All of lore.kernel.org
 help / color / mirror / Atom feed
From: Yinghai Lu <yhlu.kernel.send@gmail.com>
To: Andrew Morton <akpm@linux-foundation.org>, Ingo Molnar <mingo@elte.hu>
Cc: Christoph Lameter <clameter@sgi.com>,
	kernel list <linux-kernel@vger.kernel.org>
Subject: [PATCH 06/12] mm: fix alloc_bootmem_core to use fast searching for all nodes
Date: Wed, 19 Mar 2008 14:04:10 -0700	[thread overview]
Message-ID: <200803191404.10282.yhlu.kernel@gmail.com> (raw)
In-Reply-To: <200803181237.33861.yhlu.kernel@gmail.com>

[PATCH] mm: fix alloc_bootmem_core to use fast searching for all nodes

make the nodes other than node 0 could use bdata->last_success for fast search too.

we need to use __alloc_bootmem_core for vmemmap allocation for other nodes when
numa and sparsemem/vmemmap are enabled.

also make fail_block path increase i with incr only needed after ALIGN to avoid
extra increase when size is large than align.

Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>

Index: linux-2.6/mm/bootmem.c
===================================================================
--- linux-2.6.orig/mm/bootmem.c
+++ linux-2.6/mm/bootmem.c
@@ -238,28 +238,32 @@ __alloc_bootmem_core(struct bootmem_data
 	 * We try to allocate bootmem pages above 'goal'
 	 * first, then we try to allocate lower pages.
 	 */
-	if (goal && goal >= bdata->node_boot_start && PFN_DOWN(goal) < end_pfn) {
-		preferred = goal - bdata->node_boot_start;
+	preferred = 0;
+	if (goal && PFN_DOWN(goal) < end_pfn) {
+		if (goal > bdata->node_boot_start)
+			preferred = goal - bdata->node_boot_start;
 
 		if (bdata->last_success >= preferred)
 			if (!limit || (limit && limit > bdata->last_success))
 				preferred = bdata->last_success;
-	} else
-		preferred = 0;
+	}
 
 	preferred = PFN_DOWN(ALIGN(preferred, align)) + offset;
 	areasize = (size + PAGE_SIZE-1) / PAGE_SIZE;
 	incr = align >> PAGE_SHIFT ? : 1;
 
 restart_scan:
-	for (i = preferred; i < eidx; i += incr) {
+	for (i = preferred; i < eidx;) {
 		unsigned long j;
+
 		i = find_next_zero_bit(bdata->node_bootmem_map, eidx, i);
 		i = ALIGN(i, incr);
 		if (i >= eidx)
 			break;
-		if (test_bit(i, bdata->node_bootmem_map))
+		if (test_bit(i, bdata->node_bootmem_map)) {
+			i += incr;
 			continue;
+		}
 		for (j = i + 1; j < i + areasize; ++j) {
 			if (j >= eidx)
 				goto fail_block;
@@ -270,6 +274,8 @@ restart_scan:
 		goto found;
 	fail_block:
 		i = ALIGN(j, incr);
+		if (i == j)
+			i += incr;
 	}
 
 	if (preferred > offset) {

  parent reply	other threads:[~2008-03-19 22:27 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <200803181237.33861.yhlu.kernel@gmail.com>
     [not found] ` <200803181255.10402.yhlu.kernel@gmail.com>
2008-03-18 23:44   ` [PATCH] x86: trim mtrr don't close gap for resource allocation Yinghai Lu
2008-03-21 10:44     ` Ingo Molnar
2008-03-19 21:03 ` [PATCH 02/12] mm: fix boundary checking in free_bootmem_core fix Yinghai Lu
2008-03-19 21:03 ` [PATCH 03/12] x86_64: free_bootmem should take phys Yinghai Lu
2008-03-19 21:03 ` [PATCH 04/12] x86_64: reserve dma32 early for gart Yinghai Lu
2008-03-19 21:04 ` [PATCH 05/12] mm: make mem_map allocation continuous Yinghai Lu
2008-03-19 21:04 ` Yinghai Lu [this message]
2008-03-19 21:04 ` [PATCH 07/12] mm: offset align in alloc_bootmem v3 Yinghai Lu
2008-03-19 21:04 ` [PATCH 08/12] mm: allocate section_map for sparse_init Yinghai Lu
2008-03-19 21:04 ` [PATCH 09/12] mm: make reserve_bootmem can crossed the nodes v2 Yinghai Lu
2008-03-19 21:04 ` [PATCH 10/12] x86_64: make reserve_bootmem_generic to use new reserve_bootmem Yinghai Lu
2008-03-21 10:50   ` Ingo Molnar
2008-03-19 21:04 ` [PATCH 11/12] x86_64: do not reserve ramdisk two times Yinghai Lu
2008-03-19 21:05 ` [PATCH 12/12] x86_64: fix setup_node_bootmem to support big mem excluding with memmap Yinghai Lu
2008-03-21 10:52   ` Ingo Molnar

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=200803191404.10282.yhlu.kernel@gmail.com \
    --to=yhlu.kernel.send@gmail.com \
    --cc=akpm@linux-foundation.org \
    --cc=clameter@sgi.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    --cc=yhlu.kernel@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.