From: Lin Feng <linfeng@cn.fujitsu.com>
To: Simon Jeons <simon.jeons@gmail.com>
Cc: Jianguo Wu <wujianguo@huawei.com>,
Tang Chen <tangchen@cn.fujitsu.com>,
jiang.liu@huawei.com, hpa@zytor.com, akpm@linux-foundation.org,
wency@cn.fujitsu.com, laijs@cn.fujitsu.com, yinghai@kernel.org,
isimatu.yasuaki@jp.fujitsu.com, rob@landley.net,
kosaki.motohiro@jp.fujitsu.com, minchan.kim@gmail.com,
mgorman@suse.de, rientjes@google.com, rusty@rustcorp.com.au,
lliubbo@gmail.com, jaegeuk.hanse@gmail.com, tony.luck@intel.com,
glommer@parallels.com, linux-kernel@vger.kernel.org,
linux-mm@kvack.org, linux-doc@vger.kernel.org
Subject: Re: [PATCH v3 3/5] page_alloc: Introduce zone_movable_limit[] to keep movable limit for nodes
Date: Wed, 12 Dec 2012 09:58:49 +0800 [thread overview]
Message-ID: <50C7E4D9.30207@cn.fujitsu.com> (raw)
In-Reply-To: <1355232032.1459.2.camel@kernel.cn.ibm.com>
On 12/11/2012 09:20 PM, Simon Jeons wrote:
> On Tue, 2012-12-11 at 20:41 +0800, Jianguo Wu wrote:
>> On 2012/12/11 20:24, Simon Jeons wrote:
>>
>>> On Tue, 2012-12-11 at 11:07 +0800, Jianguo Wu wrote:
>>>> On 2012/12/11 10:33, Tang Chen wrote:
>>>>
>>>>> This patch introduces a new array zone_movable_limit[] to store the
>>>>> ZONE_MOVABLE limit from movablecore_map boot option for all nodes.
>>>>> The function sanitize_zone_movable_limit() will find out to which
>>>>> node the ranges in movable_map.map[] belongs, and calculates the
>>>>> low boundary of ZONE_MOVABLE for each node.
>>>>>
>>>>> Signed-off-by: Tang Chen <tangchen@cn.fujitsu.com>
>>>>> Signed-off-by: Jiang Liu <jiang.liu@huawei.com>
>>>>> Reviewed-by: Wen Congyang <wency@cn.fujitsu.com>
>>>>> Reviewed-by: Lai Jiangshan <laijs@cn.fujitsu.com>
>>>>> Tested-by: Lin Feng <linfeng@cn.fujitsu.com>
>>>>> ---
>>>>> mm/page_alloc.c | 77 +++++++++++++++++++++++++++++++++++++++++++++++++++++++
>>>>> 1 files changed, 77 insertions(+), 0 deletions(-)
>>>>>
>>>>> diff --git a/mm/page_alloc.c b/mm/page_alloc.c
>>>>> index 1c91d16..4853619 100644
>>>>> --- a/mm/page_alloc.c
>>>>> +++ b/mm/page_alloc.c
>>>>> @@ -206,6 +206,7 @@ static unsigned long __meminitdata arch_zone_highest_possible_pfn[MAX_NR_ZONES];
>>>>> static unsigned long __initdata required_kernelcore;
>>>>> static unsigned long __initdata required_movablecore;
>>>>> static unsigned long __meminitdata zone_movable_pfn[MAX_NUMNODES];
>>>>> +static unsigned long __meminitdata zone_movable_limit[MAX_NUMNODES];
>>>>>
>>>>> /* movable_zone is the "real" zone pages in ZONE_MOVABLE are taken from */
>>>>> int movable_zone;
>>>>> @@ -4340,6 +4341,77 @@ static unsigned long __meminit zone_absent_pages_in_node(int nid,
>>>>> return __absent_pages_in_range(nid, zone_start_pfn, zone_end_pfn);
>>>>> }
>>>>>
>>>>> +/**
>>>>> + * sanitize_zone_movable_limit - Sanitize the zone_movable_limit array.
>>>>> + *
>>>>> + * zone_movable_limit is initialized as 0. This function will try to get
>>>>> + * the first ZONE_MOVABLE pfn of each node from movablecore_map, and
>>>>> + * assigne them to zone_movable_limit.
>>>>> + * zone_movable_limit[nid] == 0 means no limit for the node.
>>>>> + *
>>>>> + * Note: Each range is represented as [start_pfn, end_pfn)
>>>>> + */
>>>>> +static void __meminit sanitize_zone_movable_limit(void)
>>>>> +{
>>>>> + int map_pos = 0, i, nid;
>>>>> + unsigned long start_pfn, end_pfn;
>>>>> +
>>>>> + if (!movablecore_map.nr_map)
>>>>> + return;
>>>>> +
>>>>> + /* Iterate all ranges from minimum to maximum */
>>>>> + for_each_mem_pfn_range(i, MAX_NUMNODES, &start_pfn, &end_pfn, &nid) {
>>>>> + /*
>>>>> + * If we have found lowest pfn of ZONE_MOVABLE of the node
>>>>> + * specified by user, just go on to check next range.
>>>>> + */
>>>>> + if (zone_movable_limit[nid])
>>>>> + continue;
>>>>> +
>>>>> +#ifdef CONFIG_ZONE_DMA
>>>>> + /* Skip DMA memory. */
>>>>> + if (start_pfn < arch_zone_highest_possible_pfn[ZONE_DMA])
>>>>> + start_pfn = arch_zone_highest_possible_pfn[ZONE_DMA];
>>>>> +#endif
>>>>> +
>>>>> +#ifdef CONFIG_ZONE_DMA32
>>>>> + /* Skip DMA32 memory. */
>>>>> + if (start_pfn < arch_zone_highest_possible_pfn[ZONE_DMA32])
>>>>> + start_pfn = arch_zone_highest_possible_pfn[ZONE_DMA32];
>>>>> +#endif
>>>>> +
>>>>> +#ifdef CONFIG_HIGHMEM
>>>>> + /* Skip lowmem if ZONE_MOVABLE is highmem. */
>>>>> + if (zone_movable_is_highmem() &&
>>>>
>>>> Hi Tang,
>>>>
>>>> I think zone_movable_is_highmem() is not work correctly here.
>>>> sanitize_zone_movable_limit
>>>> zone_movable_is_highmem <--using movable_zone here
>>>> find_zone_movable_pfns_for_nodes
>>>> find_usable_zone_for_movable <--movable_zone is specified here
>>>>
>>>
>>> Hi Jiangguo and Chen,
>>>
>>> - What's the meaning of zone_movable_is_highmem(), does it mean all zone
>>> highmem pages are zone movable pages or ....
>>
>> Hi Simon,
>>
>> zone_movable_is_highmem() means whether zone pages in ZONE_MOVABLE are taken from
>> highmem.
>>
>>> - dmesg
>>>
>>>> 0.000000] Zone ranges:
>>>> [ 0.000000] DMA [mem 0x00010000-0x00ffffff]
>>>> [ 0.000000] Normal [mem 0x01000000-0x373fdfff]
>>>> [ 0.000000] HighMem [mem 0x373fe000-0xb6cfffff]
>>>> [ 0.000000] Movable zone start for each node
>>>> [ 0.000000] Node 0: 0x97800000
>>>
>>> Why the start of zone movable is in the range of zone highmem, if all
>>> the pages of zone movable are from zone highmem? If the answer is yes,
>>
>>> zone movable and zone highmem are in the equal status or not?
>>
>> The pages of zone_movable can be taken from zone_movalbe or zone_normal,
>> if we have highmem, then zone_movable will be taken from zone_highmem,
>> otherwise zone_movable will be taken from zone_normal.
>>
>> you can refer to find_usable_zone_for_movable().
>
> Hi Jiangguo,
>
> I have 8G memory, movablecore=5G, but dmesg looks strange, what
> happended to me?
Hi Simon,
Is there any other boot parameters for memory taken besides 'movablecore=5G'?
thanks,
linfeng
>
>> [ 0.000000] Zone ranges:
>> [ 0.000000] DMA [mem 0x00010000-0x00ffffff]
>> [ 0.000000] Normal [mem 0x01000000-0x373fdfff]
>> [ 0.000000] HighMem [mem 0x373fe000-0xb6cfffff]
>> [ 0.000000] Movable zone start for each node
>> [ 0.000000] Node 0: 0xb7000000
>> [ 0.000000] Early memory node ranges
>> [ 0.000000] node 0: [mem 0x00010000-0x0009cfff]
>> [ 0.000000] node 0: [mem 0x00100000-0x1fffffff]
>> [ 0.000000] node 0: [mem 0x20200000-0x3fffffff]
>> [ 0.000000] node 0: [mem 0x40200000-0xb69cbfff]
>> [ 0.000000] node 0: [mem 0xb6a46000-0xb6a47fff]
>> [ 0.000000] node 0: [mem 0xb6b1c000-0xb6cfffff]
>> [ 0.000000] On node 0 totalpages: 748095
>> [ 0.000000] DMA zone: 32 pages used for memmap
>> [ 0.000000] DMA zone: 0 pages reserved
>> [ 0.000000] DMA zone: 3949 pages, LIFO batch:0
>> [ 0.000000] Normal zone: 1736 pages used for memmap
>> [ 0.000000] Normal zone: 219958 pages, LIFO batch:31
>> [ 0.000000] HighMem zone: 4083 pages used for memmap
>> [ 0.000000] HighMem zone: 517569 pages, LIFO batch:31
>> [ 0.000000] Movable zone: 768 pages, LIFO batch:0
>
>>
>> Thanks,
>> Jianguo Wu
>>
>>>
>>>> I think Jiang Liu's patch works fine for highmem, please refer to:
>>>> http://marc.info/?l=linux-mm&m=135476085816087&w=2
>>>>
>>>> Thanks,
>>>> Jianguo Wu
>>>>
>>>>> + start_pfn < arch_zone_lowest_possible_pfn[ZONE_HIGHMEM])
>>>>> + start_pfn = arch_zone_lowest_possible_pfn[ZONE_HIGHMEM];
>>>>> +#endif
>>>>> +
>>>>> + if (start_pfn >= end_pfn)
>>>>> + continue;
>>>>> +
>>>>> + while (map_pos < movablecore_map.nr_map) {
>>>>> + if (end_pfn <= movablecore_map.map[map_pos].start_pfn)
>>>>> + break;
>>>>> +
>>>>> + if (start_pfn >= movablecore_map.map[map_pos].end_pfn) {
>>>>> + map_pos++;
>>>>> + continue;
>>>>> + }
>>>>> +
>>>>> + /*
>>>>> + * The start_pfn of ZONE_MOVABLE is either the minimum
>>>>> + * pfn specified by movablecore_map, or 0, which means
>>>>> + * the node has no ZONE_MOVABLE.
>>>>> + */
>>>>> + zone_movable_limit[nid] = max(start_pfn,
>>>>> + movablecore_map.map[map_pos].start_pfn);
>>>>> +
>>>>> + break;
>>>>> + }
>>>>> + }
>>>>> +}
>>>>> +
>>>>> #else /* CONFIG_HAVE_MEMBLOCK_NODE_MAP */
>>>>> static inline unsigned long __meminit zone_spanned_pages_in_node(int nid,
>>>>> unsigned long zone_type,
>>>>> @@ -4358,6 +4430,10 @@ static inline unsigned long __meminit zone_absent_pages_in_node(int nid,
>>>>> return zholes_size[zone_type];
>>>>> }
>>>>>
>>>>> +static void __meminit sanitize_zone_movable_limit(void)
>>>>> +{
>>>>> +}
>>>>> +
>>>>> #endif /* CONFIG_HAVE_MEMBLOCK_NODE_MAP */
>>>>>
>>>>> static void __meminit calculate_node_totalpages(struct pglist_data *pgdat,
>>>>> @@ -4923,6 +4999,7 @@ void __init free_area_init_nodes(unsigned long *max_zone_pfn)
>>>>>
>>>>> /* Find the PFNs that ZONE_MOVABLE begins at in each node */
>>>>> memset(zone_movable_pfn, 0, sizeof(zone_movable_pfn));
>>>>> + sanitize_zone_movable_limit();
>>>>> find_zone_movable_pfns_for_nodes();
>>>>>
>>>>> /* Print out the zone ranges */
>>>>
>>>>
>>>>
>>>> --
>>>> To unsubscribe, send a message with 'unsubscribe linux-mm' in
>>>> the body to majordomo@kvack.org. For more info on Linux MM,
>>>> see: http://www.linux-mm.org/ .
>>>> Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
>>>
>>>
>>>
>>> .
>>>
>>
>>
>>
>
>
>
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
next prev parent reply other threads:[~2012-12-12 1:59 UTC|newest]
Thread overview: 28+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-12-11 2:33 [PATCH v3 0/5] Add movablecore_map boot option Tang Chen
2012-12-11 2:33 ` [PATCH v3 1/5] x86: get pg_data_t's memory from other node Tang Chen
2012-12-11 2:33 ` [PATCH v3 2/5] page_alloc: add movable_memmap kernel parameter Tang Chen
2012-12-11 2:33 ` [PATCH v3 3/5] page_alloc: Introduce zone_movable_limit[] to keep movable limit for nodes Tang Chen
2012-12-11 3:07 ` Jianguo Wu
2012-12-11 3:32 ` Tang Chen
2012-12-11 11:28 ` Simon Jeons
2012-12-12 0:49 ` Jiang Liu
2012-12-12 9:09 ` Tang Chen
2012-12-12 9:29 ` Simon Jeons
2012-12-12 10:32 ` Tang Chen
2012-12-13 0:28 ` Simon Jeons
2012-12-13 1:48 ` Tang Chen
2012-12-13 3:09 ` Simon Jeons
2012-12-11 12:24 ` Simon Jeons
2012-12-11 12:41 ` Jianguo Wu
2012-12-11 13:20 ` Simon Jeons
2012-12-12 1:57 ` Jianguo Wu
2012-12-12 2:03 ` Simon Jeons
2012-12-12 1:58 ` Lin Feng [this message]
2012-12-11 4:55 ` [PATCH v3 3/5][RESEND] " Tang Chen
2012-12-11 2:33 ` [PATCH v3 4/5] page_alloc: Make movablecore_map has higher priority Tang Chen
2012-12-11 4:56 ` [PATCH v3 4/5][RESEND] " Tang Chen
2012-12-12 1:33 ` Simon Jeons
2012-12-12 9:34 ` Tang Chen
2012-12-13 1:56 ` Simon Jeons
2012-12-11 2:33 ` [PATCH v3 5/5] page_alloc: Bootmem limit with movablecore_map Tang Chen
2012-12-11 11:33 ` [PATCH v3 0/5] Add movablecore_map boot option Simon Jeons
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=50C7E4D9.30207@cn.fujitsu.com \
--to=linfeng@cn.fujitsu.com \
--cc=akpm@linux-foundation.org \
--cc=glommer@parallels.com \
--cc=hpa@zytor.com \
--cc=isimatu.yasuaki@jp.fujitsu.com \
--cc=jaegeuk.hanse@gmail.com \
--cc=jiang.liu@huawei.com \
--cc=kosaki.motohiro@jp.fujitsu.com \
--cc=laijs@cn.fujitsu.com \
--cc=linux-doc@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=lliubbo@gmail.com \
--cc=mgorman@suse.de \
--cc=minchan.kim@gmail.com \
--cc=rientjes@google.com \
--cc=rob@landley.net \
--cc=rusty@rustcorp.com.au \
--cc=simon.jeons@gmail.com \
--cc=tangchen@cn.fujitsu.com \
--cc=tony.luck@intel.com \
--cc=wency@cn.fujitsu.com \
--cc=wujianguo@huawei.com \
--cc=yinghai@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).