Re: [PATCH v1] drm/ttm: Refcount allocated tail pages

From: Dmitry Osipenko
Date: Mon Aug 15 2022 - 07:19:46 EST


On 8/15/22 13:51, Christian König wrote:
> Am 15.08.22 um 12:47 schrieb Dmitry Osipenko:
>> On 8/15/22 13:18, Dmitry Osipenko wrote:
>>> On 8/15/22 13:14, Christian König wrote:
>>>> Am 15.08.22 um 12:11 schrieb Christian König:
>>>>> Am 15.08.22 um 12:09 schrieb Dmitry Osipenko:
>>>>>> On 8/15/22 13:05, Christian König wrote:
>>>>>>> Am 15.08.22 um 11:54 schrieb Dmitry Osipenko:
>>>>>>>> Higher order pages allocated using alloc_pages() aren't
>>>>>>>> refcounted and
>>>>>>>> they
>>>>>>>> need to be refcounted, otherwise it's impossible to map them by
>>>>>>>> KVM. This
>>>>>>>> patch sets the refcount of the tail pages and fixes the KVM memory
>>>>>>>> mapping
>>>>>>>> faults.
>>>>>>>>
>>>>>>>> Without this change guest virgl driver can't map host buffers into
>>>>>>>> guest
>>>>>>>> and can't provide OpenGL 4.5 profile support to the guest. The host
>>>>>>>> mappings are also needed for enabling the Venus driver using
>>>>>>>> host GPU
>>>>>>>> drivers that are utilizing TTM.
>>>>>>>>
>>>>>>>> Based on a patch proposed by Trigger Huang.
>>>>>>> Well I can't count how often I have repeated this: This is an
>>>>>>> absolutely
>>>>>>> clear NAK!
>>>>>>>
>>>>>>> TTM pages are not reference counted in the first place and
>>>>>>> because of
>>>>>>> this giving them to virgl is illegal.
>>>>>> A? The first page is refcounted when allocated, the tail pages are
>>>>>> not.
>>>>> No they aren't. The first page is just by coincident initialized with
>>>>> a refcount of 1. This refcount is completely ignored and not used
>>>>> at all.
>>>>>
>>>>> Incrementing the reference count and by this mapping the page into
>>>>> some other address space is illegal and corrupts the internal state
>>>>> tracking of TTM.
>>>> See this comment in the source code as well:
>>>>
>>>>          /* Don't set the __GFP_COMP flag for higher order allocations.
>>>>           * Mapping pages directly into an userspace process and
>>>> calling
>>>>           * put_page() on a TTM allocated page is illegal.
>>>>           */
>>>>
>>>> I have absolutely no idea how somebody had the idea he could do this.
>>> I saw this comment, but it doesn't make sense because it doesn't explain
>>> why it's illegal. Hence it looks like a bogus comment since the
>>> refcouting certainly works, at least to a some degree because I haven't
>>> noticed any problems in practice, maybe by luck :)
>>>
>>> I'll try to dig out the older discussions, thank you for the quick
>>> reply!
>> Are you sure it was really discussed in public previously? All I can
>> find is yours two answers to a similar patches where you're saying that
>> this it's a wrong solution without in-depth explanation and further
>> discussions.
>
> Yeah, that's my problem as well I can't find that of hand.
>
> But yes it certainly was discussed in public.

If it was only CC'd to dri-devel, then could be that emails didn't pass
the spam moderation :/

>> Maybe it was discussed privately? In this case I will be happy to get
>> more info from you about the root of the problem so I could start to
>> look at how to fix it properly. It's not apparent where the problem is
>> to a TTM newbie like me.
>>
>
> Well this is completely unfixable. See the whole purpose of TTM is to
> allow tracing where what is mapped of a buffer object.
>
> If you circumvent that and increase the page reference yourself than
> that whole functionality can't work correctly any more.

Are you suggesting that the problem is that TTM doesn't see the KVM page
faults/mappings?

--
Best regards,
Dmitry