All of lore.kernel.org
 help / color / mirror / Atom feed
From: Eric Dumazet <dada1@cosmosbay.com>
To: David Miller <davem@davemloft.net>, akpm@linux-foundation.org
Cc: dhowells@redhat.com, linux-mm@kvack.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH] MM : alloc_large_system_hash() can free some memory for non power-of-two bucketsize
Date: Sat, 19 May 2007 22:36:52 +0200	[thread overview]
Message-ID: <464F5FE4.2010607@cosmosbay.com> (raw)
In-Reply-To: <20070519.115442.30184476.davem@davemloft.net>

[-- Attachment #1: Type: text/plain, Size: 1257 bytes --]

David Miller a écrit :
> From: Eric Dumazet <dada1@cosmosbay.com>
> Date: Sat, 19 May 2007 20:07:11 +0200
> 
>> Maybe David has an idea how this can be done properly ?
>>
>> ref : http://marc.info/?l=linux-netdev&m=117706074825048&w=2
> 
> You need to use __GFP_COMP or similar to make this splitting+freeing
> thing work.
> 
> Otherwise the individual pages don't have page references, only
> the head page of the high-order page will.
> 

Oh thanks David for the hint.

I added a split_page() call and it seems to work now.


[PATCH] MM : alloc_large_system_hash() can free some memory for non 
power-of-two bucketsize

alloc_large_system_hash() is called at boot time to allocate space for several 
large hash tables.

Lately, TCP hash table was changed and its bucketsize is not a power-of-two 
anymore.

On most setups, alloc_large_system_hash() allocates one big page (order > 0) 
with __get_free_pages(GFP_ATOMIC, order). This single high_order page has a 
power-of-two size, bigger than the needed size.

We can free all pages that wont be used by the hash table.

On a 1GB i386 machine, this patch saves 128 KB of LOWMEM memory.

TCP established hash table entries: 32768 (order: 6, 393216 bytes)

Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>

[-- Attachment #2: alloc_large.patch --]
[-- Type: text/plain, Size: 823 bytes --]

diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index ae96dd8..7c219eb 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -3350,6 +3350,21 @@ void *__init alloc_large_system_hash(const char *tablename,
 			for (order = 0; ((1UL << order) << PAGE_SHIFT) < size; order++)
 				;
 			table = (void*) __get_free_pages(GFP_ATOMIC, order);
+			/*
+			 * If bucketsize is not a power-of-two, we may free
+			 * some pages at the end of hash table.
+			 */
+			if (table) {
+				unsigned long alloc_end = (unsigned long)table +
+						(PAGE_SIZE << order);
+				unsigned long used = (unsigned long)table +
+						PAGE_ALIGN(size);
+				split_page(virt_to_page(table), order);
+				while (used < alloc_end) {
+					free_page(used);
+					used += PAGE_SIZE;
+				}
+			}
 		}
 	} while (!table && size > PAGE_SIZE && --log2qty);
 

WARNING: multiple messages have this Message-ID (diff)
From: Eric Dumazet <dada1@cosmosbay.com>
To: David Miller <davem@davemloft.net>, akpm@linux-foundation.org
Cc: dhowells@redhat.com, linux-mm@kvack.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH] MM : alloc_large_system_hash() can free some memory for non power-of-two bucketsize
Date: Sat, 19 May 2007 22:36:52 +0200	[thread overview]
Message-ID: <464F5FE4.2010607@cosmosbay.com> (raw)
In-Reply-To: <20070519.115442.30184476.davem@davemloft.net>

[-- Attachment #1: Type: text/plain, Size: 1257 bytes --]

David Miller a ecrit :
> From: Eric Dumazet <dada1@cosmosbay.com>
> Date: Sat, 19 May 2007 20:07:11 +0200
> 
>> Maybe David has an idea how this can be done properly ?
>>
>> ref : http://marc.info/?l=linux-netdev&m=117706074825048&w=2
> 
> You need to use __GFP_COMP or similar to make this splitting+freeing
> thing work.
> 
> Otherwise the individual pages don't have page references, only
> the head page of the high-order page will.
> 

Oh thanks David for the hint.

I added a split_page() call and it seems to work now.


[PATCH] MM : alloc_large_system_hash() can free some memory for non 
power-of-two bucketsize

alloc_large_system_hash() is called at boot time to allocate space for several 
large hash tables.

Lately, TCP hash table was changed and its bucketsize is not a power-of-two 
anymore.

On most setups, alloc_large_system_hash() allocates one big page (order > 0) 
with __get_free_pages(GFP_ATOMIC, order). This single high_order page has a 
power-of-two size, bigger than the needed size.

We can free all pages that wont be used by the hash table.

On a 1GB i386 machine, this patch saves 128 KB of LOWMEM memory.

TCP established hash table entries: 32768 (order: 6, 393216 bytes)

Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>

[-- Attachment #2: alloc_large.patch --]
[-- Type: text/plain, Size: 823 bytes --]

diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index ae96dd8..7c219eb 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -3350,6 +3350,21 @@ void *__init alloc_large_system_hash(const char *tablename,
 			for (order = 0; ((1UL << order) << PAGE_SHIFT) < size; order++)
 				;
 			table = (void*) __get_free_pages(GFP_ATOMIC, order);
+			/*
+			 * If bucketsize is not a power-of-two, we may free
+			 * some pages at the end of hash table.
+			 */
+			if (table) {
+				unsigned long alloc_end = (unsigned long)table +
+						(PAGE_SIZE << order);
+				unsigned long used = (unsigned long)table +
+						PAGE_ALIGN(size);
+				split_page(virt_to_page(table), order);
+				while (used < alloc_end) {
+					free_page(used);
+					used += PAGE_SIZE;
+				}
+			}
 		}
 	} while (!table && size > PAGE_SIZE && --log2qty);
 

  reply	other threads:[~2007-05-19 20:38 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-05-18  9:54 [PATCH] MM : alloc_large_system_hash() can free some memory for non power-of-two bucketsize Eric Dumazet
2007-05-18  9:54 ` Eric Dumazet
2007-05-18 18:21 ` Christoph Lameter
2007-05-18 18:21   ` Christoph Lameter
2007-05-19  8:37 ` Andrew Morton
2007-05-19  8:37   ` Andrew Morton
2007-05-19 18:07   ` Eric Dumazet
2007-05-19 18:07     ` Eric Dumazet
2007-05-19 18:54     ` David Miller
2007-05-19 18:54       ` David Miller, Eric Dumazet
2007-05-19 20:36       ` Eric Dumazet [this message]
2007-05-19 20:36         ` Eric Dumazet
2007-05-19 18:21 ` William Lee Irwin III
2007-05-19 18:21   ` William Lee Irwin III
2007-05-19 18:41   ` Eric Dumazet
2007-05-19 18:41     ` Eric Dumazet
2007-05-21  8:11     ` William Lee Irwin III
2007-05-21  8:11       ` William Lee Irwin III

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=464F5FE4.2010607@cosmosbay.com \
    --to=dada1@cosmosbay.com \
    --cc=akpm@linux-foundation.org \
    --cc=davem@davemloft.net \
    --cc=dhowells@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.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 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.