linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Cody P Schafer <cody@linux.vnet.ibm.com>
To: Linux MM <linux-mm@kvack.org>
Cc: LKML <linux-kernel@vger.kernel.org>,
	Cody P Schafer <cody@linux.vnet.ibm.com>,
	Simon Jeons <simon.jeons@gmail.com>
Subject: [RFC PATCH v3 09/31] page_alloc: in move_freepages(), skip pages instead of VM_BUG on node differences.
Date: Thu,  2 May 2013 17:00:41 -0700	[thread overview]
Message-ID: <1367539263-19999-10-git-send-email-cody@linux.vnet.ibm.com> (raw)
In-Reply-To: <1367539263-19999-1-git-send-email-cody@linux.vnet.ibm.com>

With dynamic numa, pages are going to be gradully moved from one node to
another, causing the page ranges that move_freepages() examines to
contain pages that actually belong to another node.

When dynamic numa is enabled, we skip these pages instead of VM_BUGing
out on them.

This additionally moves the VM_BUG_ON() (which detects a change in node)
so that it follows the pfn_valid_within() check.

Signed-off-by: Cody P Schafer <cody@linux.vnet.ibm.com>
---
 mm/page_alloc.c | 17 ++++++++++++++---
 1 file changed, 14 insertions(+), 3 deletions(-)

diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 739b405..657f773 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -958,6 +958,7 @@ int move_freepages(struct zone *zone,
 	struct page *page;
 	unsigned long order;
 	int pages_moved = 0;
+	int zone_nid = zone_to_nid(zone);
 
 #ifndef CONFIG_HOLES_IN_ZONE
 	/*
@@ -971,14 +972,24 @@ int move_freepages(struct zone *zone,
 #endif
 
 	for (page = start_page; page <= end_page;) {
-		/* Make sure we are not inadvertently changing nodes */
-		VM_BUG_ON(page_to_nid(page) != zone_to_nid(zone));
-
 		if (!pfn_valid_within(page_to_pfn(page))) {
 			page++;
 			continue;
 		}
 
+		if (page_to_nid(page) != zone_nid) {
+#ifndef CONFIG_DYNAMIC_NUMA
+			/*
+			 * In the normal case (without Dynamic NUMA), all pages
+			 * in a pageblock should belong to the same zone (and
+			 * as a result all have the same nid).
+			 */
+			VM_BUG_ON(page_to_nid(page) != zone_nid);
+#endif
+			page++;
+			continue;
+		}
+
 		if (!PageBuddy(page)) {
 			page++;
 			continue;
-- 
1.8.2.2


  parent reply	other threads:[~2013-05-03  0:06 UTC|newest]

Thread overview: 32+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-05-03  0:00 [RFC PATCH v3 00/31] Dynamic NUMA: Runtime NUMA memory layout reconfiguration Cody P Schafer
2013-05-03  0:00 ` [RFC PATCH v3 01/31] rbtree: add postorder iteration functions Cody P Schafer
2013-05-03  0:00 ` [RFC PATCH v3 02/31] rbtree: add rbtree_postorder_for_each_entry_safe() helper Cody P Schafer
2013-05-03  0:00 ` [RFC PATCH v3 03/31] mm/memory_hotplug: factor out zone+pgdat growth Cody P Schafer
2013-05-03  0:00 ` [RFC PATCH v3 04/31] memory_hotplug: export ensure_zone_is_initialized() in mm/internal.h Cody P Schafer
2013-05-03  0:00 ` [RFC PATCH v3 05/31] mm/memory_hotplug: use {pgdat,zone}_is_empty() when resizing zones & pgdats Cody P Schafer
2013-05-03  0:00 ` [RFC PATCH v3 06/31] mm: add nid_zone() helper Cody P Schafer
2013-05-03  0:00 ` [RFC PATCH v3 07/31] mm: Add Dynamic NUMA Kconfig Cody P Schafer
2013-05-03  0:00 ` [RFC PATCH v3 08/31] page_alloc: add return_pages_to_zone() when DYNAMIC_NUMA is enabled Cody P Schafer
2013-05-03  0:00 ` Cody P Schafer [this message]
2013-05-03  0:00 ` [RFC PATCH v3 10/31] page_alloc: when dynamic numa is enabled, don't check that all pages in a block belong to the same zone Cody P Schafer
2013-05-03  0:00 ` [RFC PATCH v3 11/31] page-flags dnuma: reserve a pageflag for determining if a page needs a node lookup Cody P Schafer
2013-05-03  0:00 ` [RFC PATCH v3 12/31] memory_hotplug: factor out locks in mem_online_cpu() Cody P Schafer
2013-05-03  0:00 ` [RFC PATCH v3 13/31] mm: add memlayout & dnuma to track pfn->nid & transplant pages between nodes Cody P Schafer
2013-05-03  0:00 ` [RFC PATCH v3 14/31] mm: memlayout+dnuma: add debugfs interface Cody P Schafer
2013-05-03  0:00 ` [RFC PATCH v3 15/31] drivers/base/memory.c: alphabetize headers Cody P Schafer
2013-05-03  0:00 ` [RFC PATCH v3 16/31] drivers/base/node,memory: rename function to match interface Cody P Schafer
2013-05-03  0:00 ` [RFC PATCH v3 17/31] drivers/base/node: rename unregister_mem_blk_under_nodes() to be more acurate Cody P Schafer
2013-05-03  0:00 ` [RFC PATCH v3 18/31] drivers/base/node: add unregister_mem_block_under_nodes() Cody P Schafer
2013-05-03  0:00 ` [RFC PATCH v3 19/31] mm: memory,memlayout: add refresh_memory_blocks() for Dynamic NUMA Cody P Schafer
2013-05-03  0:00 ` [RFC PATCH v3 20/31] page_alloc: use dnuma to transplant newly freed pages in __free_pages_ok() Cody P Schafer
2013-05-03  0:00 ` [RFC PATCH v3 21/31] page_alloc: use dnuma to transplant newly freed pages in free_hot_cold_page() Cody P Schafer
2013-05-03  0:00 ` [RFC PATCH v3 22/31] page_alloc: transplant pages that are being flushed from the per-cpu lists Cody P Schafer
2013-05-03  0:00 ` [RFC PATCH v3 23/31] x86: memlayout: add a arch specific inital memlayout setter Cody P Schafer
2013-05-03  0:00 ` [RFC PATCH v3 24/31] init/main: call memlayout_global_init() in start_kernel() Cody P Schafer
2013-05-03  0:00 ` [RFC PATCH v3 25/31] dnuma: memlayout: add memory_add_physaddr_to_nid() for memory_hotplug Cody P Schafer
2013-05-03  0:00 ` [RFC PATCH v3 26/31] x86/mm/numa: when dnuma is enabled, use memlayout to handle memory hotplug's physaddr_to_nid Cody P Schafer
2013-05-03  0:00 ` [RFC PATCH v3 27/31] mm/memory_hotplug: VM_BUG if nid is too large Cody P Schafer
2013-05-03  0:01 ` [RFC PATCH v3 28/31] mm/page_alloc: in page_outside_zone_boundaries(), avoid premature decisions Cody P Schafer
2013-05-03  0:01 ` [RFC PATCH v3 29/31] mm/page_alloc: make pr_err() in page_outside_zone_boundaries() more useful Cody P Schafer
2013-05-03  0:01 ` [RFC PATCH v3 30/31] mm/page_alloc: use manage_pages instead of present pages when calculating default_zonelist_order() Cody P Schafer
2013-05-03  0:01 ` [RFC PATCH v3 31/31] mm: add a early_param "extra_nr_node_ids" to increase nr_node_ids above the minimum by a percentage Cody P Schafer

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=1367539263-19999-10-git-send-email-cody@linux.vnet.ibm.com \
    --to=cody@linux.vnet.ibm.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=simon.jeons@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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).