linux.git/drivers/usb/typec/ucsi, branch v5.8

usb: typec: ucsi: Fix the stub for ucsi_register_port_psy()

2020-04-27T11:41:45+00:00

The stub was ucsi_register_port() when it should have been
ucsi_register_port_psy().

Cc: Abhilash K V 
Fixes: 992a60ed0d5e ("usb: typec: ucsi: register with power_supply class")
Reported-by: kbuild test robot 
Signed-off-by: Heikki Krogerus 
Link: https://lore.kernel.org/r/20200427111246.4889-1-heikki.krogerus@linux.intel.com
Signed-off-by: Greg Kroah-Hartman

usb: typec: ucsi: register with power_supply class

2020-04-23T13:33:34+00:00

With this change the UCSI device will show up in
/sys/class/power_supply/. The following values are exported:
- online
- usb_type
- voltage_min
- voltage_max
- voltage_now
- current_max
- current_now

Once a PD-capable type-C power source is connected to the system, GET_PDOS
UCSI command is used to query all source capabilities. Request data object
(RDO) is used to get current values.

Signed-off-by: K V, Abhilash 
Signed-off-by: Heikki Krogerus 
Link: https://lore.kernel.org/r/20200423132058.6972-7-heikki.krogerus@linux.intel.com
Signed-off-by: Greg Kroah-Hartman

usb: typec: ucsi: save power data objects in PD mode

2020-04-23T13:33:34+00:00

When connected to a PD-capable power-source, read & save all partner
power data objects (PDOs) by using GET_PDOS UCSI command.
Also, save the current power contract in request data object (RDO)
for that connector.

Signed-off-by: K V, Abhilash 
Signed-off-by: Heikki Krogerus 
Link: https://lore.kernel.org/r/20200423132058.6972-6-heikki.krogerus@linux.intel.com
Signed-off-by: Greg Kroah-Hartman

usb: typec: ucsi: Correct bit-mask for CCI

2020-04-23T13:33:34+00:00

Bit 0 is reserved in CCI (Command Status & Connector Change Indicator)
register. So, change bit-mask for connector number field to 7..1
instead of 7..0.
There would be no functional change since we were anyways right-shifing
by 1 bit.

Signed-off-by: K V, Abhilash 
Signed-off-by: Heikki Krogerus 
Link: https://lore.kernel.org/r/20200423132058.6972-5-heikki.krogerus@linux.intel.com
Signed-off-by: Greg Kroah-Hartman

usb: typec: ucsi: replace magic numbers

2020-04-23T13:33:33+00:00

Replace magic numbers with macros in trace.h.

Signed-off-by: K V, Abhilash 
Signed-off-by: Heikki Krogerus 
Link: https://lore.kernel.org/r/20200423132058.6972-4-heikki.krogerus@linux.intel.com
Signed-off-by: Greg Kroah-Hartman

usb: typec: ucsi: Workaround for missed op_mode change

2020-04-23T13:33:33+00:00

EC firmware on Dell XPS & Latitude series does not set "Power Operation
Mode Change" bit in "Connector Status change" field of MESSAGE IN Data
while transitioning from type-C current to PD mode.

Instead the "Negotiated Power Level Change" bit is set when the "Power
Operation Mode" field shows the correct mode (i.e. PD).

This patch adds a check for this bit also, to trigger an update of
power operation mode in class driver, while handling GET_CONNECTOR_STATUS
command.

Signed-off-by: K V, Abhilash 
Signed-off-by: Heikki Krogerus 
Link: https://lore.kernel.org/r/20200423132058.6972-3-heikki.krogerus@linux.intel.com
Signed-off-by: Greg Kroah-Hartman

usb: typec: ucsi: set USB data role when partner type is power cable/ufp

2020-04-23T13:33:33+00:00

Currently UCSI framework doesn't update USB data role when partner type
is reported as power cable or power cable with ufp connected. This
results into no USB host mode functionality. This is valid usecase where
user wants to use legacy type c power cable with type a female connector
to attach different USB devices like mouse, thumb drive etc. Hence update
USB data role as host when partner type is reported as power cable or
power cable with ufp connected.

Signed-off-by: Mayank Rana 
Signed-off-by: Heikki Krogerus 
Link: https://lore.kernel.org/r/20200423132058.6972-2-heikki.krogerus@linux.intel.com
Signed-off-by: Greg Kroah-Hartman

Merge 5.6-rc7 into usb-next

2020-03-23T07:04:08+00:00

We need the USB fixes in here as well.

Signed-off-by: Greg Kroah-Hartman

usb: typec: ucsi_ccg: workaround for NVIDIA test device

2020-03-12T08:49:28+00:00

NVIDIA VirtualLink (svid 0x955) has two altmode, vdo=0x1 for
VirtualLink DP mode and vdo=0x3 for NVIDIA test mode. NVIDIA
test device FTB (Function Test Board) reports altmode list with
vdo=0x3 first and then vdo=0x1. The list is:
 SVID   VDO
0xff01  0xc05
0x28de  0x8085
0x955   0x3
0x955   0x1

Current logic to assign mode value is based on order
in altmode list. This causes a mismatch of CON and SOP altmodes
since NVIDIA GPU connector has order of vdo=0x1 first and then
vdo=0x3. Fixing this by changing the order of vdo values
reported by NVIDIA test device. the new list will be:

 SVID   VDO
0xff01  0xc05
0x28de  0x8085
0x955   0x1085
0x955   0x3

Also NVIDIA VirtualLink (svid 0x955) uses pin E for display mode.
NVIDIA test device reports vdo of 0x1 so make sure vdo values
always have pin E assignement.

Signed-off-by: Ajay Gupta 
Signed-off-by: Heikki Krogerus 
Link: https://lore.kernel.org/r/20200310121912.57879-1-heikki.krogerus@linux.intel.com
Signed-off-by: Greg Kroah-Hartman

usb: typec: ucsi: displayport: Fix a potential race during registration

2020-03-12T08:34:52+00:00

Locking the connector in ucsi_register_displayport() to make
sure that nothing can access the displayport alternate mode
before the function has finished and the alternate mode is
actually ready.

Fixes: af8622f6a585 ("usb: typec: ucsi: Support for DisplayPort alt mode")
Cc: stable@vger.kernel.org
Signed-off-by: Heikki Krogerus 
Link: https://lore.kernel.org/r/20200311130006.41288-3-heikki.krogerus@linux.intel.com
Signed-off-by: Greg Kroah-Hartman