linux-kernel - Re: Dell Latitude E6510: `dell_wmi: Dell descriptor buffer has invalid buffer length (128)`

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] [day] [month] [year] [list]

Message-ID: <20170216153220.GG22666@pali>
Date:   Thu, 16 Feb 2017 16:32:20 +0100
From:   Pali Rohár <pali.rohar@...il.com>
To:     Paul Menzel <pmenzel@...gen.mpg.de>
Cc:     Matthew Garrett <mjg59@...f.ucam.org>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: Dell Latitude E6510: `dell_wmi: Dell descriptor buffer has
 invalid buffer length (128)`

On Thursday 16 February 2017 16:15:08 Paul Menzel wrote:
> Dear Matthews, dear Pali,
> 
> 
> Installing Debian 9 (Stretch/testing) with Linux 4.9.6 on the Dell Latitude
> E6510, the warnings below are shown.
> 
> ```
> $ more /proc/version
> Linux version 4.9.0-1-amd64 (debian-kernel@...ts.debian.org) (gcc version
> 6.3.0 20170124 (Debian 6.3.0-5) ) #1 SMP Debian 4.9.6-3 (2017-01-28)
> $ sudo journalctl -k
> […]
> Feb 16 12:19:13 e6510 kernel: DMI: Dell Inc. Latitude E6510/XXXXX, BIOS A03
> 05/28/2010
> […]
> Feb 16 12:19:14 e6510 kernel: input: HDA Intel MID HDMI/DP,pcm=7 as
> /devices/pci0000:00/0000:00:1b.0/sound/card0/input16
> Feb 16 12:19:14 e6510 kernel: input: HDA Intel MID HDMI/DP,pcm=8 as
> /devices/pci0000:00/0000:00:1b.0/sound/card0/input17
> Feb 16 12:19:14 e6510 kernel: ppdev: user-space parallel port driver
> Feb 16 12:19:14 e6510 kernel: fbcon: inteldrmfb (fb0) is primary device
> Feb 16 12:19:15 e6510 kernel: logitech-djreceiver 0003:046D:C52B.0003:
> hiddev0,hidraw0: USB HID v1.11 Device [Logitech USB Receiver] on
> usb-0000:00:1d.0-1.3/input2
> Feb 16 12:19:15 e6510 kernel: dell_wmi: Dell descriptor buffer has invalid
> buffer length (128)
> Feb 16 12:19:15 e6510 kernel: dell_wmi: Detected Dell WMI interface version
> 1
> Feb 16 12:19:15 e6510 kernel: input: Dell WMI hotkeys as
> /devices/virtual/input/input18
> Feb 16 12:19:15 e6510 kernel: pcmcia_socket pcmcia_socket0: cs: memory probe
> 0x0c0000-0x0fffff:
> Feb 16 12:19:15 e6510 kernel:  excluding 0xc0000-0xfffff
> Feb 16 12:19:15 e6510 kernel: pcmcia_socket pcmcia_socket0: cs: memory probe
> 0xa0000000-0xa0ffffff:
> Feb 16 12:19:15 e6510 kernel:  excluding 0xa0000000-0xa0ffffff
> Feb 16 12:19:15 e6510 kernel: pcmcia_socket pcmcia_socket0: cs: memory probe
> 0x60000000-0x60ffffff:
> Feb 16 12:19:15 e6510 kernel:  excluding 0x60000000-0x60ffffff
> Feb 16 12:19:15 e6510 kernel: clocksource: Switched to clocksource tsc
> […]
> ```
> 
> Reading the commit description [1], it sounds like, that this model just has
> a ACPI DSDT table, that wasn’t known when writing the patch [1].
> 
> It’d be great to get rid of the warning. Please tell me what you need.
> 
> 
> Kind regards,
> 
> Paul
> 
> 
> [1] https://patchwork.kernel.org/patch/7951281/

Hi! In future please try to avoid wrapping dmesg log lines as they are
harder to read if they are wrapped.

About that warning on buffer size. That check was added based on
ACPI-WMI documentation IIRC dated back to 2009. Info is here:
https://www.spinics.net/lists/platform-driver-x86/msg07220.html

So now there is some machine which has different buffer size as
specified in that documentation. This is first time I saw such report
since introduction of that change.

But it is only check that WMI interface (which dell-wmi driver is using)
is behaving correctly according to that documentation. Warning is non
fatal and only informative.

I will think about this.

Maybe somebody from Dell could comment this?

-- 
Pali Rohár
pali.rohar@...il.com