Re: dwc3: run stop times out on dra7

[Felipe Balbi <balbi@kernel.org>]

[-- Attachment #1: Type: text/plain, Size: 2822 bytes --]


Hi Roger,

Roger Quadros <rogerq@ti.com> writes:
> on v5.5. on dra7 platform connected to PC host.
>
> root@dra7xx-evm:~# uname -a
> Linux dra7xx-evm 5.5.0 #551 SMP Tue Feb 4 17:10:23 EET 2020 armv7l armv7l armv7l GNU/Linux
> root@dra7xx-evm:~# modprobe g_zero
> [   26.176210] zero gadget: Gadget Zero, version: Cinco de Mayo 2008
> [   26.182384] zero gadget: zero ready
> root@dra7xx-evm:~# modprobe -r g_zero
> [   32.250704] ------------[ cut here ]------------
> [   32.255368] WARNING: CPU: 1 PID: 1278 at drivers/usb/gadget/composite.c:2024 __composite_unbind+0xc8/0xd8 [libcomposite]
> [   32.266330] Modules linked in: usb_f_ss_lb g_zero(-) libcomposite xhci_plat_hcd xhci_hcd usbcore dwc3 udc_core usb_common evdev spi_nor snd_soc_simple_card snd_soc_simple_card_utils leds_gpio led_class gpio_4
> [   32.322121] CPU: 1 PID: 1278 Comm: modprobe Not tainted 5.5.0 #551
> [   32.328330] Hardware name: Generic DRA74X (Flattened Device Tree)
> [   32.334467] [<c0112968>] (unwind_backtrace) from [<c010cb88>] (show_stack+0x10/0x14)
> [   32.342250] [<c010cb88>] (show_stack) from [<c08fee14>] (dump_stack+0xc0/0xe0)
> [   32.349509] [<c08fee14>] (dump_stack) from [<c013a3a8>] (__warn+0xd0/0xf8)
> [   32.356419] [<c013a3a8>] (__warn) from [<c013a478>] (warn_slowpath_fmt+0xa8/0xb8)
> [   32.363951] [<c013a478>] (warn_slowpath_fmt) from [<bf17cafc>] (__composite_unbind+0xc8/0xd8 [libcomposite])
> [   32.373872] [<bf17cafc>] (__composite_unbind [libcomposite]) from [<bf170474>] (usb_gadget_remove_driver+0x48/0x7c [udc_core])
> [   32.385329] [<bf170474>] (usb_gadget_remove_driver [udc_core]) from [<bf170540>] (usb_gadget_unregister_driver+0x98/0xd8 [udc_core])
> [   32.397303] [<bf170540>] (usb_gadget_unregister_driver [udc_core]) from [<c01d8774>] (sys_delete_module+0x1a0/0x248)
> [   32.407875] [<c01d8774>] (sys_delete_module) from [<c01011e0>] (__sys_trace_return+0x0/0x20)
> [   32.416348] Exception stack(0xecfdffa8 to 0xecfdfff0)
> [   32.421424] ffa0:                   000363d8 00036414 00036414 00000800 0000006c 00000000
> [   32.429637] ffc0: 000363d8 00036414 00000000 00000081 00000001 00000000 bea71c44 00000000
> [   32.437850] ffe0: b6e732f0 bea708b4 0001b3cf b6e732fc
> [   32.442958] ---[ end trace 17a4fad12463bfc1 ]---
>
> backtrace is because dwc3_gadget_run_stop() times out at pulldown
> and dwc3_gadget_pullup() returns -110
> So,
>
> usb_gadget_disconnect() does not call disconnect().
>
>          ret = gadget->ops->pullup(gadget, 0);
>          if (!ret) {
>                  gadget->connected = 0;
>                  gadget->udc->driver->disconnect(gadget);
>          }
>
> Any idea why controller stop might time out?

The only thing that comes to mind is that your PHY may be suspended.

Can you check that?

-- 
balbi

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 832 bytes --]