On Tue, Sep 2, 2014 at 3:58 PM, Benjamin Tissoires <benjamin.tissoires@xxxxxxxxx> wrote: > Hi, > > On 09/02/2014 01:46 PM, Dylan Alex Simon wrote: >> Whenever either disconnecting the USB device or simply rmmod'ing the module >> (even when not in use), I get a kernel panic. I haven't managed to capture a >> backtrace, but at least the first two lines were saved after an rmmod: >> >> 18:53:17 kernel: thingm 0003:27B8:01ED.0004: hidraw3: USB HID v1.01 Device [ThingM blink(1) mk2] on usb-0000:00:12.2-3.1.4/input0 >> <snip, rmmod hid-thingm:> >> 08:38:42 kernel: BUG: unable to handle kernel paging request at fffffffb8a80aaf8 >> 08:38:42 kernel: IP: [<ffffffff8106e30c>] osq_lock+0x3c/0x110 >> >> Let me know if you'd like me to try to capture more info, but this problem >> seems very reproducible (at least with a mk2 device; I never had the problem on >> older kernels with a mk1). I do direct write()s to the hidraw device, but >> don't otherwise use the driver while it's loaded. >> >> Also at https://bugzilla.kernel.org/show_bug.cgi?id=83751 > > Do you happen to see a "unsupported firmware " error when plugging your device? > > If so, then the following patch should help with the panic (but you will an other one to be able to use again your device). > > --- > > From 791297375227b91990b37f94fc9de93156a3c21a Mon Sep 17 00:00:00 2001 > From: Benjamin Tissoires <benjamin.tissoires@xxxxxxxxxx> > Date: Tue, 2 Sep 2014 15:50:43 -0400 > Subject: [PATCH] HID: thingm: set the proper error code before leaving > > In case of an unsupported firmware, the driver bails out without setting > the LEDs interfaces, but forget to set the proper error code. > err is then still equal to 0 and the hid subsytem consider the device > to be in perfect shape. > When removing it, thingm_remove() tries to unbind the rgb LEDs which > has not been created, leading to a segfault. > > Signed-off-by: Benjamin Tissoires <benjamin.tissoires@xxxxxxxxxx> > --- > drivers/hid/hid-thingm.c | 1 + > 1 file changed, 1 insertion(+) > > diff --git a/drivers/hid/hid-thingm.c b/drivers/hid/hid-thingm.c > index 134be89..f91f971 100644 > --- a/drivers/hid/hid-thingm.c > +++ b/drivers/hid/hid-thingm.c > @@ -250,6 +250,7 @@ static int thingm_probe(struct hid_device *hdev, const struct hid_device_id *id) > > if (!tdev->fwinfo) { > hid_err(hdev, "unsupported firmware %c\n", tdev->version.major); > + err = -ENODEV; > goto stop; > } > > -- > 2.1.0 > Jiri, could you still consider including this one? If Thingm decides to create a firmware with 3.X, it will fail, so I guess this is still required, even if it does not fix Dylan's problem. Cheers, Benjamin -- To unsubscribe from this list: send the line "unsubscribe linux-input" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
