From: Loic Poulain <loic.poulain@xxxxxxxxxx> Hi Loic, Thank you very much. > On Fri, 3 Nov 2023 at 11:41, Jiri Pirko <jiri@xxxxxxxxxxx> wrote: > > > > > > Mon, Sep 18, 2023 at 08:56:26AM CEST, SongJinJian@xxxxxxxxxxx wrote: > > >Tue, Sep 12, 2023 at 11:48:40AM CEST, songjinjian@xxxxxxxxxxx wrote: > > >>>Adds support for t7xx wwan device firmware flashing & coredump > > >>>collection using devlink. > > > > > >>I don't believe that use of devlink is correct here. It seems like a misfit. IIUC, what you need is to communicate with the modem. Basically a communication channel to modem. The other wwan drivers implement these channels in _ctrl.c files, using multiple protocols. Why can't you do something similar and let devlink out of this please? > > > > > >>Until you put in arguments why you really need devlink and why is it a good fit, I'm against this. Please don't send any other versions of this patchset that use devlink. > > > > > > Yes, t7xx driver need communicate with modem with a communication channel to modem. > > > I took a look at the _ctrl.c files under wwan directory, it seemed the implementation can be well integrated with QualCommon's modem, if we do like this, I think we need modem firmware change, maybe not be suitable for current MTK modem directly. > > > Except for Qualcomm modem driver, there is also an Intel wwan driver 'iosm' and it use devlink to implement firmware flash(https://www.kernel.org/doc/html/latest/networking/devlink/iosm.html), Intel and MTK design and use devlink to do this work on > > > > If that exists, I made a mistake as a gatekeeper. That usage looks > > wrong. > > > > > 'mtk_t7xx' driver and I continue to do this work. > > > > > > I think devlink framework can support this scene and if we use devlink we don't need to develop other flash tools or other user space applications, use upstream devlink commands directly. > > > > Please don't. > So this is clear that devlink should not be used for this wwan firmware upgrade, if you still want to abstract the fastboot protocol part, maybe the easier would be to move on the generic firmware framework, and especially the firmware upload API which seems to be a good fit here? https://docs.kernel.org/driver-api/firmware/fw_upload.html#firmware-upload-api 1.This api seemed fit here, but I haven't find the refer to use the API, codes in /lib/test_firmware.c shown some intruduce, I think if I'm consider how to implement ops.prepare(what to verify, it seemed modem will do that) and ops.poll_complete? And it seemed request_firmware API also can recieve the data from use space, is it a way to use sysfs to trigger request firmware to kernel? In addition to this, I may have to create sysfs interface to pass the firmware partition parameter.And find a nother way to export the coredump port to user space. 2.How about we add a new WWAN port type, abstract fastboot and dump channel, like WWAN_PORT_XXX, then use this port with WWAN framework to handle firmware ops and dump ops. Hope to get your advice, thanks very much. Regards, Jinjian
