1033:0194 Cannot read from USB 3.0 mass storage [NEC uPD720200]

[Jason Gerard DeRose]

I have 2 USB 3.0 ports on my new motherboard. I also have a Lexar USB 3.0 card reader, and a Western Digital Passport drive that's USB 3.0. When I plug these devices into a USB 2.0 port on my motherboard, they work fine. But when I try to use them on a USB 3.0 port, here is what happens:

I'm trying to run this benchmark:
http://bazaar.launchpad.net/~dmedia/filestore/trunk/view/head:/benchmark.py

Like this:

    ./benchmark.py /media/EOS_DIGITAL/

The benchmark can correctly scan the list of files (on my test card, there are 4 files, totaling 8.57 GB).

The benchmark reads in 8 MiB chunks. It appears to start reading the first chunk, but before it finishes the light goes off on card reader... almost as if power management has powered down the device.

Checking dmesg shows this:

[ 1202.020377] INFO: task udisks-daemon:1762 blocked for more than 120 seconds.
[ 1202.020382] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
[ 1202.020387] udisks-daemon D 0000000000000001 0 1762 1761 0x00000000
[ 1202.020396] ffff880227e19938 0000000000000086 ffff880227e198d8 ffffffff81046904
[ 1202.020405] ffff880227e19fd8 ffff880227e19fd8 ffff880227e19fd8 0000000000012a40
[ 1202.020414] ffff8802265544d0 ffff880225e444d0 ffff880237c52a40 7fffffffffffffff
[ 1202.020423] Call Trace:
[ 1202.020430] [<ffffffff81046904>] ? __enqueue_entity+0x84/0x90
[ 1202.020439] [<ffffffff815e5c25>] schedule_timeout+0x2a5/0x320
[ 1202.020447] [<ffffffff8104a366>] ? ttwu_do_activate.constprop.174+0x66/0x70
[ 1202.020456] [<ffffffff810329b9>] ? default_spin_lock_flags+0x9/0x10
[ 1202.020464] [<ffffffff815e568f>] wait_for_common+0xdf/0x180
[ 1202.020471] [<ffffffff810572d0>] ? try_to_wake_up+0x200/0x200
[ 1202.020479] [<ffffffff815e580d>] wait_for_completion+0x1d/0x20
[ 1202.020486] [<ffffffff8107cfae>] flush_work+0x2e/0x40
[ 1202.020495] [<ffffffff810792c0>] ? do_work_for_cpu+0x30/0x30
[ 1202.020502] [<ffffffff8107d00b>] flush_delayed_work+0x4b/0x60
[ 1202.020510] [<ffffffff812d7b04>] disk_clear_events+0x84/0x130
[ 1202.020520] [<ffffffff8119a9c7>] check_disk_change+0x37/0x80
[ 1202.020529] [<ffffffff8140107d>] sd_open+0xad/0x1e0
[ 1202.020536] [<ffffffff8119bf27>] __blkdev_get+0xc7/0x430
[ 1202.020543] [<ffffffff81178c10>] ? __pollwait+0xf0/0xf0
[ 1202.020550] [<ffffffff8119c2ee>] blkdev_get+0x5e/0x1e0
[ 1202.020556] [<ffffffff8119c4d0>] blkdev_open+0x60/0x90
[ 1202.020564] [<ffffffff81163e84>] __dentry_open+0x144/0x320
[ 1202.020571] [<ffffffff8119c470>] ? blkdev_get+0x1e0/0x1e0
[ 1202.020579] [<ffffffff811655cd>] nameidata_to_filp+0xad/0xb0
[ 1202.020588] [<ffffffff81174000>] do_last+0x3a0/0x700
[ 1202.020596] [<ffffffff8117558a>] path_openat+0xca/0x3f0
[ 1202.020604] [<ffffffff812e8397>] ? kobject_put+0x27/0x60
[ 1202.020615] [<ffffffff813c8bf7>] ? put_device+0x17/0x20
[ 1202.020623] [<ffffffff811758f2>] do_filp_open+0x42/0xa0
[ 1202.020631] [<ffffffff812f3041>] ? strncpy_from_user+0x31/0x40
[ 1202.020639] [<ffffffff815e74be>] ? _raw_spin_lock+0xe/0x20
[ 1202.020648] [<ffffffff81182ca7>] ? alloc_fd+0xf7/0x150
[ 1202.020656] [<ffffffff811656bd>] do_sys_open+0xed/0x220
[ 1202.020664] [<ffffffff81165810>] sys_open+0x20/0x30
[ 1202.020672] [<ffffffff815ef882>] system_call_fastpath+0x16/0x1b
[ 1233.940107] xhci_hcd 0000:03:00.0: Timeout while waiting for reset device command
[ 1233.940119] usb 9-1: Cannot reset HCD device state
[ 1284.210073] xhci_hcd 0000:03:00.0: Timeout while waiting for reset device command
[ 1284.210087] usb 9-1: Cannot reset HCD device state

Please let me know what additional information you need.

ProblemType: Bug
DistroRelease: Ubuntu 11.10
Package: linux-image-3.0.0-9-generic 3.0.0-9.14
ProcVersionSignature: Ubuntu 3.0.0-9.14-generic 3.0.3
Uname: Linux 3.0.0-9-generic x86_64
AlsaVersion: Advanced Linux Sound Architecture Driver Version 1.0.24.
Architecture: amd64
ArecordDevices:
 **** List of CAPTURE Hardware Devices ****
 card 0: SB [HDA ATI SB], device 0: VT1818S Analog [VT1818S Analog]
   Subdevices: 2/2
   Subdevice #0: subdevice #0
   Subdevice #1: subdevice #1
AudioDevicesInUse:
 USER PID ACCESS COMMAND
 /dev/snd/controlC0: jderose 1718 F.... pulseaudio
CRDA: Error: [Errno 2] No such file or directory
Card0.Amixer.info:
 Card hw:0 'SB'/'HDA ATI SB at 0xfe7f4000 irq 16'
   Mixer name : 'VIA VT1818S'
   Components : 'HDA:11060440,10438416,00100000'
   Controls : 33
   Simple ctrls : 19
Card1.Amixer.info:
 Card hw:1 'HDMI'/'HDA ATI HDMI at 0xfe9ec000 irq 51'
   Mixer name : 'ATI R6xx HDMI'
   Components : 'HDA:1002aa01,00aa0100,00100100'
   Controls : 4
   Simple ctrls : 1
Card1.Amixer.values:
 Simple mixer control 'IEC958',0
   Capabilities: pswitch pswitch-joined penum
   Playback channels: Mono
   Mono: Playback [on]
Date: Wed Aug 24 12:17:16 2011
EcryptfsInUse: Yes
InstallationMedia: Ubuntu 11.10 "Oneiric Ocelot" - Alpha amd64 (20110803.1)
IwConfig:
 lo no wireless extensions.

 eth0 no wireless extensions.
MachineType: System manufacturer System Product Name
ProcEnviron:
 PATH=(custom, no user)
 LANG=en_US.UTF-8
 SHELL=/bin/bash
ProcKernelCmdLine: BOOT_IMAGE=/boot/vmlinuz-3.0.0-9-generic root=UUID=f6265f37-ca88-4bcd-a927-ca36b1bc2da9 ro quiet splash vt.handoff=7
RelatedPackageVersions:
 linux-restricted-modules-3.0.0-9-generic N/A
 linux-backports-modules-3.0.0-9-generic N/A
 linux-firmware 1.60
RfKill:

SourcePackage: linux
UpgradeStatus: No upgrade log present (probably fresh install)
dmi.bios.date: 12/09/2010
dmi.bios.vendor: American Megatrends Inc.
dmi.bios.version: 1102
dmi.board.asset.tag: To Be Filled By O.E.M.
dmi.board.name: M4A87TD/USB3
dmi.board.vendor: ASUSTeK Computer INC.
dmi.board.version: Rev 1.xx
dmi.chassis.asset.tag: Asset-1234567890
dmi.chassis.type: 3
dmi.chassis.vendor: Chassis Manufacture
dmi.chassis.version: Chassis Version
dmi.modalias: dmi:bvnAmericanMegatrendsInc.:bvr1102:bd12/09/2010:svnSystemmanufacturer:pnSystemProductName:pvrSystemVersion:rvnASUSTeKComputerINC.:rnM4A87TD/USB3:rvrRev1.xx:cvnChassisManufacture:ct3:cvrChassisVersion:
dmi.product.name: System Product Name
dmi.product.version: System Version
dmi.sys.vendor: System manufacturer

[Leann Ogasawara (leannogasawara)]

Hi Jason,

Just curious if you can reproduce this with the latest upstream v3.1-rc3 mainline kernel:

http://kernel.ubuntu.com/~kernel-ppa/mainline/v3.1-rc3-oneiric/

If so, we'll likely want to have you open an upstream bug and work directly with the upstream developers on this. Thanks.

[Jason Gerard DeRose (jderose)]

Yup, same problem under 3.1.0-0301rc3-generic. Some bits from dmesg:

[ 44.712024] xhci_hcd 0000:03:00.0: WARN: Stalled endpoint
[ 44.712645] xhci_hcd 0000:03:00.0: WARN: Stalled endpoint
[ 44.713118] sd 6:0:0:0: [sdd] No Caching mode page present
[ 44.713122] sd 6:0:0:0: [sdd] Assuming drive cache: write through
[ 44.713701] xhci_hcd 0000:03:00.0: WARN: Stalled endpoint
[ 44.714392] xhci_hcd 0000:03:00.0: WARN: Stalled endpoint
[ 44.714867] sd 6:0:0:0: [sdd] No Caching mode page present
[ 44.714870] sd 6:0:0:0: [sdd] Assuming drive cache: write through
[ 44.744800] sdd: sdd1
[ 141.160127] xhci_hcd 0000:03:00.0: Timeout while waiting for a slot
[ 191.370085] xhci_hcd 0000:03:00.0: Timeout while waiting for a slot
[ 191.590072] usb 9-1: device not accepting address 2, error -62

[Jason Gerard DeRose (jderose)]

I can produce this exact same bug from my Lenovo X220 laptop. It has the same NEC USB 3.0 controller. I can open Nautilus and browse through files on the card, but shortly after I start playing one the the videos, the device hangs and the light on the card ready turns off as if it has been kicked off by power management (not say it has, this is just appearance).

[Jason Gerard DeRose (jderose)]

I'm not sure at what kernel revision this was fixed, but it's now working, on both my desktop and laptop.

Tell you what, 65 MB/s from a card reader is sweet! :^)

[Rocko (rockorequin)]

I get this occasionally with both kernel 3.0 and 3.1 on an external USB 3.0 device, particularly if it is attached via a USB 3.0 hub. Sometimes it fails during a read/write operation, but more often it fails to come up after a suspend/resume cycle with the "Device not responding to set address" message.

Removing and re-mod-probing the xhci-hcd module doesn't make any difference and nor does power-cycling the hard drive and/or hub, although sometimes the drive will respond if I plug it directly into the computer rather than via the hub. This only happens with a USB 3.0 device, not with USB 2.0 devices.

A full reboot typically fixes the problem, implying that xhci-hcd is caching something that survives removing the module.

[Jason Gerard DeRose (jderose)]

Rocko, thanks for confirming this!

Yeah, I closed this too soon I think... for a while it seemed to be reliable, but apparently the planets aren't in alignment now, and it pretty much never works.

I also filed a fresh bug for this under Precise, just in case the ubuntu-bug info includes something interesting not in this report from Oneiric:

https://bugs.launchpad.net/ubuntu/+source/linux/+bug/904412

[Rocko (rockorequin)]

So I should comment in the new bug? I'll do that. (I think you could also have used apport to attach the bug info for Precise to this report.)