From: Joel Schopp <jschopp@austin.ibm.com>
To: Andrew Morton <akpm@osdl.org>
Cc: Joel Schopp <jschopp@austin.ibm.com>,
lhms <lhms-devel@lists.sourceforge.net>,
Linux Memory Management List <linux-mm@kvack.org>,
linux-kernel@vger.kernel.org, Mel Gorman <mel@csn.ul.ie>,
Mike Kravetz <kravetz@us.ibm.com>
Subject: [PATCH 4/9] defrag helper functions
Date: Mon, 26 Sep 2005 15:09:56 -0500 [thread overview]
Message-ID: <43385594.3080303@austin.ibm.com> (raw)
In-Reply-To: <4338537E.8070603@austin.ibm.com>
[-- Attachment #1: Type: text/plain, Size: 275 bytes --]
This patch contains a handful of trivial functions, and one fairly short function
that finds an unallocated 2^MAX_ORDER-1 sized block from one type and moves it
to another type.
Signed-off-by: Mel Gorman <mel@csn.ul.ie>
Signed-off-by: Joel Schopp <jschopp@austin.ibm.com>
[-- Attachment #2: 4_defrag_helper_funcs --]
[-- Type: text/plain, Size: 2882 bytes --]
Index: 2.6.13-joel2/mm/page_alloc.c
===================================================================
--- 2.6.13-joel2.orig/mm/page_alloc.c 2005-09-20 13:45:47.%N -0500
+++ 2.6.13-joel2/mm/page_alloc.c 2005-09-20 14:16:35.%N -0500
@@ -63,7 +63,62 @@ int sysctl_lowmem_reserve_ratio[MAX_NR_Z
EXPORT_SYMBOL(totalram_pages);
EXPORT_SYMBOL(nr_swap_pages);
+static inline int need_min_fallback_reserve(struct zone *zone)
+{
+ return (zone->free_pages >> MAX_ORDER) < zone->fallback_reserve;
+}
+static inline int is_min_fallback_reserved(struct zone *zone)
+{
+ return zone->fallback_balance < 0;
+}
+static inline unsigned int get_pageblock_type(struct zone *zone,
+ struct page *page)
+{
+ unsigned long pfn = page_to_pfn(page);
+ int i, bitidx;
+ unsigned int type = 0;
+ unsigned long *usemap;
+
+ bitidx = pfn_to_bitidx(zone, pfn);
+ usemap = pfn_to_usemap(zone, pfn);
+
+ for (i=0; i < BITS_PER_RCLM_TYPE; i++) {
+ type = (type << 1);
+ type |= (!!test_bit(bitidx+i, usemap));
+ }
+
+ return type;
+}
+void assign_bit(int bit_nr, unsigned long* map, int value)
+{
+ switch (value) {
+ case 0:
+ clear_bit(bit_nr, map);
+ break;
+ default:
+ set_bit(bit_nr, map);
+ }
+}
+/*
+ * Reserve a block of pages for an allocation type & enforce function
+ * being changed if more bits are added to keep track of additional types
+ */
+BUILD_BUG_ON(BITS_PER_RCLM_TYPE > 2)
+static inline void set_pageblock_type(struct zone *zone, struct page *page,
+ int type)
+{
+ unsigned long pfn = page_to_pfn(page);
+ int bitidx;
+ unsigned long *usemap;
+
+ usemap = pfn_to_usemap(zone, pfn);
+ bitidx = pfn_to_bitidx(zone, pfn);
+
+ assign_bit(bitidx, usemap, (type & 0x1));
+ assign_bit(bitidx + 1, usemap, (type & 0x2));
+
+}
/*
* Used by page_zone() to look up the address of the struct zone whose
* id is encoded in the upper bits of page->flags
@@ -465,6 +520,41 @@ static void prep_new_page(struct page *p
kernel_map_pages(page, 1 << order, 1);
}
+/*
+ * Find a list that has a 2^MAX_ORDER-1 block of pages available and
+ * return it
+ */
+static inline struct page* steal_largepage(struct zone *zone, int alloctype)
+{
+ struct page *page;
+ struct free_area *area;
+ int i=0;
+
+ for(i = 0; i < RCLM_TYPES; i++) {
+ if(i == alloctype)
+ continue;
+
+ area = &zone->free_area_lists[i][MAX_ORDER-1];
+ if(!list_empty(&area->free_list))
+ break;
+ }
+ if (i == RCLM_TYPES) return NULL;
+
+ page = list_entry(area->free_list.next, struct page, lru);
+ area->nr_free--;
+
+ if (!is_min_fallback_reserved(zone) &&
+ need_min_fallback_reserve(zone)) {
+ alloctype = RCLM_FALLBACK;
+ }
+
+ set_pageblock_type(zone, page, alloctype);
+ dec_reserve_count(zone, i);
+ inc_reserve_count(zone, alloctype);
+
+ return page;
+}
+
/*
* Do the hard work of removing an element from the buddy allocator.
* Call me with the zone->lock already held.
next prev parent reply other threads:[~2005-09-26 20:08 UTC|newest]
Thread overview: 28+ messages / expand[flat|nested] mbox.gz Atom feed top
2005-09-26 20:01 [PATCH 0/9] fragmentation avoidance Joel Schopp
2005-09-26 20:03 ` [PATCH 1/9] add defrag flags Joel Schopp
2005-09-27 0:16 ` Kyle Moffett
2005-09-27 0:24 ` Dave Hansen
2005-09-27 0:43 ` Kyle Moffett
2005-09-27 5:44 ` Paul Jackson
2005-09-27 13:34 ` Mel Gorman
2005-09-27 16:26 ` [Lhms-devel] " Paul Jackson
2005-09-27 18:38 ` Joel Schopp
2005-09-27 19:30 ` Paul Jackson
2005-09-27 21:00 ` [Lhms-devel] " Joel Schopp
2005-09-27 21:23 ` Paul Jackson
2005-09-27 22:03 ` Joel Schopp
2005-09-27 22:45 ` Paul Jackson
2005-09-26 20:05 ` [PATCH 2/9] declare defrag structs Joel Schopp
2005-09-26 20:06 ` [PATCH 3/9] initialize defrag Joel Schopp
2005-09-26 20:09 ` Joel Schopp [this message]
2005-09-26 22:29 ` [PATCH 4/9] defrag helper functions Alex Bligh - linux-kernel
2005-09-27 16:08 ` Joel Schopp
2005-09-26 20:11 ` [PATCH 5/9] propagate defrag alloc types Joel Schopp
2005-09-26 20:13 ` [PATCH 6/9] fragmentation avoidance core Joel Schopp
2005-09-26 20:14 ` [PATCH 7/9] try harder on large allocations Joel Schopp
2005-09-27 7:21 ` Coywolf Qi Hunt
2005-09-27 16:17 ` Joel Schopp
2005-09-26 20:16 ` [PATCH 8/9] defrag fallback Joel Schopp
2005-09-26 20:17 ` [PATCH 9/9] free memory is user reclaimable Joel Schopp
2005-09-26 20:19 ` [PATCH 10/9] percpu splitout Joel Schopp
2005-09-26 21:49 ` [Lhms-devel] [PATCH 0/9] fragmentation avoidance Joel Schopp
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=43385594.3080303@austin.ibm.com \
--to=jschopp@austin.ibm.com \
--cc=akpm@osdl.org \
--cc=kravetz@us.ibm.com \
--cc=lhms-devel@lists.sourceforge.net \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mel@csn.ul.ie \
/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