SuzyQ recognized but not appearing as /dev/tty device

[David Borenstein]

I am attempting to use a SuzyQ cable for CCD debugging in order to disable cr50 write protection on my Pixelbook. I am running developer mode and CCD state is open (see output of `sudo gsctool -a -I` below). My cr50 firmware version is RW 0.3.20 (see `sudo gsctool -a -f` below), which, according to the ChromiumOS CCD documentation, should be sufficient (> 0.x.9) to use CCD with a SuzyQ cable. I'm running Version 78.0.3889.0 (Official Build) dev (64-bit) of ChromiumOS.

It is definitely not a problem of cable orientation! I have dmesg output below for both orientations, and the cable is definitely recognized for what it is. However, it appears that the tty devices are connecting and then immediately disconnecting. Consequently, they never appear in /dev/tty*.

I have seen mention of "udev rules" as an alternate path for sending debug messages. I don't know what this means. I would greatly welcome any guidance.

Output of relevant commands:

sudo gsctool -a -f

start
target running protocol version 6
keyids: RO 0xaa66150f, RW 0xde88588d
offsets: backup RO at 0x40000, backup RW at 0x44000
Current versions:
RO 0.0.10
RW 0.3.20

sudo gsctool -a -I

State: Opened
Password: None
Flags: 000000
Capabilities, current and default:
 UartGscRxAPTx Y Always
 UartGscTxAPRx Y Always
 UartGscRxECTx Y Always
 UartGscTxECRx Y IfOpened
 FlashAP Y IfOpened
 FlashEC Y IfOpened
 OverrideWP Y IfOpened
 RebootECAP Y IfOpened
 GscFullConsole Y IfOpened
 UnlockNoReboot Y Always
 UnlockNoShortPP Y Always
 OpenNoTPMWipe Y IfOpened
 OpenNoLongPP Y IfOpened
 BatteryBypassPP Y Always
 UpdateNoTPMWipe Y Always
 I2C Y IfOpened
 FlashRead Y Always
 OpenNoDevMode Y IfOpened
 OpenFromUSB Y IfOpened
 OverrideBatt Y IfOpened
CCD caps bitmap: 0xfffff
Capabilities are default.

dmesg (correct orientation)

chronos@localhost /dev $ sudo dmesg --clear
chronos@localhost /dev $ # Connecting USB-C side of cable to left port and USB-A side to right port (via adapter) now
chronos@localhost /dev $ dmesg
[ 662.267357] PDLOG 2019/09/10 15:51:21.1568130680 P0 Disconnected
[ 675.650438] usb 1-5: new high-speed USB device number 12 using xhci_hcd
[ 675.850830] usb 1-5: New USB device found, idVendor=18d1, idProduct=501f, bcdDevice=32.99
[ 675.850842] usb 1-5: New USB device strings: Mfr=10, Product=11, SerialNumber=12
[ 675.850849] usb 1-5: Product: SuzyQable
[ 675.850854] usb 1-5: Manufacturer: Google Inc.
[ 675.850859] usb 1-5: SerialNumber: 0025000E0283
[ 675.852508] hub 1-5:1.0: USB hub found
[ 675.854089] hub 1-5:1.0: 4 ports detected
[ 678.733651] usb 1-5.2: new full-speed USB device number 13 using xhci_hcd
[ 678.814200] usb 1-5.2: New USB device found, idVendor=18d1, idProduct=5014, bcdDevice= 1.00
[ 678.814212] usb 1-5.2: New USB device strings: Mfr=1, Product=2, SerialNumber=11
[ 678.814218] usb 1-5.2: Product: Cr50
[ 678.814224] usb 1-5.2: Manufacturer: Google Inc.
[ 678.814228] usb 1-5.2: SerialNumber: 02014054-92D9C357
[ 678.816661] usb_serial_simple 1-5.2:1.0: google converter detected
[ 678.817165] usb 1-5.2: google converter now attached to ttyUSB0
[ 678.818000] usb_serial_simple 1-5.2:1.1: google converter detected
[ 678.818355] usb 1-5.2: google converter now attached to ttyUSB1
[ 678.819048] usb_serial_simple 1-5.2:1.2: google converter detected
[ 678.819411] usb 1-5.2: google converter now attached to ttyUSB2
[ 679.260187] google ttyUSB0: google converter now disconnected from ttyUSB0
[ 679.260388] usb_serial_simple 1-5.2:1.0: device disconnected
[ 679.260902] google ttyUSB1: google converter now disconnected from ttyUSB1
[ 679.261070] usb_serial_simple 1-5.2:1.1: device disconnected
[ 679.261557] google ttyUSB2: google converter now disconnected from ttyUSB2
[ 679.261895] usb_serial_simple 1-5.2:1.2: device disconnected

dmesg (wrong orientation)

[ 802.821443] usb 1-5: new high-speed USB device number 14 using xhci_hcd
[ 803.022159] usb 1-5: New USB device found, idVendor=18d1, idProduct=501f, bcdDevice=32.99
[ 803.022172] usb 1-5: New USB device strings: Mfr=10, Product=11, SerialNumber=12
[ 803.022178] usb 1-5: Product: SuzyQable
[ 803.022183] usb 1-5: Manufacturer: Google Inc.
[ 803.022188] usb 1-5: SerialNumber: 0025000E0283
[ 803.024096] hub 1-5:1.0: USB hub found
[ 803.025701] hub 1-5:1.0: 4 ports detected
[ 807.316346] usb 1-5.2: new low-speed USB device number 15 using xhci_hcd
[ 807.388880] usb 1-5.2: device descriptor read/64, error -32
[ 807.573325] usb 1-5.2: device descriptor read/64, error -32
[ 807.738377] usb 1-5.2: new low-speed USB device number 16 using xhci_hcd
[ 807.810884] usb 1-5.2: device descriptor read/64, error -32
[ 807.995837] usb 1-5.2: device descriptor read/64, error -32
[ 808.096990] usb 1-5-port2: attempt power cycle
[ 808.668503] usb 1-5.2: new low-speed USB device number 17 using xhci_hcd
[ 808.670945] usb 1-5.2: Device not responding to setup address.
[ 808.873975] usb 1-5.2: Device not responding to setup address.
[ 809.074798] usb 1-5.2: device not accepting address 17, error -71
[ 809.138836] usb 1-5.2: new low-speed USB device number 18 using xhci_hcd
[ 809.141968] usb 1-5.2: Device not responding to setup address.
[ 809.346350] usb 1-5.2: Device not responding to setup address.
[ 809.547437] usb 1-5.2: device not accepting address 18, error -71
[ 809.547981] usb 1-5-port2: unable to enumerate USB device

[Daniel Verkamp]

Hi David,

This sounds like https://crbug.com/996113 ("USB serial devices (SuziQ) are disappearing rapidly after enumeration") - could you try a newer build? The initial fix linked in the bug (https://crrev.com/c/1763558) is in 78.0.3892.0 and newer.

--
--
Chromium OS Developers mailing list: chromiu...@chromium.org
View archives, change email options, or unsubscribe:
https://groups.google.com/a/chromium.org/group/chromium-os-dev

[Duncan Laurie]

Hi David,

It took me a minute to realize you are connecting the Pixelbook to
itself and not to a second machine, so I was not able to reproduce
this issue at first.

Now that I tried your setup I do see the same behavior, and I think
you may be seeing the USB device nodes get denied by a permissions
daemon.

1) check /var/log/messages after attach for userspace daemon info, I
suspect you will see them getting denied here
2) before plugging the cable try to first stop this service with "stop
permission_broker"

The result code I see is ALLOW_WITH_DETACH which is supposed to mean
it would be allowed if the kernel driver is detached from the device
before using it, but that wouldn't provide the dev nodes you are
looking for either...

-duncan

On Tue, Sep 10, 2019 at 12:30 PM David Borenstein <dbbp...@gmail.com> wrote:
>

[David Borenstein]

Duncan, you nailed it!