makedumpfile memory usage grows with system memory size
d.hatayama at jp.fujitsu.com
Wed May 16 20:21:06 EDT 2012
From: Atsushi Kumagai <kumagai-atsushi at mxc.nes.nec.co.jp>
Subject: Re: makedumpfile memory usage grows with system memory size
Date: Wed, 16 May 2012 17:02:30 +0900
> On Mon, 14 May 2012 14:44:28 +0900 (JST)
> HATAYAMA Daisuke <d.hatayama at jp.fujitsu.com> wrote:
>> Anyway, I think it better to add _mapcount information to VMCOREINFO
>> on upstream as soon as possible.
> I think it's better way to use _mapcount.
> But we don't certainly decide to use _mapcount and even if we decide to use it,
> we still have problems to use it.
> For example, the upstream kernel(v3.4-rc7) has _mapcount in union, we need
> a information to judge whether the found data is _mapcount or not.
> So, more investigation is needed and I think it's too early to send the request
> to upstream kernel.
A quick look into the other part of the union---inuse, objects,
froze---to which _mapcount belongs, they appears to be used in SLUB
allocator. The page with PG_slab appears to use the other part to
_mapcount. This means that to decide whether to use _mapcount, it's
necessary to first investigate how SLUB allocator works.
> I plan to finish working to reduce memory consumption by the end of June,
> and I will continue to discuss performance issues.
> Therefore, the request will be delayed until July or August.
I'll wait for your patch on memory consumption and feedback to
it. I'll also look into possibility of filtering free memory in
constant space on recent kernels.
More information about the kexec