Re: page allocator bug in 3.16?

[Rob Clark <robdclark@gmail.com>]

On Fri, Sep 26, 2014 at 8:34 AM, Thomas Hellstrom <thellstrom@vmware.com> wrote:
> On 09/26/2014 02:28 PM, Rob Clark wrote:
>> On Fri, Sep 26, 2014 at 6:45 AM, Thomas Hellstrom <thellstrom@vmware.com> wrote:
>>> On 09/26/2014 12:40 PM, Chuck Ebbert wrote:
>>>> On Fri, 26 Sep 2014 09:15:57 +0200
>>>> Thomas Hellstrom <thellstrom@vmware.com> wrote:
>>>>
>>>>> On 09/26/2014 01:52 AM, Peter Hurley wrote:
>>>>>> On 09/25/2014 03:35 PM, Chuck Ebbert wrote:
>>>>>>> There are six ttm patches queued for 3.16.4:
>>>>>>>
>>>>>>> drm-ttm-choose-a-pool-to-shrink-correctly-in-ttm_dma_pool_shrink_scan.patch
>>>>>>> drm-ttm-fix-handling-of-ttm_pl_flag_topdown-v2.patch
>>>>>>> drm-ttm-fix-possible-division-by-0-in-ttm_dma_pool_shrink_scan.patch
>>>>>>> drm-ttm-fix-possible-stack-overflow-by-recursive-shrinker-calls.patch
>>>>>>> drm-ttm-pass-gfp-flags-in-order-to-avoid-deadlock.patch
>>>>>>> drm-ttm-use-mutex_trylock-to-avoid-deadlock-inside-shrinker-functions.patch
>>>>>> Thanks for info, Chuck.
>>>>>>
>>>>>> Unfortunately, none of these fix TTM dma allocation doing CMA dma allocation,
>>>>>> which is the root problem.
>>>>>>
>>>>>> Regards,
>>>>>> Peter Hurley
>>>>> The problem is not really in TTM but in CMA, There was a guy offering to
>>>>> fix this in the CMA code but I guess he didn't probably because he
>>>>> didn't receive any feedback.
>>>>>
>>>> Yeah, the "solution" to this problem seems to be "don't enable CMA on
>>>> x86". Maybe it should even be disabled in the config system.
>>> Or, as previously suggested, don't use CMA for order 0 (single page)
>>> allocations....
>> On devices that actually need CMA pools to arrange for memory to be in
>> certain ranges, I think you probably do want to have order 0 pages
>> come from the CMA pool.
>
> But can the DMA subsystem or more specifically dma_alloc_coherent()
> really guarantee such things? Isn't it better for such devices to use
> CMA directly?

Well, I was thinking more specifically about a use-case that was
mentioned several times during the early CMA discussions, about video
decoders/encoders which needed Y and UV split across memory banks to
achieve sufficient bandwidth.  I assume they must use CMA directly for
this (since they'd need multiple pools per device), but not really
100% sure about that.

So perhaps, yeah, if you shunt order 0 allocations away from CMA at
the DMA layer, maybe it is ok.  If there actually is a valid use-case
for CMA on sane hardware, then maybe this is the better way, and let
the insane hw folks hack around it.

(plus, well, the use-case I was mentioning isn't really about order 0
allocations anyway)

BR,
-R


> /Thomas
>
>
>>
>> Seems like disabling CMA on x86 (where it should be unneeded) is the
>> better way, IMO
>>
>> BR,
>> -R
>>
>>
>>> /Thomas
>>>
>>> _______________________________________________
>>> dri-devel mailing list
>>> dri-devel@lists.freedesktop.org
>>> https://urldefense.proofpoint.com/v1/url?u=http://lists.freedesktop.org/mailman/listinfo/dri-devel&k=oIvRg1%2BdGAgOoM1BIlLLqw%3D%3D%0A&r=l5Ago9ekmVFZ3c4M6eauqrJWGwjf6fTb%2BP3CxbBFkVM%3D%0A&m=Uz7JXDXYXp4RlLs7G6qxMQlhOOT0trW3l78xpKg6Ass%3D%0A&s=50d6b7b3bfd093c93a228437a3d4414e49b4de817657c49c35154a115a5c2188
>