From: Jason Gunthorpe <jgg@ziepe.ca>
To: Christoph Hellwig <hch@lst.de>
Cc: "Ralph Campbell" <rcampbell@nvidia.com>,
amd-gfx@lists.freedesktop.org, nouveau@lists.freedesktop.org,
dri-devel@lists.freedesktop.org, kvm-ppc@vger.kernel.org,
"Bharata B Rao" <bharata@linux.ibm.com>,
linux-mm@kvack.org, "Jerome Glisse" <jglisse@redhat.com>,
"Ben Skeggs" <bskeggs@redhat.com>,
"Dan Williams" <dan.j.williams@intel.com>,
"Christian König" <christian.koenig@amd.com>
Subject: Re: [PATCH 3/4] mm: simplify device private page handling in hmm_range_fault
Date: Tue, 17 Mar 2020 09:47:55 -0300 [thread overview]
Message-ID: <20200317124755.GR20941@ziepe.ca> (raw)
In-Reply-To: <20200317122813.GA11866@lst.de>
On Tue, Mar 17, 2020 at 01:28:13PM +0100, Christoph Hellwig wrote:
> On Tue, Mar 17, 2020 at 01:24:45PM +0100, Christoph Hellwig wrote:
> > On Tue, Mar 17, 2020 at 09:15:36AM -0300, Jason Gunthorpe wrote:
> > > > Getting rid of HMM_PFN_DEVICE_PRIVATE seems reasonable to me since a driver can
> > > > look at the struct page but what if a driver needs to fault in a page from
> > > > another device's private memory? Should it call handle_mm_fault()?
> > >
> > > Isn't that what this series basically does?
> > >
> > > The dev_private_owner is set to the type of pgmap the device knows how
> > > to handle, and everything else is automatically faulted for the
> > > device.
> > >
> > > If the device does not know how to handle device_private then it sets
> > > dev_private_owner to NULL and it never gets device_private pfns.
> > >
> > > Since the device_private pfn cannot be dma mapped, drivers must have
> > > explicit support for them.
> >
> > No, with this series (and all actual callers before this series)
> > we never fault in device private pages.
>
> IFF we want to fault it in we'd need something like this. But I'd
> really prefer to see test cases for that first.
In general I think hmm_range_fault should have a mode that is the same
as get_user_pages in terms of when it returns a hard failure, and
generates faults. AFAIK, GUP will fault in this case?
I need this for making ODP use this API. ODP is the one that is highly
likely to see other driver's device_private pages and must have them
always fault to CPU.
> diff --git a/mm/hmm.c b/mm/hmm.c
> index b75b3750e03d..2884a3d11a1f 100644
> +++ b/mm/hmm.c
> @@ -276,7 +276,7 @@ static int hmm_vma_handle_pte(struct mm_walk *walk, unsigned long addr,
> if (!fault && !write_fault)
> return 0;
>
> - if (!non_swap_entry(entry))
> + if (!non_swap_entry(entry) || is_device_private_entry(entry))
> goto fault;
Yes, OK, makes sense.
I've been using v7 of Ralph's tester and it is working well - it has
DEVICE_PRIVATE support so I think it can test this flow too. Ralph are
you able?
This hunk seems trivial enough to me, can we include it now?
Thanks,
Jason
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel
next prev parent reply other threads:[~2020-03-17 14:17 UTC|newest]
Thread overview: 22+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <20200316193216.920734-1-hch@lst.de>
[not found] ` <20200316193216.920734-2-hch@lst.de>
2020-03-16 20:55 ` [PATCH 1/4] memremap: add an owner field to struct dev_pagemap Ralph Campbell
[not found] ` <20200316193216.920734-3-hch@lst.de>
2020-03-16 21:43 ` [PATCH 2/4] mm: handle multiple owners of device private pages in migrate_vma Ralph Campbell
[not found] ` <20200316193216.920734-4-hch@lst.de>
2020-03-16 19:59 ` [PATCH 3/4] mm: simplify device private page handling in hmm_range_fault Jason Gunthorpe
2020-03-16 22:49 ` Ralph Campbell
2020-03-17 12:15 ` Jason Gunthorpe
[not found] ` <20200317122445.GA11662@lst.de>
[not found] ` <20200317122813.GA11866@lst.de>
2020-03-17 12:47 ` Jason Gunthorpe [this message]
[not found] ` <20200317125955.GA12847@lst.de>
2020-03-17 17:32 ` Jason Gunthorpe
2020-03-17 23:14 ` Ralph Campbell
2020-03-19 18:17 ` Jason Gunthorpe
2020-03-19 22:56 ` Ralph Campbell
2020-03-20 0:03 ` Jason Gunthorpe
2020-03-20 0:14 ` Jason Gunthorpe
2020-03-20 1:33 ` Ralph Campbell
2020-03-20 12:58 ` Jason Gunthorpe
[not found] ` <20200317073454.GA5843@lst.de>
2020-03-17 22:43 ` Ralph Campbell
2020-03-19 0:28 ` ensure device private pages have an owner v2 Jason Gunthorpe
[not found] ` <20200319071633.GA32522@lst.de>
2020-03-19 11:50 ` Jason Gunthorpe
2020-03-19 18:50 ` Jason Gunthorpe
[not found] ` <20200316193216.920734-5-hch@lst.de>
2020-03-16 19:49 ` [PATCH 4/4] mm: check the device private page owner in hmm_range_fault Jason Gunthorpe
2020-03-16 23:11 ` Ralph Campbell
2020-03-20 13:41 ` Jason Gunthorpe
[not found] ` <20200321082236.GB28613@lst.de>
2020-03-21 12:38 ` Jason Gunthorpe
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20200317124755.GR20941@ziepe.ca \
--to=jgg@ziepe.ca \
--cc=amd-gfx@lists.freedesktop.org \
--cc=bharata@linux.ibm.com \
--cc=bskeggs@redhat.com \
--cc=christian.koenig@amd.com \
--cc=dan.j.williams@intel.com \
--cc=dri-devel@lists.freedesktop.org \
--cc=hch@lst.de \
--cc=jglisse@redhat.com \
--cc=kvm-ppc@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=nouveau@lists.freedesktop.org \
--cc=rcampbell@nvidia.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).