Re: [PATCH v2 05/13] iommu/rockchip: Fix error handling in init

[JeffyChen <jeffy.chen-TNX95d0MmH7DzftRWevZcw@public.gmane.org>]

Hi Robin,

Thanks for your reply.

On 01/17/2018 07:36 PM, Robin Murphy wrote:
> On 17/01/18 05:26, Tomasz Figa wrote:
>> On Tue, Jan 16, 2018 at 10:25 PM, Jeffy Chen
>> <jeffy.chen-TNX95d0MmH7DzftRWevZcw@public.gmane.org> wrote:
>>> It's hard to undo bus_set_iommu() in the error path, so move it to the
>>> end of rk_iommu_probe().
>>
>> Does this work fine now? I remember we used to need this called in an
>> early initcall for all the ARM/ARM64 DMA stuff to work.
>
> It will do once we get to patch #11 (where the IOMMU_OF_DECLARE ensures
> that masters defer until iommu_register() has set ops with a non-NULL
> .of_xlate callback); in the meantime you might end up depending on DT
> probe order as to whether the master uses the IOMMU or not. I'd say it's
> up to you guys whether you consider that a bisection-breaker or not.
>
hmmm, maybe i can just place this patch after the patch #11 ;)
> Robin.
>
>
>