[PATCH 3/3] drm/ast: Warn about unsupported TX chips

Jocelyn Falempe jfalempe at redhat.com
Mon Sep 16 13:21:28 UTC 2024


On 16/09/2024 10:25, Thomas Zimmermann wrote:
> A number of TX chips are listed in VGACRD1, but not supported by
> the ast driver. Whether any existing product uses such a chip is
> unknown. Warn if the driver encounters any. We can then add
> support as necessary.

Thanks, it looks good to me.

Reviewed-by: Jocelyn Falempe <jfalempe at redhat.com>

> 
> Signed-off-by: Thomas Zimmermann <tzimmermann at suse.de>
> Suggested-by: Jocelyn Falempe <jfalempe at redhat.com>
> ---
>   drivers/gpu/drm/ast/ast_main.c | 17 ++++++++++++++++-
>   1 file changed, 16 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/gpu/drm/ast/ast_main.c b/drivers/gpu/drm/ast/ast_main.c
> index 7289bdc6066e..bc37c65305d4 100644
> --- a/drivers/gpu/drm/ast/ast_main.c
> +++ b/drivers/gpu/drm/ast/ast_main.c
> @@ -76,7 +76,22 @@ static void ast_detect_tx_chip(struct ast_device *ast, bool need_post)
>   	};
>   
>   	struct drm_device *dev = &ast->base;
> -	u8 jreg;
> +	u8 jreg, vgacrd1;
> +
> +	/*
> +	 * Several of the listed TX chips are not explicitly supported
> +	 * by the ast driver. If these exist in real-world devices, they
> +	 * are most likely reported as VGA or SIL164 outputs. We warn here
> +	 * to get bug reports for these devices. If none come in for some
> +	 * time, we can begin to fail device probing on these values.
> +	 */
> +	vgacrd1 = ast_get_index_reg_mask(ast, AST_IO_VGACRI, 0xd1, AST_IO_VGACRD1_TX_TYPE_MASK);
> +	drm_WARN(dev, vgacrd1 == AST_IO_VGACRD1_TX_ITE66121_VBIOS,
> +		 "ITE IT66121 detected, 0x%x, Gen%lu\n", vgacrd1, AST_GEN(ast));
> +	drm_WARN(dev, vgacrd1 == AST_IO_VGACRD1_TX_CH7003_VBIOS,
> +		 "Chrontel CH7003 detected, 0x%x, Gen%lu\n", vgacrd1, AST_GEN(ast));
> +	drm_WARN(dev, vgacrd1 == AST_IO_VGACRD1_TX_ANX9807_VBIOS,
> +		 "Analogix ANX9807 detected, 0x%x, Gen%lu\n", vgacrd1, AST_GEN(ast));
>   
>   	/* Check 3rd Tx option (digital output afaik) */
>   	ast->tx_chip = AST_TX_NONE;



More information about the dri-devel mailing list