[PATCH 0/6] UCSI Power Supply Updates and Bug Fixes

Jameson Thies posted 6 patches 2 months, 1 week ago
drivers/usb/typec/ucsi/psy.c  | 41 ++++++++++++++++++++++++++++++++++-
drivers/usb/typec/ucsi/ucsi.c | 35 +++++++++++++++++++++++++-----
drivers/usb/typec/ucsi/ucsi.h |  4 ++++
3 files changed, 73 insertions(+), 7 deletions(-)
[PATCH 0/6] UCSI Power Supply Updates and Bug Fixes
Posted by Jameson Thies 2 months, 1 week ago
This series includes the following minor changes to power supply
handling by the UCSI driver.

1) Adds the "Status" property to power supplies registered by the UCSI
driver.
2) Reports power supply USB type as "DRP" when the port partner is a
DRP.
3) Updates ucsi.c to report a power supply change all power opmode
changes. Currently this only gets reported when opmode is PD.
4) Updates ucsi.c to report a power supply change when the PPM signals
a sink path change.
5) Set max current to 0 when no device is connected. Currently this
defaults to 0.1A.
6) Updates connector status after initiating a PR swap. Following the
UCSI spec, there is no reported change following a successful SET_PDR
command. This results in the stored connector status being outdated.

Abhishek Pandit-Subedi (1):
  usb: typec: ucsi: pr_swap should check connector_status

Jameson Thies (5):
  usb: typec: ucsi: psy: Add power supply status
  usb: typec: ucsi: psy: Add support for DRP USB type
  usb: typec: ucsi: Report power supply changes on power opmode changes
  usb: typec: ucsi: Report power supply change on sink path change
  usb: typec: ucsi: psy: Set max current to zero when disconnected

 drivers/usb/typec/ucsi/psy.c  | 41 ++++++++++++++++++++++++++++++++++-
 drivers/usb/typec/ucsi/ucsi.c | 35 +++++++++++++++++++++++++-----
 drivers/usb/typec/ucsi/ucsi.h |  4 ++++
 3 files changed, 73 insertions(+), 7 deletions(-)


base-commit: e40b984b6c4ce3f80814f39f86f87b2a48f2e662
-- 
2.51.0.618.g983fd99d29-goog
Re: [PATCH 0/6] UCSI Power Supply Updates and Bug Fixes
Posted by Greg KH 2 months ago
On Tue, Oct 07, 2025 at 12:00:01AM +0000, Jameson Thies wrote:
> This series includes the following minor changes to power supply
> handling by the UCSI driver.
> 
> 1) Adds the "Status" property to power supplies registered by the UCSI
> driver.
> 2) Reports power supply USB type as "DRP" when the port partner is a
> DRP.
> 3) Updates ucsi.c to report a power supply change all power opmode
> changes. Currently this only gets reported when opmode is PD.
> 4) Updates ucsi.c to report a power supply change when the PPM signals
> a sink path change.
> 5) Set max current to 0 when no device is connected. Currently this
> defaults to 0.1A.
> 6) Updates connector status after initiating a PR swap. Following the
> UCSI spec, there is no reported change following a successful SET_PDR
> command. This results in the stored connector status being outdated.
> 
> Abhishek Pandit-Subedi (1):
>   usb: typec: ucsi: pr_swap should check connector_status
> 
> Jameson Thies (5):
>   usb: typec: ucsi: psy: Add power supply status
>   usb: typec: ucsi: psy: Add support for DRP USB type
>   usb: typec: ucsi: Report power supply changes on power opmode changes
>   usb: typec: ucsi: Report power supply change on sink path change
>   usb: typec: ucsi: psy: Set max current to zero when disconnected

Please do not mix bug fixes and updates to the driver in the same patch
series, as that means the bug fixes will be delayed until the next
release, not the current one, as I can't take them all at once.

thanks,

greg k-h
Re: [PATCH 0/6] UCSI Power Supply Updates and Bug Fixes
Posted by Jameson Thies 2 months ago
Thank you for this feedback. The only patch here which is a clear bug
fix is 5/6. Currently the driver reports 0.1A max current when nothing
is connected on the port. I'll send that patch separately.
Re: [PATCH 0/6] UCSI Power Supply Updates and Bug Fixes
Posted by Greg KH 2 months ago
On Mon, Oct 13, 2025 at 01:59:23PM -0700, Jameson Thies wrote:
> Thank you for this feedback. The only patch here which is a clear bug
> fix is 5/6. Currently the driver reports 0.1A max current when nothing
> is connected on the port. I'll send that patch separately.

meta-comment, you lost all context here and I have no idea what you are
responding to, or what this is about :(

Remember, some of us get 1000+ emails a day to handle, context matters.

thanks,

greg k-h
Re: [PATCH 0/6] UCSI Power Supply Updates and Bug Fixes
Posted by Kenneth Crudup 2 months ago
On 10/6/25 17:00, Jameson Thies wrote:

> This series includes the following minor changes to power supply
> handling by the UCSI driver.
...
> base-commit: e40b984b6c4ce3f80814f39f86f87b2a48f2e662

I wanted to let you know that on my Dell XPS-9320, this patchset ended 
up spamming (i.e., hundreds) my dmesg with the following:

power_supply ucsi-source-psy-USBC000:002: driver reporting unavailable 
enum value 7

... which I believe to be POWER_SUPPLY_USB_TYPE_PD_DRP .

In my case it was coming from the call to 
power_supply_show_enum_with_available() on/around line 380 in 
.../drivers/power/supply/power_supply_sysfs.c ; I'd tried adding 
POWER_SUPPLY_USB_TYPE_PD_DRP to con->psy_desc.usb_types in 
ucsi_register_port_psy() in
.../drivers/usb/typec/ucsi/psy.c thinking that may fix it with no success.

LMK if you need any further info,

-Kenny

-- 
Kenneth R. Crudup / Sr. SW Engineer, Scott County Consulting, Orange 
County CA
Re: [PATCH 0/6] UCSI Power Supply Updates and Bug Fixes
Posted by Jameson Thies 2 months ago
Hi Kenneth,
You are correct that adding POWER_SUPPLY_USB_TYPE_PD_DRP in
ucsi_register_port_psy() is missing here. I would have expected that
adding it resolves the issue. I'll take a closer look, thanks for
raising this.
Re: [PATCH 0/6] UCSI Power Supply Updates and Bug Fixes
Posted by Kenneth Crudup 2 months ago

On 10/13/25 15:53, Jameson Thies wrote:

> You are correct that adding POWER_SUPPLY_USB_TYPE_PD_DRP in
> ucsi_register_port_psy() is missing here. I would have expected that
> adding it resolves the issue. 

The issue must be deeper than that, as I even went as far as making the 
check for "allowed types" pass all the time (as the dmesg was swamped 
making booting difficult so debugging was somewhat painful) and was 
still getting "enum" errors.

... that being said, I had a flight to catch on Friday so didn't keep 
working on a fix.

-Kenny

-- 
Kenneth R. Crudup / Sr. SW Engineer, Scott County Consulting, Orange 
County CA