All of lore.kernel.org
 help / color / mirror / Atom feed
From: Minchan Kim <minchan@kernel.org>
To: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Cc: Andrew Morton <akpm@linux-foundation.org>,
	linux-mm@kvack.org, linux-kernel@vger.kernel.org,
	Sergey Senozhatsky <sergey.senozhatsky.work@gmail.com>
Subject: Re: [RFC][PATCHv3 2/7] zsmalloc: partial page ordering within a fullness_list
Date: Mon, 29 Jun 2015 15:52:18 +0900	[thread overview]
Message-ID: <20150629065218.GC13179@bbox> (raw)
In-Reply-To: <1434628004-11144-3-git-send-email-sergey.senozhatsky@gmail.com>

On Thu, Jun 18, 2015 at 08:46:39PM +0900, Sergey Senozhatsky wrote:
> We want to see more ZS_FULL pages and less ZS_ALMOST_{FULL, EMPTY}
> pages. Put a page with higher ->inuse count first within its
> ->fullness_list, which will give us better chances to fill up this
> page with new objects (find_get_zspage() return ->fullness_list head
> for new object allocation), so some zspages will become
> ZS_ALMOST_FULL/ZS_FULL quicker.
> 
> It performs a trivial and cheap ->inuse compare which does not slow
> down zsmalloc, and in the worst case it keeps the list pages not in
> any particular order, just like we do it now.
> 
> A more expensive solution could sort fullness_list by ->inuse count.
> 
> Signed-off-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
> ---
>  mm/zsmalloc.c | 12 ++++++++++--
>  1 file changed, 10 insertions(+), 2 deletions(-)
> 
> diff --git a/mm/zsmalloc.c b/mm/zsmalloc.c
> index 7d816c2..6e2ebb6 100644
> --- a/mm/zsmalloc.c
> +++ b/mm/zsmalloc.c
> @@ -659,8 +659,16 @@ static void insert_zspage(struct page *page, struct size_class *class,
>  		return;
>  
>  	head = &class->fullness_list[fullness];
> -	if (*head)
> -		list_add_tail(&page->lru, &(*head)->lru);
> +	if (*head) {
> +		/*
> +		 * We want to see more ZS_FULL pages and less almost
> +		 * empty/full. Put pages with higher ->inuse first.
> +		 */
> +		if (page->inuse < (*head)->inuse)
> +			list_add_tail(&page->lru, &(*head)->lru);
> +		else
> +			list_add(&page->lru, &(*head)->lru);
> +	}

>  
>  	*head = page;

Why do you want to always put @page in the head?
How about this?

diff --git a/mm/zsmalloc.c b/mm/zsmalloc.c
index e8cb31c..1c5fde9 100644
--- a/mm/zsmalloc.c
+++ b/mm/zsmalloc.c
@@ -658,21 +658,25 @@ static void insert_zspage(struct page *page, struct size_class *class,
        if (fullness >= _ZS_NR_FULLNESS_GROUPS)
                return;

+       zs_stat_inc(class, fullness == ZS_ALMOST_EMPTY ?
+                       CLASS_ALMOST_EMPTY : CLASS_ALMOST_FULL, 1);
+
        head = &class->fullness_list[fullness];
-       if (*head) {
-               /*
-                * We want to see more ZS_FULL pages and less almost
-                * empty/full. Put pages with higher ->inuse first.
-                */
-               if (page->inuse < (*head)->inuse)
-                       list_add_tail(&page->lru, &(*head)->lru);
-               else
-                       list_add(&page->lru, &(*head)->lru);
+       if (!*head) {
+               *head = page;
+               return;
        }

-       *head = page;
-       zs_stat_inc(class, fullness == ZS_ALMOST_EMPTY ?
-                       CLASS_ALMOST_EMPTY : CLASS_ALMOST_FULL, 1);
+       /*
+        * We want to see more ZS_FULL pages and less almost
+        * empty/full. Put pages with higher ->inuse first.
+        */
+       list_add_tail(&page->lru, &(*head)->lru);
+       if (page->inuse >= (*head)->inuse)
+               *head = page;
 }

 /*
-- 
1.7.9.5




>  	zs_stat_inc(class, fullness == ZS_ALMOST_EMPTY ?
> -- 
> 2.4.4
> 

--
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: Minchan Kim <minchan@kernel.org>
To: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Cc: Andrew Morton <akpm@linux-foundation.org>,
	linux-mm@kvack.org, linux-kernel@vger.kernel.org,
	Sergey Senozhatsky <sergey.senozhatsky.work@gmail.com>
Subject: Re: [RFC][PATCHv3 2/7] zsmalloc: partial page ordering within a fullness_list
Date: Mon, 29 Jun 2015 15:52:18 +0900	[thread overview]
Message-ID: <20150629065218.GC13179@bbox> (raw)
In-Reply-To: <1434628004-11144-3-git-send-email-sergey.senozhatsky@gmail.com>

On Thu, Jun 18, 2015 at 08:46:39PM +0900, Sergey Senozhatsky wrote:
> We want to see more ZS_FULL pages and less ZS_ALMOST_{FULL, EMPTY}
> pages. Put a page with higher ->inuse count first within its
> ->fullness_list, which will give us better chances to fill up this
> page with new objects (find_get_zspage() return ->fullness_list head
> for new object allocation), so some zspages will become
> ZS_ALMOST_FULL/ZS_FULL quicker.
> 
> It performs a trivial and cheap ->inuse compare which does not slow
> down zsmalloc, and in the worst case it keeps the list pages not in
> any particular order, just like we do it now.
> 
> A more expensive solution could sort fullness_list by ->inuse count.
> 
> Signed-off-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
> ---
>  mm/zsmalloc.c | 12 ++++++++++--
>  1 file changed, 10 insertions(+), 2 deletions(-)
> 
> diff --git a/mm/zsmalloc.c b/mm/zsmalloc.c
> index 7d816c2..6e2ebb6 100644
> --- a/mm/zsmalloc.c
> +++ b/mm/zsmalloc.c
> @@ -659,8 +659,16 @@ static void insert_zspage(struct page *page, struct size_class *class,
>  		return;
>  
>  	head = &class->fullness_list[fullness];
> -	if (*head)
> -		list_add_tail(&page->lru, &(*head)->lru);
> +	if (*head) {
> +		/*
> +		 * We want to see more ZS_FULL pages and less almost
> +		 * empty/full. Put pages with higher ->inuse first.
> +		 */
> +		if (page->inuse < (*head)->inuse)
> +			list_add_tail(&page->lru, &(*head)->lru);
> +		else
> +			list_add(&page->lru, &(*head)->lru);
> +	}

>  
>  	*head = page;

Why do you want to always put @page in the head?
How about this?

diff --git a/mm/zsmalloc.c b/mm/zsmalloc.c
index e8cb31c..1c5fde9 100644
--- a/mm/zsmalloc.c
+++ b/mm/zsmalloc.c
@@ -658,21 +658,25 @@ static void insert_zspage(struct page *page, struct size_class *class,
        if (fullness >= _ZS_NR_FULLNESS_GROUPS)
                return;

+       zs_stat_inc(class, fullness == ZS_ALMOST_EMPTY ?
+                       CLASS_ALMOST_EMPTY : CLASS_ALMOST_FULL, 1);
+
        head = &class->fullness_list[fullness];
-       if (*head) {
-               /*
-                * We want to see more ZS_FULL pages and less almost
-                * empty/full. Put pages with higher ->inuse first.
-                */
-               if (page->inuse < (*head)->inuse)
-                       list_add_tail(&page->lru, &(*head)->lru);
-               else
-                       list_add(&page->lru, &(*head)->lru);
+       if (!*head) {
+               *head = page;
+               return;
        }

-       *head = page;
-       zs_stat_inc(class, fullness == ZS_ALMOST_EMPTY ?
-                       CLASS_ALMOST_EMPTY : CLASS_ALMOST_FULL, 1);
+       /*
+        * We want to see more ZS_FULL pages and less almost
+        * empty/full. Put pages with higher ->inuse first.
+        */
+       list_add_tail(&page->lru, &(*head)->lru);
+       if (page->inuse >= (*head)->inuse)
+               *head = page;
 }

 /*
-- 
1.7.9.5




>  	zs_stat_inc(class, fullness == ZS_ALMOST_EMPTY ?
> -- 
> 2.4.4
> 

  parent reply	other threads:[~2015-06-29  6:52 UTC|newest]

Thread overview: 48+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-06-18 11:46 [RFC][PATCH v3 0/7] introduce automatic pool compaction Sergey Senozhatsky
2015-06-18 11:46 ` Sergey Senozhatsky
2015-06-18 11:46 ` [RFC][PATCHv3 1/7] zsmalloc: drop unused variable `nr_to_migrate' Sergey Senozhatsky
2015-06-18 11:46   ` Sergey Senozhatsky
2015-06-18 11:46 ` [RFC][PATCHv3 2/7] zsmalloc: partial page ordering within a fullness_list Sergey Senozhatsky
2015-06-18 11:46   ` Sergey Senozhatsky
2015-06-18 12:13   ` Sergey Senozhatsky
2015-06-18 12:13     ` Sergey Senozhatsky
2015-06-18 12:45     ` Sergey Senozhatsky
2015-06-18 12:45       ` Sergey Senozhatsky
2015-06-18 12:48     ` Sergey Senozhatsky
2015-06-18 12:48       ` Sergey Senozhatsky
2015-06-18 14:43     ` Sergey Senozhatsky
2015-06-18 14:43       ` Sergey Senozhatsky
2015-06-29  6:52   ` Minchan Kim [this message]
2015-06-29  6:52     ` Minchan Kim
2015-06-29 23:41     ` Sergey Senozhatsky
2015-06-29 23:41       ` Sergey Senozhatsky
2015-06-30  0:42       ` Minchan Kim
2015-06-30  0:42         ` Minchan Kim
2015-06-18 11:46 ` [RFC][PATCHv3 3/7] zsmalloc: always keep per-class stats Sergey Senozhatsky
2015-06-18 11:46   ` Sergey Senozhatsky
2015-06-29  6:40   ` Minchan Kim
2015-06-29  6:40     ` Minchan Kim
2015-06-29  9:06     ` Sergey Senozhatsky
2015-06-29  9:06       ` Sergey Senozhatsky
2015-06-18 11:46 ` [RFC][PATCHv3 4/7] zsmalloc: introduce zs_can_compact() function Sergey Senozhatsky
2015-06-18 11:46   ` Sergey Senozhatsky
2015-06-29  6:45   ` Minchan Kim
2015-06-29  6:45     ` Minchan Kim
2015-06-29  8:58     ` Sergey Senozhatsky
2015-06-29  8:58       ` Sergey Senozhatsky
2015-06-18 11:46 ` [RFC][PATCHv3 5/7] zsmalloc: cosmetic compaction code adjustments Sergey Senozhatsky
2015-06-18 11:46   ` Sergey Senozhatsky
2015-06-18 11:46 ` [RFC][PATCHv3 6/7] zsmalloc/zram: move `num_migrated' to zs_pool Sergey Senozhatsky
2015-06-18 11:46   ` Sergey Senozhatsky
2015-06-18 11:46 ` [RFC][PATCHv3 7/7] zsmalloc: register a shrinker to trigger auto-compaction Sergey Senozhatsky
2015-06-18 11:46   ` Sergey Senozhatsky
2015-06-29  7:07   ` Minchan Kim
2015-06-29  7:07     ` Minchan Kim
2015-06-29  8:57     ` Sergey Senozhatsky
2015-06-29  8:57       ` Sergey Senozhatsky
2015-06-29 13:39       ` Minchan Kim
2015-06-29 13:39         ` Minchan Kim
2015-06-29 23:36         ` Sergey Senozhatsky
2015-06-29 23:36           ` Sergey Senozhatsky
2015-06-18 12:17 ` [RFC][PATCH v3 0/7] introduce automatic pool compaction Sergey Senozhatsky
2015-06-18 12:17   ` Sergey Senozhatsky

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=20150629065218.GC13179@bbox \
    --to=minchan@kernel.org \
    --cc=akpm@linux-foundation.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=sergey.senozhatsky.work@gmail.com \
    --cc=sergey.senozhatsky@gmail.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.