Re: [PATCH 11/21] x86: get pg_data_t's memory from other node

2013-07-24 Thread Tejun Heo
Hello, Tang. On Wed, Jul 24, 2013 at 11:52:53AM +0800, Tang Chen wrote: > The node data should be on local, I agree with that. I'm not saying I > won't do it. Just for now, it will be complicated to fix memory hot-remove > path. So I think pushing this patch for now, and do the local node things >

Re: [PATCH 11/21] x86: get pg_data_t's memory from other node

2013-07-23 Thread Tang Chen
On 07/24/2013 04:09 AM, Tejun Heo wrote: On Fri, Jul 19, 2013 at 03:59:24PM +0800, Tang Chen wrote: From: Yasuaki Ishimatsu If system can create movable node which all memory of the node is allocated as ZONE_MOVABLE, setup_node_data() cannot allocate memory for the node's pg_data_t. So, use mem

Re: [PATCH 11/21] x86: get pg_data_t's memory from other node

2013-07-23 Thread Tejun Heo
On Fri, Jul 19, 2013 at 03:59:24PM +0800, Tang Chen wrote: > From: Yasuaki Ishimatsu > > If system can create movable node which all memory of the > node is allocated as ZONE_MOVABLE, setup_node_data() cannot > allocate memory for the node's pg_data_t. > So, use memblock_alloc_try_nid() instead o

[PATCH 11/21] x86: get pg_data_t's memory from other node

2013-07-19 Thread Tang Chen
From: Yasuaki Ishimatsu If system can create movable node which all memory of the node is allocated as ZONE_MOVABLE, setup_node_data() cannot allocate memory for the node's pg_data_t. So, use memblock_alloc_try_nid() instead of memblock_alloc_nid() to retry when the first allocation fails. Otherw