kern/63431: motherboard going to suspend mode stops system clock after 4.9-R

[Jin Guojun[VFF]]

>Number:         63431
>Category:       kern
>Synopsis:       motherboard going to suspend mode stops system clock after 4.9-R
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    freebsd-bugs
>State:          open
>Quarter:        
>Keywords:       
>Date-Required:
>Class:          sw-bug
>Submitter-Id:   current-users
>Arrival-Date:   Thu Feb 26 20:40:06 PST 2004
>Closed-Date:
>Last-Modified:
>Originator:     Jin Guojun[VFF]
>Release:        FreeBSD 4.9-RELEASE i386
>Organization:
>Environment:
System: FreeBSD spode.pacbell.net 4.9-RELEASE FreeBSD 4.9-RELEASE #0: Sat Nov 1 18:00:32 PST 2003 root at spode.pacbell.net:/usr/src/sys/compile/MinMax i386


	FreeBSD 4.9-RELEASE and later.
	May be for all motherboards, but at least for listed below:
		ASUS A7V600
		TYAN S2466N
		SuperMicro P4DPE
		ASUS P2B-DS

>Description:
	If BIOS of a motherboard is set for suspend/stand-by mode when
	system is idle for a certain time, system will go into stand-by
	or suspend mode.
	I do not remember 4.8-R or earlier has problem on the system clock.
	But system clock is off when system is in suspend mode in 4.9-R and 5.2-R.
	After wake up, system did not syn its clock with the RTC.
	Two ways to correct system clock:
		use date command to update the time manually
		reboot the system (this means RTC is still correct)
		
>How-To-Repeat:
	Try to enable "suspend mode" in power management in BIOS.
	
>Fix:

	When system is suspended, only keyboard can wake it up
	by default. Maybe some motherborad can be waked up by LAN.
	I do not know if system can detect a wake up signal.
	If so, it should check if system clock is off from RTC.

>Release-Note:
>Audit-Trail:
>Unformatted: