On Wed, May 22, 2019 at 10:55:17AM +0000, Biju Das wrote: > Hi Chunfeng Yun, > > Thanks for the feedback. > > > Subject: RE: [PATCH v5 4/6] usb: roles: add API to get usb_role_switch by > > node > > > > Hi Biju, > > On Wed, 2019-05-22 at 08:05 +0000, Biju Das wrote: > > > Hi Heikki, > > > > > > Thanks for the feedback. > > > > > > > Subject: Re: [PATCH v5 4/6] usb: roles: add API to get > > > > usb_role_switch by node > > > > > > > > On Mon, May 20, 2019 at 09:45:46AM +0000, Biju Das wrote: > > > > > > > > > > > > > > > Hi Heikki, > > > > > > > > > > Thanks for the feedback. > > > > > > > > > > > Subject: Re: [PATCH v5 4/6] usb: roles: add API to get > > > > > > usb_role_switch by node > > > > > > > > > > > > On Mon, May 20, 2019 at 08:06:41AM +0000, Biju Das wrote: > > > > > > > Hi Heikki, > > > > > > > > > > > > > > > Subject: Re: [PATCH v5 4/6] usb: roles: add API to get > > > > > > > > usb_role_switch by node > > > > > > > > > > > > > > > > On Mon, May 20, 2019 at 10:39:11AM +0800, Chunfeng Yun wrote: > > > > > > > > > Hi, > > > > > > > > > On Fri, 2019-05-17 at 16:05 +0300, Heikki Krogerus wrote: > > > > > > > > > > Hi, > > > > > > > > > > > > > > > > > > > > On Fri, May 17, 2019 at 01:37:36PM +0300, Heikki Krogerus > > wrote: > > > > > > > > > > > On Tue, May 14, 2019 at 04:47:21PM +0800, Chunfeng Yun > > > > wrote: > > > > > > > > > > > > Add fwnode_usb_role_switch_get() to make easier to > > > > > > > > > > > > get usb_role_switch by fwnode which register it. > > > > > > > > > > > > It's useful when there is not device_connection > > > > > > > > > > > > registered between two drivers and only knows the > > > > > > > > > > > > fwnode which register usb_role_switch. > > > > > > > > > > > > > > > > > > > > > > > > Signed-off-by: Chunfeng Yun > > > > > > > > > > > > > > > > > > > > > > > > Tested-by: Biju Das > > > > > > > > > > > > > > > > > > > > > > Acked-by: Heikki Krogerus > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > Hold on. I just noticed Rob's comment on patch 2/6, > > > > > > > > > > where he points out that you don't need to use device > > > > > > > > > > graph since the controller is the parent of the > > > > > > > > > > connector. Doesn't that mean you don't really need this API? > > > > > > > > > No, I still need it. > > > > > > > > > The change is about the way how to get fwnode; when use > > > > > > > > > device graph, get fwnode by of_graph_get_remote_node(); > > > > > > > > > but now will get fwnode by of_get_parent(); > > > > > > > > > > > > > > > > OK, I get that, but I'm still not convinced about if > > > > > > > > something like this function is needed at all. I also have > > > > > > > > concerns regarding how you are using the function. I'll > > > > > > > > explain in comment to the patch 5/6 in this > > > > > > series... > > > > > > > > > > > > > > FYI, Currently I am also using this api in my patch series. > > > > > > > https://patchwork.kernel.org/patch/10944637/ > > > > > > > > > > > > Yes, and I have the same question for you I jusb asked in > > > > > > comment I added to the patch 5/6 of this series. Why isn't > > > > > > usb_role_switch_get() > > > > enough? > > > > > > > > > > Currently no issue. It will work with this api as well, since the > > > > > port node is > > > > part of controller node. > > > > > For eg:- > > > > > https://patchwork.kernel.org/patch/10944627/ > > > > > > > > > > However if any one adds port node inside the connector node, then > > > > > this > > > > api may won't work as expected. > > > > > Currently I get below error > > > > > > > > > > [ 2.299703] OF: graph: no port node found in > > > > /soc/i2c@e6500000/hd3ss3220@47 > > > > > > > > We need to understand why is that happening? > > > > > > > > > > Form the stack trace the parent node is "parent_node=hd3ss3220@47" , > > instead of the "connector" node. > > > That is the reason for the above error. > > > > > > [ 2.442429] of_graph_get_next_endpoint.part.0+0x28/0x168 > > > [ 2.447889] of_fwnode_graph_get_next_endpoint+0x5c/0xb0 > > > [ 2.453267] fwnode_graph_get_next_endpoint+0x20/0x30 > > > [ 2.458374] device_connection_find_match+0x74/0x1a0 > > > [ 2.463399] usb_role_switch_get+0x20/0x28 > > > [ 2.467542] hd3ss3220_probe+0xc4/0x218 > > > > > > The use case is > > > > > > &i2c0 { > > > hd3ss3220@47 { > > > compatible = "ti,hd3ss3220"; > > > > > > usb_con: connector { > > > compatible = "usb-c-connector"; > > > port { > > > hd3ss3220_ep: endpoint { > > > remote-endpoint = > > <&usb3_role_switch>; > > > }; > > > }; > > > }; > > > }; > > > }; > > > > > > &usb3_peri0 { > > > companion = <&xhci0>; > > > usb-role-switch; > > > > > > port { > > > usb3_role_switch: endpoint { > > > remote-endpoint = <&hd3ss3220_ep>; > > > }; > > > }; > > > }; > > > > > > Q1) How do we modify the usb_role_switch_get() function to search > > > Child(connector) and child's endpoint? > > How about firstly finding connector node in fwnode_graph_devcon_match(), > > then search each endpoint? > > I have done a quick prototyping with the changes you suggested and it works. > > - struct fwnode_handle *ep; > + struct fwnode_handle *ep,*child,*tmp = fwnode; > > - fwnode_graph_for_each_endpoint(fwnode, ep) { > + child = fwnode_get_named_child_node(fwnode, "connector"); > + if (child) > + tmp = child; > + > + fwnode_graph_for_each_endpoint(tmp, ep) { > > Form the stack trace the parent node is "parent_node= connector" . > > [ 2.440922] of_graph_get_next_endpoint.part.0+0x28/0x168 > [ 2.446381] of_fwnode_graph_get_next_endpoint+0x5c/0xb0 > [ 2.451758] fwnode_graph_get_next_endpoint+0x20/0x30 > [ 2.456866] device_connection_find_match+0x84/0x1c0 > [ 2.461888] usb_role_switch_get+0x20/0x28 > > Heikki, > Are you ok with the above changes? Doesn't that mean that if we made fwnode_usb_role_switch_get() the way I proposed, there is no problem? You just find the "connector" child node in your driver, and pass that to fwnode_usb_role_switch_get(): struct fwnode_handle *connector; ... connector = device_get_named_child_node(&client->dev, "connector"); if (IS_ERR(connector)) hd3ss3220->role_sw = fwnode_usb_role_switch_get(connector); ... The difference is that instead of just converting a device node of an usb role switch to the usb role switch, it works just like usb_role_switch_get(), just taking fwnode instead of device entry as parameter. I prepared the patches implementing fwnode_usb_role_switch_get() the way I though it needs to work for my own tests. Please find the patches attached. thanks, -- heikki