Re: [HMM 03/15] mm/unaddressable-memory: new type of ZONE_DEVICE for unaddressable memory

From	Jon Masters <jcm@jonmasters.org>
Newsgroups	linux.kernel
Subject	Re: [HMM 03/15] mm/unaddressable-memory: new type of ZONE_DEVICE for unaddressable memory
Date	2017-04-25 09:10 +0200
Message-ID	<tA2LM-1un-29@gated-at.bofh.it> (permalink)
References	(1 earlier) <tyUdA-64D-1@gated-at.bofh.it> <tyVW2-7fF-5@gated-at.bofh.it> <tz7Nv-6cn-5@gated-at.bofh.it> <tzpAJ-yN-1@gated-at.bofh.it> <tzAmt-7Jg-5@gated-at.bofh.it>
Organization	World Organi{s,z}ation Of Broken Dreams

Show all headers | View raw

On 04/23/2017 08:39 PM, John Hubbard wrote:

> Actually, MEMORY_DEVICE_PRIVATE / _PUBLIC seems like a good choice to 
> me, because the memory may not remain CPU-unaddressable in the future. 
> By that, I mean that I know of at least one company (ours) that is 
> working on products that will support hardware-based memory coherence 
> (and access counters to go along with that). If someone were to enable 
> HMM on such a system, then the device memory would be, in fact, directly 
> addressable by a CPU--thus exactly contradicting the "unaddressable" name.

I'm expecting similar with CCIX-like coherently attached accelerators
running within FPGAs and as discrete devices as well. Everyone and their
dog is working on hardware based coherence as a programming convenience
and so the notion of ZONE_DEVICE as it stood is going to rapidly evolve
over the next 18 months, maybe less. Short term anyway.

Jon.

Thread

[HMM 00/15] HMM (Heterogeneous Memory Management) v20 Jérôme Glisse <jglisse@redhat.com> - 2017-04-22 05:40 +0200
  [HMM 03/15] mm/unaddressable-memory: new type of ZONE_DEVICE for unaddressable memory Jérôme Glisse <jglisse@redhat.com> - 2017-04-22 05:50 +0200
    Re: [HMM 03/15] mm/unaddressable-memory: new type of ZONE_DEVICE for  unaddressable memory Dan Williams <dan.j.williams@intel.com> - 2017-04-22 07:40 +0200
      Re: [HMM 03/15] mm/unaddressable-memory: new type of ZONE_DEVICE for  unaddressable memory Jerome Glisse <jglisse@redhat.com> - 2017-04-22 20:20 +0200
        Re: [HMM 03/15] mm/unaddressable-memory: new type of ZONE_DEVICE for  unaddressable memory Dan Williams <dan.j.williams@intel.com> - 2017-04-23 15:20 +0200
          Re: [HMM 03/15] mm/unaddressable-memory: new type of ZONE_DEVICE for  unaddressable memory John Hubbard <jhubbard@nvidia.com> - 2017-04-24 02:50 +0200
            Re: [HMM 03/15] mm/unaddressable-memory: new type of ZONE_DEVICE for  unaddressable memory Jon Masters <jcm@jonmasters.org> - 2017-04-25 09:10 +0200
  [HMM 01/15] mm, memory_hotplug: introduce add_pages Jérôme Glisse <jglisse@redhat.com> - 2017-04-22 05:50 +0200
  [HMM 07/15] mm/hmm: heterogeneous memory management (HMM for short) v2 Jérôme Glisse <jglisse@redhat.com> - 2017-04-22 06:10 +0200
  [HMM 08/15] mm/hmm/mirror: mirror process address space on device with HMM helpers v2 Jérôme Glisse <jglisse@redhat.com> - 2017-04-22 06:10 +0200
  [HMM 14/15] mm/hmm/devmem: dummy HMM device for ZONE_DEVICE memory v3 Jérôme Glisse <jglisse@redhat.com> - 2017-04-22 07:00 +0200

csiph-web