>Number:         9573
>Category:       kern
>Synopsis:       fxp0 hangs across hibernation on sony vaio z505he
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    kern-bug-people (Kernel Bug People)
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Wed Mar  8 02:09:01 2000
>Last-Modified:
>Originator:     John Hawkinson
>Organization:
	MIT
>Release:        NetBSD 1.4.1
>Environment:
	
System: NetBSD zorkmid.mit.edu 1.4.1 NetBSD 1.4.1 (ZORKMID) #62: Wed Mar 8 03:59:10 EST 2000 jhawk@zorkmid.mit.edu:/usr/src/sys/arch/i386/compile/ZORKMID i386


>Description:
	I strongly suspect this happens under -current.
This is quite similar to kern/9571 ("fxp0 hangs across apm
suspend on sony vaio z505he").

The fxp on-board ethernet on the Sony VAIO Z505-series laptops
(specifically Z505HE) goes out to lunch when the laptop is
APM hibernated and then resumed. Hibernation is when the BIOS writes
all memory to disk and then completely powers off the machine.
The fix for the APM suspend problem in kern/9571 does not serve here,
nor does combining it with my fix for kern/9370 and trying to move
the interface from ACPI D3 to ACPI D0 states.

Worse, it seems to hang the machine entire, probably because of
kern/9572 ("polled DMA can hang the machine (fxp)"), but I've
not verified.

I do not have the time or resources to try to debug this particular
problem at this juncture, so here is a PR without a fix.
>How-To-Repeat:
	Buy a new Sony VAIO laptop.
	Fire up the ethernet interface.
	APM hibernate it (Fn + F12)
	Resume (frob power switch)
	Watch the ethernet hang, and then subsequently the entire
machine hangs. One "SCB timed out" and one "device timeout"
printf() are seen.

>Fix:
WORKAROUND: (not)
	Be thankful that APM sleep works and use that instead.
Note that the BIOS seems to sometimes fail to read all the
memory from disk and just hang halfway through in the midst
of it's cheery graphical ANSI-color-style display, so perhaps
this feature isn't as useful as you might hope. On the other hand,
perhaps NetBSD is causing it to fail somehow. Go figure. Not enough
info for a PR yet on that.
>Audit-Trail:
>Unformatted: