public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: gurudas pai <gurudas.pai@oracle.com>
To: akpm@linux-foundation.org
Cc: Joe Jin <joe.jin@oracle.com>,
	bill.irwin@oracle.com, linux-kernel@vger.kernel.org,
	"GURUDAS,PAI" <gurudas.pai@oracle.com>
Subject: Re: [PATCH] Add nid sanity on alloc_pages_node
Date: Fri, 13 Jul 2007 13:34:29 +0530	[thread overview]
Message-ID: <4697320D.1060703@oracle.com> (raw)
In-Reply-To: <20070712234938.c77f3a48.akpm@linux-foundation.org>



Andrew Morton wrote:
> On Fri, 13 Jul 2007 14:40:04 +0800 Joe Jin <joe.jin@oracle.com> wrote:
>
>> On 2007-07-12 22:18, Andrew Morton wrote:
>>> On Fri, 13 Jul 2007 10:45:07 +0800 Joe Jin <joe.jin@oracle.com> wrote:
>>>
>>> Something like this?
>>>
>>> --- a/mm/hugetlb.c~a
>>> +++ a/mm/hugetlb.c
>>> @@ -105,13 +105,20 @@ static void free_huge_page(struct page *
>>>  
>>>  static int alloc_fresh_huge_page(void)
>>>  {
>>> -	static int nid = 0;
>>> +	static int prev_nid;
>>> +	static DEFINE_SPINLOCK(nid_lock);
>>>  	struct page *page;
>>> -	page = alloc_pages_node(nid, htlb_alloc_mask|__GFP_COMP|__GFP_NOWARN,
>>> -					HUGETLB_PAGE_ORDER);
>>> -	nid = next_node(nid, node_online_map);
>>> +	int nid;
>>> +
>>> +	spin_lock(&nid_lock);
>>> +	nid = next_node(prev_nid, node_online_map);
>>>  	if (nid == MAX_NUMNODES)
>>>  		nid = first_node(node_online_map);
>>> +	prev_nid = nid;
>>> +	spin_unlock(&nid_lock);
>>> +
>>> +	page = alloc_pages_node(nid, htlb_alloc_mask|__GFP_COMP|__GFP_NOWARN,
>>> +					HUGETLB_PAGE_ORDER);
>>>  	if (page) {
>>>  		set_compound_page_dtor(page, free_huge_page);
>>>  		spin_lock(&hugetlb_lock);
>>> _
>>>
I think this will never get pages from node 0 ? Because nid = 
next_node(prev_node,node_online_map) and even if prev_node = 0, nid will 
become 1.

How about this patch ?


--- linux-2.6.22.orig/mm/hugetlb.c 2007-07-08 16:32:17.000000000 -0700
+++ linux-2.6.22-devel/mm//hugetlb.c 2007-07-13 00:26:27.000000000 -0700
@@ -103,11 +103,18 @@
{
static int nid = 0;
struct page *page;
- page = alloc_pages_node(nid, GFP_HIGHUSER|__GFP_COMP|__GFP_NOWARN,
- HUGETLB_PAGE_ORDER);
+ int cur_nid;
+ static DEFINE_SPINLOCK(nid_lock);
+
+ spin_lock(&nid_lock);
+ cur_nid = nid;
nid = next_node(nid, node_online_map);
if (nid == MAX_NUMNODES)
nid = first_node(node_online_map);
+ spin_unlock(&nid_lock);
+
+ page = alloc_pages_node(cur_nid, GFP_HIGHUSER|__GFP_COMP|__GFP_NOWARN,
+ HUGETLB_PAGE_ORDER);
if (page) {
set_compound_page_dtor(page, free_huge_page);
spin_lock(&hugetlb_lock);



Thanks,
-Guru



  parent reply	other threads:[~2007-07-13  8:05 UTC|newest]

Thread overview: 37+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-07-13  2:45 [PATCH] Add nid sanity on alloc_pages_node Joe Jin
2007-07-13  5:18 ` Andrew Morton
2007-07-13  6:40   ` Joe Jin
2007-07-13  6:49     ` Andrew Morton
2007-07-13  6:57       ` Andrew Morton
2007-07-13  8:29         ` Paul Jackson
2007-07-13  8:38           ` Andrew Morton
2007-07-13  8:43             ` Paul Jackson
2007-07-13  8:49               ` Andrew Morton
2007-07-13  8:54                 ` Paul Jackson
2007-07-13 12:48                   ` Benjamin Herrenschmidt
2007-07-13  8:03       ` Joe Jin
2007-07-13  8:15         ` Andrew Morton
2007-07-13 12:18           ` Joe Jin
2007-07-13 12:42             ` Paul Jackson
2007-07-14 17:40             ` Nish Aravamudan
2007-07-14 18:04               ` Andrew Morton
2007-07-14 20:47                 ` Nish Aravamudan
2007-07-13  8:04       ` gurudas pai [this message]
2007-07-13  8:19         ` Andrew Morton
2007-07-13 12:37           ` gurudas pai
2007-07-13  8:37       ` Joe Jin
2007-07-13  8:44         ` Andrew Morton
2007-07-17 15:04   ` Hugh Dickins
2007-07-17 16:32     ` Andrew Morton
2007-07-17 17:26       ` Hugh Dickins
2007-07-17 18:58         ` Andrew Morton
2007-07-17 19:49           ` Hugh Dickins
2007-07-17 20:01             ` Andrew Morton
2007-07-17 20:35               ` Hugh Dickins
2007-07-18  1:40                 ` Joe Jin
2007-07-18  4:49                   ` Hugh Dickins
2007-07-18  5:45                     ` Andrew Morton
2007-07-18  7:34                       ` Joe Jin
2007-07-18  6:32                     ` Joe Jin
2007-07-18  8:09                     ` Joe Jin
2007-07-18  8:35                       ` Hugh Dickins

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=4697320D.1060703@oracle.com \
    --to=gurudas.pai@oracle.com \
    --cc=akpm@linux-foundation.org \
    --cc=bill.irwin@oracle.com \
    --cc=joe.jin@oracle.com \
    --cc=linux-kernel@vger.kernel.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox