I/O errors on mmc partitions

[Vianney Bajart]

Running live USB of the latest Vivid daily build on Lenovo Thinkpad 10 with Samsung eMMC MDGAGC 116 GiB:

Ubuntu 15.04 (vivid-desktop-amd64.iso 2015-02-16 8:05)
Linux 3.18.0-13-generic
systemd 218-10ubuntu1

Extract of dmesg (full log in attachment):

[ 12.337332] mmc0: BKOPS_EN bit is not set
[ 12.348788] mmc0: new HS200 MMC card at address 0001
...
[ 12.369112] mmcblk0: mmc0:0001 MDGAGC 116 GiB
[ 12.369786] mmcblk0boot0: mmc0:0001 MDGAGC partition 1 4.00 MiB
[ 12.370414] mmcblk0boot1: mmc0:0001 MDGAGC partition 2 4.00 MiB
[ 12.370976] mmcblk0rpmb: mmc0:0001 MDGAGC partition 3 4.00 MiB
[ 12.376682] mmcblk0: p1 p2 p3 p4

I get multiple errors when trying to access to mmc partitions:

[ 21.950514] sdhci: Timeout waiting for Buffer Read Ready interrupt during tuning procedure, falling back to fixed sampling clock
[ 21.950537] mmc0: Got data interrupt 0x00600000 even though no data operation was in progress.
[ 21.952904] mmc0: Got data interrupt 0x00000002 even though no data operation was in progress.
[ 21.955002] mmcblk0: error -110 sending stop command, original cmd response 0x900, card status 0x400900
[ 21.955007] mmcblk0: error -84 transferring data, sector 244277120, nr 8, cmd response 0x900, card status 0x0
[ 21.955011] mmcblk0: retrying using single block read
...
[ 22.197516] mmcblk0boot0: error -84 transferring data, sector 8071, nr 1, cmd response 0x900, card status 0x0
[ 22.199814] mmc0: Got data interrupt 0x00000002 even though no data operation was in progress.
[ 22.201911] mmcblk0boot1: error -110 sending stop command, original cmd response 0x900, card status 0x400900
[ 22.201914] mmcblk0boot1: error -84 transferring data, sector 8064, nr 8, cmd response 0x900, card status 0x0
[ 22.201917] mmcblk0boot1: retrying using single block read
...
[ 22.241280] Buffer I/O error on dev mmcblk0boot0, logical block 1008, async page read
[ 22.244301] mmc0: Got data interrupt 0x00000002 even though no data operation was in progress.
[ 22.246331] mmcblk0boot1: error -110 sending stop command, original cmd response 0x900, card status 0x400900
[ 22.246335] mmcblk0boot1: error -84 transferring data, sector 8064, nr 8, cmd response 0x900, card status 0x0
[ 22.246337] mmcblk0boot1: retrying using single block read
[ 22.248553] mmcblk0boot1: error -84 transferring data, sector 8064, nr 8, cmd response 0x900, card status 0x0
...
[ 22.265044] Buffer I/O error on dev mmcblk0boot1, logical block 1008, async page read

Ubiquity reports "Input/output errors" consequently, I'm unable to install Ubuntu on mmcblk0 device.

Content of /sys/block/mmcblk0:

alignment_offset
bdi
capability
dev
device
discard_alignment
ext_range
force_ro
holders
inflight
mmcblk0boot0
mmcblk0boot1
mmcblk0p1
mmcblk0p2
mmcblk0p3
mmcblk0p4
mmcblk0rpmb
power
queue
range
removable
ro
size
slaves
stat
subsystem
trace
uevent

I don't know if it's related, but I also get a warning from libparted:

sudo parted -l
Warning: The driver descriptor says the physical block size is 2048 bytes, but
Linux says it is 512 bytes.
Ignore/Cancel? i

Error: /dev/mmcblk0rpmb: unrecognised disk label
Model: Generic SD/MMC Storage Card (sd/mmc)
Disk /dev/mmcblk0rpmb: 4194kB
Sector size (logical/physical): 512B/512B
Partition Table: unknown
Disk Flags:

Error: /dev/mmcblk0boot0: unrecognised disk label
Model: Generic SD/MMC Storage Card (sd/mmc)
Disk /dev/mmcblk0boot0: 4194kB
Sector size (logical/physical): 512B/512B
Partition Table: unknown
Disk Flags:

Error: /dev/mmcblk0boot1: unrecognised disk label
Model: Generic SD/MMC Storage Card (sd/mmc)
Disk /dev/mmcblk0boot1: 4194kB
Sector size (logical/physical): 512B/512B
Partition Table: unknown
Disk Flags:

Model: MMC MDGAGC (sd/mmc)
Disk /dev/mmcblk0: 125GB
Sector size (logical/physical): 512B/512B
Partition Table: gpt
Disk Flags:

Number Start End Size File system Name Flags
 1 1049kB 274MB 273MB fat32 EFI system partition boot, esp
 2 274MB 408MB 134MB Microsoft reserved partition msftres
 3 408MB 113GB 113GB ntfs Basic data partition msftdata
 4 113GB 125GB 12.1GB ntfs Basic data partition hidden, diag

Let me know if you need additional information.
---
ApportVersion: 2.16.1-0ubuntu2
Architecture: amd64
AudioDevicesInUse: Error: command ['fuser', '-v', '/dev/snd/seq', '/dev/snd/timer'] failed with exit code 1:
CasperVersion: 1.351
CurrentDesktop: Unity
DistroRelease: Ubuntu 15.04
IwConfig:
 usb0 no wireless extensions.

 lo no wireless extensions.
LiveMediaBuild: Ubuntu 15.04 "Vivid Vervet" - Alpha amd64 (20150216)
MachineType: LENOVO 20C1CTO1WW
Package: linux (not installed)
ProcEnviron:
 TERM=xterm
 PATH=(custom, no user)
 XDG_RUNTIME_DIR=<set>
 LANG=en_US.UTF-8
 SHELL=/bin/bash
ProcFB: 0 inteldrmfb
ProcKernelCmdLine: BOOT_IMAGE=/casper/vmlinuz.efi file=/cdrom/preseed/username.seed boot=casper quiet splash --
ProcVersionSignature: Ubuntu 3.18.0-13.14-generic 3.18.5
RelatedPackageVersions:
 linux-restricted-modules-3.18.0-13-generic N/A
 linux-backports-modules-3.18.0-13-generic N/A
 linux-firmware 1.141
RfKill:
 0: LNV4752:00: GPS
  Soft blocked: no
  Hard blocked: no
Tags: vivid
Uname: Linux 3.18.0-13-generic x86_64
UpgradeStatus: No upgrade log present (probably fresh install)
UserGroups: adm cdrom dip lpadmin plugdev sambashare sudo
_MarkForUpload: True
dmi.bios.date: 10/17/2014
dmi.bios.vendor: LENOVO
dmi.bios.version: GWET25WW (1.25)
dmi.board.asset.tag: Base Board Asset Tag
dmi.board.name: 20C1CTO1WW
dmi.board.vendor: LENOVO
dmi.board.version: SKG18 I
dmi.chassis.asset.tag: No Asset Information
dmi.chassis.type: 11
dmi.chassis.vendor: LENOVO
dmi.chassis.version: 0.1
dmi.modalias: dmi:bvnLENOVO:bvrGWET25WW(1.25):bd10/17/2014:svnLENOVO:pn20C1CTO1WW:pvrThinkPad10:rvnLENOVO:rn20C1CTO1WW:rvrSKG18I:cvnLENOVO:ct11:cvr0.1:
dmi.product.name: 20C1CTO1WW
dmi.product.version: ThinkPad 10
dmi.sys.vendor: LENOVO

[Martin Pitt (pitti)]

> I get multiple errors when trying to access to mmc partitions:

Ah, so this happens if you actually try and mount the card, and try to access it manually? Or does this happen automatically by merely placing a card into the reader?

[Vianney Bajart (vianney-bajart)]

> Or does this happen automatically by merely placing a card into the reader?

In my case, /dev/mmcblk0 is not a removable SD card, it is the internal storage of the machine that comes in the form of a embedded MMC (eMMC).

Here is the result of lsblk command:

NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINT
sda 8:0 1 29.4G 0 disk
├─sda1 8:1 1 1.1G 0 part
└─sda2 8:2 1 2.2M 0 part
loop0 7:0 0 1G 1 loop /rofs
mmcblk0rpmb 179:24 0 4M 0 disk
mmcblk0boot0 179:8 0 4M 1 disk
mmcblk0boot1 179:16 0 4M 1 disk
mmcblk0 179:0 0 116.5G 0 disk
├─mmcblk0p1 179:1 0 260M 0 part
├─mmcblk0p2 179:2 0 128M 0 part
├─mmcblk0p3 179:3 0 28G 0 part /media/ubuntu/4EBA142BBA1411D7
├─mmcblk0p4 179:4 0 11.2G 0 part
├─mmcblk0p5 179:5 0 68G 0 part
└─mmcblk0p6 179:6 0 8G 0 part

sda is the vivid live USB
mmcblk0 is the eMMC
- p1 is the EFI partition
- p2 is a partition containing Microsoft stuffs (recovery tools or something else, whatever)
- p3 is the partition where Windows 8.1 is installed (I have successfully resized it despite the I/O errors)
- p4 is the partition containing images for reinstallation of Windows 8.1
- p5 is the partition where I want to install Ubuntu
- p6 is the swap partition

I can mount p3 and p5 without any error. The first errors appear during boot. Then, I get the same kind of errors when launching gparted or trying to install ubuntu.

When I run GParted > Refresh devices, I get a dialog box titled "Libparted Bug Found!" with the message "Input/output error during read on /dev/mmcblk0" and error below appears in dmesg:

[ 392.191741] sdhci: Timeout waiting for Buffer Read Ready interrupt during tuning procedure, falling back to fixed sampling clock
[ 392.194120] mmc0: Got data interrupt 0x00000002 even though no data operation was in progress.
[ 392.196461] mmcblk0: error -110 sending stop command, original cmd response 0x900, card status 0x400900
[ 392.196480] mmcblk0: error -84 transferring data, sector 0, nr 8, cmd response 0x900, card status 0x0
[ 392.196493] mmcblk0: retrying using single block read
[ 392.199701] mmcblk0: error -84 transferring data, sector 0, nr 8, cmd response 0x900, card status 0x0
[ 392.199725] blk_update_request: 1088 callbacks suppressed
[ 392.199737] blk_update_request: I/O error, dev mmcblk0, sector 0
[ 392.202048] mmcblk0: error -84 transferring data, sector 1, nr 7, cmd response 0x900, card status 0x0
[ 392.202071] blk_update_request: I/O error, dev mmcblk0, sector 1
[ 392.204640] mmcblk0: error -84 transferring data, sector 2, nr 6, cmd response 0x900, card status 0x0
[ 392.204660] blk_update_request: I/O error, dev mmcblk0, sector 2
[ 392.207004] mmcblk0: error -84 transferring data, sector 3, nr 5, cmd response 0x900, card status 0x0
[ 392.207022] blk_update_request: I/O error, dev mmcblk0, sector 3
[ 392.209275] mmcblk0: error -84 transferring data, sector 4, nr 4, cmd response 0x900, card status 0x0
[ 392.209308] blk_update_request: I/O error, dev mmcblk0, sector 4
[ 392.211919] mmcblk0: error -84 transferring data, sector 5, nr 3, cmd response...

[Martin Pitt (pitti)]

Assigning to the kernel for now, as that looks like a block device driver issue at first sight.

[Brad Figg (brad-figg)]

This bug is missing log files that will aid in diagnosing the problem. From a terminal window please run:

apport-collect 1422338

and then change the status of the bug to 'Confirmed'.

If, due to the nature of the issue you have encountered, you are unable to run this command, please add a comment stating that fact and change the bug status to 'Confirmed'.

This change has been made by an automated script, maintained by the Ubuntu Kernel Team.

[Vianney Bajart (vianney-bajart)]

apport information

[Vianney Bajart (vianney-bajart)]

apport information

[Vianney Bajart (vianney-bajart)]

apport information

[Vianney Bajart (vianney-bajart)]

apport information

[Vianney Bajart (vianney-bajart)]

apport information

[Vianney Bajart (vianney-bajart)]

apport information

[Vianney Bajart (vianney-bajart)]

apport information

[Vianney Bajart (vianney-bajart)]

apport information

[Vianney Bajart (vianney-bajart)]

apport information

[Vianney Bajart (vianney-bajart)]

apport information

[Vianney Bajart (vianney-bajart)]

apport information

[Vianney Bajart (vianney-bajart)]

apport information

[Vianney Bajart (vianney-bajart)]

apport information

[Vianney Bajart (vianney-bajart)]

apport information provided

[Martin Pitt (pitti)]

Please note that it's very unlikely that the computer's BIOS has anyhing to do with this. It's the linux kernel which exposes these RPMB devices, and this issue is well understood. It's certainly not "incomplete" from a "we don't know enough to understand this problem" view.

[Vianney Bajart (vianney-bajart)]

I have just tried the 04-02-2015 vivid build with latest bios version and the problem is still present.

Which additional information would be useful to understand the problem?

[Jérôme Pouiller (jezz)]

For information, I have an Atom E3845 and I observe same problem.

This issue may be related to erratas:

  - For Z36xx and Z37xx, errata VLT7 described here:
    http://www.intel.fr/content/dam/www/public/us/en/documents/specification-updates/atom-Z36xxx-Z37xxx-spec-update.pdf

  - For E38xx, errata VLI6 described here:
    http://www.intel.fr/content/dam/www/public/us/en/documents/specification-updates/atom-e3800-family-spec-update.pdf

[Ingo Siegert (therealbigear)]

I encounter the bug on my think pad x250. as hard drive I use a Crucial MX100.

[Pascal VITOUX (vitoux-pascal)]

I confirm the problem on the Atom Z37xxx. (comment #23)

 ( 00:00.0 Host bridge [0600]: Intel Corporation Atom Processor Z36xxx/Z37xxx Series SoC Transaction Register [8086:0f00] (rev 0f) )

I get these errors while installing the system on a baytrail tablet with eMMC (bug https://bugs.launchpad.net/ubuntu/+source/grub-installer/+bug/1451473 )

[Jérôme Pouiller (jezz)]

I made some investigations:

In some cases system work correctly until it reboot. Applying this patch http://permalink.gmane.org/gmane.linux.kernel.mmc/28281 help in this case.

I also find that opening eMMC ith O_SYNC (or mounting partition with "sync" flag) improve results.

Finally, I noticed this problem depends on eMMC chip. Samsung eMMC works while ones from Kingstom don't.

[Tsung-Han Lin (tsunghan-tw)]

I also have this long boot time issue with an eMMC device on my trusty machine (3.13.0-45 kernel).
Simply upgrading to the new systemd packages didn't help at all,
but w/ the patch mentioned in comment #26, it is working now.

[Dan Parslow (djp-ubuntu)]

The MMC I/O errors as detailed in the bug description continue to occur in Vivid. This prevents installation, and also will eventually occur with any intensive write activity, for example if attempting to copy a working system to the eMMC with rsync.

Voyo Minipc, wih Atom CPU Z3735F, and Samsung MCGAFB 64GB eMMC

The problem is not present however, in Ubuntu 14.04-2. Can repeatably, successfuly install Lubuntu in this release with none of the errors or messages present in Vivid. So the problem has been introduced since Kernel 3.16.0-30-generic x86_64 #40-14.04.1-Ubuntu.

[penalvch (penalvch)]

Dan Parslow, it will help immensely if you filed a new report via a terminal:
ubuntu-bug linux

Please feel free to subscribe me to it.

[Jérôme Pouiller (jezz)]

There are multiple cause to this bug.

However, patch in attachement may for bug described by Dan Parslow.

It is result of a bisect between a working kernel 3.14 and a non working kernel 3.17.

[penalvch (penalvch)]

Jérôme Pouiller, just to advise, your attachment wouldn't be a patch that is immediately integrable into Ubuntu.

Despite this, in order to focus on the hardware you have, and your bisect results, please file a new report via a terminal:
ubuntu-bug linux

Please feel free to subscribe me to it.

[Dan Parslow (djp-ubuntu)]

I retract the latter part of comment #28, where I state that Ubuntu 14.04-2 is repeatably installable on this device. I have just attempted reinstallation of Ubuntu 14.04-2 and it failed with repeated block write errors. This indicates problems were present in the earlier version and were not introduced since.

[Launchpad Janitor (janitor)]

[Expired for linux (Ubuntu) because there has been no activity for 60 days.]

[Lukas (lukas-wringer)]

trying to install ubuntu wily on MMC of my Convertible results in dozens of errors cause of MMC I/O problems. i tried ubiquity and debootstrap, different partition-formats (e.g.e ext 2-4, fat 32, btrfs) are also not working except that btrfs throws more errors.

[penalvch (penalvch)]

Lukas, it will help immensely if you filed a new report with the Ubuntu repository kernel (not mainline/upstream) via a terminal:
ubuntu-bug linux

Please feel free to subscribe me to it.

For more on why this is helpful, please see https://wiki.ubuntu.com/ReportingBugs.

[Lukas (lukas-wringer)]

What new report? i just commented that i'm affected to and said what
i've done to run into the bug, i were not able to do anything within the
livesystem after the error occured, it just crashed down so that even
getty-Terminals (F1-F6) won't open.

--
Lukas Wringer .................... http://www.rz.uni-augsburg.de/zebra
Universität Augsburg, Rechenzentrum ............. Tel. [phone]
86135 Augsburg .................................. Fax. [phone]

Am 25.04.2016 um 02:00 schrieb Christopher M. Penalver:
> Lukas, it will help immensely if you filed a new report with the Ubuntu repository kernel (not mainline/upstream) via a terminal:
> ubuntu-bug linux
>
> Please feel free to subscribe me to it.
>
> For more on why this is helpful, please see
> https://wiki.ubuntu.com/ReportingBugs.
>