From: David Rientjes <rientjes@google.com>
To: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
Cc: linux-mm@kvack.org, linux-numa@vger.kernel.org,
Andrew Morton <akpm@linux-foundation.org>,
Mel Gorman <mel@csn.ul.ie>,
Randy Dunlap <randy.dunlap@oracle.com>,
Nishanth Aravamudan <nacc@us.ibm.com>,
Andi Kleen <andi@firstfloor.org>, Adam Litke <agl@us.ibm.com>,
Andy Whitcroft <apw@canonical.com>,
eric.whitney@hp.com
Subject: Re: [PATCH 7/12] hugetlb: add per node hstate attributes
Date: Mon, 12 Oct 2009 19:09:50 -0700 (PDT) [thread overview]
Message-ID: <alpine.DEB.1.00.0910121906590.26949@chino.kir.corp.google.com> (raw)
In-Reply-To: <1255362064.4344.105.camel@useless.americas.hpqcorp.net>
On Mon, 12 Oct 2009, Lee Schermerhorn wrote:
> > Hmm, does this really work for memory hot-remove? If all memory is
> > removed from a nid, does node_hstates[nid]->hstate_objs[] get updated
> > appropriately? I assume we'd never pass that particular kobj to
> > kobj_to_node_hstate() anymore, but I'm wondering if the pointer would
> > remain in the hstate_kobjs[] table.
>
> Patch 11 is intended to address this. The hotplug notifier, added by
> that patch, will call hugetlb_unregister_node() in the event all memory
> is removed from a node. hugetlb_unregister_node() NULLs out the per
> node hstate_kobjs[] after freeing them. This patch [7/12] handles node
> hot-plug--as opposed to memory hot-plug that transitions the node
> to/from the memoryless state.
>
Ahh, I see it done in hugetlb_register_node(), thanks.
There's probably not much of a need to unregister the attributes if all
memory is removed, anyway, subsequent allocation attempts on its node
should simply fail. It looks like your patches address node hotplug well,
thanks for the clarification.
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
next prev parent reply other threads:[~2009-10-13 2:10 UTC|newest]
Thread overview: 31+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-10-08 16:24 [PATCH 0/12] hugetlb: V10 numa control of persistent huge pages alloc/free Lee Schermerhorn
2009-10-08 16:25 ` [PATCH 1/12] nodemask: make NODEMASK_ALLOC more general Lee Schermerhorn
2009-10-08 20:17 ` David Rientjes
2009-10-08 16:25 ` [PATCH 2/12] hugetlb: rework hstate_next_node_* functions Lee Schermerhorn
2009-10-08 16:25 ` [PATCH 3/12] hugetlb: add nodemask arg to huge page alloc, free and surplus adjust fcns Lee Schermerhorn
2009-10-08 20:32 ` David Rientjes
2009-10-08 16:25 ` [PATCH 4/12] hugetlb: factor init_nodemask_of_node Lee Schermerhorn
2009-10-08 20:20 ` David Rientjes
2009-10-08 16:25 ` [PATCH 5/12] hugetlb: derive huge pages nodes allowed from task mempolicy Lee Schermerhorn
2009-10-08 21:22 ` [patch] mm: add gfp flags for NODEMASK_ALLOC slab allocations David Rientjes
2009-10-09 1:01 ` KAMEZAWA Hiroyuki
2009-10-08 16:25 ` [PATCH 6/12] hugetlb: add generic definition of NUMA_NO_NODE Lee Schermerhorn
2009-10-08 20:16 ` Christoph Lameter
2009-10-08 20:26 ` David Rientjes
2009-10-27 21:44 ` [patch -mm] acpi: remove NID_INVAL David Rientjes
2009-10-28 14:53 ` Cyrill Gorcunov
2009-10-29 18:40 ` Christoph Lameter
2009-10-08 16:25 ` [PATCH 7/12] hugetlb: add per node hstate attributes Lee Schermerhorn
2009-10-08 20:42 ` David Rientjes
2009-10-09 12:57 ` Lee Schermerhorn
2009-10-09 22:10 ` David Rientjes
2009-10-09 13:49 ` Lee Schermerhorn
2009-10-09 22:18 ` David Rientjes
2009-10-12 15:41 ` Lee Schermerhorn
2009-10-13 2:09 ` David Rientjes [this message]
2009-10-08 16:25 ` [PATCH 8/12] hugetlb: update hugetlb documentation for NUMA controls Lee Schermerhorn
2009-10-08 16:25 ` [PATCH 9/12] hugetlb: use only nodes with memory for huge pages Lee Schermerhorn
2009-10-08 16:26 ` [PATCH 10/12] mm: clear node in N_HIGH_MEMORY and stop kswapd when all memory is offlined Lee Schermerhorn
2009-10-08 20:19 ` David Rientjes
2009-10-08 16:26 ` [PATCH 11/12] hugetlb: handle memory hot-plug events Lee Schermerhorn
2009-10-08 16:26 ` [PATCH 12/12] hugetlb: offload per node attribute registrations Lee Schermerhorn
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=alpine.DEB.1.00.0910121906590.26949@chino.kir.corp.google.com \
--to=rientjes@google.com \
--cc=Lee.Schermerhorn@hp.com \
--cc=agl@us.ibm.com \
--cc=akpm@linux-foundation.org \
--cc=andi@firstfloor.org \
--cc=apw@canonical.com \
--cc=eric.whitney@hp.com \
--cc=linux-mm@kvack.org \
--cc=linux-numa@vger.kernel.org \
--cc=mel@csn.ul.ie \
--cc=nacc@us.ibm.com \
--cc=randy.dunlap@oracle.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