[deliverable/linux.git] / Documentation / ABI / testing / pstore

Where:		/dev/pstore/...
Date:		March 2011
Kernel Version: 2.6.39
Contact:	tony.luck@intel.com
Description:	Generic interface to platform dependent persistent storage.

		Platforms that provide a mechanism to preserve some data
		across system reboots can register with this driver to
		provide a generic interface to show records captured in
		the dying moments.  In the case of a panic the last part
		of the console log is captured, but other interesting
		data can also be saved.

		# mount -t pstore -o kmsg_bytes=8000 - /dev/pstore

		$ ls -l /dev/pstore
		total 0
		-r--r--r-- 1 root root 7896 Nov 30 15:38 dmesg-erst-1

		Different users of this interface will result in different
		filename prefixes.  Currently two are defined:

		"dmesg"	- saved console log
		"mce"	- architecture dependent data from fatal h/w error

		Once the information in a file has been read, removing
		the file will signal to the underlying persistent storage
		device that it can reclaim the space for later re-use.

		$ rm /dev/pstore/dmesg-erst-1

		The expectation is that all files in /dev/pstore
		will be saved elsewhere and erased from persistent store
		soon after boot to free up space ready for the next
		catastrophe.

		The 'kmsg_bytes' mount option changes the target amount of
		data saved on each oops/panic. Pstore saves (possibly
		multiple) files based on the record size of the underlying
		persistent storage until at least this amount is reached.
		Default is 10 Kbytes.

		Pstore only supports one backend at a time. If multiple
		backends are available, the preferred backend may be
		set by passing the pstore.backend= argument to the kernel at
		boot time.
Commit	Line	Data
ca01d6dd	1	Where: /dev/pstore/...
366f7e7a LT	2	Date: March 2011
366f7e7a LT	3	Kernel Version: 2.6.39
ca01d6dd TL	4	Contact: tony.luck@intel.com
	5	Description: Generic interface to platform dependent persistent storage.
	6
	7	Platforms that provide a mechanism to preserve some data
	8	across system reboots can register with this driver to
	9	provide a generic interface to show records captured in
	10	the dying moments. In the case of a panic the last part
	11	of the console log is captured, but other interesting
	12	data can also be saved.
	13
366f7e7a	14	# mount -t pstore -o kmsg_bytes=8000 - /dev/pstore
ca01d6dd TL	15
	16	$ ls -l /dev/pstore
	17	total 0
	18	-r--r--r-- 1 root root 7896 Nov 30 15:38 dmesg-erst-1
	19
	20	Different users of this interface will result in different
	21	filename prefixes. Currently two are defined:
	22
	23	"dmesg" - saved console log
	24	"mce" - architecture dependent data from fatal h/w error
	25
	26	Once the information in a file has been read, removing
	27	the file will signal to the underlying persistent storage
	28	device that it can reclaim the space for later re-use.
	29
	30	$ rm /dev/pstore/dmesg-erst-1
	31
	32	The expectation is that all files in /dev/pstore
	33	will be saved elsewhere and erased from persistent store
	34	soon after boot to free up space ready for the next
	35	catastrophe.
366f7e7a LT	36
	37	The 'kmsg_bytes' mount option changes the target amount of
	38	data saved on each oops/panic. Pstore saves (possibly
	39	multiple) files based on the record size of the underlying
	40	persistent storage until at least this amount is reached.
	41	Default is 10 Kbytes.
dee28e72 MG	42
	43	Pstore only supports one backend at a time. If multiple
	44	backends are available, the preferred backend may be
	45	set by passing the pstore.backend= argument to the kernel at
	46	boot time.
	47