Hello Sascha,

On Thu, Jan 04, 2024 at 03:17:45PM +0100, Sascha Hauer wrote:
> diff --git a/drivers/mci/imx-esdhc-pbl.c b/drivers/mci/imx-esdhc-pbl.c
> index c13830d726..2c74d101e4 100644
> --- a/drivers/mci/imx-esdhc-pbl.c
> +++ b/drivers/mci/imx-esdhc-pbl.c
> @@ -9,6 +9,8 @@
>  #include <asm/sections.h>
>  #include <asm/cache.h>
>  #include <mach/imx/xload.h>
> +#include <firmware.h>
> +#include <asm/atf_common.h>
>  #ifdef CONFIG_ARCH_IMX
>  #include <mach/imx/atf.h>
>  #include <mach/imx/imx6-regs.h>
> @@ -319,7 +321,8 @@ static int layerscape_esdhc_load_image(struct fsl_esdhc_host *host, void *adr, u
>  	val |= div_val;
>  	sdhci_write32(&host->sdhci, SDHCI_CLOCK_CONTROL__TIMEOUT_CONTROL__SOFTWARE_RESET, val);
>  
> -	sdhci_write32(&host->sdhci, ESDHC_DMA_SYSCTL, ESDHC_SYSCTL_DMA_SNOOP);
> +	sdhci_write32(&host->sdhci, ESDHC_DMA_SYSCTL,
> +		      ESDHC_SYSCTL_DMA_SNOOP | ESDHC_SYSCTL_PERIPHERAL_CLK_SEL);

This affects ls1046, too. Is this intended?

Best regards
Uwe

>  	ret = esdhc_read_blocks(host, adr, size);
>  	if (ret) {

-- 
Pengutronix e.K.                           | Uwe Kleine-König            |
Industrial Linux Solutions                 | https://www.pengutronix.de/ |