Re: [RFC PATCH v3 0/7] Introduce automatic DMA configuration for IOMMU masters

[Will Deacon <will.deacon-5wv7dgnIgG8@public.gmane.org>]

On Tue, Sep 16, 2014 at 12:40:27PM +0100, Robin Murphy wrote:
> On 12/09/14 17:34, Will Deacon wrote:
> > Here is version three of the RFC I've previously posted here:
> >
> >    RFCv1: http://lists.infradead.org/pipermail/linux-arm-kernel/2014-August/283023.html
> >    RFCv2: http://lists.infradead.org/pipermail/linux-arm-kernel/2014-September/283752.html
> >
> > Changes since RFCv2 include:
> >
> >    - Put the iommu_ops in iommu_data so of_iommu_configure can avoid using
> >      the bus_type
> >    - Initialise the offset and DMA masks on the dev in of_dma_configure
> >      instead of in the arch callback (as this would cause a regression on
> >      some architectures)
> >    - Added deconfigure/teardown code based on ref counting the iommu_dma_mapping
> >    - A bunch of small fixes (_OF_DECLARE,  some code shuffling, fix multiple
> >      IOMMU parsing)
> >
> > All feedback welcome. Hopefully this is now at a point where people can
> > start looking to port dma-mapping and/or IOMMU drivers to it.
> >
> 
> What about AMBA devices? Playing with this on Juno and wondering why my 
> PL330 doesn't get any of_xlate callbacks, I see that 
> of_amba_device_create doesn't call of_dma_configure or anything from 
> that path. It's easy to work around by removing the arm,primecell 
> compatible, but that feels pretty dirty.

Yeah, that's just a bug in mainline. I'll look at fixing it if you don't
beat me to it.

Will