正文文本
Server
Lenovo OEM/Extended SEL Events
Design Specification
Revision: 0.1
Initial Writer: Lawrence Li
Initial Date: 1/3, 2017
Document Control Information
Role Name Date Company / E-mail Address
Dept
Author
PD Development
SW
PA
Approver
Lenovo OEM/Extended SEL Events Design Specification
Copy Rights
Copyright © 2001 – 2014 Lenovo Group
All Rights Reserved.
Lenovo Group
No.6 Chuang Ye Road, Haidian District,Beijing China 100085
This publication contains proprietary information which is protected by copyright. No part of this publication may be reproduced,
transcribed, stored in a retrieval system, translated into any language or computer language, or transmitted in any form whatsoever
without the prior written consent of the publisher, Lenovo Group and American Megatrends, Inc.
Trademark and Copyright Acknowledgments
All product names used in this publication are for identification purposes only and are trademarks of their respective Companies.
2
Lenovo OEM/Extended SEL Events Design Specification
Revision History
Revision Date Author Description of Changes
0.1 2017/1/3 Lawrence Initial Release for Purley
3
Lenovo OEM/Extended SEL Events Design Specification
Table of Contents
COPY RIGHTS ...................................................................................................................................................................... 2
REVISION HISTORY .............................................................................................................................................................. 3
TABLE OF CONTENTS ........................................................................................................................................................... 4
1. POST ERROR LOG WITH LENOVO EXTENDED SEL EVENT FORMATS ................................................................................ 5
1.1 QPI RC ERRORS ..................................................................................................................................................................... 5
1.2 MRC ERRORS ........................................................................................................................................................................ 6
1.3 PCIE PERFORMANCE DEGRADATION .......................................................................................................................................... 7
2. POST ERROR LOG WITH STANDARD SEL EVENT FORMATS ............................................................................................. 8
2.1 MICROCODE ERRORS .............................................................................................................................................................. 8
2.2 BMC REQUESTED NMI ........................................................................................................................................................... 8
2.3 CPU EXCEPTION .................................................................................................................................................................... 8
2.4 NO BOOTABLE DEVICE FOUND .................................................................................................................................................. 8
2.5 CMOS CLEARED OR RTC RESET .............................................................................................................................................. 9
2.6 ME VERVISON MISMATCH DETECTED ......................................................................................................................................... 9
3. RUNTIME ERROR LOGGING AND FRU ISOLATING ........................................................................................................ 10
3.1 MEMORY ERROR .................................................................................................................................................................. 10
3.2 AER (PCIE DEVICES) ............................................................................................................................................................ 11
3.3 RUNTIME PROCESSOR ERRORS ............................................................................................................................................ 13
3.4 IIO CORE ............................................................................................................................................................................ 14
3.5 IIO-COHERENCE INTERFACE (IIO QPI)...................................................................................................................................... 15
3.6 IIO-VTD ............................................................................................................................................................................. 16
4. SYSTEM FAULT LED .................................................................................................................................................... 17
4
Lenovo OEM/Extended SEL Events Design Specification
1. POST ERROR LOG WITH LENOVO EXTENDED SEL EVENT FORMATS
This section describes the SEL formats of different Lenovo Extended SEL events. SEL viewer utility
development teams can use this document.
1.1 QPI RC ERRORS
SEL:
Sensor Type 0x7 Processor
Sensor Number Per BMC spec design
Event Dir/Type 0x6F
Event data 1 Bit[7:4] = 0xA (Indicates that OEMEvData2 and OEMEvData3 are OEM value)
Bit[3:0] = Error Type (PROCESSOR CONFIGURATION ERROR)
Event data 2 Bit[7:2] = Reserved
Bit[1:0] =
0 – QPI Warning
1 – QPI Non-Fatal Error
2 – QPI Fatal Error
Event data 3 Bit[7:4] = CPU Socket Number
Bit[3:0] = QPI Port number
ExtSEL:
Sensor Type 0x7 Processor
Sensor Number Per BMC spec design
Event Dir/Type 0x70 – OEM Event
Event Data 1-3 Filled by BMC
Extended SEL 0x5 - Error Type (PROCESSOR CONFIGURATION ERROR)
Data 1
Extended SEL Bit[7:2] = Reserved
Data 2 Bit[1:0] =
0 – QPI Warning
1 – QPI Non-Fatal Error
2 – QPI Fatal Error
Extended SEL Bit[7:4] = CPU Socket Number
Data 3 Bit[3:0] = QPI Port number
Extended SEL Bit[7:0] = Major Code based on Purley RC User Guide.
5
Lenovo OEM/Extended SEL Events Design Specification
Data 4
Extended SEL Bit[7:0] = Minor Code based on Purley RC User Guide.
Data 5
1.2 MRC ERRORS
SEL:
Sensor Type 0xC Memory
Sensor Number Per BMC spec design
Event Dir/Type 0x6F
Event data 1 Bit[7:4] = 0xA (Indicates that OEMEvData2 and OEMEvData3 are OEM value)
Bit[3:0] = 0x7 - Error Type (MEMORY CONFIGURATION ERROR)
Event data 2 Bit[7:4] = CPU Socket Number
Bit[3:0] = Memory Channel number
Event data 3 Bit[7:4] = DIMM slot number
Bit[3:0] = DIMM Rank number
ExtSEL:
Sensor Type 0xC Memory
Sensor Number Per BMC spec design
Event Dir/Type 0x70 – OEM Event
Event Data 1-3 Filled by BMC
Extended SEL 0x7 - Error Type (MEMORY CONFIGURATION ERROR)
Data 1
Extended SEL Bit[7:2] = Reserved
Data 2 Bit[1:0] =
0 – MRC Warning
1 – MRC Non-Fatal Error
2 – MRC Fatal Error
Extended SEL Bit[7:4] = CPU Socket Number
Data 3 Bit[3:0] = Memory Channel number
Extended SEL Bit[7:4] = DIMM slot number
Data 4 Bit[3:0] = DIMM Rank number
6
Lenovo OEM/Extended SEL Events Design Specification
Extended SEL Bit[7:0] = Major Code based on Purley RC User Guide.
Data 5
Extended SEL Bit[7:0] = Minor Code based on Purley RC User Guide.
Data 6
1.3 PCIE PERFORMANCE DEGRADATION
Extended SEL format:
Sensor Type 0x21 Slot / Connector
Sensor Number Per BMC spec design
Event data 1 Bit[7:4] = 0xA (Indicates that OEMEvData2 and OEMEvData3 are OEM value)
Bit [3:0] = 0 (Error Type - PCIE Performance Degradation)
Event data 2 Bit[7-3] = Reserved
Bit[2:0] = Degradation Type
BIT 0 – Link speed degraded
BIT 1 – Link width degraded
BIT 2 – Link training failed
Event data 3 Bit[7:0] = Slot Number
7
Lenovo OEM/Extended SEL Events Design Specification
2. POST ERROR LOG WITH STANDARD SEL EVENT FORMATS
These messages are covered under the IPMI spec, but may have additional OEM data included.
2.1 MICROCODE ERRORS
Sensor Type 0x7 Processor Microcode
Sensor Number Per BMC spec design
Event data 1 Bit[7:4] = 0xF
Bit[3:0] = 0x5
Event data 2 Bit[7:6] = Reserved
Bit[5:0] = CPU number
Event data 3 0x2
2.2 BMC REQUESTED NMI
Sensor Type 0x13 Critical Interrupt
Sensor Number Per BMC spec design
Event Type 0x6f Sensor Specific Offset
Event data 1 Bit[7:4] = 0x0 (Indicates that OEMEvData2 and OEMEvData3 are unspecified)
Bit[3:0] = 0x03 Software SMI
Event data 2 Bit [7:0] = 0
Event data 3 Bit [7:0] = 0
2.3 CPU EXCEPTION
Sensor Type 0x0F System Firmware Progress
Sensor Number Per BMC spec design
Event Type 0x6F
Event data 1 Bit[7:4] = 0xA (Indicates that OEMEvData2 and OEMEvData3 is OEM values)
Bit [3:0] = 0x1 System Firmware Hang
Event data 2 Bit[7:0] = 0 unspecified
Event data 3 Bit[7:0] = Exception Number
2.4 NO BOOTABLE DEVICE FOUND
Sensor Type 0x1E Boot Error
8
Lenovo OEM/Extended SEL Events Design Specification
Sensor Number Per BMC spec design
Event Type 0x6f sensor specific offset
Event data 1 Bit[7:4] = 0x0 (Indicates that OEMEvData2 and OEMEvData3 are unspecified)
Bit[3:0] = 0x00 No Bootable media
Event data 2 Bit [7:0] = 0xFF (unspecified)
Event data 3 Bit [7:0] = 0xFF (unspecified)
2.5 CMOS CLEARED OR RTC RESET
Record ID Record ID
Record Type 02 – System Event Record
Timestamp Timestamp
Generator ID 0x01
EvmRev 4
Sensor Type 0x29 Battery
Sensor Number Per BMC spec design
Event Dir | 0x6F
EventType
Event data 1 Bit[7:4] - 0x0 - Indicates byte 2 and byte3 are reserved.
Bit[3:0] – 0x1 - Battery failed.
Event data 2 Reserved
Event data 3 Reserved
Note: For runtime errors current BIOS supports logging both SEL and extended SEL events, as Lenovo current web
utilities does not support reading the extended SEL events.
2.6 ME VERVISON MISMATCH DETECTED
Record ID Record ID
Record Type 02 – System Event Record
Timestamp Timestamp
Generator ID 0x01
EvmRev 4
Sensor Type 0x2B Version Change
Sensor Number Per BMC spec design
Event Dir | 0x6F
EventType
Event data 1 Bit[7:4] - 0x0 - Indicates byte 2 and byte3 are reserved.
Bit[3:0] – 0x3 - Firmware or software incompatibility detected with associated Entity.
Event data 2 Reserved
9
Lenovo OEM/Extended SEL Events Design Specification
Event data 3 Reserved
3. RUNTIME ERROR LOGGING AND FRU ISOLATING
3.1 MEMORY ERROR
Lenovo Correctable error Logging Policy:
1. If system encounters 160 correctable errors in a DIMM rank in 10 seconds or 2560 errors in 24 hours, BIOS
will log one SEL log to BMC.
2. If system encounters 1000 correctable errors in a DIMM rank in 10 seconds, then disable the SMI of the
memory controller. The memory connected to this controller will not report error until next reset.
3. Memory Corrected/Uncorrected error are in same formats.
SEL:
Byte Field Description
1-2 Record ID Record ID
3 Record Type 02 – System Event Record
4-7 Timestamp Timestamp
8-9 Generator ID 0x21
10 EvmRev 4
11 Sensor Type 0xC Memory
12 Sensor Number Per BMC spec design
13 Event Dir | EventType 0x6F
14 Event data 1 Bit[7:4] = 0xA (Indicates that OEMEvData2 and OEMEvData3 are
OEM values)
Bit[3:0] = Error Type
Error Type:
Correctable Memory Error: 0x00
Uncorrectable Memory Error: 0x01
Parity Error: 0x02
Correctable Memory Error Logging Limit Reached: 0x05
15 Event data 2 Bit[0] = Current/Last Boot Error
Current Boot : 0
Last Boot : 1
Bit[2:1] = Memory Controller Number
16 Event data 3 Bit[7:6]: CPU Socket Number.
Bit[5:4]: Channel Number.
Bit[3:0]: DIMM Number.
*Note: This SEL will turn on fault LED.
ExtSEL
Byte Field Description
1-2 Record ID Record ID
3 Record Type 0xDF – OEM System Event Record
4-7 Timestamp Timestamp
8-9 Generator ID 0x21
10 EvmRev 0
10
Lenovo OEM/Extended SEL Events Design Specification
11 Sensor Type 0xC Memory
12 Sensor Number Per BMC spec design
13 Event Dir | EventType 0x70
14 Reserved 1Byte (modified as 0xAA by BMC)
15 Reserved 1Byte (modified by BMC)
16 Resreved 1Byte (modified by BMC)
17 Ext Event data 1 Bit[7:4] = 0xA (Indicates that OEMEvData2 and OEMEvData3
are OEM values)
Bit[3:0] = Error Type
Error Type:
Correctable Memory Error: 0x00
Uncorrectable Memory Error: 0x01
Parity Error: 0x02
Correctable Memory Error Logging Limit Reached:
0x05
18 Ext Event data 2 Bit[0] = Current/Last Boot Error
Current Boot : 0
Last Boot : 1
Bit[2:1] = Memory Controller Number
19 Ext Event data 3 Bit[7:6]: CPU Socket Number.
Bit[5:4]: Channel Number.
Bit[3:0]: DIMM Number.
20-27 Ext Event data 4-11 MC Bank Status Register
28-35 Ext Event data 12-19 MC Bank Address Register
36-43 Ext Event data 20-27 MC Bank Misc Register
44 Ext Event data 28 MC Bank number
3.2 AER (PCIE DEVICES)
SEL:
Byte Field Description
1-2 Record ID Record ID
3 Record Type 02 – System Event Record
4-7 Timestamp Timestamp
8-9 Generator ID 0x21
10 EvmRev 4
11 Sensor Type SEL_SENS_TYPE_CRIT_ERR <0x13>
12 Sensor Number
Per BMC spec design
13 Event Dir | EventType 0x6F
14 Event data 1 Bit[7:4] = 0xA (Indicates that OEMEvData2 and OEMEvData3 are OEM
11
Lenovo OEM/Extended SEL Events Design Specification
values)
Bit[3:0] = Error Type
Error Type
PCI PERR - 0x04
PCI SERR - 0x05
Bus Correctable Error - 0x07
Bus Uncorrectable Error - 0x08
Bus Fatal Error - 0x0A
15 Event data 2 Bit[7:0] = PCI bus number for failed device.
16 Event data 3 Bit[7:3] = PCI device number for failed device.
Bit[2:0] = PCI function number for failed device.
ExtSEL
Byte Field Description
1-2 Record ID Record ID
3 Record Type 0xDF – OEM System Event Record
4-7 Timestamp Timestamp
8-9 Generator ID 0x21
10 EvmRev 0
11 Sensor Type SEL_SENS_TYPE_CRIT_ERR <0x13>
12 Sensor Number Per BMC spec design
13 Event Dir | EventType 0x70
14 Reserved 1Byte (modified as 0xAA by BMC)
15 Reserved 1Byte (modified by BMC)
16 Resreved 1Byte (modified by BMC)
17 Ext Event data 1 Bit[7:4] = 0xA (Indicates that OEMEvData2 and OEMEvData3
are OEM values)
Bit[3:0] = Error Type
Error Type
PCI PERR - 0x04
PCI SERR - 0x05
Bus Correctable Error - 0x07
Bus Uncorrectable Error - 0x08
Bus Fatal Error - 0x0A
18 Ext Event data 2 Bit[7:0] = PCI bus number for failed device.
19 Ext Event data 3 Bit[7:3] = PCI device number for failed device.
Bit[2:0] = PCI function number for failed device.
20-21 Ext Event data 4-5 Slot Infomation(Number)
22 Ext Event data 6 IsBridge
23-24 Ext Event data 7-8 SecondaryStatus
25-26 Ext Event Data Byte[9-10] DeviceStatus
27-30 Ext Event Data Byte[11-14] UncorrErrStatus Register
12
Lenovo OEM/Extended SEL Events Design Specification
31-34 Ext Event Data Byte[15-18] CorrErrStatus Register
35-38 Ext Event Data Byte[19-22] RootErrStatus Register
3.3 RUNTIME PROCESSOR ERRORS
For MCA Banks IFU, DCU, DTLB and MLC Errors
SEL:
Byte Field Description
1-2 Record ID Record ID
3 Record Type 02 – System Event Record
4-7 Timestamp Timestamp
8-9 Generator ID 0x21
10 EvmRev 4
11 Sensor Type 0x7 PROCESSOR_SENSOR_TYPE
12 Sensor Number Per BMC senor number Spec Design
13 Event Dir | EventType 0x6F
14 Event data 1 Bit[7:4] = 0xA (Indicates that OEMEvData2 and OEMEvData3
are OEM values)
Bit[3:0] = Error Type (5-PROCESSOR
CONFIGURATION_ERROR)
15 Event data 2 Bit[7:4] = Socket Number
Bit[3:0] = Bank Type (Please see below)
Bank Type:
0 (0000) - None
1 (0001) - IFU
2 (0010) - DCU
3 (0011) - DTLB
4 (0100) - MLC
5 (0101) - PCU
6 (0110) - IIO
7 (0111) - CHA
8 (1000) - KTI
16 Event data 3 Bit[7:4] = Processor Error Type
(UNKNOWN/Cache/TLB/BUS/Micro Arch)
Bit[3:0] = Error Severity (Correctable/Uncorrectable)
Error Severity:
00 - Correctable Error
01 - Fatal Error
02 - Corrected Error
Extended SEL
Byte Field Description
1-2 Record ID Record ID
3 Record Type 0xDF – OEM System Event Record
4-7 Timestamp Timestamp
8-9 Generator ID 0x21
10 EvmRev 0
11 Sensor Type 0x7 PROCESSOR_SENSOR_TYPE
12 Sensor Number Per BMC senor number Spec Design
13
Lenovo OEM/Extended SEL Events Design Specification
13 Event Dir | EventType 0x70
14 Reserved 1Byte (modified as 0xAA by BMC)
15 Reserved 1Byte (modified by BMC)
16 Resreved 1Byte (modified by BMC)
17 Event data 1 Bit[7:4] = 0xA (Indicates that OEMEvData2 and OEMEvData3
are OEM values)
Bit[3:0] = Error Type (5-PROCESSOR
CONFIGURATION_ERROR)
18 Event data 2 Bit[7:4] = Socket Number
Bit[3:0] = Bank Type (Please see below)
Bank Type:
0 (0000) - None
1 (0001) - IFU
2 (0010) - DCU
3 (0011) - DTLB
4 (0100) - MLC
5 (0101) - PCU
6 (0110) - IIO
7 (0111) - CHA
8 (1000) - KTI
19 Event data 3 Bit[7:4] = Processor Error Type
(UNKNOWN/Cache/TLB/BUS/Micro Arch)
Bit[3:0] = Error Severity (Correctable/Uncorrectable)
Error Severity:
00 - Correctable Error
01 - Fatal Error
02 - Corrected Error
20-27 Ext Event data 4-11 MC Bank Status Register
28-35 Ext Event data 12-19 MC Bank Address Register
36-43 Ext Event data 20-27 MC Bank Misc Register
44 Ext Event data 28-29 MC BANK number
3.4 IIO CORE
SEL:
Byte Field Description
1-2 Record ID Record ID
3 Record Type 0x2 –System Event Record
4-7 Timestamp Timestamp
8-9 Generator ID 0x21
10 EvmRev 4
11 Sensor Type 0x07 PROCESSOR_SENSOR_TYPE
14
Lenovo OEM/Extended SEL Events Design Specification
12 Sensor Number
Per BMC spec design
13 Event Dir | EventType 0x6F
14 Event data 1 Bit[7:4] = 0xA (Indicates that OEMEvData2 and OEMEvData3 are OEM
values)
Bit[3:0] = Error Type (5-PROCESSOR CONFIGURATION_ERROR)
15 Event data 2 Bit[7:4] = CPU Socket Number.
Bit[3:0] = Reserved
16 Event data 3 Error Sub Type
c4h - Master abort address error
c5h - Completor abort address error
c6h - FIFO Overflow/Underflow error
3.5 IIO-COHERENCE INTERFACE (IIO QPI)
SEL:
Byte Field Description
1-2 Record ID Record ID
3 Record Type 0x2 –System Event Record
4-7 Timestamp Timestamp
8-9 Generator ID 0x21
10 EvmRev 4
11 Sensor Type 0x07 PROCESSOR_SENSOR_TYPE
12 Sensor Number Per BMC spec design
13 Event Dir | 0x6F
EventType
14 Event data 1 Bit[7:4] = 0xA (Indicates that OEMEvData2 and OEMEvData3 are OEM values)
Bit[3:0] = Error Type (5-PROCESSOR CONFIGURATION_ERROR)
15 Event data 2 Bit[3:0] = QPI Port number.
Bit[7:4] = CPU Socket Number.
16 Event data 3 Coherent Interface (IRP) Local Group Error code as per IvyBridge-EX 4S Processor
External Design Specification (EDS) specification.
B4h - Write Cache Detected ECC Correctable Error.
C1h - Coherency Interface Protocol Layer Received Poisoned packet
C2h - I/O module Write Cache uncorrectable Data ECC error
C3h - I/O module CSR access crossing 32-bit boundary.
D7h - Coherent Interface Protocol Layer Received Unexpected or Illegal Response/Completion
DAh - Coherent Interface Protocol Layer Queue/Table Overflow or Underflow
DBh - Coherent Interface Protocol Parity Error.
15
Lenovo OEM/Extended SEL Events Design Specification
3.6 IIO-VTD
SEL
Byte Field Description
1-2 Record ID Record ID
3 Record Type 0x2 –System Event Record
4-7 Timestamp Timestamp
8-9 Generator ID 0x21
10 EvmRev 4
11 Sensor Type 0x07 PROCESSOR_SENSOR_TYPE
12 Sensor Per BMC spec design
Number
13 Event Dir | 0x6F
EventType
14 Event data 1 Bit[7:4] = 0xA (Indicates that OEMEvData2 and OEMEvData3 are OEM values)
Bit[3:0] = Error Type (5-PROCESSOR CONFIGURATION_ERROR)
15 Event data 2 CPU Socket Number.
16 Event data 3 Intel VT-d Local Group error code as per IvyBridge-EX 4S Processor
External Design Specification (EDS)
90h - Data Parity Error during Context Cache Lookup
91h - Data Parity Error during L1 Lookup
92h - Data Parity Error during L2 Lookup
93h - Data Parity Error during L3 Lookup
94h - TLB0 Data Parity Error
95h - TLB1 Data Parity Error
96h - Unsuccessful completion status received in the coherent interface
97h - Illegal request to 0xFEE
98h - Protected Memory region space violated status
A0h - Intel VT-d spec defined errors
16
Lenovo OEM/Extended SEL Events Design Specification
4. SYSTEM FAULT LED
When BIOS detected listed system errors, BIOS or BMC will turn on system fault LED.
• QPI RC errors.
• MRC errors.
• Microcode Errors
• CPU Exception
• Runtime Processor Errors
• Uncorrected Memory Error
• IIO Core/ IIO-Coherence interface(IIO QPI)/IIO-Vtd Error
17
