On Thu, Feb 25, 2016 at 03:38:55PM +0100, [email protected] wrote: > On Thu, Feb 18, 2016 at 04:16:26PM +0000, Stuart Yoder wrote: > > #define IOMMU_READ (1 << 0) > > #define IOMMU_WRITE (1 << 1) > > -#define IOMMU_CACHE (1 << 2) /* DMA cache coherency */ > > +#define IOMMU_CACHE_COHERENT (1 << 2) /* cacheable and coherent */ > > #define IOMMU_NOEXEC (1 << 3) > > #define IOMMU_MMIO (1 << 4) /* e.g. things like MSI doorbells */ > > +#define IOMMU_CACHEABLE (1 << 5) /* cacheable, not coherent */ > > +#define IOMMU_CACHE_ALLOCATE (1 << 6) /* hint to allocate in the cache */ > > Hmm, this might be a valid use-case to introduce the > map_attr()/unmap_attr() calls to the IOMMU-API. I have been resistant to > these functions until now because the proposed use-cases were not > convincing, but for hardware-specific map-attributes like this one it > could make sense.
So this particular use-case is completely insane, but something like read/write allocation hints could make sense for arm-smmu systems. Will _______________________________________________ iommu mailing list [email protected] https://lists.linuxfoundation.org/mailman/listinfo/iommu
