Alcor Micro AU9540 keeps powering down when card is present

[Jonathan Davies]

I have smart card reader built-into this laptop, however, whenever I insert a card - it'll read it for a few seconds, then the USB connection to the reader gets powered down and thus the card disappears in the systems eyes.

After a few seconds, it comes back and everything's happy.

Then the cycle repeats again and again and again.

ProblemType: Bug
DistroRelease: Ubuntu 14.04
Package: linux-image-3.13.0-23-generic 3.13.0-23.45 [modified: boot/vmlinuz-3.13.0-23-generic]
ProcVersionSignature: Ubuntu 3.13.0-23.45-generic 3.13.8
Uname: Linux 3.13.0-23-generic x86_64
ApportVersion: 2.14.1-0ubuntu1
Architecture: amd64
AudioDevicesInUse:
 USER PID ACCESS COMMAND
 /dev/snd/controlC1: jpds 3933 F.... pulseaudio
 /dev/snd/controlC0: jpds 3933 F.... pulseaudio
CurrentDesktop: Unity
Date: Sun Apr 6 18:37:24 2014
HibernationDevice: RESUME=UUID=05490d4c-6800-46a1-b5ba-8b32bdc57e89
InstallationDate: Installed on 2014-04-06 (0 days ago)
InstallationMedia: Ubuntu 14.04 LTS "Trusty Tahr" - Daily amd64 (20140406)
MachineType: Hewlett-Packard HP EliteBook Folio 1040 G1
ProcFB: 0 inteldrmfb
ProcKernelCmdLine: BOOT_IMAGE=/vmlinuz-3.13.0-23-generic.efi.signed root=/dev/mapper/ubuntu--vg-root ro quiet splash vt.handoff=7
RelatedPackageVersions:
 linux-restricted-modules-3.13.0-23-generic N/A
 linux-backports-modules-3.13.0-23-generic N/A
 linux-firmware 1.127
SourcePackage: linux
UpgradeStatus: No upgrade log present (probably fresh install)
dmi.bios.date: 02/09/2014
dmi.bios.vendor: Hewlett-Packard
dmi.bios.version: L83 Ver. 01.05
dmi.board.name: 213E
dmi.board.vendor: Hewlett-Packard
dmi.board.version: KBC Version 24.2A
dmi.chassis.type: 10
dmi.chassis.vendor: Hewlett-Packard
dmi.modalias: dmi:bvnHewlett-Packard:bvrL83Ver.01.05:bd02/09/2014:svnHewlett-Packard:pnHPEliteBookFolio1040G1:pvrA3009DD18303:rvnHewlett-Packard:rn213E:rvrKBCVersion24.2A:cvnHewlett-Packard:ct10:cvr:
dmi.product.name: HP EliteBook Folio 1040 G1
dmi.product.version: A3009DD18303
dmi.sys.vendor: Hewlett-Packard

[Brad Figg (brad-figg)]

This change was made by a bot.

[Chris J Arges (arges)]

The dmesg corresponding to device removal and power on was related to a BIOS option that does "Power on if card is present" vs "Always powered on". If the BIOS option is switched to "Always powered on" these messages are not present.

[Launchpad Janitor (janitor)]

Status changed to 'Confirmed' because the bug affects multiple users.

[Jonathan Davies (jpds)]

I ran 'opensc-tool' three times while running this debug.

[Chris J Arges (arges)]

For context I found the following thread which looked related:
http://ludovicrousseau.blogspot.fr/2010/10/card-auto-power-on-and-off.html

From bug 1161882. So I built the following packages with DISABLE_ON_DEMAND_POWER_ON defined:
http://people.canonical.com/~arges/lp1307404/

This also failed.

[Jonathan Davies (jpds)]

I ran 'opensc-tool' three times during this debug run. First two times, it reported that there was a card in the system; third 'No'.

[Chris J Arges (arges)]

An example of a non-working log:
http://pastebin.ubuntu.com/7250315/

An example of a working log:
http://pastebin.ubuntu.com/7250304/

The main differences I see start after here in the log:
00000003 winscard.c:261:SCardConnect() Attempting Connect to Alcor Micro AU9540 00 00 using protocol: 3
00000002 readerfactory.c:771:RFReaderInfo() RefReader() count was: 1

[Ludovic Rousseau (ludovic-rousseau-gmail)]

Why does powering down a present card is a problem?

A problem would be if the reader reports no card inserted when a card is inserted. Is that what happens?

[Jonathan Davies (jpds)]

Ludovic: That is exactly what happens. I'm not touching the card and as shown in comment #3, it keeps appearing and disappearing.

[Ludovic Rousseau (ludovic-rousseau-gmail)]

Please generates a pcscd log as in #2 but without starting any application. Just insert a smart card and wait.

[Jonathan Davies (jpds)]

That output above is what happens without touching the card or running an application - it just sits there with Card present, as soon as I do something like 'opensc-tool -l' - the card is "absent" and then it eventually loops back to present.

[Ludovic Rousseau (ludovic-rousseau-gmail)]

The readers report card present and then card absent. The driver can't do much to know this is not correct and must trust the reader.

I will try to find a Alcor Micro AU9540 reader and to tests. But it looks like it is a reader bug.

[Jonathan Davies (jpds)]

It comes built-into the HP EliteBook Folio 1040 G1, otherwise, I'm happy to do any tests on my laptop.

[Ludovic Rousseau (ludovic-rousseau-gmail)]

Maybe you can find a reader firmware update from Alcor Micro or HP.

[Jonathan Davies (jpds)]

This is a brand new laptop and HP appear to have shipped it with their latest firmware.

[Jonathan Davies (jpds)]

OK, in comments #22 and #23, we have outputs with ccid compiled with the patch to add some more debug in #24.

Comment #22 features a working Lenovo laptop.

Comment #23 features the HP laptop.

For some reason, the HP laptop hits the IFDHICCPresence() function's CCID_ICC_PRESENT_INACTIVE case and drops to the else statement, while this doesn't happen on the Lenovo.

[Ludovic Rousseau (ludovic-rousseau-gmail)]

What is strange is the response RDR_to_PC_SlotStatus (first byte 81h) when a card is present and powered.
The normal CCID frame should be:
81 00 00 00 00 00 xx _00_ 00 00
00 is: An ICC is present and active (power is on and stable, RST is inactive)

But in the case of the HP we have:
81 00 00 00 00 00 xx _01_ 00 01
01 is: An ICC is present and inactive (not activated or shut down by hardware error)
The last 01 is: Clock stopped in state L

Something (the reader? the BIOS?) decided to stop the reader clock.

I need more information about your Alcor Micro reader.
Please follow http://pcsclite.alioth.debian.org/ccid.html#CCID_compliant

[Ludovic Rousseau (ludovic-rousseau-gmail)]

Your reader firmware is newer than the one I have at http://pcsclite.alioth.debian.org/ccid/readers/AU9540.txt

I confirm the reader firmware is bogus according to the output.txt you attached in #27.
  bNumDataRatesSupported: 138
   IFD does not support GET_DATA_RATES request: Resource temporarily unavailable

You should get a list of 138 values, not an error.

[Chris J Arges (arges)]

So I see that get_data_rates is returning an error, are there alternative Control USB bytes we need to send? Or do we think this failure is because of something else (reader clock being stopped.)

[Ludovic Rousseau (ludovic-rousseau-gmail)]

It may be possible to make the driver work with this reader.
I am not sure how to come back from the state "An ICC is present and inactive". Without such a reader at hand it will be hard to debug the driver.
Maybe you can offer me a "Hewlett-Packard HP EliteBook Folio 1040 G1"? :-)

[Chris McDonough (chrism-plope)]

I have the same problem, using pcscd 1.8.14-1ubuntu1.16.04.1 on an IBM Thinkpad P50, which has the same card reader.

However, interestingly, it only happens when I use a Feitian/Entersafe A22CR card. I tried two of them with the same result, so it's not a flaky card either.

When I use a PIVKey C910 card, it does not "flap" like it does with the entersafe card. The PIVKey card works fine in the reader.

[Chris McDonough (chrism-plope)]

I'll also note that the A22CR card works fine (does not exhibit the flapping behavior) in an external Feitian reader.

Interestingly, the A22CR card + Alcor combibation doesn't "flap" under Windows either when running opensc-tool in a loop. Just under Linux.

[Chris McDonough (chrism-plope)]

Similar output of ccid conformance as James' from a few years back.

[Chris McDonough (chrism-plope)]

Another data point: an Aventra MyEID PKI card (http://www.aventra.fi/webshop/index.php?route=product/product&product_id=53 ; it's apparently a "version 4", on the invoice it's identified as SKU: MYEID and MPN: TH#130) works fine in this Alcor reader under OpenSC on Ubuntu. The only card that has this flapping problem with this reader under Linux so far is the Feitian/Entersafe A22CR, and again, the A22CR+Alcor combo doesn't flap the same way under Windows.

[Ray (raymonster)]

I personally couldn't waste anymore time troubleshooting this problem.

Ensure you are using a Smartcard Reader that is compatible with the CCID (Chip/Smart Card Interface Devices) and ICCD (Integrated Circuit(s) Card Devices) driver.

Supported CCID readers/ICCD tokens: https://ccid.apdu.fr/ccid/supported.html

Here are some Smart Card Readers that were intermittently working in Linux:
[ray@ThinkPad ~]$ lsusb
Bus 005 Device 021: ID 0bda:0165 Realtek Semiconductor Corp. Smart Card Reader Interface
Bus 003 Device 003: ID 058f:9540 Alcor Micro Corp. AU9540 Smartcard Reader

The Realtek Semiconductor Corp. Smart Card Reader Interface (0bda:0165) is actually a Stanley Global SGT111 Smart Card Reader.
The Alcor Micro Corp. AU9540 Smartcard Reader (058f:9540) is a smart card reader that I added when I bought my Lenovo ThinkPad T14 (not knowing that it had compatibility issues with Linux).

After running pcsc_scan, I could see the both of the above card readers fluctuate between:
Card state: Card removed AND Card state: Card inserted. This would cause my VMware Horizon Client to randomly disconnect from Desktop Anywhere (very annoying).

Here are the Smart Card Readers that I ended up buying on Amazon and eBay because they are fully supported by the CCID driver:

SCM Microsystems Inc. SCR 3310 USB Smart Card Reader: https://ccid.apdu.fr/ccid/supported.html#0x04E60x5116

CHERRY SmartTerminal ST-1144: https://ccid.apdu.fr/ccid/supported.html#0x046A0x002D
(Features: “One-hand operation“ possible due to heavy weight and stable base.)

[ray@ThinkPad ~]$ lsusb
Bus 005 Device 020: ID 04e6:5116 SCM Microsystems, Inc. SCR331-LC1 / SCR3310 SmartCard Reader
Bus 005 Device 022: ID 046a:002d Cherry GmbH SmartTerminal XX44

I haven't had any issues with the using the above supported Card Readers. I use the CHERRY SmartTerminal ST-1144 when I'm at my desk, and the Identiv SCR3310v2.0 USB Smart Card Reader when I have to travel for work.