Re: [PATCH V2] iommu/arm-smmu-v2: ThunderX mis-extends 64bit registers

2015-08-06 Thread Robin Murphy
On 06/08/15 17:16, Will Deacon wrote: Hi Tirumalesh, I think this looks pretty good now, just one small comment below. On Wed, Aug 05, 2015 at 05:54:28PM +0100, Tirumalesh Chalamarla wrote: [...] -#define TTBRn_HI_ASID_SHIFT16 +#define TTBRn_ASID_SHIFT 48 #define

Re: [PATCH V2] iommu/arm-smmu-v2: ThunderX mis-extends 64bit registers

2015-08-06 Thread Will Deacon
Hi Tirumalesh, I think this looks pretty good now, just one small comment below. On Wed, Aug 05, 2015 at 05:54:28PM +0100, Tirumalesh Chalamarla wrote: > The SMMU architecture defines two different behaviors when 64-bit > registers are written with 32-bit writes. The first behavior causes > zero

[PATCH V2] iommu/arm-smmu-v2: ThunderX mis-extends 64bit registers

2015-08-05 Thread Tirumalesh Chalamarla
The SMMU architecture defines two different behaviors when 64-bit registers are written with 32-bit writes. The first behavior causes zero extension into the upper 32-bits. The second behavior splits a 64-bit register into "normal" 32-bit register pairs. On some buggy implementations, registers