Re: [PATCH] dmaengine: ste_dma40: allow COMPILE_TEST

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Low] kconfig: STE_DMA40 missing HAS_IOMEM dependency for COMPILE_TEST
--

commit f5b43b7cb15b71f44f87940150c9d52817cad404
Author: Rosen Penev <rosenp@gmail.com>

dmaengine: ste_dma40: allow COMPILE_TEST

This commit loosens the Kconfig dependencies for STE_DMA40 to allow building
on other architectures via COMPILE_TEST. It aims to increase compile coverage
by exposing the driver to more build configurations.

> diff --git a/drivers/dma/Kconfig b/drivers/dma/Kconfig
> index ae6a682c9f769..f16bd4059d84a 100644
> --- a/drivers/dma/Kconfig
> +++ b/drivers/dma/Kconfig
> @@ -559,7 +559,7 @@ config SOPHGO_CV1800B_DMAMUX
>  
>  config STE_DMA40
>  	bool "ST-Ericsson DMA40 support"
> -	depends on ARCH_U8500
> +	depends on ARCH_U8500 || COMPILE_TEST
>  	select DMA_ENGINE
>  	select SRAM

[Severity: Low]
Does this also need a dependency on HAS_IOMEM?

By allowing COMPILE_TEST in drivers/dma/Kconfig for STE_DMA40 and keeping the
select for SRAM, architectures without HAS_IOMEM (like User-Mode Linux or
some s390 configurations) will trigger unmet dependency build warnings for
SRAM when COMPILE_TEST is enabled.

This can also lead to compilation failures on those architectures due to
missing I/O memory accessors like readl or writel that SRAM might depend on.

Could this be resolved by making STE_DMA40 depend on HAS_IOMEM?

>  	help

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260916225241.280606-1-rosenp@gmail.com?part=1