[LTCTest][OPAL][OP930] Machine hangs after injecting the Machine Check Error
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
The Ubuntu-power-systems project |
Fix Released
|
Critical
|
Canonical Kernel Team | ||
linux (Ubuntu) |
Fix Released
|
Critical
|
Manoj Iyer | ||
Bionic |
Fix Released
|
Critical
|
Manoj Iyer | ||
Disco |
Fix Released
|
Critical
|
Manoj Iyer | ||
Eoan |
Fix Released
|
Critical
|
Manoj Iyer |
Bug Description
[IMPACT]
MCE test renders the system unresponsive on P9 open power hardware (Withersoon)
[TEST]
A test kernel is available in ppa:ubuntu-
[FIX]
IBM has identified the following patch that fixes this issue:
commit 99ead78afd1128b
Author: Balbir Singh <email address hidden>
Date: Tue Aug 20 13:43:47 2019 +0530
powerpc/mce: Fix MCE handling for huge pages
[REGRESSION POTENTIAL]
The patch is applicable the powerpc architecture and limited in scope to MCE handling for huge pages. Patch does not touch any generic code. Regression if any is limited to powerpc MCE handling.
[OTHER]
== Comment: #0 - PAVAMAN SUBRAMANIYAM <email address hidden> - 2019-05-07 23:31:20 ==
Install a P9 Open Power Hardware with the latest OP930 Firmware images built from the upstream op-build git tree.
root@witherspoon:~# cat /etc/os-release
ID="openbmc-
NAME="Phosphor OpenBMC (Phosphor OpenBMC Project Reference Distro)"
VERSION="ibm-v2.3"
VERSION_
PRETTY_
BUILD_ID=
root@witherspoon:~# cat /var/lib/
open-power-
occ-58e422d
sbe-249671d
Then enable sw xstop manually by using below command:
root@ltc-
root@ltc-
"ibm,skiboot" Partition
-------
experimental-
snarf-mode=noooooo
opal-sw-
Then from the Linux HOST injected the MCE UE Error on the machine as follows:
root@ltc-
CHIP ID: 0 CORE ID: 0 THREADS: 4 CPUs: 0 1 2 3
CHIP ID: 0 CORE ID: 1 THREADS: 4 CPUs: 4 5 6 7
CHIP ID: 0 CORE ID: 2 THREADS: 4 CPUs: 8 9 10 11
CHIP ID: 0 CORE ID: 3 THREADS: 4 CPUs: 12 13 14 15
CHIP ID: 0 CORE ID: 6 THREADS: 4 CPUs: 16 17 18 19
CHIP ID: 0 CORE ID: 7 THREADS: 4 CPUs: 20 21 22 23
CHIP ID: 0 CORE ID: 8 THREADS: 4 CPUs: 24 25 26 27
CHIP ID: 0 CORE ID: 9 THREADS: 4 CPUs: 28 29 30 31
CHIP ID: 0 CORE ID: 10 THREADS: 4 CPUs: 32 33 34 35
CHIP ID: 0 CORE ID: 11 THREADS: 4 CPUs: 36 37 38 39
CHIP ID: 0 CORE ID: 12 THREADS: 4 CPUs: 40 41 42 43
CHIP ID: 0 CORE ID: 13 THREADS: 4 CPUs: 44 45 46 47
CHIP ID: 0 CORE ID: 16 THREADS: 4 CPUs: 48 49 50 51
CHIP ID: 0 CORE ID: 17 THREADS: 4 CPUs: 52 53 54 55
CHIP ID: 0 CORE ID: 18 THREADS: 4 CPUs: 56 57 58 59
CHIP ID: 0 CORE ID: 19 THREADS: 4 CPUs: 60 61 62 63
CHIP ID: 0 CORE ID: 20 THREADS: 4 CPUs: 64 65 66 67
CHIP ID: 0 CORE ID: 21 THREADS: 4 CPUs: 68 69 70 71
CHIP ID: 8 CORE ID: 6 THREADS: 4 CPUs: 72 73 74 75
CHIP ID: 8 CORE ID: 7 THREADS: 4 CPUs: 76 77 78 79
CHIP ID: 8 CORE ID: 8 THREADS: 4 CPUs: 80 81 82 83
CHIP ID: 8 CORE ID: 9 THREADS: 4 CPUs: 84 85 86 87
CHIP ID: 8 CORE ID: 10 THREADS: 4 CPUs: 88 89 90 91
CHIP ID: 8 CORE ID: 11 THREADS: 4 CPUs: 92 93 94 95
CHIP ID: 8 CORE ID: 12 THREADS: 4 CPUs: 96 97 98 99
CHIP ID: 8 CORE ID: 13 THREADS: 4 CPUs: 100 101 102 103
CHIP ID: 8 CORE ID: 14 THREADS: 4 CPUs: 104 105 106 107
CHIP ID: 8 CORE ID: 15 THREADS: 4 CPUs: 108 109 110 111
CHIP ID: 8 CORE ID: 16 THREADS: 4 CPUs: 112 113 114 115
CHIP ID: 8 CORE ID: 17 THREADS: 4 CPUs: 116 117 118 119
CHIP ID: 8 CORE ID: 18 THREADS: 4 CPUs: 120 121 122 123
CHIP ID: 8 CORE ID: 19 THREADS: 4 CPUs: 124 125 126 127
CHIP ID: 8 CORE ID: 20 THREADS: 4 CPUs: 128 129 130 131
CHIP ID: 8 CORE ID: 21 THREADS: 4 CPUs: 132 133 134 135
CHIP ID: 8 CORE ID: 22 THREADS: 4 CPUs: 136 137 138 139
CHIP ID: 8 CORE ID: 23 THREADS: 4 CPUs: 140 141 142 143
-------
p[0]
eq[0,1,2,3,4,5]
ex[0,
c[0,
p[8]
eq[1,2,3,4,5]
ex[3,
c[6,
-------
----------Processor Layout-
p[0]
|EX-0 C0 | |EX-4 C8 | |EX-8 C16|
+ - - - - - + + - - - - - + + - - - - - +
|EX-0 C1 | |EX-4 C9 | |EX-8 C17|
+ - - - - - + + - - - - - + + - - - - - +
|EX-1 C2 | |EX-5 C10| |EX-9 C18|
+ - - - - - + + - - - - - + + - - - - - +
|EX-1 C3 | |EX-5 C11| |EX-9 C19|
| | |EX-6 C12| |EX-10 C20|
+ - - - - - + + - - - - - + + - - - - - +
| | |EX-6 C13| |EX-10 C21|
+ - - - - - + + - - - - - + + - - - - - +
|EX-3 C6 | | | | |
+ - - - - - + + - - - - - + + - - - - - +
|EX-3 C7 | | | | |
p[8]
| | |EX-4 C8 | |EX-8 C16|
+ - - - - - + + - - - - - + + - - - - - +
| | |EX-4 C9 | |EX-8 C17|
+ - - - - - + + - - - - - + + - - - - - +
| | |EX-5 C10| |EX-9 C18|
+ - - - - - + + - - - - - + + - - - - - +
| | |EX-5 C11| |EX-9 C19|
| | |EX-6 C12| |EX-10 C20|
+ - - - - - + + - - - - - + + - - - - - +
| | |EX-6 C13| |EX-10 C21|
+ - - - - - + + - - - - - + + - - - - - +
|EX-3 C6 | |EX-7 C14| |EX-11 C22|
+ - - - - - + + - - - - - + + - - - - - +
|EX-3 C7 | |EX-7 C15| |EX-11 C23|
root@ltc-
./statedisable.sh: line 10: /sys/devices/
./statedisable.sh: line 11: /sys/devices/
root@ltc-
CPUidle driver: powernv_idle
CPUidle governor: menu
analyzing CPU 0:
Number of idle states: 7
Available idle states: snooze stop0_lite stop0 stop1 stop2 stop4 stop5
snooze (DISABLED) :
Flags/Description: snooze
Latency: 0
Usage: 81861
Duration: 29748269
stop0_lite (DISABLED) :
Flags/Description: stop0_lite
Latency: 1
Usage: 70
Duration: 1982345
stop0 (DISABLED) :
Flags/Description: stop0
Latency: 2
Usage: 274
Duration: 125896
stop1 (DISABLED) :
Flags/Description: stop1
Latency: 5
Usage: 36
Duration: 4922
stop2 (DISABLED) :
Flags/Description: stop2
Latency: 10
Usage: 3745
Duration: 88300041
stop4 (DISABLED) :
Flags/Description: stop4
Latency: 100
Usage: 65
Duration: 1048951
stop5 (DISABLED) :
Flags/Description: stop5
Latency: 200
Usage: 30377
Duration: 61977191643
root@ltc-
root@ltc-
EQ[ 3]: 0x1301080c
EX[ 7]: 0x13010c0c
C[15]: 0x3f01080c
root@ltc-
0000000000000000
root@ltc-
0c00000000000000
root@ltc-
0c00000000000000
After injecting the Machine check error, the HOST Linux stops pinging and the console access to the machine also gets lost.
But still the Open BMC shell and GUI still shows that the HOST is in Running state.
== Comment: #1 - PAVAMAN SUBRAMANIYAM <email address hidden> - 2019-05-07 23:33:31 ==
The machine is installed with the Ubuntu 18.04 Linux OS.
root@ltc-
Linux ltc-wspoon11 4.15.0-48-generic #51-Ubuntu SMP Wed Apr 3 08:26:19 UTC 2019 ppc64le ppc64le ppc64le GNU/Linux
root@ltc-
NAME="Ubuntu"
VERSION="18.04.2 LTS (Bionic Beaver)"
ID=ubuntu
ID_LIKE=debian
PRETTY_NAME="Ubuntu 18.04.2 LTS"
VERSION_ID="18.04"
HOME_URL="https:/
SUPPORT_URL="https:/
BUG_REPORT_URL="https:/
PRIVACY_
VERSION_
UBUNTU_
root@ltc-
cpu : POWER9, altivec supported
clock : 2300.000000MHz
revision : 2.3 (pvr 004e 1203)
timebase : 512000000
platform : PowerNV
model : 8335-GTH
machine : PowerNV 8335-GTH
firmware : OPAL
MMU : Radix
root@ltc-
Version of System Firmware :
Product Name : OpenPOWER Firmware
Product Version : witherspoon-
Product Extra : skiboot-v6.3-rc2
Product Extra : bmc-firmware-
Product Extra : occ-58e422d
Product Extra : hostboot-19a436e
Product Extra : buildroot-
Product Extra : capp-ucode-
Product Extra : machine-xml-a6f4df3
Product Extra : hostboot-
Product Extra : sbe-249671d
Product Extra : hcode-hw040319a.940
Product Extra : petitboot-v1.10.3
Product Extra : linux-5.
== Comment: #3 - PAVAMAN SUBRAMANIYAM <email address hidden> - 2019-05-07 23:42:35 ==
I quickly tested MCE on op930 build ( IBM-witherspoon
== Comment: #9 - Application Cdeadmin <email address hidden> - 2019-05-22 06:45:07 ==
==== State: Working by: jayeshp on 22 May 2019 06:37:27 ====
Any update?
== Comment: #11 - MAHESH J. SALGAONKAR <email address hidden> - 2019-09-19 04:44:01 ==
The hang issues should go away with below patch.
commit 99ead78afd1128b
Author: Balbir Singh <email address hidden>
Date: Tue Aug 20 13:43:47 2019 +0530
powerpc/mce: Fix MCE handling for huge pages
The current code would fail on huge pages addresses, since the shift would
be incorrect. Use the correct page shift value returned by
__find_
generic and can handle both regular and compound pages.
Fixes: ba41e1e1ccb9 ("powerpc/mce: Hookup derror (load/store) UE errors")
Signed-off-by: Balbir Singh <email address hidden>
[<email address hidden>: Fixup pseries_
Signed-off-by: Reza Arbab <email address hidden>
Tested-by: Mahesh Salgaonkar <email address hidden>
Signed-off-by: Santosh Sivaraj <email address hidden>
Cc: <email address hidden> # v4.15+
Signed-off-by: Michael Ellerman <email address hidden>
Link: https://<email address hidden>
affects: | kernel-package (Ubuntu) → linux (Ubuntu) |
Changed in ubuntu-power-systems: | |
status: | New → Triaged |
importance: | Undecided → Critical |
assignee: | nobody → Canonical Kernel Team (canonical-kernel-team) |
Changed in linux (Ubuntu): | |
assignee: | Ubuntu on IBM Power Systems Bug Triage (ubuntu-power-triage) → Canonical Kernel Team (canonical-kernel-team) |
importance: | Undecided → Critical |
Changed in linux (Ubuntu Eoan): | |
assignee: | Canonical Kernel Team (canonical-kernel-team) → Manoj Iyer (manjo) |
description: | updated |
Changed in linux (Ubuntu Eoan): | |
status: | New → In Progress |
Changed in ubuntu-power-systems: | |
status: | Triaged → In Progress |
Changed in ubuntu-power-systems: | |
status: | Incomplete → In Progress |
Changed in linux (Ubuntu Disco): | |
status: | Incomplete → In Progress |
Changed in linux (Ubuntu Bionic): | |
status: | Incomplete → In Progress |
Changed in linux (Ubuntu Bionic): | |
status: | Fix Committed → Fix Released |
Changed in linux (Ubuntu Disco): | |
status: | Fix Committed → Fix Released |
Changed in linux (Ubuntu Eoan): | |
status: | Fix Committed → Fix Released |
Changed in linux (Ubuntu): | |
status: | Fix Committed → Fix Released |
Changed in ubuntu-power-systems: | |
status: | Fix Committed → Fix Released |
Default Comment by Bridge