ath11k/mhi backtraces on shutdown with linux-next

[Kalle Valo]

Kalle Valo <kvalo at codeaurora.org> writes:

> Dmitry Baryshkov <dmitry.baryshkov at linaro.org> writes:
>
>> On Sat, 12 Dec 2020 at 08:11, Kalle Valo <kvalo at codeaurora.org> wrote:
>>>
>>> Dmitry Baryshkov <dmitry.baryshkov at linaro.org> writes:
>>>
>>> > Hello,
>>> >
>>> > I've noticed the following backtrace during shutdown stage when
>>> > rebooting RB5 board,
>>> > using linux-next/master:
>>> >
>>> > [   31.060483] ath11k_pci 0000:01:00.0: shutdown
>>> > [   31.067844] ------------[ cut here ]------------
>>> > [   31.073111] WARNING: CPU: 4 PID: 101 at
>>> > drivers/iommu/io-pgtable-arm.c:583 __arm_lpae_unmap+0x39c/0x550
>>>
>>> Do you have CONFIG_DMA_API_DEBUG disabled? There is one commit fixing a
>>> DMA API debug warning but that should be in linux-next already:
>>>
>>> cd6181ff7e93 ath11k: dp_rx: fix monitor status dma unmap direction
>>
>> I have enabled CONFIG_DMA_API_DEBUG (and _SG too). Got no additional
>> traces during boot and the same trace during reboot.
>
> On x86 I see this during rmmod when iommu is enabled:
>
> [  +0.012066] DMAR: DRHD: handling fault status reg 2
> [ +0.000022] DMAR: [DMA Write] Request device [06:00.0] PASID ffffffff
> fault addr 0 [fault reason 05] PTE Write access is not set
> [  +0.014657] DMAR: DRHD: handling fault status reg 2
> [ +0.000017] DMAR: [DMA Write] Request device [06:00.0] PASID ffffffff
> fault addr 0 [fault reason 05] PTE Write access is not set
>
> Is that the similar issue you are seeing, but just the printouts are
> different due to different iommu implementation?

BTW, this patch should fix this:

https://patchwork.kernel.org/project/linux-wireless/patch/20210913180246.193388-5-jouni@codeaurora.org/

-- 
https://patchwork.kernel.org/project/linux-wireless/list/

https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches