Re: [HMM 12/15] mm/migrate: new memory migration helper for use with device memory v4

From: Balbir Singh
Date: Wed Jun 07 2017 - 05:02:19 EST


On Fri, Jun 2, 2017 at 8:35 AM, Jerome Glisse <jglisse@xxxxxxxxxx> wrote:
> On Wed, May 31, 2017 at 01:59:54PM +1000, Balbir Singh wrote:
>> On Wed, 24 May 2017 13:20:21 -0400
>> JÃrÃme Glisse <jglisse@xxxxxxxxxx> wrote:
>>
>> > This patch add a new memory migration helpers, which migrate memory
>> > backing a range of virtual address of a process to different memory
>> > (which can be allocated through special allocator). It differs from
>> > numa migration by working on a range of virtual address and thus by
>> > doing migration in chunk that can be large enough to use DMA engine
>> > or special copy offloading engine.
>> >
>> > Expected users are any one with heterogeneous memory where different
>> > memory have different characteristics (latency, bandwidth, ...). As
>> > an example IBM platform with CAPI bus can make use of this feature
>> > to migrate between regular memory and CAPI device memory. New CPU
>> > architecture with a pool of high performance memory not manage as
>> > cache but presented as regular memory (while being faster and with
>> > lower latency than DDR) will also be prime user of this patch.
>> >
>> > Migration to private device memory will be useful for device that
>> > have large pool of such like GPU, NVidia plans to use HMM for that.
>> >
>>
>> It is helpful, for HMM-CDM however we would like to avoid the downsides
>> of MIGRATE_SYNC_NOCOPY
>
> What are the downside you are referring too ?

IIUC, MIGRATE_SYNC_NO_COPY is for anonymous memory only.

Balbir Singh.