[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <05bcdcd7-d17b-9092-1a9c-759fb0eab728@nvidia.com>
Date: Mon, 22 Jun 2020 09:58:54 -0700
From: Ralph Campbell <rcampbell@...dia.com>
To: Jason Gunthorpe <jgg@...lanox.com>
CC: <nouveau@...ts.freedesktop.org>, <linux-rdma@...r.kernel.org>,
<linux-mm@...ck.org>, <linux-kselftest@...r.kernel.org>,
<linux-kernel@...r.kernel.org>, Jerome Glisse <jglisse@...hat.com>,
"John Hubbard" <jhubbard@...dia.com>,
Christoph Hellwig <hch@....de>,
Ben Skeggs <bskeggs@...hat.com>,
Andrew Morton <akpm@...ux-foundation.org>,
Shuah Khan <shuah@...nel.org>
Subject: Re: [PATCH 00/16] mm/hmm/nouveau: THP mapping and migration
On 6/22/20 5:39 AM, Jason Gunthorpe wrote:
> On Fri, Jun 19, 2020 at 02:56:33PM -0700, Ralph Campbell wrote:
>> These patches apply to linux-5.8.0-rc1. Patches 1-3 should probably go
>> into 5.8, the others can be queued for 5.9. Patches 4-6 improve the HMM
>> self tests. Patch 7-8 prepare nouveau for the meat of this series which
>> adds support and testing for compound page mapping of system memory
>> (patches 9-11) and compound page migration to device private memory
>> (patches 12-16). Since these changes are split across mm core, nouveau,
>> and testing, I'm guessing Jason Gunthorpe's HMM tree would be appropriate.
>
> You need to break this up into parts that go where they need to
> go. Nouveau rc changes should go to DRM or some series needs to
> explain the linkage
>
>> Ralph Campbell (16):
>> mm: fix migrate_vma_setup() src_owner and normal pages
>> nouveau: fix migrate page regression
>> nouveau: fix mixed normal and device private page migration
>> mm/hmm: fix test timeout on slower machines
>> mm/hmm/test: remove redundant page table invalidate
>> mm/hmm: test mixed normal and device private migrations
>> nouveau: make nvkm_vmm_ctor() and nvkm_mmu_ptp_get() static
>> nouveau/hmm: fault one page at a time
>> mm/hmm: add output flag for compound page mapping
>> nouveau/hmm: support mapping large sysmem pages
>> hmm: add tests for HMM_PFN_COMPOUND flag
>> mm/hmm: optimize migrate_vma_setup() for holes
>
> Order things so it is hmm, test, noeveau
>
>> mm: support THP migration to device private memory
>> mm/thp: add THP allocation helper
>> mm/hmm/test: add self tests for THP migration
>> nouveau: support THP migration to private memory
>
> This is another series, you should split it even if it has to go
> through the hmm tree
>
> Jason
Thanks. I thought there was probably a better way to submit this but
I posted everything so people could see how it all fit together.
Powered by blists - more mailing lists