Macbook pro freezes with WRITE FPDMA QUEUED failed

[Balmung]

Hi,

I have installed Fedora 19 on a Macbook Pro Retina with a fixed 512 GB SSD.

This SSD makes Problems, the computer freezes from time to time.

Any idea what to do?

Here is the log:

[ 1.026305] ata1.00: unexpected _GTF length (8)
[ 1.026876] ata1.00: ATA-8: APPLE SSD SM0512F, UXM2JA1Q, max UDMA/133
[ 1.026880] ata1.00: 977105060 sectors, multi 16: LBA48 NCQ (depth 31/32), AA
[ 1.026901] Magic number: 1:120:396
[ 1.027030] rtc_cmos 00:05: setting system clock to 2013-10-30 19:21:31 UTC (1383160891)
[ 1.027172] ata1.00: unexpected _GTF length (8)
[ 1.027257] ata1.00: configured for UDMA/133


[ 68.086619] ata1.00: exception Emask 0x0 SAct 0x7fffffff SErr 0x0 action 0x6 frozen
[ 68.086624] ata1.00: failed command: WRITE FPDMA QUEUED
[ 68.086628] ata1.00: cmd 61/20:00:30:ba:03/00:00:2a:00:00/40 tag 0 ncq 16384 out
res 40/00:00:00:4f:c2/00:00:00:00:00/00 Emask 0x4 (timeout)
[ 68.086630] ata1.00: status: { DRDY }

....

[ 68.086786] ata1.00: failed command: WRITE FPDMA QUEUED
[ 68.086789] ata1.00: cmd 61/08:f0:20:ba:03/00:00:2a:00:00/40 tag 30 ncq 4096 out
res 40/00:00:00:00:00/00:00:00:00:00/00 Emask 0x4 (timeout)
[ 68.086790] ata1.00: status: { DRDY }
[ 68.086792] ata1: hard resetting link
[ 68.393377] ata1: SATA link up 6.0 Gbps (SStatus 133 SControl 300)
[ 68.393756] ata1.00: unexpected _GTF length (8)
[ 68.394215] ata1.00: unexpected _GTF length (8)
[ 68.394396] ata1.00: configured for UDMA/133
[ 68.394412] ata1.00: device reported invalid CHS sector 0
[ 68.394414] ata1.00: device reported invalid CHS sector 0
[ 68.394416] ata1.00: device reported invalid CHS sector 0
[ 68.394417] ata1.00: device reported invalid CHS sector 0
[ 68.394418] ata1.00: device reported invalid CHS sector 0
[ 68.394420] ata1.00: device reported invalid CHS sector 0
[ 68.394421] ata1.00: device reported invalid CHS sector 0
[ 68.394422] ata1.00: device reported invalid CHS sector 0
[ 68.394424] ata1.00: device reported invalid CHS sector 0
[ 68.394425] ata1.00: device reported invalid CHS sector 0
[ 68.394426] ata1.00: device reported invalid CHS sector 0
[ 68.394427] ata1.00: device reported invalid CHS sector 0
[ 68.394429] ata1.00: device reported invalid CHS sector 0
[ 68.394430] ata1.00: device reported invalid CHS sector 0
[ 68.394431] ata1.00: device reported invalid CHS sector 0
[ 68.394432] ata1.00: device reported invalid CHS sector 0
[ 68.394434] ata1.00: device reported invalid CHS sector 0
[ 68.394435] ata1.00: device reported invalid CHS sector 0
[ 68.394436] ata1.00: device reported invalid CHS sector 0
[ 68.394437] ata1.00: device reported invalid CHS sector 0
[ 68.394439] ata1.00: device reported invalid CHS sector 0
[ 68.394440] ata1.00: device reported invalid CHS sector 0
[ 68.394441] ata1.00: device reported invalid CHS sector 0
[ 68.394442] ata1.00: device reported invalid CHS sector 0
[ 68.394443] ata1.00: device reported invalid CHS sector 0
[ 68.394445] ata1.00: device reported invalid CHS sector 0
[ 68.394446] ata1.00: device reported invalid CHS sector 0
[ 68.394447] ata1.00: device reported invalid CHS sector 0
[ 68.394449] ata1.00: device reported invalid CHS sector 0
[ 68.394450] ata1.00: device reported invalid CHS sector 0
[ 68.394451] ata1.00: device reported invalid CHS sector 0
[ 68.394466] ata1: EH complete

[ 259.014808] ata1.00: exception Emask 0x0 SAct 0x7fffffff SErr 0x0 action 0x6 frozen
[ 259.014813] ata1.00: failed command: WRITE FPDMA QUEUED
[ 259.014818] ata1.00: cmd 61/78:00:c8:72:00/00:00:34:00:00/40 tag 0 ncq 61440 out
res 40/00:00:00:4f:c2/00:00:00:00:00/00 Emask 0x4 (timeout)
[ 259.014820] ata1.00: status: { DRDY }

[ 259.014978] ata1.00: status: { DRDY }
[ 259.014981] ata1: hard resetting link
[ 259.321569] ata1: SATA link up 6.0 Gbps (SStatus 133 SControl 300)
[ 259.321939] ata1.00: unexpected _GTF length (8)
[ 259.322261] ata1.00: unexpected _GTF length (8)
[ 259.322355] ata1.00: configured for UDMA/133
[ 259.322360] ata1.00: device reported invalid CHS sector 0
[ 259.322362] ata1.00: device reported invalid CHS sector 0
[ 259.322363] ata1.00: device reported invalid CHS sector 0
[ 259.322364] ata1.00: device reported invalid CHS sector 0
[ 259.322366] ata1.00: device reported invalid CHS sector 0
[ 259.322367] ata1.00: device reported invalid CHS sector 0
[ 259.322368] ata1.00: device reported invalid CHS sector 0
[ 259.322369] ata1.00: device reported invalid CHS sector 0
[ 259.322371] ata1.00: device reported invalid CHS sector 0
[ 259.322372] ata1.00: device reported invalid CHS sector 0
[ 259.322373] ata1.00: device reported invalid CHS sector 0
[ 259.322374] ata1.00: device reported invalid CHS sector 0
[ 259.322376] ata1.00: device reported invalid CHS sector 0
[ 259.322377] ata1.00: device reported invalid CHS sector 0
[ 259.322378] ata1.00: device reported invalid CHS sector 0
[ 259.322380] ata1.00: device reported invalid CHS sector 0
[ 259.322381] ata1.00: device reported invalid CHS sector 0
[ 259.322382] ata1.00: device reported invalid CHS sector 0
[ 259.322383] ata1.00: device reported invalid CHS sector 0
[ 259.322385] ata1.00: device reported invalid CHS sector 0
[ 259.322386] ata1.00: device reported invalid CHS sector 0
[ 259.322387] ata1.00: device reported invalid CHS sector 0
[ 259.322388] ata1.00: device reported invalid CHS sector 0
[ 259.322390] ata1.00: device reported invalid CHS sector 0
[ 259.322391] ata1.00: device reported invalid CHS sector 0
[ 259.322392] ata1.00: device reported invalid CHS sector 0
[ 259.322394] ata1.00: device reported invalid CHS sector 0
[ 259.322395] ata1.00: device reported invalid CHS sector 0
[ 259.322396] ata1.00: device reported invalid CHS sector 0
[ 259.322397] ata1.00: device reported invalid CHS sector 0
[ 259.322399] ata1.00: device reported invalid CHS sector 0
[ 259.322411] ata1: EH complete

Oct 30 20:34:17 localhost kernel: [ 766.718890] ata1.00: exception Emask 0x0 SAct 0x7fffffff SErr 0x0 action 0x6 frozen
Oct 30 20:34:17 localhost kernel: [ 766.718896] ata1.00: failed command: WRITE FPDMA QUEUED
Oct 30 20:34:17 localhost kernel: [ 766.718900] ata1.00: cmd 61/08:00:d8:72:40/00:00:39:00:00/40 tag 0 ncq 4096 out
Oct 30 20:34:17 localhost kernel: [ 766.718900] res 40/00:00:00:4f:c2/00:00:00:00:00/00 Emask 0x4 (timeout)
Oct 30 20:34:17 localhost kernel: [ 766.718902] ata1.00: status: { DRDY }
Oct 30 20:34:17 localhost kernel: [ 766.718904] ata1.00: failed command: WRITE FPDMA QUEUED

Oct 30 20:36:20 localhost kernel: [ 889.662663] ata1.00: exception Emask 0x0 SAct 0x7fffffff SErr 0x0 action 0x6 frozen
Oct 30 20:36:20 localhost kernel: [ 889.662669] ata1.00: failed command: WRITE FPDMA QUEUED
Oct 30 20:36:20 localhost kernel: [ 889.662673] ata1.00: cmd 61/08:00:40:b8:03/00:00:2a:00:00/40 tag 0 ncq 4096 out
Oct 30 20:36:20 localhost kernel: [ 889.662673] res 40/00:00:00:4f:c2/00:00:00:00:00/00 Emask 0x4 (timeout)
Oct 30 20:36:20 localhost kernel: [ 889.662675] ata1.00: status: { DRDY }

Oct 30 20:44:05 localhost kernel: [ 1354.379985] ata1.00: exception Emask 0x0 SAct 0x7fc00003 SErr 0x0 action 0x6 frozen
Oct 30 20:44:05 localhost kernel: [ 1354.379990] ata1.00: failed command: WRITE FPDMA QUEUED
Oct 30 20:44:05 localhost kernel: [ 1354.379995] ata1.00: cmd 61/10:00:00:70:00/00:00:2a:00:00/40 tag 0 ncq 8192 out
Oct 30 20:44:05 localhost kernel: [ 1354.379995] res 40/00:00:00:4f:c2/00:00:00:00:00/00 Emask 0x4 (timeout)
Oct 30 20:44:05 localhost kernel: [ 1354.379997] ata1.00: status: { DRDY }

[Gareth Jones]

Looks like dodgy hardware to me. Does it work with any other OS?

[Balmung]

Works fine with OS X Mavericks and Windows 8.1.

dodgy? No idea what this means.

Searching the internet for this error gives the impression, that this error is rather close to the hardware,
I found the advice to check the cables, to change the ports and so on.

Unfortunately the ssd is fixed in the macbook pro.

Could it help to report this problem to some kernel guys?

[Gareth Jones]

(Moved to the Mac section.)

Works fine with OS X Mavericks and Windows 8.1.

dodgy? No idea what this means.

Sorry. By “dodgy” I meant that it looks like there’s something not right about the hardware. That’s still possible even if it works in other OSes—sometimes drivers on different OSes are more or less tolerant of hardware faults. Or it could be a driver bug, but that seems unlikely with something as fundamental as SATA.

Could it help to report this problem to some kernel guys?

I’d wait a bit longer to see if anyone else has any suggestions, and search around and experiment a bit more, before reporting it as a bug. Search for the problem with this specific model if you can, to see if anyone else is having that problem. If not, then it’s more likely to be bad hardware. If it’s a driver bug, you probably won’t have much luck getting it fixed unless there are several others seeing it, or you fix it yourself.

[Balmung]

Thanks, I'll follow your advice.

[Balmung]

Now I also experience READ FPDMA QUEUED

[ 288.540797] ata1.00: exception Emask 0x0 SAct 0x1 SErr 0x0 action 0x6 frozen
[ 288.540803] ata1.00: failed command: READ FPDMA QUEUED
[ 288.540807] ata1.00: cmd 60/08:00:00:59:6f/00:00:37:00:00/40 tag 0 ncq 4096 in
[ 288.540807] res 40/00:00:00:4f:c2/00:00:00:00:00/00 Emask 0x4 (timeout)
[ 288.540809] ata1.00: status: { DRDY }
[ 288.540812] ata1: hard resetting link
[ 288.847595] ata1: SATA link up 6.0 Gbps (SStatus 133 SControl 300)
[ 288.847952] ata1.00: unexpected _GTF length (8)
[ 288.848421] ata1.00: unexpected _GTF length (8)
[ 288.848566] ata1.00: configured for UDMA/133
[ 288.848573] ata1.00: device reported invalid CHS sector 0
[ 288.848582] ata1: EH complete

[Balmung]

smartd finds no problems:

# smartctl -l selftest /dev/sda
smartctl 6.2 2013-07-26 r3841 [x86_64-linux-3.11.9-200.fc19.x86_64] (local build)
Copyright (C) 2002-13, Bruce Allen, Christian Franke, www.smartmontools.org

=== START OF READ SMART DATA SECTION ===
SMART Self-test log structure revision number 1
Num Test_Description Status Remaining LifeTime(hours) LBA_of_first_error
# 1 Extended offline Completed without error 00% 0 -
# 2 Short offline Completed without error 00% 0 -

[codacarlson]

In order to fix the freezing problem add the following to /etc/rc.d/rc.local. If the file doesn't exist create it and make sure its executable: chmod 755 /etc/rc.d/rc.local. What is happening is native command queing is enabled which should only be enabled on hard drives not ssds. Please let me know if that fixes your issue. Also make sure you change sda to whatever your drive is.

echo 1 > /sys/block/sda/device/queue_depth

[chrismurphy]

Unexpected _GTF length (8) MacBook Air 2013
http://www.spinics.net/lists/linux-ide/msg46306.html

[Balmung]

In order to fix the freezing problem add the following to /etc/rc.d/rc.local. If the file doesn't exist create it and make sure its executable: chmod 755 /etc/rc.d/rc.local. What is happening is native command queing is enabled which should only be enabled on hard drives not ssds. Please let me know if that fixes your issue. Also make sure you change sda to whatever your drive is.

echo 1 > /sys/block/sda/device/queue_depth

Thanks, codacarlson. I followed your suggestion and it looks like the WRITE FPDMA QUEUED and
with it the freezes are gone.

Thanks again!

BTW: I had to add a first line

#!/bin/sh

to rc.local to fix this error

rc-local.service: control process exited, code=exited status=203

.