Re: [PATCH 6/6] platform/chrome: cros_ec_typec: Store cable plug type

[Greg Kroah-Hartman]

On Fri, Nov 06, 2020 at 12:59:07AM -0800, Prashant Malani wrote:
> Hi Greg,
> 
> Thanks for looking at the patch.
> 
> On Fri, Nov 06, 2020 at 08:20:59AM +0100, Greg Kroah-Hartman wrote:
> > On Thu, Nov 05, 2020 at 05:28:03PM -0800, Prashant Malani wrote:
> > > Use the PD VDO Type C cable plug type macro to retrieve and store the
> > > cable plug type in the cable descriptor.
> > > 
> > > Cc: Heikki Krogerus <heikki.krogerus@xxxxxxxxxxxxxxx>
> > > Cc: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
> > > Signed-off-by: Prashant Malani <pmalani@xxxxxxxxxxxx>
> > > ---
> > >  drivers/platform/chrome/cros_ec_typec.c | 21 ++++++++++++++++++++-
> > >  1 file changed, 20 insertions(+), 1 deletion(-)
> > 
> > Where are the first 5 patches in this series?
> 
> I believe I'd cc-ed you (and linux-usb) on the following patches in the
> series(modified git_cc[1] which adds individual patch get_maintainers.pl to the cover
> letter, patch 1 and patch 2):
> cover letter:
> https://lore.kernel.org/lkml/20201106012758.525472-1-pmalani@xxxxxxxxxxxx/
> patch 1/6 (cros-ec-typec cleanup):
> https://lore.kernel.org/lkml/20201106012758.525472-2-pmalani@xxxxxxxxxxxx/
> patch 2/6 (cros-ec-typec cleanup):
> https://lore.kernel.org/lkml/20201106012758.525472-2-pmalani@xxxxxxxxxxxx/
> patch 5/6 (PD VDO header patch):
> https://lore.kernel.org/lkml/20201106012758.525472-6-pmalani@xxxxxxxxxxxx/
> patch 6/6 (this patch):
> https://lore.kernel.org/lkml/20201106012758.525472-7-pmalani@xxxxxxxxxxxx/
> 
> Did you not receive these?

Ah, I got 1, 2, and 5, and now 6.  That's confusing, think about if you
were to receive such a series, what would you think to do with it?

> > > diff --git a/drivers/platform/chrome/cros_ec_typec.c b/drivers/platform/chrome/cros_ec_typec.c
> > > index 5e7f0b4ebbec..0a2a8b0f8115 100644
> > > --- a/drivers/platform/chrome/cros_ec_typec.c
> > > +++ b/drivers/platform/chrome/cros_ec_typec.c
> > > @@ -709,6 +709,7 @@ static int cros_typec_handle_sop_prime_disc(struct cros_typec_data *typec, int p
> > >  		.port = port_num,
> > >  		.partner_type = TYPEC_PARTNER_SOP_PRIME,
> > >  	};
> > > +	uint32_t cable_plug_type;
> > 
> > u32 please, this isn't userspace code :)
> 
> Will fix this in the next version.
> > 
> > >  	int ret = 0;
> > >  
> > >  	memset(disc, 0, EC_PROTO2_MAX_RESPONSE_SIZE);
> > > @@ -722,8 +723,26 @@ static int cros_typec_handle_sop_prime_disc(struct cros_typec_data *typec, int p
> > >  	/* Parse the PD identity data, even if only 0s were returned. */
> > >  	cros_typec_parse_pd_identity(&port->c_identity, disc);
> > >  
> > > -	if (disc->identity_count != 0)
> > > +	if (disc->identity_count != 0) {
> > > +		cable_plug_type = VDO_TYPEC_CABLE_TYPE(port->c_identity.vdo[0]);
> > > +		switch (cable_plug_type) {
> > > +		case CABLE_ATYPE:
> > > +			desc.type = USB_PLUG_TYPE_A;
> > > +			break;
> > > +		case CABLE_BTYPE:
> > > +			desc.type = USB_PLUG_TYPE_B;
> > > +			break;
> > > +		case CABLE_CTYPE:
> > > +			desc.type = USB_PLUG_TYPE_C;
> > > +			break;
> > > +		case CABLE_CAPTIVE:
> > > +			desc.type = USB_PLUG_CAPTIVE;
> > > +			break;
> > > +		default:
> > > +			desc.type = USB_PLUG_NONE;
> > > +		}
> > >  		desc.active = PD_IDH_PTYPE(port->c_identity.id_header) == IDH_PTYPE_ACABLE;
> > > +	}
> > 
> > So you save it but what happens with the value?
> 
> The type C connector class framework exposes it via syfs to user-space when we
> register the cable via typec_register_cable() in patch 4/6 [2].

So you added a new sysfs file and api without updating
Documentation/ABI/?  That's not good :(

> I'll go ahead and CC
> you and linux-usb on the entire series for the next version.

Please do, thanks!

greg k-h