On Thu, Nov 20, 2025 at 6:41 AM Manivannan Sadhasivam [off-list ref] wrote:

On Wed, Nov 19, 2025 at 02:08:33PM +0100, Loic Poulain wrote:

quoted

On Wed, Nov 19, 2025 at 12:27 PM Dmitry Baryshkov
[off-list ref] wrote:

quoted

On Wed, Nov 19, 2025 at 06:56:15PM +0800, Slark Xiao wrote:

quoted

T99W760 is designed based on Qualcomm SDX35 chip. It use similar
architechture with SDX72/SDX75 chip. So we need to assign initial
link id for this device to make sure network available.

Signed-off-by: Slark Xiao <redacted>
---
 drivers/net/wwan/mhi_wwan_mbim.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wwan/mhi_wwan_mbim.c b/drivers/net/wwan/mhi_wwan_mbim.c
index c814fbd756a1..a142af59a91f 100644
--- a/drivers/net/wwan/mhi_wwan_mbim.c
+++ b/drivers/net/wwan/mhi_wwan_mbim.c

@@ -98,7 +98,8 @@ static struct mhi_mbim_link *mhi_mbim_get_link_rcu(struct mhi_mbim_context *mbim
 static int mhi_mbim_get_link_mux_id(struct mhi_controller *cntrl)
 {
      if (strcmp(cntrl->name, "foxconn-dw5934e") == 0 ||
-         strcmp(cntrl->name, "foxconn-t99w515") == 0)
+         strcmp(cntrl->name, "foxconn-t99w515") == 0 ||
+         strcmp(cntrl->name, "foxconn-t99w760") == 0)

Can we replace this list of strinc comparisons with some kind of device
data, being set in the mhi-pci-generic driver?

If we move this MBIM-specific information into mhi-pci-generic, we
should consider using a software node (e.g. via
device_add_software_node) so that these properties can be accessed
through the generic device-property API.

MHI has to business in dealing with MBIM specific information as we already
concluded. So it should be handled within the WWAN subsystem.

it doesn’t make sense to include MBIM-specific fields in a generic MHI
structure. However, attaching fwnode properties could be reasonable
since the MHI PCI driver is responsible for device enumeration, and
that would keep device model specific handling fully covered in that
driver.

It’s fine to keep device-specific handling within WWAN/MBIM. However,
next time, please introduce a dedicated device data structure for the
mux-id instead of adding another strcmp.

Regards,
Loic