Re: [RFC PATCH 2/7] mmc: sdhci: Add quirk to indicate controller supports ADMA2

[Adrian Hunter]

On 07/08/17 19:01, Kishon Vijay Abraham I wrote:
> TI's implementation of sdhci controller used in DRA7 SoC's doesn't
> have SDHCI_CAN_DO_ADMA2 set in CAPA register though it supports
> ADMA2. Add quirk to support using ADMA2 even if the controller reports
> incorrect capability in CAPA.

A quirk is not needed for this.  Just call sdhci_read_caps() and change
host->caps before calling sdhci_add_host().

> 
> Signed-off-by: Kishon Vijay Abraham I <kishon@xxxxxx>
> ---
>  drivers/mmc/host/sdhci.c | 3 ++-
>  drivers/mmc/host/sdhci.h | 2 ++
>  2 files changed, 4 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/mmc/host/sdhci.c b/drivers/mmc/host/sdhci.c
> index ece3751d2a25..fff0baadbc3e 100644
> --- a/drivers/mmc/host/sdhci.c
> +++ b/drivers/mmc/host/sdhci.c
> @@ -3264,7 +3264,8 @@ int sdhci_setup_host(struct sdhci_host *host)
>  	}
>  
>  	if ((host->version >= SDHCI_SPEC_200) &&
> -		(host->caps & SDHCI_CAN_DO_ADMA2))
> +		((host->caps & SDHCI_CAN_DO_ADMA2) ||
> +		(host->quirks2 & SDHCI_QUIRK2_FORCE_ADMA)))
>  		host->flags |= SDHCI_USE_ADMA;
>  
>  	if ((host->quirks & SDHCI_QUIRK_BROKEN_ADMA) &&
> diff --git a/drivers/mmc/host/sdhci.h b/drivers/mmc/host/sdhci.h
> index 6905131f603d..d778034e324d 100644
> --- a/drivers/mmc/host/sdhci.h
> +++ b/drivers/mmc/host/sdhci.h
> @@ -437,6 +437,8 @@ struct sdhci_host {
>  #define SDHCI_QUIRK2_CLOCK_DIV_ZERO_BROKEN		(1<<15)
>  /* Controller does not have CRC stripped in Command Response */
>  #define SDHCI_QUIRK2_NO_CRC_STRIPPING			(1<<16)
> +/* Controller has bad caps bits, but really supports DMA */
> +#define SDHCI_QUIRK2_FORCE_ADMA				(1<<17)
>  
>  	int irq;		/* Device IRQ */
>  	void __iomem *ioaddr;	/* Mapped address */
>