SR Changelog

	bugfix: onboard ethernet ports can link up at 10Mb due to hardware error

	bugfix: raid6 corruption when writing during rebuilding with two disk failures (introduced in 20090728)
	SSD sata2 preference mode updates

	bugfix: when output broken for raid1/raid10 disk rebuilds
	bugfix: write performance drop for multiple raid levels (introduced in 20090622)
	add rebuild/init percent complete to when output

	bugfix: potential for slow parity init rates
	bugfix: devices failing during parity init do not fail raid

	add support for SR821T
	add raid6rs raid level
	add ability to rebuild multiple disks concurrently for raid6rs, raid10
	update when, list to display rebuild of individual disks
	enable flow control for 10G dual port nics

	bugfix: erroneous disk failures every 49 days of boot time due to rollover of 32 bit millisecond counter
	bugfix: recovering raids sustaining double failures don't mark failures properly
	force luns with failed raids offline

	add support for AoEr11 command additions
	add reslist command to show/clear lun reservations

	add aggressive sata2 connection attempt for SSD disks
	add -m flag to show
	add resetdisk command

	bugfix: multiple disks in recover state in raid10 leads to bad raid state on reboot

	bugfix: SMART "threshhold exceeded" (TE) misspelling
	bugfix: SR4[23]1 does not issue syslog on SMART TE due to misspelling
	bugfix: raw LUN type can write to its disk incorrectly
	bugfix: failed LUN can cause other LUNs to not respond to broadcast query config
	add better handling for SATA Phy errors

	bugfix: syslog messages can be lost due to windows syslog support workarounds in 20080815

	bugfix: slot led state of previously spared rebuild disk returns to spare after rebuild
	bugfix: panic in rare cases using dual port CX4
	bugfix: "82598: no buffers" error for 10G dual port CX4

	bugfix: panic in rare cases on issuing reboot during heavy I/O
	improved support for windows syslog daemons
	add better error handling for drives with smart disabled
	update show -s output to show reason for "unknown" status
	add support for SR2421

	bugfix: spare, rmspare issue bad diagnostic on invalid input
	add periodic check of spare disks to warn on "silent" failures

	update ses support for hardware variation in LD2461 backplane

	bugfix: "corrupt ilock" panic in very rare cases
	add support for LD2461
	add support for SR431
	add support for unique, persistent ATA serial id on LUNs
	replace use of term "lblade" with term "lun"

	bugfix: link states in ifstat can be reported incorrectly for CX4 interfaces

	update ses support for hardware variation in SR2461 backplane

	bugfix: syslog -c can exit abnormally on backspace
	add ses features for SR16xx and SR24xx
	add ses commands fans, alarm, ledslot, and temp
	add ifstat command
	add release label to release command output

	bugfix: replace using disk already in use can cause panic
	bugfix: when command rates can show incorrect results in rare cases
	bugfix: raw device offset error causing raw device to not be truly raw
	GbE flow control workarounds for switch compatibility

	bugfix: release command shows new release after update (and before reboot) instead of the running release
	update SR421 sata driver to better handle sata communication faults

	bugfix: disk failure during raid5 parity init allocates replacement disk instead of failing array
	bugfix: online of failed lblade issues diagnostic "qcget short read"
	bugfix: raid devices in failed state change across reboot, eventually become missing
	bugfix: SR152x shows 16 disk ports
	bugfix: replace of replaced disk results in incorrect action
	bugfix: various sata disk communication error cases causing lockups and I/O failures
	bugfix: show -s reports removed disks as "disabled" instead of "unknown"
	bugfix: reboot during heavy I/O causes panic, reboots with lblades missing disks

	bugfix: SR will rebuild raid5 parity on boot despite being halted cleanly

	update cec server to permit connection persistence across reboot
	add help usage to default output of help
	add eject command
	add restore command
	add smart status support
	add -s flag to show command for displaying smart status
	add smartenable and smartdisable commands to turn on/off specific drives
	add smart monitor daemon to monitor disk smart status
	add raw raid type
	add flash signature support
	add sigcheck command
	add raid guard to catch user errors
	throughput enhancements
	permit cecon/cecoff to take ethernet name without path
	remove shutdown command
	syslog settable local access interface

	enable flash signature setting in preparation for new, faster releases

	bugfix: sufficiently long mask lists corrupt disk config strings
	bugfix: duplicate mask additions result in duplicate mask entries
	bugfix: aoe protocol errors do not set the aoe error flag
	add support for aoe mac mask list protocol extension