When selecting a crypto cipher, we also need to select the
subsystem itself:
WARNING: unmet direct dependencies detected for CRYPTO_SHA1
Depends on [m]: CRYPTO [=m]
Selected by [y]:
- TEE [=y] && (HAVE_ARM_SMCCC [=n] || COMPILE_TEST [=y] || CPU_SUP_AMD [=y])
Selected by [m]:
- CRYPTO_DEV_QAT [=m] && CRYPTO [=m] && CRYPTO_HW [=y]
- CRYPTO_DEV_MEDIATEK [=m] && CRYPTO [=m] && CRYPTO_HW [=y] && (ARM && ARCH_MEDIATEK || COMPILE_TEST [=y])
- CRYPTO_DEV_SAFEXCEL [=m] && CRYPTO [=m] && CRYPTO_HW [=y] && (OF [=y] || PCI [=y] || COMPILE_TEST [=y]) && HAS_IOMEM [=y]
- CRYPTO_DEV_CCREE [=m] && CRYPTO [=m] && CRYPTO_HW [=y] && OF [=y] && HAS_DMA [=y]
- CRYPTO_DEV_SP_CCP [=y] && CRYPTO [=m] && CRYPTO_HW [=y] && CRYPTO_DEV_CCP [=y] && CRYPTO_DEV_CCP_DD [=m] && DMADEVICES [=y]
Fixes: e33bcbab16d1 ("tee: add support for session's client UUID generation")
Signed-off-by: Arnd Bergmann <[email protected]>
---
The regression was introduced in the soc tree, I'd pick this patch
up directly into that unless someone sees a problem
---
drivers/tee/Kconfig | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/tee/Kconfig b/drivers/tee/Kconfig
index 806eb87d4da0..e99d840c2511 100644
--- a/drivers/tee/Kconfig
+++ b/drivers/tee/Kconfig
@@ -3,6 +3,7 @@
config TEE
tristate "Trusted Execution Environment support"
depends on HAVE_ARM_SMCCC || COMPILE_TEST || CPU_SUP_AMD
+ select CRYPTO
select CRYPTO_SHA1
select DMA_SHARED_BUFFER
select GENERIC_ALLOCATOR
--
2.26.2
Hi Arnd,
On Wed, May 27, 2020 at 3:39 PM Arnd Bergmann <[email protected]> wrote:
>
> When selecting a crypto cipher, we also need to select the
> subsystem itself:
>
> WARNING: unmet direct dependencies detected for CRYPTO_SHA1
> Depends on [m]: CRYPTO [=m]
> Selected by [y]:
> - TEE [=y] && (HAVE_ARM_SMCCC [=n] || COMPILE_TEST [=y] || CPU_SUP_AMD [=y])
> Selected by [m]:
> - CRYPTO_DEV_QAT [=m] && CRYPTO [=m] && CRYPTO_HW [=y]
> - CRYPTO_DEV_MEDIATEK [=m] && CRYPTO [=m] && CRYPTO_HW [=y] && (ARM && ARCH_MEDIATEK || COMPILE_TEST [=y])
> - CRYPTO_DEV_SAFEXCEL [=m] && CRYPTO [=m] && CRYPTO_HW [=y] && (OF [=y] || PCI [=y] || COMPILE_TEST [=y]) && HAS_IOMEM [=y]
> - CRYPTO_DEV_CCREE [=m] && CRYPTO [=m] && CRYPTO_HW [=y] && OF [=y] && HAS_DMA [=y]
> - CRYPTO_DEV_SP_CCP [=y] && CRYPTO [=m] && CRYPTO_HW [=y] && CRYPTO_DEV_CCP [=y] && CRYPTO_DEV_CCP_DD [=m] && DMADEVICES [=y]
>
> Fixes: e33bcbab16d1 ("tee: add support for session's client UUID generation")
> Signed-off-by: Arnd Bergmann <[email protected]>
> ---
> The regression was introduced in the soc tree, I'd pick this patch
> up directly into that unless someone sees a problem
Thanks for taking care of this, please go ahead.
Cheers,
Jens
> ---
> drivers/tee/Kconfig | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/drivers/tee/Kconfig b/drivers/tee/Kconfig
> index 806eb87d4da0..e99d840c2511 100644
> --- a/drivers/tee/Kconfig
> +++ b/drivers/tee/Kconfig
> @@ -3,6 +3,7 @@
> config TEE
> tristate "Trusted Execution Environment support"
> depends on HAVE_ARM_SMCCC || COMPILE_TEST || CPU_SUP_AMD
> + select CRYPTO
> select CRYPTO_SHA1
> select DMA_SHARED_BUFFER
> select GENERIC_ALLOCATOR
> --
> 2.26.2
>
Hi Arnd & Jens,
On 2020-05-27 18:07, Jens Wiklander wrote:
> Hi Arnd,
>
> On Wed, May 27, 2020 at 3:39 PM Arnd Bergmann <[email protected]> wrote:
>>
>> When selecting a crypto cipher, we also need to select the
>> subsystem itself:
>>
>> WARNING: unmet direct dependencies detected for CRYPTO_SHA1
>> Depends on [m]: CRYPTO [=m]
>> Selected by [y]:
>> - TEE [=y] && (HAVE_ARM_SMCCC [=n] || COMPILE_TEST [=y] || CPU_SUP_AMD [=y])
>> Selected by [m]:
>> - CRYPTO_DEV_QAT [=m] && CRYPTO [=m] && CRYPTO_HW [=y]
>> - CRYPTO_DEV_MEDIATEK [=m] && CRYPTO [=m] && CRYPTO_HW [=y] && (ARM && ARCH_MEDIATEK || COMPILE_TEST [=y])
>> - CRYPTO_DEV_SAFEXCEL [=m] && CRYPTO [=m] && CRYPTO_HW [=y] && (OF [=y] || PCI [=y] || COMPILE_TEST [=y]) && HAS_IOMEM [=y]
>> - CRYPTO_DEV_CCREE [=m] && CRYPTO [=m] && CRYPTO_HW [=y] && OF [=y] && HAS_DMA [=y]
>> - CRYPTO_DEV_SP_CCP [=y] && CRYPTO [=m] && CRYPTO_HW [=y] && CRYPTO_DEV_CCP [=y] && CRYPTO_DEV_CCP_DD [=m] && DMADEVICES [=y]
>>
>> Fixes: e33bcbab16d1 ("tee: add support for session's client UUID generation")
>> Signed-off-by: Arnd Bergmann <[email protected]>
>> ---
>> The regression was introduced in the soc tree, I'd pick this patch
>> up directly into that unless someone sees a problem
>
> Thanks for taking care of this, please go ahead.
>
> Cheers,
> Jens
I am also OK for the change. Sorry that we did not notice this during
the review.
Feel free to merge it to the original commit or leave as separate commit.
Reviewed-by: Vesa Jääskeläinen <[email protected]>
>
>> ---
>> drivers/tee/Kconfig | 1 +
>> 1 file changed, 1 insertion(+)
>>
>> diff --git a/drivers/tee/Kconfig b/drivers/tee/Kconfig
>> index 806eb87d4da0..e99d840c2511 100644
>> --- a/drivers/tee/Kconfig
>> +++ b/drivers/tee/Kconfig
>> @@ -3,6 +3,7 @@
>> config TEE
>> tristate "Trusted Execution Environment support"
>> depends on HAVE_ARM_SMCCC || COMPILE_TEST || CPU_SUP_AMD
>> + select CRYPTO
>> select CRYPTO_SHA1
>> select DMA_SHARED_BUFFER
>> select GENERIC_ALLOCATOR
>> --
>> 2.26.2
>>