From mboxrd@z Thu Jan 1 00:00:00 1970 From: Sha Zhengju Subject: Re: [PATCH 7/7] memcg: print more detailed info while memcg oom happening Date: Wed, 04 Jul 2012 19:20:49 +0800 Message-ID: <4FF42711.50303@gmail.com> References: <1340880885-5427-1-git-send-email-handai.szj@taobao.com> <1340881609-5935-1-git-send-email-handai.szj@taobao.com> <4FF3FED6.9010700@jp.fujitsu.com> Mime-Version: 1.0 Content-Transfer-Encoding: 7bit Return-path: DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=message-id:date:from:user-agent:mime-version:to:cc:subject :references:in-reply-to:content-type:content-transfer-encoding; bh=2jrv1AFmcozevJSzxQuR90AC630qvzUpTd4ezO1FXKM=; b=TXNQK/qChyDv1FlmmLwC2QqA1nK8LbnB1k3cFz6FjxOzuxpahtfxBQcAuLWsNdVvgF em73hes9HNLFQnuCPJvx1gfvJmfgAz9+EUT42RQA/m1VXnsgJ3C1Z/6xXXJ7XubYLXC0 FgGVyAuuuDGTotbXp1q4LZAidkBk+cCXKmyytooBS4X2Hzd/J6zsWEwWJLloCIU/4hOk pnpDsOECgP3H0bUyrjSyW7XZUhkogtmh7IiWbwaCtALoxCjlh6qZAIAj6hZZNG/eqa2x tZCgj3IKMyNT2C/zCKtbs5vkZaU0UCdxAqgTk8An1o7aueuQnsgwRRWEDH8V2MCMQkGD Djdw== In-Reply-To: <4FF3FED6.9010700-+CUm20s59erQFUHtdCDX3A@public.gmane.org> Sender: cgroups-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org List-ID: Content-Type: text/plain; charset="us-ascii" To: Kamezawa Hiroyuki Cc: linux-mm-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org, cgroups-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, gthelen-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org, yinghan-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org, akpm-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org, mhocko-AlSwsSmVLrQ@public.gmane.org, linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, Sha Zhengju On 07/04/2012 04:29 PM, Kamezawa Hiroyuki wrote: > (2012/06/28 20:06), Sha Zhengju wrote: >> From: Sha Zhengju >> >> While memcg oom happening, the dump info is limited, so add this >> to provide memcg page stat. >> >> Signed-off-by: Sha Zhengju > Could you split this into a different series ? > seems good to me in general but...one concern is hierarchy handling. > > IIUC, the passed 'memcg' is the root of hierarchy which gets OOM. > So, the LRU info, which is local to the root memcg, may not contain any good > information. I think you should visit all memcg under the tree. > Yes, you're right! I did not handle hierarchy here, and just now I make a test case to prove this. I'll split it to another series later. Thanks for reviewing! Thanks, Sha