Re: [PATCH] x86/mm: Don't force unencrypted DMA for IOMMU-backed devices

[Aneesh Kumar K.V <aneesh.kumar@kernel.org>]

Alexey Kardashevskiy <aik@amd.com> writes:

> On 8/9/26 21:32, Aneesh Kumar K.V (Arm) wrote:
>> Commit 8277a12d0d60 ("dma-pool: track decrypted atomic pools and select
>> them via attrs") exposed an issue with force_dma_unencrypted() on
>> systems using host memory encryption.
>> 
>> force_dma_unencrypted() checks whether the device DMA mask can address
>> the encryption bit and, if not, requires DMA allocations to use
>> unencrypted memory. However, this check is not applicable when the
>> device is using the IOMMU. In that case, the device DMA mask constrains
>> the IOVA seen by the device, not the backing physical address, so it
>> does not need to cover the C-bit.
>> 
>> This currently causes dma_alloc_attrs() to set
>> __DMA_ATTR_ALLOC_CC_SHARED for such devices. iommu_dma_alloc() does not
>> support that attribute and rejects the allocation, causing DMA
>> allocations to fail.
>> 
>> Do not force DMA allocations to be unencrypted when the device is using
>> the IOMMU. This allows the IOMMU to map the encrypted physical pages as
>> before and avoids incorrectly requesting CC_SHARED allocations.
>> 
>> Fixes: 8277a12d0d60 ("dma-pool: track decrypted atomic pools and select them via attrs")
>> Reported-by: Timo Witte <timo.witte@gmail.com>
>> Link: https://lore.kernel.org/all/CANB4YXR7h8V5Xp=MXVZeSdvw9UiriSagp=E+ju5RRDNghoPHLQ@mail.gmail.com
>> Signed-off-by: Aneesh Kumar K.V (Arm) <aneesh.kumar@kernel.org>
>> ---
>>   arch/x86/mm/mem_encrypt.c | 3 ++-
>>   1 file changed, 2 insertions(+), 1 deletion(-)
>> 
>> diff --git a/arch/x86/mm/mem_encrypt.c b/arch/x86/mm/mem_encrypt.c
>> index 912f22ca838f..a349d8d21569 100644
>> --- a/arch/x86/mm/mem_encrypt.c
>> +++ b/arch/x86/mm/mem_encrypt.c
>> @@ -13,6 +13,7 @@
>>   #include <linux/cc_platform.h>
>>   #include <linux/mem_encrypt.h>
>>   #include <linux/virtio_anchor.h>
>> +#include <linux/iommu-dma.h>
>>   
>>   #include <asm/sev.h>
>>   
>> @@ -30,7 +31,7 @@ bool force_dma_unencrypted(struct device *dev)
>>   	 * device does not support DMA to addresses that include the
>>   	 * encryption mask.
>>   	 */
>> -	if (cc_platform_has(CC_ATTR_HOST_MEM_ENCRYPT)) {
>> +	if (cc_platform_has(CC_ATTR_HOST_MEM_ENCRYPT) && !use_dma_iommu(dev)) {
>>   		u64 dma_enc_mask = DMA_BIT_MASK(__ffs64(sme_me_mask));
>>   		u64 dma_dev_mask = min_not_zero(dev->coherent_dma_mask,
>>   						dev->bus_dma_limit);
>
> if we drop this whole "if (cc_platform_has(CC_ATTR_HOST_MEM_ENCRYPT))"
> chunk, won't dma_alloc_attrs()&co be able to figure out this from
> dev->bus_dma_limit and other DMA masks we have all over the place? I
> should probably just try it... Thanks,
>


That would imply that, with host memory encryption, no device requires
unencrypted DMA. IIUC, that is not true. There are three cases:

1. Host memory encryption with an IOMMU: unencrypted DMA is not required.
2. Host memory encryption without an IOMMU, where the device's DMA addressing constraints cannot cover the C-bit: unencrypted DMA is required.
3. Host memory encryption without an IOMMU, where the device can address the C-bit: unencrypted DMA is not required.

-aneesh