linux-kernel - Re: usb/hid: slab-out-of-bounds read in usbhid

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]

Message-ID: <Pine.LNX.4.44L0.1709201149420.1563-100000@iolanthe.rowland.org>
Date:   Wed, 20 Sep 2017 11:50:46 -0400 (EDT)
From:   Alan Stern <stern@...land.harvard.edu>
To:     Kim Jaejoong <climbbb.kim@...il.com>
cc:     Andrey Konovalov <andreyknvl@...gle.com>,
        Jiri Kosina <jikos@...nel.org>,
        Benjamin Tissoires <benjamin.tissoires@...hat.com>,
        USB list <linux-usb@...r.kernel.org>,
        <linux-input@...r.kernel.org>, LKML <linux-kernel@...r.kernel.org>,
        syzkaller <syzkaller@...glegroups.com>,
        Dmitry Vyukov <dvyukov@...gle.com>,
        Kostya Serebryany <kcc@...gle.com>
Subject: Re: usb/hid: slab-out-of-bounds read in usbhid_parse

On Wed, 20 Sep 2017, Kim Jaejoong wrote:

> To. usb & input guys.
> 
> While dig this report, i was wondering about bNumDescriptors in HID descriptor.
> HID document from usb.org said, 'this number must be at least one (1)
> as a Report descriptor will always be present.'
> 
> There is no mention of the order of class descriptors. Suppose you
> have a HID device with a report descriptor and a physical descriptor.
> 
> If you have the following hid descriptor in this case,
> HID descriptor
>    bLength: 12
>    bDescriptor Type: HID
>    .. skip
>    bNumDescriptors: 2
>    bDescriptorType: physical
>    bDescriptorLength: any
>    bDescriptorType: Report
>    bDescriptorLength: any
> 
> If the order of the report descriptor is the second as above,
> usbhid_parse () will fail because my patch is only check the first
> bDescriptor Type.
> But If the order of the report descriptor is always first, there is no
> problem. How do you think this?

The descriptors can appear in any order.  You should not assume that 
the report descriptor will always come first.

Alan Stern