[PATCH v3 12/33] iommu/mediatek: Always tlb_flush_all when each PM resume

[Dafna Hirschfeld]

On 30.11.21 09:39, Yong Wu wrote:
> On Sat, 2021-11-27 at 12:11 +0200, Dafna Hirschfeld wrote:
>>
>> On 10.11.21 09:50, Yong Wu wrote:
>>> On Wed, 2021-11-10 at 07:29 +0200, Dafna Hirschfeld wrote:
>>>>
>>>> On 10.11.21 04:20, Yong Wu wrote:
>>>>> On Tue, 2021-11-09 at 14:21 +0200, Dafna Hirschfeld wrote:
>>>>>> Hi
>>>>>> This patch is needed in order to update the tlb when a device
>>>>>> is
>>>>>> powered on.
>>>>>> Could you send this patch alone without the whole series so
>>>>>> it
>>>>>> get
>>>>>> accepted easier?
>>>>>
>>>>> Which SoC are you testing on? In previous SoC, the IOMMU HW
>>>>> don't
>>>>> have
>>>>> power-domain, and we have a "has_pm"[1] in the tlb function for
>>>>> that
>>>>> case. The "has_pm" should be always 0 for the previous SoC like
>>>>> mt8173,
>>>>> it should always tlb synchronize.
>>>>>
>>>>> thus, Could you help share more about your issue? In which case
>>>>> it
>>>>> lack
>>>>> the necessary tlb operation. At least, We need confirm if it
>>>>> needs
>>>>> a
>>>>> "Fixes" tags if sending this patch alone.
>>>>
>>>> Hi,
>>>> I work with the mtk-vcodec driver on mt8173. As you wrote, the
>>>> iommu
>>>> doesn't
>>>> have a power-domain and so when allocating buffers before the
>>>> device
>>>> is powered
>>>> on, there is the warning
>>>> "Partial TLB flush timed out, falling back to full flush"
>>>> flooding the log buf.
>>>
>>> oh. Thanks very much for your information. Get it now.
>>>
>>> This issue should be introduced by the:
>>>
>>> b34ea31fe013 ("iommu/mediatek: Always enable the clk on resume")
>>
>> Hi, reverting this commit didn't solve those warnings,
>> I think this is because in the function mtk_iommu_attach_device
>> the first call to pm_runtime_resume_and_get does not turn the clks on
>> since m4u_dom is not yet initialize. And then mtk_iommu_attach_device
>> calls pm_runtime_put right after mtk_iommu_hw_init is called
>> (where the clks are turned on)
> 
> oh. Right. this is also related with the patch of "Add power-domain
> operation".
> 
> The current problem is that there is a redundant log of "Partial TLB
> flush timed out" in mt8173. We need fix this issue firstly. Are you
> going to prepare the patches again? If not, I could help this. You
> could help confirm them if you are free.

Hi,
I already sent a patch last week: https://lore.kernel.org/linux-iommu/afb46ad6ca9477a2bf71233858406caa6ccb1588.camel@mediatek.com/T/
could you please review it?

Thanks,
Dafna



> 
> Thanks.
> 
>>
>> thanks,
>> Dafna
>>
>>
>>>
>>> tlb failed due to the bclk is not enabled. Could you help try that
>>> after reverting this?
>>>
>>>>
>>>> Sebastian Reichel suggested to remove the 'if(has_pm)' check to
>>>> avoid
>>>> this warning,
>>>> and avoid flushing the tlb if the device is off:
>>>>
>>>> [1] http://ix.io/3Eyr
>>>>
>>>> This fixes the warning, but then the tlb is not flushed in sync,
>>>> Therefore the tlb should be flushed when the device is resumed.
>>>>
>>>> So the two patches (the one suggested in the link [1] and this
>>>> patch)
>>>> should be sent together as a 2-patch series.
>>>
>>> then this is reasonable. You could help this into a new patchset if
>>> you
>>> are free(add Fixes tag).
>>>
>>> Thanks.
>>>
>>>>
>>>> Thanks,
>>>> Dafna
>>>>
>