PEF Configuration Parameters 相关表项详情
更新时间：2025/06/26
在Gitcode上查看源码修订记录
| openUBMC版本号 | 修订日期 | 修订人 | 修订内容 |
|---|---|---|---|
| 25.06 | 2025/06/26 | pengqiang-gs | 初稿，新增 |
PEF Configuration Parameters
| Parameter | # | Parameter Data |
|---|---|---|
| Set In Progress(volatile) | 0 | data 1 - This parameter is used to indicate when any of the following parameters are being updated, and when the updates are completed. The bit is primarily provided to alert software than some other software or utility is in the process of making changes to the data. An implementation can also elect to provide a ‘rollback’ feature that uses this information to decide whether to ‘roll back’ to the previous configuration information, or to accept the configuration change. If used, the roll back shall restore all parameters to their previous state. Otherwise, the change shall take effect when the write occurs. [7:2] - reserved [1:0] 00b = set complete. If a system reset or transition to powered down state occurs while ‘set in progress’ is active, the BMC will go to the ‘set complete’ state. If rollback is implemented, going directly to ‘set complete’ without first doing a ‘commit write’ will cause any pending write data to be discarded. 01b = set in progress. This flag indicates that some utility or other software is presently doing writes to parameter data. It is a notification flag only, it is not a resource lock. The BMC does not provide any interlock mechanism that would prevent other software from writing parameter data while. 10b = commit write (optional). This is only used if a rollback is implemented. The BMC will save the data that has been written since the last time the ‘set in progress’ and then go to the ‘set in progress’ state. An error completion code will be returned if this option is not supported. 11b = reserved |
| PEF control (non-volatile) | 1 | data 1 [7:4] - reserved [3] - PEF Alert Startup Delay disable. (optional) 1b = enable PEF Alert Startup delay 0b = disable PEF startup delay. [2] - PEF Startup Delay disable. (optional) An implementation that supports this bit should also provide a mechanism that allows the user to Disable PEF in case the filter entries are programmed to cause an ‘infinite loop’ of PEF actions (such as system resets or power cycles) when the PEF startup delay is disabled. If this bit is not implemented the PEF startup delay must always be enabled. 1b = enable PEF startup delay on manual (pushbutton) system power-ups (from S4/S5) and system resets (including system resets initiated by PEF). 0b = disable PEF startup delay. [1] 1b = enable event messages for PEF actions. If this bit is set, each action triggered by a filter will generate an event message for the action. These allow the occurrence of PEF triggered actions to be logged (if event logging is enabled). The events are logged as System Event Sensor 12h, offset 04h. See Sensor Type Codes.) These event messages are also subject to PEF. 0b = disable event messages for PEF actions. [0] 1b = enable PEF. 0b = disable PEF. |
| PEF Action global control (non-volatile) | 2 | data 1 [7:6] - reserved [5] - 1b = enable diagnostic interrupt [4] - 1b = enable OEM action [3] - 1b = enable power cycle action (No effect if power is already off) [2] - 1b = enable reset action [1] - 1b = enable power down action [0] - 1b = enable Alert action |
| PEF Startup Delay (optional, non-volatile) | 3 | data 1 - time to delay PEF after a system power-ups (from S4/S5) and resets. Default = 60 seconds. If this parameter is not provided, the default PEF Startup Delay must be implemented. Enable/disable of the delay is configured using the PEF Control parameter, above. If this parameter is supported, a 00h value can also be used to disable the delay if necessary. See Section 17.4, PEF Startup Delay, for more information. Note: An implementation that supports this parameter should also provide a mechanism that allows the user to Disable PEF in case the filter entries are programmed to cause an ‘infinite loop’ of PEF actions under the situation where this parameter is set to too short an interval to allow a user to locally disable PEF. An implementation is allowed to force this parameter to a minimum, non-zero value. PEF Startup Delay [7:0] - PEF Startup Delay in seconds, +/- 10%. 1-based. 00h = no delay. |
| PEF Alert Startup Delay (optional, non-volatile) | 4 | data 1 - time to delay Alerts after system power-ups (from S4/S5) and resets. Default = platform-specific. 60-seconds typical, though may be longer on systems that require more startup time before user can take action to disable PEF. If this parameter is not provided, a default PEF Startup Delay, appropriate for the platform, must be implemented. Enable/disable of the delay can also be optionally configured using the PEF Control parameter, above. An implementation can separately implement this parameter and/or the enable/disable bit. PEF Alert Delay [7:0] - PEF Alert Startup Delay in seconds, +/- 10%. 1-based. 00h = no delay. |
| Number of Event Filters (READ ONLY) | 5 | Number of event filters supported. 1-based. This parameter does not need to be supported if Alerting is not supported. [7:0] - number of event filter entries. 0 = alerting not supported. |
| Event Filter Table, (nonvolatile) | 6 | data 1 - Set Selector = filter number. [7:0] - Filter number. 1-based. 00h = reserved. data 2:21 - filter data |
| Event Filter Table Data 1 (non-volatile) | 7 | This parameter provides an aliased access to the first byte of the event filter data. This is provided to simplify the act of enabling and disabling individual filters by avoiding the need to do a read-modify-write of the entire filter data. data 1 - Set Selector = filter number [7:0] - Filter number. 1-based. 00h = reserved. data 2 - data byte 1 of event filter data |
| Number of Alert Policy Entries (READ ONLY) | 8 | Number of alert policy entries supported. 1-based. This parameter does not need to be supported if Alerting is not supported. [7] - reserved [6:0] - number of alert policy entries. 0 = alerting not supported. |
| Alert Policy Table (non-volatile) | 9 | data 1 - Set Selector = entry number [7] - reserved [6:0] - alert policy entry number. 1-based. data 2:4 - entry data |
| System GUID (non-volatile) | 10 | data 1 Used to fill in the GUID field in a PET Trap. Stored per GUID Format. [7:1] -reserved [0] 1b = BMC uses following value in PET Trap. 0b = BMC ignores following value and uses value returned from Get System GUID command instead. 2:17 - System GUID |
| Number of Alert Strings (READ ONLY) | 11 | Number of alert strings supported in addition to Alert String 0. 1-based. This parameter does not need to be supported if Alerting is not supported. [7] - reserved [6:0] - number of alert strings. |
| Alert String Keys (volatile) & (non-volatile) | 12 | Sets the keys used to look up Alert String data in PEF. This parameter does not need to be supported if Alerting is not supported. data 1 - Set Selector = Alert string selector. [7] - reserved. [6:0] - string selector. 0 = selects volatile string parameters 01h-7Fh = non-volatile string selectors PEF uses the following Event Filter Number and the Alert String Key fields to look up the string associated with a particular event. String 0 is a special, volatile string reserved for use by the Alert Immediate command. The following two fields are used by PEF to look up a particular Alert String based on information obtained from the alert policy entry. The fields should typically be set to 0’s (unspecified) for string selector 0. PEF will scan the values for string 0 when doing a look up, so the string 0 values can be set to non-zero values for PEF testing/debug purposes in order to avoid writes to non-volatile storage. data 2 - Event Filter Number [7] - reserved. [6:0] - Filter number. 1-based. 00h = unspecified. data 3 - Alert String Set [7] - reserved [6:0] - Set number for string. 1-based. 00h = unspecified. |
| Alert Strings (volatile) & (non-volatile) | 13 | Sets the Alert String data. The string data that should be used is dependent on the Channel and Alert Type. This parameter does not need to be supported if Alerting is not supported. For Dial paging, the BMC automatically follows the string with a <CR>(carriage return) character when sending it to the modem. For TAP paging the string corresponds to ‘Field 2’, the Pager Message. Note that while the string accepts 8-bit ASCII data, the TAP implementation only supports 7-bit ASCII. The BMC shall automatically zero the 8th bit when transmitting the string during TAP paging. String 0 is a special, volatile string reserved for use by the Alert Immediate command. data 1 - Set Selector = string selector. [7] - reserved. [6:0] - string selector. 0 = selects volatile string 01h-7Fh = non-volatile string selectors data 2 - Block Selector = string block number to set, 1 based. Blocks are 16 bytes. data 3:N - String data. Null terminated 8-bit ASCII string. 16-bytes max. per block. |
| Number of Group Control Table entries (READ ONLY) (optional. Present if BMC supports automatic ICMB Group Power Control. See ICMB specification for details.) | 14 | data 1 - Number of group control table entries. 1-based (4 min, 8 max) |
| Group Control Table (optional, non-volatile. Present if BMC supports automatic ICMB Group Power Control. See ICMB specification for details.) | 15 | data 1 - Set Selector = group control table entry selector. [7] - reserved. [6:0] - group control table entry selector. data 2 - [7:6] - reserved [5] - Request/Force 0b = request control operation. A requested operation will only complete once the same operation has been requested for all control groups and all enabled control members for the given chassis. 1b = force control operation. A forced operation will occur regardless of whether the same operation has been requested for all control groups and all enabled control membership for the given chassis. [4] - Immediate/Delayed. Selects whether the BMC requests an immediate or delayed control operation. Note: whether this operation is initiated at the time the command is received is dependent on the request/force bit, see above. 0b = immediate control. BMC sends command that requests an immediate control operation. 1b = delayed control. BMC sends control command to request a delayed control operation. This is conditioned by the request/force bit. [3:0] - Channel Number (channel number for ICMB that group control operation is to be delivered over) data 3: Group ID 0 (1-based) 00h = unspecified FFh = all groups data 4: Member ID 0 (0-based) [7:5] - reserved [4] 0b = enable member ID check. 1b = disable member ID check[1]. [3:0] - member ID. ID of this chassis within specified group. (value is ignored if Group ID 0 = FFh) data 5: Group ID 1 (1-based) 00h = unspecified FFh = all groups data 6: Member ID 1 (0-based) [7:5] - reserved [4] 0b = enable member ID check. 1b = disable member ID check[1]. [3:0] - member ID. ID of this chassis within specified group. (value is ignored if Group ID 1 = FFh) data 7: Group ID 2 (1-based) 00h = unspecified FFh = all groups data 8: Member ID 2 (0-based) [7:5] - reserved [4] 0b = enable member ID check. 1b = disable member ID check[1]. [3:0] - member ID. ID of this chassis within specified group. (value is ignored if Group ID 2 = FFh) data 9: Group ID 3 (1-based) 00h = unspecified FFh = all groups data 10: Member ID 3 (0-based) [7:5] - reserved [4] 0b = enable member ID check. 1b = disable member ID check[1]. [3:0] - member ID. ID of this chassis within specified group. (value is ignored if Group ID 3 = FFh) data 11: - Retries and Operation [7] - reserved [6:4] - number of times to retry sending the command to perform the group operation [For ICMB, the BMC broadcasts a Group Chassis Control command] (1-based) [3:0] - operation 0h = power down. Force system into soft off (S4/S45) state. This is for ‘emergency’ management power down actions. The command does not initiate a clean shut-down of the operating system prior to powering down the system. 1h = power up. 2h = power cycle (optional). This command provides a power off interval of at least 1 second. 3h = hard reset. Some systems may accept this option even if the system is in a state (e.g. powered down) where resets are unavailable. 4h = pulse Diagnostic Interrupt. (optional) Pulse a version of a diagnostic interrupt that goes directly to the processor(s). This is typically used to cause the operating system to do a diagnostic dump (OS dependent). The interrupt is commonly an NMI on IA-32 systems and an INIT on Intel®Itanium™ processor based systems. 5h = Initiate a soft-shutdown of OS via ACPI by emulating a fatal overtemperature. (optional) |
| OEM Parameters (optional. Non-volatile or volatile as specified by OEM) | 96:127 | This range is available for special OEM configuration parameters. The OEM is identified according to the Manufacturer ID field returned by the Get Device ID command. |