Re: [RFC PATCH 0/2] usb: Link USB devices with their USB Type-C partner counterparts

[Heikki Krogerus <heikki.krogerus@linux.intel.com>]

On Tue, Aug 22, 2023 at 05:24:44PM -0700, Benson Leung wrote:
> Hi Heikki,
> 
> On Tue, Aug 22, 2023 at 6:32 AM Heikki Krogerus
> <heikki.krogerus@linux.intel.com> wrote:
> >
> > Hi Benson,
> >
> > RFC for now. I can't test these properly. If you guys could take over
> > this, I would much appreciated. I hope this is at least close to your
> > proposal.
> >
> > With this (or something like it) you should be able to get
> > notification about USB connections and disconnections to your port
> > driver by implementing the new "attach" and "deattach" callbacks in
> > struct typec_partner_desc. The typec partner devices will also have
> > symlinks to the enumerated USB devices and vise versa.
> >
> > I took a little shortcut and did not implement a proper device list.
> > Instead there is now only a member for the USB2 device and a member
> > for the USB3 device in struct typec_port, so with this only USB is
> > supported. But the API does not deal with struct usb_device, so
> > extending this to support other devices (TBT, Displayport, etc.) by
> > adding the actual device list should be fairly easy.
> >
> > thanks,
> >
> > Heikki Krogerus (2):
> >   usb: typec: Link enumerated USB devices with Type-C partner
> >   usb: Inform the USB Type-C class about enumerated devices
> >
> >  drivers/usb/core/hub.c          |   4 ++
> >  drivers/usb/core/hub.h          |   3 +
> >  drivers/usb/core/port.c         |  19 +++++-
> >  drivers/usb/typec/class.c       | 108 +++++++++++++++++++++++++++++---
> >  drivers/usb/typec/class.h       |  16 +++++
> >  drivers/usb/typec/port-mapper.c |   9 ++-
> >  include/linux/usb/typec.h       |  37 +++++++++++
> >  7 files changed, 184 insertions(+), 12 deletions(-)
> >
> > --
> > 2.40.1
> >
> 
> Tested-by: Benson Leung <bleung@chromium.org>
> 
> 
> I picked these two changes back to my Brya/Redrix Chromebook which has
> the PLD changes to link subsystems.
> 
> First I plugged in a USB-C to USB-A receptacle adapter with a USB3
> thumbdrive into port0, and went to the port0-partner path.
> 
> redrix-rev3 /sys/class/typec/port0-partner # ls -lh
> total 0
> lrwxrwxrwx. 1 root root    0 Aug 22 17:16 2-1 ->
> ../../../../../../../0000:00:0d.0/usb2/2-1
> -r--r--r--. 1 root root 4.0K Aug 22 17:14 accessory_mode
> lrwxrwxrwx. 1 root root    0 Aug 22 17:14 device -> ../../port0
> drwxr-xr-x. 2 root root    0 Aug 22 17:14 identity
> drwxr-xr-x. 2 root root    0 Aug 22 17:14 power
> lrwxrwxrwx. 1 root root    0 Aug 22 17:14 subsystem ->
> ../../../../../../../../../class/typec
> -r--r--r--. 1 root root 4.0K Aug 22 17:14 supports_usb_power_delivery
> -r--r--r--. 1 root root 4.0K Aug 22 17:14 type
> -rw-r--r--. 1 root root 4.0K Aug 22 17:14 uevent
> -r--r--r--. 1 root root 4.0K Aug 22 17:14 usb_power_delivery_revision
> 
> 2-1 symlink appears, which is the SuperSpeed usb device associated
> with the thumbdrive.
> Unplugging the USB3 thumbdrive without unplugging the C-to-A adapter,
> and then plugging in a USB2.0 security key:
> 
> redrix-rev3 /sys/class/typec/port0-partner # ls -lh
> total 0
> lrwxrwxrwx. 1 root root    0 Aug 22 17:19 3-1 ->
> ../../../../../../../0000:00:14.0/usb3/3-1
> -r--r--r--. 1 root root 4.0K Aug 22 17:14 accessory_mode
> lrwxrwxrwx. 1 root root    0 Aug 22 17:14 device -> ../../port0
> drwxr-xr-x. 2 root root    0 Aug 22 17:14 identity
> drwxr-xr-x. 2 root root    0 Aug 22 17:14 power
> lrwxrwxrwx. 1 root root    0 Aug 22 17:14 subsystem ->
> ../../../../../../../../../class/typec
> -r--r--r--. 1 root root 4.0K Aug 22 17:14 supports_usb_power_delivery
> -r--r--r--. 1 root root 4.0K Aug 22 17:14 type
> -rw-r--r--. 1 root root 4.0K Aug 22 17:14 uevent
> -r--r--r--. 1 root root 4.0K Aug 22 17:14 usb_power_delivery_revision
> 
> 2-1 node disappears. 3-1 appears
> 
> Unplugging the adapter, plugging in a USB4 hub:
> redrix-rev3 /sys/class/typec/port0-partner # ls -lh
> total 0
> lrwxrwxrwx. 1 root root    0 Aug 22 17:21 2-1 ->
> ../../../../../../../0000:00:0d.0/usb2/2-1
> lrwxrwxrwx. 1 root root    0 Aug 22 17:21 3-1 ->
> ../../../../../../../0000:00:14.0/usb3/3-1
> -r--r--r--. 1 root root 4.0K Aug 22 17:21 accessory_mode
> lrwxrwxrwx. 1 root root    0 Aug 22 17:21 device -> ../../port0
> drwxr-xr-x. 2 root root    0 Aug 22 17:21 identity
> -r--r--r--. 1 root root 4.0K Aug 22 17:21 number_of_alternate_modes
> drwxr-xr-x. 5 root root    0 Aug 22 17:21 pd0
> drwxr-xr-x. 4 root root    0 Aug 22 17:21 port0-partner.0
> drwxr-xr-x. 2 root root    0 Aug 22 17:21 power
> lrwxrwxrwx. 1 root root    0 Aug 22 17:21 subsystem ->
> ../../../../../../../../../class/typec
> -r--r--r--. 1 root root 4.0K Aug 22 17:21 supports_usb_power_delivery
> -r--r--r--. 1 root root 4.0K Aug 22 17:21 type
> -rw-r--r--. 1 root root 4.0K Aug 22 17:21 uevent
> lrwxrwxrwx. 1 root root    0 Aug 22 17:21 usb_power_delivery -> pd0
> -r--r--r--. 1 root root 4.0K Aug 22 17:21 usb_power_delivery_revision
> 
> Both 2-1 and 3-1 are linked.
> 
> Thanks so much for this, Heikki! I can look a little closer at the
> attach and deattach callbacks in our typec port driver in a little
> while.

Cool! Thank you!

-- 
heikki