From mboxrd@z Thu Jan  1 00:00:00 1970
From: jszhang@marvell.com (Jisheng Zhang)
Date: Wed, 3 Dec 2014 19:02:20 +0800
Subject: [RFC] implement arm64_adjust_dma_zone as arm32 does?
In-Reply-To: <20141203102431.GA7373@e104818-lin.cambridge.arm.com>
References: <20141203175855.114d6717@xhacker>
 <20141203102431.GA7373@e104818-lin.cambridge.arm.com>
Message-ID: <20141203190220.4415c21a@xhacker>
To: linux-arm-kernel@lists.infradead.org
List-Id: linux-arm-kernel.lists.infradead.org

Dear Catalin,

On Wed, 3 Dec 2014 02:24:32 -0800
Catalin Marinas <catalin.marinas@arm.com> wrote:

> On Wed, Dec 03, 2014 at 09:58:55AM +0000, Jisheng Zhang wrote:
> > If one platform has one limitation only some banks are DMA-able, for
> > example 0-2GB. Under arm32, we can set the dma_zone_size, then
> > arm_adjust_dma_zone() will set the correct dma zone for us. But under
> > arm64, we can't do it.
> 
> Do you really have plans for such platform (and it won't have an IOMMU)
> or it's just theoretical? Currently ZONE_DMA is limited to the bottom
> 32-bit of RAM. The arm32 dma_zone_size relies on the machine descriptors
> which we don't have on arm64.

Thanks for your detailed clarification. Yes, there's plan for such platform and
there's no IOMMU.

> 
> > Or set dma-range in dts, then use swiotlb to bounce?
> 
> The problem is that swiotlb is only guaranteed to live in the ZONE_DMA,
> so it doesn't buy you much if dma-ranges specify something smaller than
> ZONE_DMA.

oh Yeah!! Thanks for pointing this out which I didn't noticed.

> 
> Similarly for CMA (coherent allocations), the reserved CMA area is
> limited to ZONE_DMA (currently 32-bit mask).

Yes, I also considered CMA. The problem with the CMA is not all device
drivers' DMA buffer are allocated from CMA, so swiotlb is the only solution,
right?

> 
> > Could you kindly please give one suggestion about how to handle this
> > situation?
> 
> First of all I need to know if that's a real case and we don't look into
> it for fun.
> 
> What we could do is re-introduce ZONE_DMA32 for 32-bit dma masks and a
> ZONE_DMA for smaller masks. The problem is describing such ZONE_DMA mask
> via DT but very early before we set up the zones. A quick hack would be
> to limit ZONE_DMA to the minimum of 32-bit limit or the end of the first
> memblock, so you describe it as such in DT (but then we penalise
> platforms that can do DMA on all memblocks).
> 

We can keep ZONE_DMA32 out from arm64, just describe the ZONE_DMA mask via.
DT as you pointed out, then setup the matching dma zone.

I'm wondering whether setting up ZONE_DMA mask via. DT is acceptable or not?

Thanks,
Jisheng