From: "Thomas Hellström" <thomas.hellstrom@linux.intel.com>
To: Jason Gunthorpe <jgg@ziepe.ca>, Daniel Vetter <daniel@ffwll.ch>
Cc: linux-samsung-soc <linux-samsung-soc@vger.kernel.org>,
Jan Kara <jack@suse.cz>, KVM list <kvm@vger.kernel.org>,
Pawel Osciak <pawel@osciak.com>,
John Hubbard <jhubbard@nvidia.com>,
LKML <linux-kernel@vger.kernel.org>,
DRI Development <dri-devel@lists.freedesktop.org>,
Tomasz Figa <tfiga@chromium.org>,
Christoph Hellwig <hch@infradead.org>,
Linux MM <linux-mm@kvack.org>,
J??r??me Glisse <jglisse@redhat.com>,
"open list:DMA BUFFER SHARING FRAMEWORK"
<linux-media@vger.kernel.org>,
Daniel Vetter <daniel.vetter@intel.com>,
Kyungmin Park <kyungmin.park@samsung.com>,
Andrew Morton <akpm@linux-foundation.org>,
Mauro Carvalho Chehab <mchehab@kernel.org>,
Dan Williams <dan.j.williams@intel.com>,
Linux ARM <linux-arm-kernel@lists.infradead.org>,
Marek Szyprowski <m.szyprowski@samsung.com>
Subject: Re: [PATCH v5 05/15] mm/frame-vector: Use FOLL_LONGTERM
Date: Mon, 09 Nov 2020 09:44:02 +0100 [thread overview]
Message-ID: <504d77b87c81b7027157e0c7b5286e17123c59d9.camel@linux.intel.com> (raw)
In-Reply-To: <20201106125505.GO36674@ziepe.ca>
On Fri, 2020-11-06 at 08:55 -0400, Jason Gunthorpe wrote:
> On Fri, Nov 06, 2020 at 11:27:59AM +0100, Daniel Vetter wrote:
> > On Fri, Nov 6, 2020 at 11:01 AM Daniel Vetter <daniel@ffwll.ch>
> > wrote:
> > > On Fri, Nov 6, 2020 at 5:08 AM John Hubbard <jhubbard@nvidia.com>
> > > wrote:
> > > > On 11/5/20 4:49 AM, Jason Gunthorpe wrote:
> > > > > On Thu, Nov 05, 2020 at 10:25:24AM +0100, Daniel Vetter
> > > > > wrote:
> > > > > > > /*
> > > > > > > * If we can't determine whether or not a pte is
> > > > > > > special, then fail immediately
> > > > > > > * for ptes. Note, we can still pin HugeTLB and THP as
> > > > > > > these are guaranteed not
> > > > > > > * to be special.
> > > > > > > *
> > > > > > > * For a futex to be placed on a THP tail page,
> > > > > > > get_futex_key requires a
> > > > > > > * get_user_pages_fast_only implementation that can pin
> > > > > > > pages. Thus it's still
> > > > > > > * useful to have gup_huge_pmd even if we can't operate
> > > > > > > on ptes.
> > > > > > > */
> > > > > >
> > > > > > We support hugepage faults in gpu drivers since recently,
> > > > > > and I'm not
> > > > > > seeing a pud_mkhugespecial anywhere. So not sure this
> > > > > > works, but probably
> > > > > > just me missing something again.
> > > > >
> > > > > It means ioremap can't create an IO page PUD, it has to be
> > > > > broken up.
> > > > >
> > > > > Does ioremap even create anything larger than PTEs?
> > >
> > > gpu drivers also tend to use vmf_insert_pfn* directly, so we can
> > > do
> > > on-demand paging and move buffers around. From what I glanced for
> > > lowest level we to the pte_mkspecial correctly (I think I
> > > convinced
> > > myself that vm_insert_pfn does that), but for pud/pmd levels it
> > > seems
> > > just yolo.
> >
> > So I dug around a bit more and ttm sets PFN_DEV | PFN_MAP to get
> > past
> > the various pft_t_devmap checks (see e.g.
> > vmf_insert_pfn_pmd_prot()).
> > x86-64 has ARCH_HAS_PTE_DEVMAP, and gup.c seems to handle these
> > specially, but frankly I got totally lost in what this does.
>
> The fact vmf_insert_pfn_pmd_prot() has all those BUG_ON's to prevent
> putting VM_PFNMAP pages into the page tables seems like a big red
> flag.
>
> The comment seems to confirm what we are talking about here:
>
> /*
> * If we had pmd_special, we could avoid all these
> restrictions,
> * but we need to be consistent with PTEs and architectures
> that
> * can't support a 'special' bit.
> */
>
> ie without the ability to mark special we can't block fast gup and
> anyone who does O_DIRECT on these ranges will crash the kernel when
> it
> tries to convert a IO page into a struct page.
>
> Should be easy enough to directly test?
>
> Putting non-struct page PTEs into a VMA without setting VM_PFNMAP
> just
> seems horribly wrong to me.
Although core mm special huge-page support is currently quite limited,
some time ago, I extended the pre-existing vma_is_dax() to
vma_is_special_huge():
/**
* vma_is_special_huge - Are transhuge page-table entries considered
special?
* @vma: Pointer to the struct vm_area_struct to consider
*
* Whether transhuge page-table entries are considered "special"
following
* the definition in vm_normal_page().
*
* Return: true if transhuge page-table entries should be considered
special,
* false otherwise.
*/
static inline bool vma_is_special_huge(const struct vm_area_struct
*vma)
{
return vma_is_dax(vma) || (vma->vm_file &&
(vma->vm_flags & (VM_PFNMAP |
VM_MIXEDMAP)));
}
meaning that currently all transhuge page-table-entries in a PFNMAP or
MIXEDMAP vma are considered "special". The number of calls to this
function (mainly in the page-splitting code) is quite limited so
replacing it with a more elaborate per-page-table-entry scheme would, I
guess, definitely be possible. Although all functions using it would
need to require a fallback path for architectures not supporting it.
/Thomas
>
> Jason
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel
next prev parent reply other threads:[~2020-11-09 8:52 UTC|newest]
Thread overview: 59+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-10-30 10:08 [PATCH v5 00/15] follow_pfn and other iomap races Daniel Vetter
2020-10-30 10:08 ` [PATCH v5 01/15] drm/exynos: Stop using frame_vector helpers Daniel Vetter
2020-10-30 10:08 ` [PATCH v5 02/15] drm/exynos: Use FOLL_LONGTERM for g2d cmdlists Daniel Vetter
2020-10-30 10:08 ` [PATCH v5 03/15] misc/habana: Stop using frame_vector helpers Daniel Vetter
2020-10-30 10:08 ` [PATCH v5 04/15] misc/habana: Use FOLL_LONGTERM for userptr Daniel Vetter
2020-10-30 10:08 ` [PATCH v5 05/15] mm/frame-vector: Use FOLL_LONGTERM Daniel Vetter
2020-10-30 14:11 ` Tomasz Figa
2020-10-30 14:37 ` Daniel Vetter
2020-11-02 18:19 ` Tomasz Figa
2020-10-31 2:55 ` John Hubbard
2020-10-31 14:45 ` Daniel Vetter
2020-11-01 5:22 ` John Hubbard
2020-11-01 10:30 ` Daniel Vetter
2020-11-01 21:13 ` John Hubbard
2020-11-01 22:50 ` Daniel Vetter
2020-11-04 14:00 ` Jason Gunthorpe
2020-11-04 15:54 ` Daniel Vetter
[not found] ` <20201104162125.GA13007@infradead.org>
2020-11-04 16:26 ` Daniel Vetter
[not found] ` <20201104163758.GA17425@infradead.org>
[not found] ` <20201104164119.GA18218@infradead.org>
2020-11-04 18:17 ` Jason Gunthorpe
2020-11-04 18:44 ` John Hubbard
2020-11-04 19:02 ` Jason Gunthorpe
2020-11-05 9:25 ` Daniel Vetter
2020-11-05 12:49 ` Jason Gunthorpe
2020-11-06 4:08 ` John Hubbard
2020-11-06 10:01 ` Daniel Vetter
2020-11-06 10:27 ` Daniel Vetter
2020-11-06 12:55 ` Jason Gunthorpe
2020-11-09 8:44 ` Thomas Hellström [this message]
2020-11-09 20:19 ` Jason Gunthorpe
2020-11-06 12:58 ` Jason Gunthorpe
2020-10-30 10:08 ` [PATCH v5 06/15] media: videobuf2: Move frame_vector into media subsystem Daniel Vetter
2020-10-30 10:08 ` [PATCH v5 07/15] mm: Close race in generic_access_phys Daniel Vetter
2020-10-30 10:08 ` [PATCH v5 08/15] mm: Add unsafe_follow_pfn Daniel Vetter
[not found] ` <20201102072931.GA16419@infradead.org>
2020-11-02 12:56 ` Daniel Vetter
2020-11-02 13:01 ` Jason Gunthorpe
2020-11-02 13:23 ` Daniel Vetter
2020-11-02 15:52 ` Jason Gunthorpe
2020-11-02 16:42 ` Daniel Vetter
2020-11-02 17:16 ` Jason Gunthorpe
2020-10-30 10:08 ` [PATCH v5 09/15] media/videbuf1|2: Mark follow_pfn usage as unsafe Daniel Vetter
2020-10-30 10:08 ` [PATCH v5 10/15] vfio/type1: Mark follow_pfn " Daniel Vetter
2020-10-30 10:08 ` [PATCH v5 11/15] PCI: Obey iomem restrictions for procfs mmap Daniel Vetter
2020-11-03 21:28 ` Bjorn Helgaas
2020-11-03 22:09 ` Dan Williams
2020-11-04 8:44 ` Daniel Vetter
2020-11-04 16:50 ` Bjorn Helgaas
2020-11-04 20:12 ` Dan Williams
2020-11-05 9:34 ` Daniel Vetter
2020-10-30 10:08 ` [PATCH v5 12/15] /dev/mem: Only set filp->f_mapping Daniel Vetter
2020-10-30 10:08 ` [PATCH v5 13/15] resource: Move devmem revoke code to resource framework Daniel Vetter
2020-10-31 6:36 ` John Hubbard
2020-10-31 14:40 ` Daniel Vetter
2020-11-03 6:06 ` [resource] 22b17dc667: Kernel panic - not syncing: Fatal exception lkp
2020-11-03 6:15 ` John Hubbard
2020-11-03 10:10 ` Daniel Vetter
2020-10-30 10:08 ` [PATCH v5 14/15] sysfs: Support zapping of binary attr mmaps Daniel Vetter
2020-10-30 10:08 ` [PATCH v5 15/15] PCI: Revoke mappings like devmem Daniel Vetter
2020-10-30 19:22 ` Dan Williams
2020-11-03 21:30 ` Bjorn Helgaas
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=504d77b87c81b7027157e0c7b5286e17123c59d9.camel@linux.intel.com \
--to=thomas.hellstrom@linux.intel.com \
--cc=akpm@linux-foundation.org \
--cc=dan.j.williams@intel.com \
--cc=daniel.vetter@intel.com \
--cc=daniel@ffwll.ch \
--cc=dri-devel@lists.freedesktop.org \
--cc=hch@infradead.org \
--cc=jack@suse.cz \
--cc=jgg@ziepe.ca \
--cc=jglisse@redhat.com \
--cc=jhubbard@nvidia.com \
--cc=kvm@vger.kernel.org \
--cc=kyungmin.park@samsung.com \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-media@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=linux-samsung-soc@vger.kernel.org \
--cc=m.szyprowski@samsung.com \
--cc=mchehab@kernel.org \
--cc=pawel@osciak.com \
--cc=tfiga@chromium.org \
/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).