netdev - RE: NULL pointer dereference when removing xmm7360 PCI device

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:   Fri, 28 Apr 2023 09:47:40 +0000
From:   Samuel Wein PhD <sam@...wein.com>
To:     "Kumar, M Chetan" <m.chetan.kumar@...el.com>
Cc:     Jakub Kicinski <kuba@...nel.org>,
        "netdev@...r.kernel.org" <netdev@...r.kernel.org>,
        linuxwwan <linuxwwan@...el.com>
Subject: RE: NULL pointer dereference when removing xmm7360 PCI device

I've added the full set of logs to the gist. The sequence was:
boot PC
suspend PC
resume PC
repeat the following several (it varies) times until the error:
`sudo echo 1 > /sys/devices/pci0000:00/0000:00:1c.0/0000:02:00.0/remove`
`echo 1 > /sys/devices/pci0000:00/0000:00:1c.0/rescan`

The XMM7360 card doesn't resume properly from suspend, with the AT devices not responding to any commands. I've been working off of some of Shane Parslow's code to try to get ModemManager control of the device and was basically trying a bunch of ways to try to reset the card, this one clearly didn't work, but it also clearly exposed some weirdness in how the kernel handled the attempt. The fact that there is not a proper MBIM interface for this card really makes managing it difficult.


------- Original Message -------
On Friday, April 28th, 2023 at 11:21 AM, Kumar, M Chetan <m.chetan.kumar@...el.com> wrote:


> 
> 
> > -----Original Message-----
> 
> > From: Jakub Kicinski kuba@...nel.org
> > Sent: Friday, April 28, 2023 2:38 AM
> > To: Kumar, M Chetan m.chetan.kumar@...el.com
> > Cc: Samuel Wein PhD sam@...wein.com; netdev@...r.kernel.org;
> > linuxwwan linuxwwan@...el.com
> > Subject: Re: NULL pointer dereference when removing xmm7360 PCI device
> > 
> > On Thu, 27 Apr 2023 10:31:29 +0000 Samuel Wein PhD wrote:
> > 
> > > Hi Folks,
> > > I've been trying to get the xmm7360 working with IOSM and the
> > > ModemManager. This has been what my highschool advisor would call a
> > > "learning process".
> > > When trying `echo 1 > /sys/devices/pci0000:00/0000:00:1c.0/0000:02:00.0/remove` I get a
> > > variety of errors. One of these is a kernel error
> > > `2023-04-27T12:23:38.937223+02:00 Nase kernel: [ 587.997430] BUG: kernel NULL pointer dereference, address: 0000000000000048 2023-04-27T12:23:38.937237+02:00 Nase kernel: [ 587.997447] #PF: supervisor read access in kernel mode 2023-04-27T12:23:38.937238+02:00 Nase kernel: [ 587.997455] #PF: error_code(0x0000) - not-present page 2023-04-27T12:23:38.937241+02:00 Nase kernel: [ 587.997463] PGD 0 P4D 0 2023-04-27T12:23:38.937242+02:00 Nase kernel: [ 587.997476] Oops: 0000 [#1] PREEMPT SMP NOPTI 2023-04-27T12:23:38.937242+02:00 Nase kernel: [ 587.997489] CPU: 1 PID: 4767 Comm: bash Not tainted 6.3.0-060300-generic #202304232030 ...`
> > > the full log is available at
> > > https://gist.github.com/poshul/0c5ffbde6106a71adcbc132d828dbcd7
> > > 
> > > Steps to reproduce: Boot device with xmm7360 installed and in PCI mode,
> > > place into suspend. Resume, and start issuing reset/remove commands to
> > > the PCI interface (without properly unloading the IOSM module first).
> > > 
> > > I'm not sure how widely applicable this is but wanted to at least report it.
> > 
> > Intel folks, PTAL.
> 
> 
> I tried reproducing the issue by following the steps you mentioned but so far could not
> reproduce it. Could you please share the logs from boot-up and procedure you carried out in steps.
> 
> Once you boot-up the laptop, driver will be in working condition why device removal ?