On 6/9/2019 8:13 PM, Larry Finger wrote: > On 6/9/19 4:57 PM, Michael Büsch wrote: >> On Sun, 9 Jun 2019 17:44:10 -0400 >> H Buus <ubuntu@xxxxxxxxx> wrote: >> >>> I have an old 32 bit laptop with a BCM4401-B0 100Base-TX ethernet >>> controller. For every kernel from 4.19-rc1 going forward, I get a >>> warning and call trace within a few seconds of start up (see dmesg >>> snippet below). I have traced it to a specific commit (see commit >>> below). On the face of it, I would think it is a regression, but it >>> doesn't seem to cause a problem, since networking over ethernet is >>> working. >> >> >> This warning is not a problem. The commit just exposes a warning, that >> has always been there. >> I suggest we just remove the WARN_ON from ssb_gpio_init and >> ssb_gpio_unregister. >> I don't see a reason to throw a warning in that case. > > Michael, > > I agree. Do you want to prepare the patch, or should I? > > Larry Let me know if you would like me to verify a patch when/if it is available. Since kernels 4.19 thru 5.1 are unstable on this laptop, I would either need to test with a 4.18 or older kernel, or limit my testing to recovery mode & verifying that the b44 module is initialized without causing a warning & call trace. Unless I get lucky and figure out what commit is making newer kernels unstable on this laptop.
