From: "Christian König" <ckoenig.leichtzumerken@gmail.com>
To: "Daniel Vetter" <daniel.vetter@ffwll.ch>,
"Thomas Hellström (Intel)" <thomas_os@shipmail.org>
Cc: "Intel Graphics Development" <intel-gfx@lists.freedesktop.org>,
"DRI Development" <dri-devel@lists.freedesktop.org>,
"moderated list:DMA BUFFER SHARING FRAMEWORK"
<linaro-mm-sig@lists.linaro.org>,
"Jason Gunthorpe" <jgg@ziepe.ca>,
"John Stultz" <john.stultz@linaro.org>,
"Matthew Wilcox" <willy@infradead.org>,
"Daniel Vetter" <daniel.vetter@intel.com>,
"Suren Baghdasaryan" <surenb@google.com>,
"Christian König" <christian.koenig@amd.com>,
"open list:DMA BUFFER SHARING FRAMEWORK"
<linux-media@vger.kernel.org>
Subject: Re: [Intel-gfx] [Linaro-mm-sig] [PATCH 1/2] dma-buf: Require VM_PFNMAP vma for mmap
Date: Thu, 25 Feb 2021 11:28:31 +0100 [thread overview]
Message-ID: <b30dacb0-edea-0a3c-6163-0f329e58ba61@gmail.com> (raw)
In-Reply-To: <CAKMK7uE2UrOruQPWG9KPBQ781f9Bq9xpVRNserAC9BZ2VzDutQ@mail.gmail.com>
Am 24.02.21 um 10:31 schrieb Daniel Vetter:
> On Wed, Feb 24, 2021 at 10:16 AM Thomas Hellström (Intel)
> <thomas_os@shipmail.org> wrote:
>>
>> On 2/24/21 9:45 AM, Daniel Vetter wrote:
>>> On Wed, Feb 24, 2021 at 8:46 AM Thomas Hellström (Intel)
>>> <thomas_os@shipmail.org> wrote:
>>>> On 2/23/21 11:59 AM, Daniel Vetter wrote:
>>>>> tldr; DMA buffers aren't normal memory, expecting that you can use
>>>>> them like that (like calling get_user_pages works, or that they're
>>>>> accounting like any other normal memory) cannot be guaranteed.
>>>>>
>>>>> Since some userspace only runs on integrated devices, where all
>>>>> buffers are actually all resident system memory, there's a huge
>>>>> temptation to assume that a struct page is always present and useable
>>>>> like for any more pagecache backed mmap. This has the potential to
>>>>> result in a uapi nightmare.
>>>>>
>>>>> To stop this gap require that DMA buffer mmaps are VM_PFNMAP, which
>>>>> blocks get_user_pages and all the other struct page based
>>>>> infrastructure for everyone. In spirit this is the uapi counterpart to
>>>>> the kernel-internal CONFIG_DMABUF_DEBUG.
>>>>>
>>>>> Motivated by a recent patch which wanted to swich the system dma-buf
>>>>> heap to vm_insert_page instead of vm_insert_pfn.
>>>>>
>>>>> v2:
>>>>>
>>>>> Jason brought up that we also want to guarantee that all ptes have the
>>>>> pte_special flag set, to catch fast get_user_pages (on architectures
>>>>> that support this). Allowing VM_MIXEDMAP (like VM_SPECIAL does) would
>>>>> still allow vm_insert_page, but limiting to VM_PFNMAP will catch that.
>>>>>
>>>>> From auditing the various functions to insert pfn pte entires
>>>>> (vm_insert_pfn_prot, remap_pfn_range and all it's callers like
>>>>> dma_mmap_wc) it looks like VM_PFNMAP is already required anyway, so
>>>>> this should be the correct flag to check for.
>>>>>
>>>> If we require VM_PFNMAP, for ordinary page mappings, we also need to
>>>> disallow COW mappings, since it will not work on architectures that
>>>> don't have CONFIG_ARCH_HAS_PTE_SPECIAL, (see the docs for vm_normal_page()).
>>> Hm I figured everyone just uses MAP_SHARED for buffer objects since
>>> COW really makes absolutely no sense. How would we enforce this?
>> Perhaps returning -EINVAL on is_cow_mapping() at mmap time. Either that
>> or allowing MIXEDMAP.
>>
>>>> Also worth noting is the comment in ttm_bo_mmap_vma_setup() with
>>>> possible performance implications with x86 + PAT + VM_PFNMAP + normal
>>>> pages. That's a very old comment, though, and might not be valid anymore.
>>> I think that's why ttm has a page cache for these, because it indeed
>>> sucks. The PAT changes on pages are rather expensive.
>> IIRC the page cache was implemented because of the slowness of the
>> caching mode transition itself, more specifically the wbinvd() call +
>> global TLB flush.
Yes, exactly that. The global TLB flush is what really breaks our neck
here from a performance perspective.
>>> There is still an issue for iomem mappings, because the PAT validation
>>> does a linear walk of the resource tree (lol) for every vm_insert_pfn.
>>> But for i915 at least this is fixed by using the io_mapping
>>> infrastructure, which does the PAT reservation only once when you set
>>> up the mapping area at driver load.
>> Yes, I guess that was the issue that the comment describes, but the
>> issue wasn't there with vm_insert_mixed() + VM_MIXEDMAP.
>>
>>> Also TTM uses VM_PFNMAP right now for everything, so it can't be a
>>> problem that hurts much :-)
>> Hmm, both 5.11 and drm-tip appears to still use MIXEDMAP?
>>
>> https://elixir.bootlin.com/linux/latest/source/drivers/gpu/drm/ttm/ttm_bo_vm.c#L554
> Uh that's bad, because mixed maps pointing at struct page wont stop
> gup. At least afaik.
Hui? I'm pretty sure MIXEDMAP stops gup as well. Otherwise we would have
already seen tons of problems with the page cache.
Regards,
Christian.
> Christian, do we need to patch this up, and maybe fix up ttm fault
> handler to use io_mapping so the vm_insert_pfn stuff is fast?
> -Daniel
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
next prev parent reply other threads:[~2021-02-25 14:05 UTC|newest]
Thread overview: 42+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-02-23 10:59 [Intel-gfx] [PATCH 1/2] dma-buf: Require VM_PFNMAP vma for mmap Daniel Vetter
2021-02-23 10:59 ` [Intel-gfx] [PATCH 2/2] drm/vgem: use shmem helpers Daniel Vetter
2021-02-23 11:19 ` Thomas Zimmermann
2021-02-23 11:51 ` [Intel-gfx] [PATCH] " Daniel Vetter
2021-02-23 14:21 ` [Intel-gfx] [PATCH 2/2] " kernel test robot
2021-02-23 15:07 ` kernel test robot
2021-02-25 10:23 ` [Intel-gfx] [PATCH] " Daniel Vetter
2021-02-26 9:19 ` Thomas Zimmermann
2021-02-26 13:30 ` Daniel Vetter
2021-02-26 13:51 ` Thomas Zimmermann
2021-02-26 14:04 ` Daniel Vetter
2021-02-23 11:19 ` [Intel-gfx] ✗ Fi.CI.BUILD: failure for series starting with [1/2] dma-buf: Require VM_PFNMAP vma for mmap Patchwork
2021-02-23 13:11 ` [Intel-gfx] ✗ Fi.CI.BUILD: failure for series starting with [1/2] dma-buf: Require VM_PFNMAP vma for mmap (rev2) Patchwork
2021-02-24 7:46 ` [Intel-gfx] [Linaro-mm-sig] [PATCH 1/2] dma-buf: Require VM_PFNMAP vma for mmap Thomas Hellström (Intel)
2021-02-24 8:45 ` Daniel Vetter
2021-02-24 9:15 ` Thomas Hellström (Intel)
2021-02-24 9:31 ` Daniel Vetter
2021-02-25 10:28 ` Christian König [this message]
2021-02-25 10:44 ` Daniel Vetter
2021-02-25 15:49 ` Daniel Vetter
2021-02-25 16:53 ` Christian König
2021-02-26 9:41 ` Thomas Hellström (Intel)
2021-02-26 13:28 ` Daniel Vetter
2021-02-27 8:06 ` Thomas Hellström (Intel)
2021-03-01 8:28 ` Daniel Vetter
2021-03-01 8:39 ` Thomas Hellström (Intel)
2021-03-01 9:05 ` Daniel Vetter
2021-03-01 9:21 ` Thomas Hellström (Intel)
2021-03-01 10:17 ` Christian König
2021-03-01 14:09 ` Daniel Vetter
2021-03-11 10:22 ` Thomas Hellström (Intel)
2021-03-11 13:00 ` Daniel Vetter
2021-03-11 13:12 ` Thomas Hellström (Intel)
2021-03-11 13:17 ` Daniel Vetter
2021-03-11 15:37 ` Thomas Hellström (Intel)
2021-03-12 7:51 ` Christian König
[not found] ` <20210224184657.GR2643399@ziepe.ca>
2021-02-25 10:30 ` Christian König
2021-02-25 10:45 ` Daniel Vetter
2021-02-25 10:38 ` [Intel-gfx] ✗ Fi.CI.CHECKPATCH: warning for series starting with [1/2] dma-buf: Require VM_PFNMAP vma for mmap (rev3) Patchwork
2021-02-25 11:19 ` [Intel-gfx] ✗ Fi.CI.BAT: failure " Patchwork
2021-02-26 3:57 ` [Intel-gfx] [PATCH 1/2] dma-buf: Require VM_PFNMAP vma for mmap John Stultz
2021-03-11 10:58 ` [Intel-gfx] ✗ Fi.CI.BUILD: failure for series starting with [Linaro-mm-sig,1/2] dma-buf: Require VM_PFNMAP vma for mmap (rev4) Patchwork
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=b30dacb0-edea-0a3c-6163-0f329e58ba61@gmail.com \
--to=ckoenig.leichtzumerken@gmail.com \
--cc=christian.koenig@amd.com \
--cc=daniel.vetter@ffwll.ch \
--cc=daniel.vetter@intel.com \
--cc=dri-devel@lists.freedesktop.org \
--cc=intel-gfx@lists.freedesktop.org \
--cc=jgg@ziepe.ca \
--cc=john.stultz@linaro.org \
--cc=linaro-mm-sig@lists.linaro.org \
--cc=linux-media@vger.kernel.org \
--cc=surenb@google.com \
--cc=thomas_os@shipmail.org \
--cc=willy@infradead.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).