On 3/18/26 12:15 AM, Val Packett wrote: > > On 3/16/26 12:23 AM, Dmitry Baryshkov wrote: >> On Sat, Mar 14, 2026 at 10:10:26PM -0300, Val Packett wrote: >>> On 3/14/26 9:51 PM, Val Packett wrote: >>>> On 3/13/26 10:09 PM, Dmitry Baryshkov wrote: >>>>> Currently, all HPD interrupt handling must go through the HPD state >>>>> machine. >>>>> >>>>> This has caused many issues where the DRM framework assumes that DP is >>>>> in one state while the state machine is stuck in another state. >>>>> >>>>> As discussed here [1], this series:
[...] > [ 3362.364893] typec_displayport port1-partner.1: failed to create symlinks > [ 3362.364902] typec_displayport port1-partner.1: probe with driver > typec_displayport failed with error -17 I believe this happens when the PAN_ACK doesn't arrive in time i.e. if pmic_glink_altmode.c doesn't write ALTMODE_PAN_ACK to the Type-C service, which is influenced by the time it takes for all the typec_mux_set() and typec_retimer_set() callbacks in the chain to complete (which are called by the various flavors of pmic_glink_altmode_enable_xyz() in pmic_glink_altmode_worker() right before the ack) Konrad
