Bug 177108 - [mvs] mvs driver with MV88SX6081 have timeout waiting for write DRQ [regression]
Summary: [mvs] mvs driver with MV88SX6081 have timeout waiting for write DRQ [regression]
Status: Open
Alias: None
Product: Base System
Classification: Unclassified
Component: kern (show other bugs)
Version: 9.1-PRERELEASE
Hardware: Any Any
: Normal Affects Only Me
Assignee: freebsd-bugs (Nobody)
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2013-03-19 13:50 UTC by Goran Lowkrantz
Modified: 2018-05-20 23:50 UTC (History)
1 user (show)

See Also:


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description Goran Lowkrantz 2013-03-19 13:50:00 UTC
When upgrading from FreeBSD 8.2-STABLE #0: Sun Sep 25 03:26:06 CEST 2011
to FreeBSD 9.1-PRERELEASE #0 r243780M: Sun Dec  2 22:57:58 UTC 2012 we
found that the a number of devices over two SUPERMICRO AOC-SAT2-MV8 64-bit
PCI-X133MHz SATA II (3.0Gb/s) Controller Card started to experience
timeouts. They work just fine under 8.2 but are unusable due to the
timeouts in 9.1.

Disks are direct attached, no port multiplier.

Problem devices are ada1, ada2, ada8, ada9, ada10, ada11.

Errors are a mix of these two on the affected devices.
mvsch8: Wait status 50
mvsch8: timeout waiting for write DRQ
(ada8:mvsch8:0:0:0): WRITE_MUL48. ACB: 39 00 a8 49 06 40 95 00 00 00 01 00
(ada8:mvsch8:0:0:0): CAM status: Command timeout
(ada8:mvsch8:0:0:0): Retrying command
mvsch2: Wait status 50
mvsch2: timeout waiting for write DRQ
(ada2:mvsch2:0:0:0): WRITE_MUL. ACB: c5 00 64 73 05 4c 00 00 00 00 03 00
(ada2:mvsch2:0:0:0): CAM status: Command timeout
(ada2:mvsch2:0:0:0): Retrying command


camcontrol devlist -v:
scbus1 on mvsch0 bus 0:
<ST31500341AS CC1H>                at scbus1 target 0 lun 0 (pass2,ada0)
scbus2 on mvsch1 bus 0:
<ST31500341AS CC1H>                at scbus2 target 0 lun 0 (pass3,ada1)
scbus3 on mvsch2 bus 0:
<WDC WD2003FYYS-02W0B0 01.01D01>   at scbus3 target 0 lun 0 (pass4,ada2)
scbus4 on mvsch3 bus 0:
<WDC WD2003FYYS-02W0B0 01.01D01>   at scbus4 target 0 lun 0 (pass5,ada3)
scbus5 on mvsch4 bus 0:
<ST2000DM001-1CH164 CC43>          at scbus5 target 0 lun 0 (pass6,ada4)
scbus6 on mvsch5 bus 0:
<ST2000DM001-1CH164 CC43>          at scbus6 target 0 lun 0 (pass7,ada5)
scbus7 on mvsch6 bus 0:
<ST2000DM001-1CH164 CC43>          at scbus7 target 0 lun 0 (pass8,ada6)
scbus8 on mvsch7 bus 0:
<ST3000DM001-9YN166 CC4H>          at scbus8 target 0 lun 0 (pass9,ada7)
scbus9 on mvsch8 bus 0:
<ST31500341AS CC1H>                at scbus9 target 0 lun 0 (pass10,ada8)
scbus10 on mvsch9 bus 0:
<ST31500341AS CC1H>                at scbus10 target 0 lun 0 (pass11,ada9)
scbus11 on mvsch10 bus 0:
<WDC WD2003FYYS-02W0B0 01.01D01>   at scbus11 target 0 lun 0 (pass12,ada10)
scbus12 on mvsch11 bus 0:
<WDC WD2003FYYS-02W0B0 01.01D01>   at scbus12 target 0 lun 0 (pass13,ada11)
scbus13 on mvsch12 bus 0:
<ST2000DM001-1CH164 CC43>          at scbus13 target 0 lun 0 (pass14,ada12)
scbus14 on mvsch13 bus 0:
<ST2000DM001-1CH164 CC43>          at scbus14 target 0 lun 0 (pass15,ada13)
scbus15 on mvsch14 bus 0:
<ST2000DM001-1CH164 CC43>          at scbus15 target 0 lun 0 (pass16,ada14)
scbus16 on mvsch15 bus 0:
<INTEL SSDSA2M160G2GC 2CV102HD>    at scbus16 target 0 lun 0 (pass17,ada15)

Controllers:
mvs0: <Marvell 88SX6081 SATA controller> port 0xa800-0xa8ff mem 0xfb800000-0xfb8fffff,0xfb400000-0xfb7fffff irq 28 at device 2.0 on pci6
mvs0: Gen-II, 8 3Gbps ports, Port Multiplier supported
mvs0: [ITHREAD]
mvsch0: <Marvell SATA channel> at channel 0 on mvs0
mvsch0: [ITHREAD]
mvsch1: <Marvell SATA channel> at channel 1 on mvs0
mvsch1: [ITHREAD]
mvsch2: <Marvell SATA channel> at channel 2 on mvs0
mvsch2: [ITHREAD]
mvsch3: <Marvell SATA channel> at channel 3 on mvs0
mvsch3: [ITHREAD]
mvsch4: <Marvell SATA channel> at channel 4 on mvs0
mvsch4: [ITHREAD]
mvsch5: <Marvell SATA channel> at channel 5 on mvs0
mvsch5: [ITHREAD]
mvsch6: <Marvell SATA channel> at channel 6 on mvs0
mvsch6: [ITHREAD]
mvsch7: <Marvell SATA channel> at channel 7 on mvs0
mvsch7: [ITHREAD]
pcib11: <ACPI PCI-PCI bridge> at device 0.2 on pci4
mvs1: <Marvell 88SX6081 SATA controller> port 0x9800-0x98ff mem 0xfa700000-0xfa7fffff,0xfac00000-0xfaffffff irq 48 at device 1.0 on pci5
mvs1: Gen-II, 8 3Gbps ports, Port Multiplier supported
mvs1: [ITHREAD]
mvsch8: <Marvell SATA channel> at channel 0 on mvs1
mvsch8: [ITHREAD]
mvsch9: <Marvell SATA channel> at channel 1 on mvs1
mvsch9: [ITHREAD]
mvsch10: <Marvell SATA channel> at channel 2 on mvs1
mvsch10: [ITHREAD]
mvsch11: <Marvell SATA channel> at channel 3 on mvs1
mvsch11: [ITHREAD]
mvsch12: <Marvell SATA channel> at channel 4 on mvs1
mvsch12: [ITHREAD]
mvsch13: <Marvell SATA channel> at channel 5 on mvs1
mvsch13: [ITHREAD]
mvsch14: <Marvell SATA channel> at channel 6 on mvs1
mvsch14: [ITHREAD]
mvsch15: <Marvell SATA channel> at channel 7 on mvs1
mvsch15: [ITHREAD]


/glz

How-To-Repeat: Boot 9.1 image.
Access file systems on any of the devices.
Comment 1 Joseph Brower 2015-06-29 13:29:55 UTC
We are also affected by this bug, in this case we're on 9.3.  It would be awesome to see some motion on this since we'd like not needing to swap out hardware that is perfectly good.  We can easily provide access to hardware that is exhibiting this behaviour along with making it easily reproduced (takes about 5 minutes from bootup to start seeing this happen.)
Comment 2 Eitan Adler freebsd_committer freebsd_triage 2018-05-20 23:50:22 UTC
For bugs matching the following conditions:
- Status == In Progress
- Assignee == "bugs@FreeBSD.org"
- Last Modified Year <= 2017

Do
- Set Status to "Open"