linux-kernel - Re: crash by cdc_acm driver in kernels 4.8-rc1/5

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Tue, 15 Nov 2016 12:07:03 +0100
From:   Oliver Neukum <oneukum@...e.com>
To:     wim@....tudelft.nl
Cc:     linux-kernel@...r.kernel.org, linux-usb@...r.kernel.org
Subject: Re: crash by cdc_acm driver in kernels 4.8-rc1/5

On Tue, 2016-11-15 at 01:16 +0100, Wim Osterholt wrote:

Hi,

> Whell, yes.
> The only thing that appears you'll have to do is unset 'CONFIG_SMP'.

OK. I haven't tested that, nor would I ever considered it.
 
> My machines didn't have the luxury of multicore processors (until recently),
> so there never has been any reason to deliberately switch these options on!
> 
> In the process of searching, many options may have changed. The crash/OOPS
> has now mitigated into just a WARNING with a call trace.
> (Or it could be a totally different bug?)
> After the call trace the device is working normally and a shutdown
> completes to the end now.
> That is with the config given here:
> http://webserver.djo.tudelft.nl/.config-4.9-rc4.OK (CONFIG_SMP=y)
> http://webserver.djo.tudelft.nl/WARNING-4.9-rc4    (call trace for C_S unset)
> 
> Tests on other machines with (slightly) different configs all seem to
> confirm that the problems are gone when CONFIG_SMP is set.

OK, something extremely strange is going on. And I think it is time to get
the big hammer out. I made an extremely stupid debugging patch. Could
you test with it?

	Regards
		Oliver


View attachment "0001-acm-insane-debugging.patch" of type "text/x-patch" (5444 bytes)