Re: [PATCH v7 11/14] arm64/numa: support HAVE_MEMORYLESS_NODES

2016-08-28 Thread Leizhen (ThunderTown)
On 2016/8/27 19:05, Leizhen (ThunderTown) wrote: > > > On 2016/8/26 23:43, Will Deacon wrote: >> On Wed, Aug 24, 2016 at 03:44:50PM +0800, Zhen Lei wrote: >>> Some numa nodes may have no memory. For example: >>> 1. cpu0 on node0 >>> 2. cpu1 on node1 >>> 3. device0 access the momory from node0 a

Re: [PATCH v7 11/14] arm64/numa: support HAVE_MEMORYLESS_NODES

2016-08-27 Thread Leizhen (ThunderTown)
On 2016/8/26 23:43, Will Deacon wrote: > On Wed, Aug 24, 2016 at 03:44:50PM +0800, Zhen Lei wrote: >> Some numa nodes may have no memory. For example: >> 1. cpu0 on node0 >> 2. cpu1 on node1 >> 3. device0 access the momory from node0 and node1 take the same time. >> >> So, we can not simply class

Re: [PATCH v7 11/14] arm64/numa: support HAVE_MEMORYLESS_NODES

2016-08-26 Thread Will Deacon
On Wed, Aug 24, 2016 at 03:44:50PM +0800, Zhen Lei wrote: > Some numa nodes may have no memory. For example: > 1. cpu0 on node0 > 2. cpu1 on node1 > 3. device0 access the momory from node0 and node1 take the same time. > > So, we can not simply classify device0 to node0 or node1, but we can > defi

[PATCH v7 11/14] arm64/numa: support HAVE_MEMORYLESS_NODES

2016-08-24 Thread Zhen Lei
Some numa nodes may have no memory. For example: 1. cpu0 on node0 2. cpu1 on node1 3. device0 access the momory from node0 and node1 take the same time. So, we can not simply classify device0 to node0 or node1, but we can define a node2 which distances to node0 and node1 are the same. Signed-off-