[PATCH v6 8/8] arm: dma-mapping: plumb our iommu mapping ops into arch_setup_dma_ops

[Thierry Reding]

On Wed, Jan 14, 2015 at 07:17:50PM +0000, Will Deacon wrote:
> On Wed, Jan 14, 2015 at 01:51:36PM +0000, Heiko Stübner wrote:
> > Am Mittwoch, 14. Januar 2015, 10:46:10 schrieb Will Deacon:
> > > On Wed, Jan 14, 2015 at 09:00:24AM +0000, Alexandre Courbot wrote:
> > > > On 12/02/2014 01:57 AM, Will Deacon wrote:
> > > > > This patch plumbs the existing ARM IOMMU DMA infrastructure (which isn't
> > > > > actually called outside of a few drivers) into arch_setup_dma_ops, so
> > > > > that we can use IOMMUs for DMA transfers in a more generic fashion.
> > > > > 
> > > > > Since this significantly complicates the arch_setup_dma_ops function,
> > > > > it is moved out of line into dma-mapping.c. If CONFIG_ARM_DMA_USE_IOMMU
> > > > > is not set, the iommu parameter is ignored and the normal ops are used
> > > > > instead.
> > > > 
> > > > A series for IOMMU support with Tegra/Nouveau ceased to work after this
> > > > patch.
> > > 
> > > Which series? This code shouldn't even be executed unless you start using
> > > of_xlate and the generic bindings.
> > > 
> > > > The Tegra IOMMU is not registered by the time the DT is parsed,
> > > > and thus all devices end up without the proper DMA ops set up because
> > > > the phandle to the IOMMU cannot be resolved.
> > > 
> > > You might want to look at the patches posted for the exynos, renesas and ARM
> > > SMMUs for some hints in how to use the new API.
> > > 
> > > > Subsequently calling arm_iommu_create_mapping() and
> > > > arm_iommu_attach_device() from the driver (as I used to do until 3.18)
> > > > does not help since the call to set_dma_ops() has been moved out of
> > > > arm_iommu_attach_device(). Therefore there seems to be no way for a device
> > > > to gets its correct DMA ops unless the IOMMU is ready by the time the DT
> > > > is parsed.
> > > > 
> > > > Also potentially affected by this are the Rockchip DRM and OMAP3 ISP
> > > > drivers, which follow the same pattern.
> > > 
> > > I don't understand why any code currently in mainline should be affected.
> > > Please can you elaborate on the failure case?
> > 
> > As Alexandre suspected the new Rockchip drm code seems to be affected by
> > this. I hadn't played with the drm code before last weekend and was then
> > stumbling over different iommu related issues. As I hadn't to much contact
> > with iommus till now I didn't get very far.
> > 
> > But with Alexandre's bandaid patch of adding
> > 	set_dma_ops(dev, &iommu_ops);
> > to arm_iommu_attach_device both problems go away.
> > 
> > 
> > So to elaborate on the two failure cases:
> 
> Aha, I see what you mean now -- the issue is that attaching to an IOMMU
> domain no longer swizzles the DMA ops. Furthermore, we also need to take
> into account the coherency of the device, which we didn't do before (and
> assumedly "worked" because all of the users happened to be non-coherent).
> 
> Maybe we just need to add something like arm_iommu_swizzle_dma_ops, so
> that direct users of the arm_iommu_* API can manage things manually?

Why does this even have to be an ARM-specific API? Couldn't this equally
well be generic across all platforms so that we get unified handling of
IOMMU through the DMA API?

Thierry
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 819 bytes
Desc: not available
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20150115/93c23635/attachment.sig>