Hi Dmitry,

On Tue, Oct 19, 2021 at 07:24:41PM +0300, Dmitry Baryshkov wrote:

On Tue, 27 Jul 2021 at 03:41, Matthias Kaehlcke [off-list ref] wrote:

quoted

This series adds:
- the onboard_usb_hub_driver
- glue in the xhci-plat driver to create and destroy the
  onboard_usb_hub platform devices if needed
- a device tree binding for the Realtek RTS5411 USB hub controller
- device tree changes that add RTS5411 entries for the QCA SC7180
  based boards trogdor and lazor
- a couple of stubs for platform device functions to avoid
  unresolved symbols with certain kernel configs

The main issue the driver addresses is that a USB hub needs to be
powered before it can be discovered. For discrete onboard hubs (an
example for such a hub is the Realtek RTS5411) this is often solved
by supplying the hub with an 'always-on' regulator, which is kind
of a hack. Some onboard hubs may require further initialization
steps, like changing the state of a GPIO or enabling a clock, which
requires even more hacks. This driver creates a platform device
representing the hub which performs the necessary initialization.
Currently it only supports switching on a single regulator, support
for multiple regulators or other actions can be added as needed.
Different initialization sequences can be supported based on the
compatible string.

I have the feeling that you might want to check if you can use pwrseq
subsystem being proposed at
https://lore.kernel.org/linux-arm-msm/20211006035407.1147909-1-dmitry.baryshkov@linaro.org/ (local).
It has been created for exactly the same reason of handling complex
power up/down requirements in a bus-neutral way. So instead of
creating an onboard-usb-hub, you might want to populate the hub node
with the reference to pwrseq device and make usb core call into
pwrseq. How does that sound to you?

Thanks for the pointer, it's good to see another attempt to sort out
power sequencing.

The pwrseq framework could potentially be used by the onboard_usb_hub
driver, but it probably can't replace it completely. Besides powering
the USB hub on before enumeration the driver also can optionally power
it off during system suspend when no wakeup capable USB devices are
connected, which can result in signifcant power savings on battery
powered devices. For this the driver needs knowledge about the USB
(hub) devices that are provided by a hub chip. That part is probably
best implemented by a driver under drivers/usb/.

It might be an option to have the USB core and the onboard_usb_hub
driver use the pwrseq framework, though we'd have to ensure that it
isn't a problem that the USB core turns power on (before
enumeration) and the onboard_usb_hub driver turns it off during
system suspend (and on again on resume).