[alsa-devel] [PATCH v2] ASoC: bytcr_rt5640: Allow quirk set via module option

[Pierre-Louis Bossart]

On 04/21/2017 03:38 PM, Takashi Iwai wrote:
> The bytcr-rt5640 driver has a few quirk setups depending on the board,
> where the quirk value is set by DMI matching.  When you have a new
> device to add the support, you often experience to try the different
> quirk by trial-and-error.  Or, you may have a development model that
> still has no proper DMI string.  In either case, you'd need to compile
> the driver at each time.
>
> This patch introduces a module option to override the quirk value on
> the fly.  User can boot like snd-soc-sst-bytcr-rt5640.quirk=0x4004 to
> override the default value without recompilation.  It's a raw value,
> so user needs to check the source code for the meaning of each bit.
>
> Signed-off-by: Takashi Iwai <tiwai at suse.de>
> ---
> v1->v2: override the quirk value completely
>
>   sound/soc/intel/boards/bytcr_rt5640.c | 6 ++++++
>   1 file changed, 6 insertions(+)
>
> diff --git a/sound/soc/intel/boards/bytcr_rt5640.c b/sound/soc/intel/boards/bytcr_rt5640.c
> index 5c7219fb3aa8..23f5ffdb3731 100644
> --- a/sound/soc/intel/boards/bytcr_rt5640.c
> +++ b/sound/soc/intel/boards/bytcr_rt5640.c
> @@ -19,6 +19,7 @@
>   
>   #include <linux/init.h>
>   #include <linux/module.h>
> +#include <linux/moduleparam.h>
>   #include <linux/platform_device.h>
>   #include <linux/acpi.h>
>   #include <linux/device.h>
> @@ -58,6 +59,9 @@ struct byt_rt5640_private {
>   };
>   
>   static unsigned long byt_rt5640_quirk = BYT_RT5640_MCLK_EN;
> +static unsigned int quirk_override;
> +module_param_named(quirk, quirk_override, int, 0444);
> +MODULE_PARM_DESC(quirk, "Board-specific quirk override");
>   
>   static void log_quirks(struct device *dev)
>   {
> @@ -806,6 +810,8 @@ static int snd_byt_rt5640_mc_probe(struct platform_device *pdev)
>   
>   	/* check quirks before creating card */
>   	dmi_check_system(byt_rt5640_quirk_table);
> +	if (quirk_override)
> +		byt_rt5640_quirk = quirk_override;
>   	log_quirks(&pdev->dev);
sounds fine.
I would also change this log_quirks function to signal if the bits set 
by the user aren't aligned with the driver known quirks or are inconsistent.
I can add a patch if you want.

>   
>   	if ((byt_rt5640_quirk & BYT_RT5640_SSP2_AIF2) ||