Re: [PATCH 1/2] venus: kconfig: Fix compile-testing on x86 platforms

[Bjorn Andersson <andersson@kernel.org>]

On Wed, Jul 13, 2022 at 02:03:50PM +0300, Stanimir Varbanov wrote:
> Fix Venus driver COMPILE_TEST compilation on x86 platform by
> adding a dependacy on V4L_PLATFORM_DRIVERS and select QCOM_SMEM
> instead of depending on it.
> 
> Signed-off-by: Stanimir Varbanov <stanimir.varbanov@linaro.org>
> ---
>  drivers/media/platform/qcom/venus/Kconfig | 4 +++-
>  1 file changed, 3 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/media/platform/qcom/venus/Kconfig b/drivers/media/platform/qcom/venus/Kconfig
> index bfd50e8f3421..95814b175c3e 100644
> --- a/drivers/media/platform/qcom/venus/Kconfig
> +++ b/drivers/media/platform/qcom/venus/Kconfig
> @@ -1,12 +1,14 @@
>  config VIDEO_QCOM_VENUS
>  	tristate "Qualcomm Venus V4L2 encoder/decoder driver"
> +	depends on V4L_PLATFORM_DRIVERS
>  	depends on V4L_MEM2MEM_DRIVERS
> -	depends on VIDEO_DEV && QCOM_SMEM
> +	depends on VIDEO_DEV
>  	depends on (ARCH_QCOM && IOMMU_DMA) || COMPILE_TEST
>  	select QCOM_MDT_LOADER if ARCH_QCOM
>  	select QCOM_SCM
>  	select VIDEOBUF2_DMA_CONTIG
>  	select V4L2_MEM2MEM_DEV
> +	select QCOM_SMEM

QCOM_SMEM is user selectable, and it's a fairly broadly used API. So I
would prefer that we introduce stub functions in linux/soc/qcom/smem.h.

Regards,
Bjorn

>  	help
>  	  This is a V4L2 driver for Qualcomm Venus video accelerator
>  	  hardware. It accelerates encoding and decoding operations
> -- 
> 2.25.1
>