On 11/14/2018 08:52 PM, Laurent Pinchart wrote:
> Hi Hans,
>
> On Tuesday, 13 November 2018 17:43:48 EET Hans Verkuil wrote:
>> On 11/13/18 16:06, Philipp Zabel wrote:
>>> From: John Sheu <[email protected]>
>>>
>>> Videobuf2 presently does not allow VIDIOC_REQBUFS to destroy outstanding
>>> buffers if the queue is of type V4L2_MEMORY_MMAP, and if the buffers are
>>> considered "in use". This is different behavior than for other memory
>>> types and prevents us from deallocating buffers in following two cases:
>>>
>>> 1) There are outstanding mmap()ed views on the buffer. However even if
>>> we put the buffer in reqbufs(0), there will be remaining references,
>>> due to vma .open/close() adjusting vb2 buffer refcount appropriately.
>>> This means that the buffer will be in fact freed only when the last
>>> mmap()ed view is unmapped.
>>>
>>> 2) Buffer has been exported as a DMABUF. Refcount of the vb2 buffer
>>> is managed properly by VB2 DMABUF ops, i.e. incremented on DMABUF
>>> get and decremented on DMABUF release. This means that the buffer
>>> will be alive until all importers release it.
>>>
>>> Considering both cases above, there does not seem to be any need to
>>> prevent reqbufs(0) operation, because buffer lifetime is already
>>> properly managed by both mmap() and DMABUF code paths. Let's remove it
>>> and allow userspace freeing the queue (and potentially allocating a new
>>> one) even though old buffers might be still in processing.
>>>
>>> To let userspace know that the kernel now supports orphaning buffers
>>> that are still in use, add a new V4L2_BUF_CAP_SUPPORTS_ORPHANED_BUFS
>>> to be set by reqbufs and create_bufs.
>>
>> Looks good, but I have some questions:
>>
>> 1) does v4l2-compliance together with vivid (easiest to test) still work?
>> I don't think I have a proper test for this in v4l2-compliance, but
>> I'm not 100% certain. If it fails with this patch, then please provide
>> a fix for v4l2-compliance as well.
>>
>> 2) I would like to see a new test in v4l2-compliance for this: i.e. if
>> the capability is set, then check that you can call REQBUFS(0) before
>> unmapping all buffers. Ditto with dmabuffers.
>>
>> I said during the media summit that I wanted to be more strict about
>> requiring compliance tests before adding new features, so you're the
>> unlucky victim of that :-)
>
> Do you have plans to refactor and document the v4l2-compliance internals to
> make it easier ?
Yes. I hope to be able to set aside one or two days for that in the next two
weeks.
Regards,
Hans
