On Fri, Feb 19, 2016 at 11:05 PM, Fabio Estevam <festevam@xxxxxxxxx> wrote: > On Fri, Feb 19, 2016 at 12:13 PM, Felipe Balbi <balbif@xxxxxxxxx> wrote: > >> ooooh, okay. So you have another problem, actually. Seems like ci_hdrc >> just shouldn't call your phy->suspend(), or your phy->suspend() >> shouldn't do anything. How about below? >> >> @@ -370,6 +370,9 @@ static int mxs_phy_suspend(struct usb_phy *x, int suspend) >> struct mxs_phy *mxs_phy = to_mxs_phy(x); >> bool low_speed_connection, vbus_is_on; >> >> + if (mxs_phy->data->flags & MXS_PHY_ABNORMAL_IN_SUSPEND) >> + return 0; /* or should we return -EBUSY ? */ >> + > > Tested with 0 and also -EBUSY. This code is called now, but it does > not fix the problem. Same log: > > [ 3.005881] hub 1-1:1.0: USB hub found > [ 3.010341] hub 1-1:1.0: 2 ports detected > [ 3.536362] usb 1-1: USB disconnect, device number 2 > [ 3.582732] usb usb1-port1: cannot reset (err = -32) > [ 3.588289] usb usb1-port1: cannot reset (err = -32) > [ 3.593546] usb usb1-port1: cannot reset (err = -32) > [ 3.598929] usb usb1-port1: cannot reset (err = -32) > [ 3.604188] usb usb1-port1: cannot reset (err = -32) > [ 3.609339] usb usb1-port1: Cannot enable. Maybe the USB cable is bad? > -- Fabio, Felipe is correct. The mx23 and mx28 should NOT call mxs phy's suspend API since the runtime suspend is not enabled for mx23 and mx28 at chipidea driver. Would you please check if CI_HDRC_SUPPORTS_RUNTIME_PM is set wrongly for mx23/mx28 at ci_hdrc_imx.c? If it does not been set, would please add WARN_ON(1) at mxs_phy_suspend to show call stack? -- BR, Peter Chen -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
