All of lore.kernel.org
 help / color / mirror / Atom feed
From: Xishi Qiu <qiuxishi@huawei.com>
To: Minchan Kim <minchan@kernel.org>, Mel Gorman <mgorman@suse.de>
Cc: Andrew Morton <akpm@linux-foundation.org>,
	riel@redhat.com, aquini@redhat.com, linux-mm@kvack.org,
	LKML <linux-kernel@vger.kernel.org>,
	Xishi Qiu <qiuxishi@huawei.com>
Subject: Re: [PATCH] mm: skip the page buddy block instead of one page
Date: Thu, 15 Aug 2013 15:45:11 +0800	[thread overview]
Message-ID: <520C8707.4000100@huawei.com> (raw)
In-Reply-To: <20130815042434.GA3139@gmail.com>

On 2013/8/15 12:24, Minchan Kim wrote:

>> Please read full thread in detail.
>>
>> Mel suggested following as
>>
>> if (PageBuddy(page)) {
>>         int nr_pages = (1 << page_order(page)) - 1;
>>         if (PageBuddy(page)) {
>>                 nr_pages = min(nr_pages, MAX_ORDER_NR_PAGES - 1);
>>                 low_pfn += nr_pages;
>>                 continue;
>>         }
>> }
>>
>> min(nr_pages, xxx) removes your concern but I think Mel's version
>> isn't right. It should be aligned with pageblock boundary so I 
>> suggested following.
>>
>> if (PageBuddy(page)) {
>> #ifdef CONFIG_MEMORY_ISOLATION
>> 	unsigned long order = page_order(page);
>> 	if (PageBuddy(page)) {
>> 		low_pfn += (1 << order) - 1;
>> 		low_pfn = min(low_pfn, end_pfn);
>> 	}
>> #endif
>> 	continue;
>> }
>>

Hi Minchan,

I understand now, but why use "end_pfn" here? 
Maybe like this:

if (PageBuddy(page)) {
	/*
	 * page_order is racy without zone->lock but worst case
	 * by the racing is just skipping pageblock_nr_pages.
	 */
	unsigned long nr_pages = 1 << page_order(page);
	if (likely(PageBuddy(page))) {
		nr_pages = min(nr_pages, MAX_ORDER_NR_PAGES);

		/* Align with pageblock boundary */
		if ((low_pfn & (pageblock_nr_pages - 1)) + nr_pages >
		    pageblock_nr_pages)
			low_pfn = ALIGN(low_pfn + 1, pageblock_nr_pages) - 1;
		else
			low_pfn += nr_pages - 1;
	}
	continue;
}

Thanks,
Xishi Qiu

>> so worst case is (pageblock_nr_pages - 1).
>> but we don't need to add CONFIG_MEMORY_ISOLATION so my suggestion
>> is following as.
>>
>> if (PageBuddy(page)) {
>> 	unsigned long order = page_order(page);
>> 	if (PageBuddy(page)) {
>> 		low_pfn += (1 << order) - 1;
>> 		low_pfn = min(low_pfn, end_pfn);
> 
> Maybe it should be low_pfn = min(low_pfn, end_pfn - 1).
> 
> 
>> 	}
>> 	continue;
>> }
>>
>>
> 



--
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>

WARNING: multiple messages have this Message-ID (diff)
From: Xishi Qiu <qiuxishi@huawei.com>
To: Minchan Kim <minchan@kernel.org>, Mel Gorman <mgorman@suse.de>
Cc: Andrew Morton <akpm@linux-foundation.org>, <riel@redhat.com>,
	<aquini@redhat.com>, <linux-mm@kvack.org>,
	LKML <linux-kernel@vger.kernel.org>,
	Xishi Qiu <qiuxishi@huawei.com>
Subject: Re: [PATCH] mm: skip the page buddy block instead of one page
Date: Thu, 15 Aug 2013 15:45:11 +0800	[thread overview]
Message-ID: <520C8707.4000100@huawei.com> (raw)
In-Reply-To: <20130815042434.GA3139@gmail.com>

On 2013/8/15 12:24, Minchan Kim wrote:

>> Please read full thread in detail.
>>
>> Mel suggested following as
>>
>> if (PageBuddy(page)) {
>>         int nr_pages = (1 << page_order(page)) - 1;
>>         if (PageBuddy(page)) {
>>                 nr_pages = min(nr_pages, MAX_ORDER_NR_PAGES - 1);
>>                 low_pfn += nr_pages;
>>                 continue;
>>         }
>> }
>>
>> min(nr_pages, xxx) removes your concern but I think Mel's version
>> isn't right. It should be aligned with pageblock boundary so I 
>> suggested following.
>>
>> if (PageBuddy(page)) {
>> #ifdef CONFIG_MEMORY_ISOLATION
>> 	unsigned long order = page_order(page);
>> 	if (PageBuddy(page)) {
>> 		low_pfn += (1 << order) - 1;
>> 		low_pfn = min(low_pfn, end_pfn);
>> 	}
>> #endif
>> 	continue;
>> }
>>

Hi Minchan,

I understand now, but why use "end_pfn" here? 
Maybe like this:

if (PageBuddy(page)) {
	/*
	 * page_order is racy without zone->lock but worst case
	 * by the racing is just skipping pageblock_nr_pages.
	 */
	unsigned long nr_pages = 1 << page_order(page);
	if (likely(PageBuddy(page))) {
		nr_pages = min(nr_pages, MAX_ORDER_NR_PAGES);

		/* Align with pageblock boundary */
		if ((low_pfn & (pageblock_nr_pages - 1)) + nr_pages >
		    pageblock_nr_pages)
			low_pfn = ALIGN(low_pfn + 1, pageblock_nr_pages) - 1;
		else
			low_pfn += nr_pages - 1;
	}
	continue;
}

Thanks,
Xishi Qiu

>> so worst case is (pageblock_nr_pages - 1).
>> but we don't need to add CONFIG_MEMORY_ISOLATION so my suggestion
>> is following as.
>>
>> if (PageBuddy(page)) {
>> 	unsigned long order = page_order(page);
>> 	if (PageBuddy(page)) {
>> 		low_pfn += (1 << order) - 1;
>> 		low_pfn = min(low_pfn, end_pfn);
> 
> Maybe it should be low_pfn = min(low_pfn, end_pfn - 1).
> 
> 
>> 	}
>> 	continue;
>> }
>>
>>
> 




  reply	other threads:[~2013-08-15  7:45 UTC|newest]

Thread overview: 66+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-08-14  4:45 [PATCH] mm: skip the page buddy block instead of one page Xishi Qiu
2013-08-14  4:45 ` Xishi Qiu
2013-08-14  7:07 ` Minchan Kim
2013-08-14  7:07   ` Minchan Kim
2013-08-14  8:57 ` Mel Gorman
2013-08-14  8:57   ` Mel Gorman
2013-08-14  9:14   ` Xishi Qiu
2013-08-14  9:14     ` Xishi Qiu
2013-08-14 15:52   ` Minchan Kim
2013-08-14 15:52     ` Minchan Kim
2013-08-14 16:16     ` Mel Gorman
2013-08-14 16:16       ` Mel Gorman
2013-08-14 16:39       ` Minchan Kim
2013-08-14 16:39         ` Minchan Kim
2013-08-14 18:00         ` Mel Gorman
2013-08-14 18:00           ` Mel Gorman
2013-08-14 19:11           ` Minchan Kim
2013-08-14 19:11             ` Minchan Kim
2013-08-15  2:32           ` Xishi Qiu
2013-08-15  2:32             ` Xishi Qiu
2013-08-15  2:44             ` Minchan Kim
2013-08-15  2:44               ` Minchan Kim
2013-08-15  3:46               ` Xishi Qiu
2013-08-15  3:46                 ` Xishi Qiu
2013-08-15  3:59                 ` Wanpeng Li
2013-08-15  3:59                 ` Wanpeng Li
2013-08-15  4:17                 ` Minchan Kim
2013-08-15  4:17                   ` Minchan Kim
2013-08-15  4:24                   ` Minchan Kim
2013-08-15  4:24                     ` Minchan Kim
2013-08-15  7:45                     ` Xishi Qiu [this message]
2013-08-15  7:45                       ` Xishi Qiu
2013-08-15  9:51                       ` Wanpeng Li
2013-08-15 11:15                         ` Xishi Qiu
2013-08-15 11:15                           ` Xishi Qiu
2013-08-15 11:23                           ` Wanpeng Li
2013-08-15 11:23                           ` Wanpeng Li
2013-08-15 11:17                         ` Xishi Qiu
2013-08-15 11:17                           ` Xishi Qiu
2013-08-15  9:51                       ` Wanpeng Li
2013-08-15  6:38                   ` Xishi Qiu
2013-08-15  6:38                     ` Xishi Qiu
2013-08-15 11:30                   ` Mel Gorman
2013-08-15 11:30                     ` Mel Gorman
2013-08-15 13:19                     ` Minchan Kim
2013-08-15 13:19                       ` Minchan Kim
2013-08-15 13:42                       ` Mel Gorman
2013-08-15 13:42                         ` Mel Gorman
2013-08-15 14:16                         ` Minchan Kim
2013-08-15 14:16                           ` Minchan Kim
2013-08-14 20:26     ` Andrew Morton
2013-08-14 20:26       ` Andrew Morton
2013-08-14 22:22       ` Mel Gorman
2013-08-14 22:22         ` Mel Gorman
2014-01-17 14:32         ` [PATCH] mm: Improve documentation of page_order Mel Gorman
2014-01-17 14:32           ` Mel Gorman
2014-01-17 18:40           ` Rafael Aquini
2014-01-17 18:40             ` Rafael Aquini
2014-01-17 18:53           ` Laura Abbott
2014-01-17 18:53             ` Laura Abbott
2014-01-17 19:59             ` Mel Gorman
2014-01-17 19:59               ` Mel Gorman
2014-01-21 11:05               ` [PATCH] mm: Improve documentation of page_order v2 Mel Gorman
2014-01-21 11:05                 ` Mel Gorman
2014-01-20  6:12           ` [PATCH] mm: Improve documentation of page_order Minchan Kim
2014-01-20  6:12             ` Minchan Kim

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=520C8707.4000100@huawei.com \
    --to=qiuxishi@huawei.com \
    --cc=akpm@linux-foundation.org \
    --cc=aquini@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mgorman@suse.de \
    --cc=minchan@kernel.org \
    --cc=riel@redhat.com \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.