Re: [PATCH 12/14] dma-direct: handle the memory encryption bit in common code

[Christoph Hellwig]

On Mon, Mar 19, 2018 at 06:01:41PM +0000, Catalin Marinas wrote:
> I don't particularly like maintaining an arm64-specific dma-direct.h
> either but arm64 seems to be the only architecture that needs to
> potentially force a bounce when cache_line_size() > ARCH_DMA_MINALIGN
> and the device is non-coherent.

mips is another likely candidate, see all the recent drama about
dma_get_alignmet().  And I'm also having major discussion about even
exposing the cache line size architecturally for RISC-V, so changes
are high it'll have to deal with this mess sooner or later as they
probably can't agree on a specific cache line size.

> Note that lib/swiotlb.c doesn't even
> deal with non-coherent DMA (e.g. map_sg doesn't have arch callbacks for
> cache maintenance), so not disrupting lib/swiotlb.c seems to be the
> least intrusive option.

No yet.  I have patches to consolidate the various swiotlb ops
that deal with cache flushing or barriers.  I was hoping to get them
in for this merge window, but it probably is too late now given that
I have a few other fires to fight.  But they are going to be out
early for the next merge window.

> > Nevermind that the commit should at least be three different patches:
> > 
> >  (1) revert the broken original commit
> >  (2) increase the dma min alignment
> 
> Reverting the original commit could, on its own, break an SoC which
> expects ARCH_DMA_MINALIGN == 128. So these two should be a single commit
> (my patch only reverts the L1_CACHE_BYTES change rather than
> ARCH_DMA_MINALIGN, the latter being correct as 128).

It would revert to the state before this commit.

> As I said above, adding a check in swiotlb.c for
> !is_device_dma_coherent(dev) && (ARCH_DMA_MINALIGN < cache_line_size())
> feels too architecture specific.

And what exactly is architecture specific about that?  It is a totally
generic concept, which at this point also seems entirely theoretical
based on the previous mail in this thread.