On Thu, Nov 15, 2018 at 09:46:26AM +0200, Felipe Balbi wrote: > > Hi Heikki, > > kbuild test robot <lkp@xxxxxxxxx> writes: > > > tree: https://git.kernel.org/pub/scm/linux/kernel/git/balbi/usb.git testing/next > > head: bad6c1502dac79c80ad5a7149fa308849c0191dd > > commit: 24e2238d8c102f242ece57f45fbeb4014929aad4 [17/22] usb: dwc3: drd: Register a USB role switch > > config: x86_64-randconfig-s4-11151335 (attached as .config) > > compiler: gcc-7 (Debian 7.3.0-1) 7.3.0 > > reproduce: > > git checkout 24e2238d8c102f242ece57f45fbeb4014929aad4 > > # save the attached .config to linux build tree > > make ARCH=x86_64 > > > > All errors (new ones prefixed by >>): > > > > drivers/usb/dwc3/drd.o: In function `dwc3_drd_exit': > >>> drivers/usb/dwc3/drd.c:604: undefined reference to `usb_role_switch_unregister' > > drivers/usb/dwc3/drd.o: In function `dwc3_drd_init': > >>> drivers/usb/dwc3/drd.c:563: undefined reference to `usb_role_switch_register' > >>> drivers/usb/dwc3/drd.c:563: undefined reference to `usb_role_switch_register' > > care to fix this? Have you picked the RFC patch that registers the drd switch in drivers/usb/dwc3/drd.c? I have not even tried to compile that code :-) I really did not want that patch to be picked by you. I was just presenting the idea to Chen Yu. It was probable wrong of me to send it as a patch, even though I did use the magic "RFC" letters. Sorry for that. Is it possible for you to drop the patch, or do you prefer that we just fix it? I don't think reverting the patch is a good idea. If dropping the patch at this stage is not possible anymore, then I think it's probable better to just fix the situation. thanks, -- heikki
