Re: [PATCH 8/8] Per-container pages reclamation

From: Pavel Emelianov <xemul@openvz.org>
To: balbir@linux.vnet.ibm.com,
	Vaidyanathan Srinivasan <svaidy@linux.vnet.ibm.com>
Cc: Andrew Morton <akpm@linux-foundation.org>,
	Paul Menage <menage@google.com>, Balbir Singh <balbir@in.ibm.com>,
	Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
	devel@openvz.org, Kirill Korotaev <dev@openvz.org>
Subject: Re: [PATCH 8/8] Per-container pages reclamation
Date: Fri, 01 Jun 2007 13:27:24 +0400	[thread overview]
Message-ID: <465FE67C.2040206@openvz.org> (raw)
In-Reply-To: <465E93D7.2020506@linux.vnet.ibm.com>

Balbir Singh wrote:

[snip]

>>>> +	for_each_online_node(node) {
>>>> +#ifdef CONFIG_HIGHMEM
>>>> +		zones = NODE_DATA(node)->node_zonelists[ZONE_HIGHMEM].zones;
>>>> +		if (do_try_to_free_pages(zones, sc.gfp_mask, &sc))
>>>> +			return 1;
>>>> +#endif
>>>> +		zones = NODE_DATA(node)->node_zonelists[ZONE_NORMAL].zones;
>>>> +		if (do_try_to_free_pages(zones, sc.gfp_mask, &sc))
>>>> +			return 1;
>>>> +	}
>>> Definitely need to handle ZONE_DMA32 and ZONE_DMA (some architectures put
>>> all memory into ZONE_DMA (or they used to))
> 
> Won't the node_zonelists[ZONE_NORMAL].zones contain all lower zones (including
> ZONE_DMA and ZONE_DMA32) from various nodes organized by cost?
> 
> I need to go lookup the node_zonelists creation.

OK, I've got it. The node's zonelist indexed with ZONE_XXX contains
the zone specified and all the zones beyond it. Thus shrinking from
ZONE_HIGHMEM (ZONE_NORMAL if the latter does not exist) is enough.

Thanks,
Pavel