| Message ID | d80a42ceae3a39ef7c593a5181a1742a84c65533.1459854346.git.robin.murphy@arm.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
On Tue, Apr 05, 2016 at 12:39:30PM +0100, Robin Murphy wrote: > On some platforms, MMIO regions might need slightly different treatment > compared to mapping regular memory; add the notion of MMIO mappings to > the IOMMU API's memory type flags, so that callers can let the IOMMU > drivers know to do the right thing. > > Signed-off-by: Robin Murphy <robin.murphy@arm.com> Applied the series, thanks.
diff --git a/include/linux/iommu.h b/include/linux/iommu.h index a5c539f..34b6432 100644 --- a/include/linux/iommu.h +++ b/include/linux/iommu.h @@ -30,6 +30,7 @@ #define IOMMU_WRITE (1 << 1) #define IOMMU_CACHE (1 << 2) /* DMA cache coherency */ #define IOMMU_NOEXEC (1 << 3) +#define IOMMU_MMIO (1 << 4) /* e.g. things like MSI doorbells */ struct iommu_ops; struct iommu_group;
On some platforms, MMIO regions might need slightly different treatment compared to mapping regular memory; add the notion of MMIO mappings to the IOMMU API's memory type flags, so that callers can let the IOMMU drivers know to do the right thing. Signed-off-by: Robin Murphy <robin.murphy@arm.com> --- Hi all, Since this is a prerequisite for both MSI remapping and slave DMA handling to function properly, and the latter still needs some design work to reach a consensus[1], I thought it worth reposting this separately from Niklas' series. I've also split it up now that there's short-descriptor format to cover as well. Robin. [1]:http://lists.linuxfoundation.org/pipermail/iommu/2016-March/016059.html include/linux/iommu.h | 1 + 1 file changed, 1 insertion(+)