Thu, 3 Mar 2011 14:24:45 +0900
Subject: [PATCH 6/6] fix virsh commands' message for memtune'swap_hard_limit.

cgroup's /cgroup/memory/memory.memsw.limit_in_bytes is not
for limitinit 'swap' but for 'memory+swap' (then, it's memsw)
(So, this number cannot be smaller than memory.limit_in_bytes)

Yes, that's what BZ about.


Note:
If other hypervisors than Linux support this and meaning is
not same as memory+swap, the name swap_hard_limit will have confusion.

Currently, only LXC and QEMU driver support "swap_hard_limit"
using cgroup, ESX just support setting "min_guarantee", so
perhaps renaming "swap_hard_limit" to something like
"memswap_hard_limit" is a good idea?


Yes, I think it's better. Should I prepare patches ? or you'll do ?

Let's see other guys's opinions before doing it, :)

Regards
Osier


[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]