Hi,
Agreed on the main finding - skb->data cannot be NULL there. Same
conclusion as in the thread with Loic:
https://lore.kernel.org/netdev/20261102105307.11263-1-fangxy@xiaopeng.com/ (local)
I'll drop this version and send a cleanup that removes the
unreachable check instead, without a Fixes: tag so stable does not
pick it up as a leak fix.
The pre-existing findings on the missing skb->len checks and the
unchecked next_cmd_index walk in ipc_mux_dl_acb_decode() are being
addressed in a series I'm preparing.
pw-bot: cr