Documentation/driver-api/ipmi.rst

41 properly provides the SMBIOS info for IPMI, the driver will detect it
43 will generally be either "KCS", "SMIC", or "BT", consult your hardware
46 support this, but it is unknown if it will work on every board.  For
48 figuring to see if it will work on your system if the SMBIOS/APCI
74 'IPMI Poweroff' to do this.  The driver will auto-detect if the system
226 will have no place to put the message.
228 Messages coming up from the message handler in kernelland will come in
244 	/* Call this when done with the message.  It will presumably free
285 piece of data, the handler_data, that will be passed back to you on
318 If the message cannot fit into the data you provide, you will get an
319 EMSGSIZE error and the driver will leave the data in the receive
324 the netfn per the IPMI spec) on the IPMB bus, the driver will
327 seconds, it will generate a response automatically saying the command
329 seconds, for instance), that response will be ignored.
332 MUST call ipmi_free_recv_msg() on it, or you will leak messages.  Note
350 these, you must register for them, they will not automatically be sent
355 since startup will immediately be delivered to the first user that
357 for events, they will all receive all events that come in.
416 If you specify addrs as non-zero for an interface, the driver will
420 If you specify ports as non-zero for an interface, the driver will
423 If you specify irqs as non-zero for an interface, the driver will
460 If unload_when_empty is set to 1, the driver will be unloaded if it
481 SMIC interface, the IPMI driver will start a kernel thread for the
484 is in progress.  The force_kipmid module parameter will all the user to
486 interrupts, the driver will run VERY slowly.  Don't blame me,
493 microseconds, that kipmid will spin before sleeping for a tick.  This
495 tuned to your needs.  Maybe, someday, auto-tuning will be added, but
529 system.  By default, the driver will only register with something it
555 Setting dbg_probe to 1 will enable debugging of the probing and
563 alert will be enabled on supported hardware.
599 system interface addressed messages to it will cause the message to go
604 the bus and they will be handled through the normal received command
613 Loading the module will not result in the driver automatcially
678 is the amount of seconds before the reset that the pre-timeout panic will
679 occur (if pretimeout is zero, then pretimeout will not be enabled).  Note
682 will occur in 40 second (10 seconds before the timeout). The panic_wdt_timeout
705 If start_now is set to 1, the watchdog timer will start running as
708 If nowayout is set to 1, the watchdog timer will not stop when the
725 The watchdog will panic and start a 120 second reset timeout if it
726 gets a pre-action.  During a panic or a reboot, the watchdog will
732 otherwise unhandled NMI, it must be from IPMI and it will panic
736 device to close it, or the timer will not stop.  This is a new semantic
746 'Generate a panic event to all BMCs on a panic' option, you will get
748 the 'Generate OEM events containing the panic string' option, you will
764 always sent to the local management controller.  It will handle routing
774   is longer than 11 bytes, multiple messages will be sent with increasing
778 function will attempt to find an SEL and add the events there.  It
779 will first query the capabilities of the local management controller.
780 If it has an SEL, then they will be stored in the SEL of the local
783 controller will be queried and the events sent to the SEL on that
791 If the poweroff capability is selected, the IPMI driver will install
794 it will send the proper IPMI commands to do this.  This is supported on
800 ipmi_poweroff.poweroff_control=x will do the same thing on the kernel
803 does not support power cycling, it will always do the power off.
809 Note that if you have ACPI enabled, the system will prefer using ACPI to