[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <e85da2ef-88bd-18f9-8b92-b64875d98a8b@intel.com>
Date: Thu, 6 Jul 2017 16:30:08 -0700
From: Dave Hansen <dave.hansen@...el.com>
To: Jerome Glisse <jglisse@...hat.com>,
Ross Zwisler <ross.zwisler@...ux.intel.com>
Cc: linux-kernel@...r.kernel.org,
"Anaczkowski, Lukasz" <lukasz.anaczkowski@...el.com>,
"Box, David E" <david.e.box@...el.com>,
"Kogut, Jaroslaw" <Jaroslaw.Kogut@...el.com>,
"Lahtinen, Joonas" <joonas.lahtinen@...el.com>,
"Moore, Robert" <robert.moore@...el.com>,
"Nachimuthu, Murugasamy" <murugasamy.nachimuthu@...el.com>,
"Odzioba, Lukasz" <lukasz.odzioba@...el.com>,
"Rafael J. Wysocki" <rafael.j.wysocki@...el.com>,
"Rafael J. Wysocki" <rjw@...ysocki.net>,
"Schmauss, Erik" <erik.schmauss@...el.com>,
"Verma, Vishal L" <vishal.l.verma@...el.com>,
"Zheng, Lv" <lv.zheng@...el.com>,
Andrew Morton <akpm@...ux-foundation.org>,
Dan Williams <dan.j.williams@...el.com>,
Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
Len Brown <lenb@...nel.org>,
Tim Chen <tim.c.chen@...ux.intel.com>, devel@...ica.org,
linux-acpi@...r.kernel.org, linux-mm@...ck.org,
linux-nvdimm@...ts.01.org
Subject: Re: [RFC v2 0/5] surface heterogeneous memory performance information
On 07/06/2017 04:08 PM, Jerome Glisse wrote:
>> So, for applications that need to differentiate between memory ranges based
>> on their performance, what option would work best for you? Is the local
>> (initiator,target) performance provided by patch 5 enough, or do you
>> require performance information for all possible (initiator,target)
>> pairings?
>
> Am i right in assuming that HBM or any faster memory will be relatively small
> (1GB - 8GB maybe 16GB ?) and of fix amount (ie size will depend on the exact
> CPU model you have) ?
For HBM, that's certainly consistent with the Xeon Phi MCDRAM.
But, please remember that this patch set is for fast memory *and* slow
memory (vs. plain DRAM).
> If so i am wondering if we should not restrict NUMA placement policy for such
> node to vma only. Forbid any policy that would prefer those node globally at
> thread/process level. This would avoid wide thread policy to exhaust this
> smaller pool of memory.
You would like to take the NUMA APIs and bifurcate them? Make some of
them able to work on this memory, and others not? So, set_mempolicy()
would work if you passed it one of these "special" nodes with
MPOL_F_ADDR, but would fail otherwise?
> Drawback of doing so would be that existing applications would not benefit
> from it. So workload where is acceptable to exhaust such memory wouldn't
> benefit until their application are updated.
I think the guys running 40-year-old fortran binaries might not be so
keen on this restriction. I bet there are a pretty substantial number
of folks out there that would love to get new hardware and just do:
numactl --membind=fast-node ./old-binary
If I were working for a hardware company, I'd sure like to just be able
to sell somebody some fancy new hardware and have their existing
software "just work" with a minimal wrapper.
Powered by blists - more mailing lists