Open advisoft opened 8 years ago
Having a named stack trace would be a good step. Any possibility of running a newer kernel? The current version of net/wireless/core.c contains no BUG() statements, only WARN() lines. In the old days, some authors went overboard and improperly used BUG() when WARN() was appropriate. An out-of=order shutdown sequence for a device is not usually a good reason to crash the system.
It would be good to move to a newer kernel, but too much time pressure at this point in time. I guess the best approach would be to look at the latest version of that function, and patch the current one if everything makes sense, and be done with it. If further investigation is of benefit to you, I am happy to look into it further.
Thanks again for your quick response.
I have upgraded the kernel to 4.6.5, and the crash is swapped for a warning. However it still hangs on shutdown if the interface associated with the driver is up. None of the other wifi devices seem to do this. Is this something fixable? Cheers.
@advisoft hi, you should probably try again. See also #34, #36 and the recent commits.
Hello lwfinger,
I have run into an issue, when unloading the driver I get the following segfault:
Here is some information about the system.
It is on a TI AM3352 using a LM811 USB IC. I have looked at the line of the code raising the error, and understanding the potential cause is beyond me at this point in time.
Are you able to suggest a potential starting point of investigation?
Thanks, Phil