Re: [PATCH v1 2/2] dma-mapping-common: add DMA attribute - DMA_ATTR_IOMMU_BYPASS

[Christoph Hellwig <hch@lst.de>]

On Tue, Nov 03, 2015 at 10:08:13AM +1100, Benjamin Herrenschmidt wrote:
> On Mon, 2015-11-02 at 22:45 +0100, Arnd Bergmann wrote:
> > > Then I would argue for naming this differently. Make it an optional
> > > hint "DMA_ATTR_HIGH_PERF" or something like that. Whether this is
> > > achieved via using a bypass or other means in the backend not the
> > > business of the driver.
> > > 
> > 
> > With a name like that, who wouldn't pass that flag? ;-)
> 
> xHCI for example, vs. something like 10G ethernet... but yes I agree it
> sucks. I don't like that sort of policy anywhere in drivers. On the
> other hand the platform doesn't have much information to make that sort
> of decision either.

Mabye because it should simply use what's optimal?  E.g. passthrough
whenever possible, where arguments against possible are:  dma_mask, vfio
requirements, kernel command line option.  This is what a lot of
architectures already do, I remember the SGI Origin / Altix code has the
same behavior as well.  Those IOMMUs already had the 64 bit passthrough
and 32-bit sliding window in addition to the real IOMMU 10 years ago.