[Intel-gfx] [PATCH v2 0/9] Prereqs for TTM accelerated migration

[Thomas Hellström]

On 6/9/21 3:08 PM, Thomas Hellström wrote:
>
> On 6/9/21 2:20 PM, Matthew Auld wrote:
>> On 09/06/2021 13:16, Thomas Hellström wrote:
>>>
>>> On 6/9/21 1:48 PM, Daniel Vetter wrote:
>>>> On Wed, Jun 09, 2021 at 08:34:27AM +0200, Thomas Hellström wrote:
>>>>> A couple of patches from Chris which implement pipelined migration 
>>>>> and
>>>>> clears by atomically writing the PTEs in place before performing the
>>>>> actual blit.
>>>>>
>>>>> Some ww utilities mainly for the accompanying selftests added by 
>>>>> Thomas,
>>>>> as well as modified the above patches for ww locking- and lmem 
>>>>> support.
>>>>>
>>>>> The actual hook up to the i915 ttm backend is being worked on and not
>>>>> included yet, so this is considered to be an early review 
>>>>> opportunity.
>>>>>
>>>>> v2:
>>>>> - A couple of minor style fixes pointed out by Matthew Auld
>>>>> - Export and use intel_engine_destroy_pinned_context() to address a
>>>>> ls   CI warning / failure.
>>>> Just to check my understanding of the plan: These are the new windowed
>>>> clear/blt functions which we plan to use everywhere, because less 
>>>> nasty
>>>> locking implications? And the clear/blt we currently have in 
>>>> upstream will
>>>> be replaced?
>>>
>>> Yes. These are for LMEM clearing and migration. It looks to me like 
>>> the other ones we have in upstream are actually unused except for 
>>> selftests. We're actually using CPU clearing for now.
>>>
>>>>
>>>> If so would be nice if this patch set includes that replacement 
>>>> work (I
>>>> think right now all we have is the clear for lmem), including 
>>>> updating of
>>>> selftests and stuff like that. Just to avoid having two ways to do the
>>>> same thing in the driver.
>>>
>>> OK, I'll have a look at stripping the existing code. If we need it 
>>> moving forward, we can always re-add.
>>
>> Some low hanging fruit would be all the client_blt stuff which can 
>> for sure be deleted now.
>
> What has the client_blt stuff been used for?
>
> In any case, I need 25 commits for commit access to be granted.
>
> /Thomas
>
>
Series on trybot now. It's the dg1 live memory_region selftest that 
changes sligthly to use the new migration fill blt.

/Thomas