From: Dmitry Osipenko <dmitry.osipenko@collabora.com>
To: "Christian König" <christian.koenig@amd.com>,
"Thomas Hellström (Intel)" <thomas_os@shipmail.org>
Cc: David Airlie <airlied@linux.ie>,
dri-devel@lists.freedesktop.org,
virtualization@lists.linux-foundation.org,
Thierry Reding <thierry.reding@gmail.com>,
Gerd Hoffmann <kraxel@redhat.com>,
Dmitry Osipenko <digetx@gmail.com>,
kernel@collabora.com, Sumit Semwal <sumit.semwal@linaro.org>,
Mauro Carvalho Chehab <mchehab@kernel.org>,
Steven Price <steven.price@arm.com>,
Gustavo Padovan <gustavo.padovan@collabora.com>,
Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>,
linux-media@vger.kernel.org, intel-gfx@lists.freedesktop.org,
linaro-mm-sig@lists.linaro.org,
Rodrigo Vivi <rodrigo.vivi@intel.com>,
linux-tegra@vger.kernel.org,
Gurchetan Singh <gurchetansingh@chromium.org>,
Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>,
Daniel Almeida <daniel.almeida@collabora.com>,
amd-gfx@lists.freedesktop.org,
Tomeu Vizoso <tomeu.vizoso@collabora.com>,
Gert Wollny <gert.wollny@collabora.com>,
"Pan, Xinhui" <Xinhui.Pan@amd.com>,
linux-kernel@vger.kernel.org, Tomasz Figa <tfiga@chromium.org>,
Qiang Yu <yuq825@gmail.com>,
Thomas Zimmermann <tzimmermann@suse.de>,
Alex Deucher <alexander.deucher@amd.com>,
Robin Murphy <robin.murphy@arm.com>,
Emil Velikov <"emil.l.veli kov"@gmail.com>
Subject: Re: [Linaro-mm-sig] Re: [PATCH v6 02/22] drm/gem: Move mapping of imported dma-bufs to drm_gem_mmap_obj()
Date: Tue, 5 Jul 2022 01:44:20 +0300 [thread overview]
Message-ID: <3e9fbf88-c1bd-d336-b0e0-86fb208a952e@collabora.com> (raw)
In-Reply-To: <0335814b-dc11-4a54-9f24-bb4e9dbef35c@amd.com>
On 7/4/22 15:33, Christian König wrote:
> Am 30.06.22 um 01:06 schrieb Dmitry Osipenko:
>> On 6/29/22 11:43, Thomas Hellström (Intel) wrote:
>>> On 6/29/22 10:22, Dmitry Osipenko wrote:
>>>> On 6/29/22 09:40, Thomas Hellström (Intel) wrote:
>>>>> On 5/27/22 01:50, Dmitry Osipenko wrote:
>>>>>> Drivers that use drm_gem_mmap() and drm_gem_mmap_obj() helpers don't
>>>>>> handle imported dma-bufs properly, which results in mapping of
>>>>>> something
>>>>>> else than the imported dma-buf. For example, on NVIDIA Tegra we get a
>>>>>> hard
>>>>>> lockup when userspace writes to the memory mapping of a dma-buf that
>>>>>> was
>>>>>> imported into Tegra's DRM GEM.
>>>>>>
>>>>>> To fix this bug, move mapping of imported dma-bufs to
>>>>>> drm_gem_mmap_obj().
>>>>>> Now mmaping of imported dma-bufs works properly for all DRM drivers.
>>>>> Same comment about Fixes: as in patch 1,
>>>>>> Cc: stable@vger.kernel.org
>>>>>> Signed-off-by: Dmitry Osipenko <dmitry.osipenko@collabora.com>
>>>>>> ---
>>>>>> drivers/gpu/drm/drm_gem.c | 3 +++
>>>>>> drivers/gpu/drm/drm_gem_shmem_helper.c | 9 ---------
>>>>>> drivers/gpu/drm/tegra/gem.c | 4 ++++
>>>>>> 3 files changed, 7 insertions(+), 9 deletions(-)
>>>>>>
>>>>>> diff --git a/drivers/gpu/drm/drm_gem.c b/drivers/gpu/drm/drm_gem.c
>>>>>> index 86d670c71286..7c0b025508e4 100644
>>>>>> --- a/drivers/gpu/drm/drm_gem.c
>>>>>> +++ b/drivers/gpu/drm/drm_gem.c
>>>>>> @@ -1038,6 +1038,9 @@ int drm_gem_mmap_obj(struct drm_gem_object
>>>>>> *obj,
>>>>>> unsigned long obj_size,
>>>>>> if (obj_size < vma->vm_end - vma->vm_start)
>>>>>> return -EINVAL;
>>>>>> + if (obj->import_attach)
>>>>>> + return dma_buf_mmap(obj->dma_buf, vma, 0);
>>>>> If we start enabling mmaping of imported dma-bufs on a majority of
>>>>> drivers in this way, how do we ensure that user-space is not blindly
>>>>> using the object mmap without calling the needed DMA_BUF_IOCTL_SYNC
>>>>> which is needed before and after cpu access of mmap'ed dma-bufs?
>>>>>
>>>>> I was under the impression (admittedly without looking) that the few
>>>>> drivers that actually called into dma_buf_mmap() had some private
>>>>> user-mode driver code in place that ensured this happened.
>>>> Since it's a userspace who does the mapping, then it should be a
>>>> responsibility of userspace to do all the necessary syncing.
>>> Sure, but nothing prohibits user-space to ignore the syncing thinking
>>> "It works anyway", testing those drivers where the syncing is a NOP. And
>>> when a driver that finally needs syncing is tested it's too late to fix
>>> all broken user-space.
>>>
>>>> I'm not
>>>> sure whether anyone in userspace really needs to map imported dma-bufs
>>>> in practice. Nevertheless, this use-case is broken and should be fixed
>>>> by either allowing to do the mapping or prohibiting it.
>>>>
>>> Then I'd vote for prohibiting it, at least for now. And for the future
>>> moving forward we could perhaps revisit the dma-buf need for syncing,
>>> requiring those drivers that actually need it to implement emulated
>>> coherent memory which can be done not too inefficiently (vmwgfx being
>>> one example).
>> Alright, I'll change it to prohibit the mapping. This indeed should be a
>> better option.
>
> Oh, yes please. But I would expect that some people start screaming.
>
> Over time I've got tons of TTM patches because people illegally tried to
> mmap() imported DMA-bufs in their driver.
>
> Anyway this is probably the right thing to do and we can work on fixing
> the fallout later on.
I already sent out the patch [1] that prohibits the mapping. Would be
great if you all could take a look and give a r-b, thanks in advance.
[1] https://patchwork.freedesktop.org/patch/492148/
--
Best regards,
Dmitry
next prev parent reply other threads:[~2022-07-05 11:34 UTC|newest]
Thread overview: 62+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-05-26 23:50 [PATCH v6 00/22] Add generic memory shrinker to VirtIO-GPU and Panfrost DRM drivers Dmitry Osipenko
2022-05-26 23:50 ` [PATCH v6 01/22] drm/gem: Properly annotate WW context on drm_gem_lock_reservations() error Dmitry Osipenko
2022-06-28 20:12 ` Thomas Hellström (Intel)
2022-06-29 8:23 ` Dmitry Osipenko
2022-05-26 23:50 ` [PATCH v6 02/22] drm/gem: Move mapping of imported dma-bufs to drm_gem_mmap_obj() Dmitry Osipenko
2022-06-29 6:40 ` Thomas Hellström (Intel)
2022-06-29 8:22 ` Dmitry Osipenko
2022-06-29 8:43 ` Thomas Hellström (Intel)
2022-06-29 23:06 ` Dmitry Osipenko
2022-07-04 12:33 ` [Linaro-mm-sig] " Christian König
2022-07-04 22:44 ` Dmitry Osipenko [this message]
2022-05-26 23:50 ` [PATCH v6 03/22] drm/panfrost: Put mapping instead of shmem obj on panfrost_mmu_map_fault_addr() error Dmitry Osipenko
2022-05-26 23:50 ` [PATCH v6 04/22] drm/panfrost: Fix shrinker list corruption by madvise IOCTL Dmitry Osipenko
2022-05-27 15:56 ` Alyssa Rosenzweig
2022-05-30 9:41 ` Steven Price
2022-05-26 23:50 ` [PATCH v6 05/22] drm/virtio: Correct drm_gem_shmem_get_sg_table() error handling Dmitry Osipenko
2022-05-26 23:50 ` [PATCH v6 06/22] drm/virtio: Check whether transferred 2D BO is shmem Dmitry Osipenko
2022-05-26 23:50 ` [PATCH v6 07/22] drm/virtio: Unlock reservations on virtio_gpu_object_shmem_init() error Dmitry Osipenko
2022-05-26 23:50 ` [PATCH v6 08/22] drm/virtio: Unlock reservations on dma_resv_reserve_fences() error Dmitry Osipenko
2022-06-29 6:43 ` Thomas Hellström (Intel)
2022-05-26 23:50 ` [PATCH v6 09/22] drm/virtio: Use appropriate atomic state in virtio_gpu_plane_cleanup_fb() Dmitry Osipenko
2022-05-26 23:50 ` [PATCH v6 10/22] drm/shmem-helper: Add missing vunmap on error Dmitry Osipenko
2022-05-26 23:50 ` [PATCH v6 11/22] drm/shmem-helper: Correct doc-comment of drm_gem_shmem_get_sg_table() Dmitry Osipenko
2022-05-26 23:50 ` [PATCH v6 12/22] drm/virtio: Simplify error handling of virtio_gpu_object_create() Dmitry Osipenko
2022-05-26 23:50 ` [PATCH v6 13/22] drm/virtio: Improve DMA API usage for shmem BOs Dmitry Osipenko
2022-05-26 23:50 ` [PATCH v6 14/22] dma-buf: Introduce new locking convention Dmitry Osipenko
2022-05-30 6:50 ` Christian König
2022-05-30 13:26 ` Dmitry Osipenko
2022-05-30 13:41 ` Christian König
2022-05-30 13:57 ` Dmitry Osipenko
2022-06-28 21:26 ` Thomas Hellström (Intel)
2022-07-01 10:43 ` Dmitry Osipenko
2022-07-04 22:38 ` Dmitry Osipenko
2022-07-05 10:52 ` Dmitry Osipenko
2022-05-26 23:50 ` [PATCH v6 15/22] drm/shmem-helper: Don't use vmap_use_count for dma-bufs Dmitry Osipenko
2022-05-26 23:50 ` [PATCH v6 16/22] drm/shmem-helper: Use reservation lock Dmitry Osipenko
2022-05-26 23:50 ` [PATCH v6 17/22] drm/shmem-helper: Add generic memory shrinker Dmitry Osipenko
2022-06-05 16:47 ` Daniel Vetter
2022-06-05 18:32 ` Rob Clark
2022-06-05 18:45 ` Daniel Vetter
2022-06-06 10:57 ` Christian König
2022-06-07 15:33 ` Dmitry Osipenko
2022-06-19 17:53 ` Rob Clark
2022-06-20 14:08 ` Dmitry Osipenko
2022-06-20 15:18 ` Rob Clark
2022-06-24 20:23 ` Daniel Vetter
2022-06-24 20:21 ` Daniel Vetter
2022-06-20 15:37 ` Rob Clark
2022-06-21 16:39 ` Dmitry Osipenko
2022-05-26 23:50 ` [PATCH v6 18/22] drm/gem: Add drm_gem_pin_unlocked() Dmitry Osipenko
2022-05-26 23:50 ` [PATCH v6 19/22] drm/virtio: Support memory shrinking Dmitry Osipenko
2022-05-26 23:50 ` [PATCH v6 20/22] drm/virtio: Use dev_is_pci() Dmitry Osipenko
2022-05-26 23:50 ` [PATCH v6 21/22] drm/virtio: Return proper error codes instead of -1 Dmitry Osipenko
2022-05-26 23:50 ` [PATCH v6 22/22] drm/panfrost: Switch to generic memory shrinker Dmitry Osipenko
2022-05-27 11:48 ` Alyssa Rosenzweig
2022-06-10 14:40 ` [PATCH v6 00/22] Add generic memory shrinker to VirtIO-GPU and Panfrost DRM drivers Dmitry Osipenko
2022-06-28 12:31 ` Robin Murphy
2022-06-28 12:50 ` Dmitry Osipenko
2022-06-28 16:48 ` Rob Clark
2022-06-28 23:11 ` Dmitry Osipenko
2022-06-28 12:51 ` Dmitry Osipenko
2022-06-28 13:11 ` Dmitry Osipenko
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=3e9fbf88-c1bd-d336-b0e0-86fb208a952e@collabora.com \
--to=dmitry.osipenko@collabora.com \
--cc="emil.l.veli kov"@gmail.com \
--cc=Xinhui.Pan@amd.com \
--cc=airlied@linux.ie \
--cc=alexander.deucher@amd.com \
--cc=alyssa.rosenzweig@collabora.com \
--cc=amd-gfx@lists.freedesktop.org \
--cc=christian.koenig@amd.com \
--cc=daniel.almeida@collabora.com \
--cc=digetx@gmail.com \
--cc=dri-devel@lists.freedesktop.org \
--cc=gert.wollny@collabora.com \
--cc=gurchetansingh@chromium.org \
--cc=gustavo.padovan@collabora.com \
--cc=intel-gfx@lists.freedesktop.org \
--cc=kernel@collabora.com \
--cc=kraxel@redhat.com \
--cc=linaro-mm-sig@lists.linaro.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-media@vger.kernel.org \
--cc=linux-tegra@vger.kernel.org \
--cc=mchehab@kernel.org \
--cc=robin.murphy@arm.com \
--cc=rodrigo.vivi@intel.com \
--cc=steven.price@arm.com \
--cc=sumit.semwal@linaro.org \
--cc=tfiga@chromium.org \
--cc=thierry.reding@gmail.com \
--cc=thomas_os@shipmail.org \
--cc=tomeu.vizoso@collabora.com \
--cc=tvrtko.ursulin@linux.intel.com \
--cc=tzimmermann@suse.de \
--cc=virtualization@lists.linux-foundation.org \
--cc=yuq825@gmail.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).