Subject: port-i386/4386: apm problems in -current
To: None <gnats-bugs@gnats.netbsd.org>
From: None <ksulliva@psc.edu>
List: netbsd-bugs
Date: 10/29/1997 12:23:45
>Number:         4386
>Category:       port-i386
>Synopsis:       apm problems in -current
>Confidential:   no
>Severity:       critical
>Priority:       high
>Responsible:    gnats-admin (GNATS administrator)
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Wed Oct 29 09:35:02 1997
>Last-Modified:
>Originator:     Kevin Sullivan
>Organization:
Pittsburgh Supercomputing Center
	
>Release:        NetBSD 1.3 ALPHA
>Environment:
	
System: NetBSD balrog 1.3A NetBSD 1.3A (BALROG) #1: Tue Oct 28 22:50:16 EST 1997 ksulliva@sullivan.pgh.pa.us:/extra/src/sys/arch/i386/compile/BALROG i386


>Description:
APM used to work fine on my laptop (an IBM Thinkpad 701cs).  Lately
though it has been having problems.  It goes to sleep okay, but when I
wake it up the hard disk doesn't work at first.  The first time I access
the disk the machine freezes for 60 seconds; the disk light blinks at
about twice per second.  After 60 seconds the machine continues; however
it seems that the first read returns a hard error since I often get wierd
errors about filesystem corruption or I cannot run whichever program
first accessed the disk.  When I reboot the machine the errors generally
go away (though once it trashed about 10% of my filesystem).

This started happening around the time that atapi-CD support went into
-current.  Also around that time, I switched from John Kohl's pcmcia
patches to the marc-pcmcia branch.

I went into DDB during the 60-seconds and got this:

db> trace
[ various seemingly unimportant stuff; let me know if I need to copy it ]
--- interrupt ---
_idle(80000000,0,0,0,f028bf7c) at _idle+0x12
bpendtsleep(f02532a0,4,f01d0e6c,0) at bpendtsleep
_scheduler(f07f9400,f011ecd4,289008,289000,2a8000) at _scheduler+0x64
_main(0,0,0,0,0) at _main+0x51e

>How-To-Repeat:
On a Thinkpad 701cs, suspend and wakeup under -current. 
>Fix:
>Audit-Trail:
>Unformatted: