Re: [PATCH v4 3/9] bus: mhi: Match devices exposing the protocol on the SAHARA channel

On 3/19/2026 12:31 AM, Kishore Batta wrote:
> Some Qualcomm devices expose the Sahara protocol on a generic SAHARA MHI
> channel rather than a QAIC specific channel name. As a result, the sahara

"Sahara"

> driver does not currently bind to such devices and never probes.
> 
> Extend the MHI device ID match table to also match the SAHARA channel
> name. This allows the Sahara protocol driver to bind to devices that
> expose the protocol directly on a standard sahara MHI channel.

"Sahara"

> 
> Signed-off-by: Kishore Batta <[email protected]>
> ---
>   drivers/bus/mhi/sahara/sahara.c | 2 ++
>   1 file changed, 2 insertions(+)
> 
> diff --git a/drivers/bus/mhi/sahara/sahara.c b/drivers/bus/mhi/sahara/sahara.c
> index 8ff7b6425ac5423ef8f32117151dca10397686a8..e3499977e7c6b53bc624a8eb00d0636f2ea63307 100644
> --- a/drivers/bus/mhi/sahara/sahara.c
> +++ b/drivers/bus/mhi/sahara/sahara.c
> @@ -911,8 +911,10 @@ static void sahara_mhi_dl_xfer_cb(struct mhi_device *mhi_dev, struct mhi_result
>   
>   static const struct mhi_device_id sahara_mhi_match_table[] = {
>   	{ .chan = "QAIC_SAHARA", },
> +	{ .chan = "SAHARA"},

This doesn't work and breaks bisect. At this point Sahara will bind to 
QDU100, ath12k, or something else but not know how to drive one of those 
devices.

Just add this when you add QDU100 support, which looks to be patch 5.

>   	{},
>   };
> +MODULE_DEVICE_TABLE(mhi, sahara_mhi_match_table);
>   
>   static struct mhi_driver sahara_mhi_driver = {
>   	.id_table = sahara_mhi_match_table,
>