[Search for users] [Overall Top Noters] [List of all Conferences] [Download this site]

Conference smurf::dec_mls_plus

Title:	dec_mls_plus

Moderator:	SMURF::BAT

Created:	Mon Nov 29 1993
Last Modified:	Fri Jun 06 1997
Last Successful Update:	Fri Jun 06 1997
Number of topics:	534
Total number of notes:	2544

480.0. "arpoutput causing panic" by ADISSW::FERRARA () Thu Apr 17 1997 13:16

    
    I've been working on a VME-FDDI device driver, up to this point
    on Digital UNIX 4.0b, all works fine.
    
    I now install and statically configure the driver on MLS+ 4.0a 
    and I get the following system panic during system boot.
    
    The panic stems from the ARP-output call...
    
    Any ideas?
    
    Thanks,
     BobF
    
.
.
.
Configuring network
hostname: nruclone
rf0 rf_init: Host Buffer VME Address = 0x8002000
arpoutput: can't handle type 65535

trap: invalid memory read access from kernel mode

    faulting virtual address:     0x0000000000000000
    pc of faulting instruction:   0xfffffc00003ef670
    ra contents at time of fault: 0xfffffc00002d1d00
    sp contents at time of fault: 0xffffffff840537e0

panic (cpu 0): kernel memory fault
syncing disks...
DUMP: 376832 blocks available for dumping.
DUMP: 16113 required for a partial dump.
DUMP: 0x800c01 is the primary swap with 376831, start our last 16112
    : of dump at 360719, going to end (real end is one more, for header)
device string for dump = SCSI 0 2 0 3 300 0 0.
DUMP.prom: dev SCSI 0 2 0 3 300 0 0, block 155648
DUMP: Header to 0x800c01 at 376831 (0x5bfff)
device string for dump = SCSI 0 2 0 3 300 0 0.
DUMP.prom: dev SCSI 0 2 0 3 300 0 0, block 155648
DUMP: Dump to 0x800c01: .......: End 0x800c01
device string for dump = SCSI 0 2 0 3 300 0 0.
DUMP.prom: dev SCSI 0 2 0 3 300 0 0, block 155648
DUMP: Header to 0x800c01 at 376831 (0x5bfff)
succeeded

T.R	Title	User	Personal Name	Date	Lines
480.1	or tell us where we can copy it from	SMURF::BAT	Segui la tua beatitudine	`Thu Apr 17 1997 21:16`	2
	Can you send or post the /var/adm/crash/crash_data file?
480.2	from mark	SMURF::BAT	Segui la tua beatitudine	`Thu Apr 17 1997 21:34`	5
	Boot genvmunix to get the crash data file... Also, do you have a CDROM with DIGITAL UNIX V4.0A, and can you install it and test your driver? We may have one if you don't. That would eliminate that variable.
480.3		ADISSW::FERRARA		`Tue Apr 22 1997 12:43`	4
	Thanks for the suggestions...I need to unpack in my new office here in ZK2...stay tuned... -Bob
480.4	Ok on UNIX V4.0a	ADISSW::FERRARA		`Wed Apr 23 1997 13:25`	8
	I loaded Digital UNIX V4.0a and installed/configured my device driver...all worked fine. I will now re-load MLS+ v4.0a and try again... -BobF
480.5	System Panics under MLS+ 4.0A	ADISSW::FERRARA		`Wed Apr 23 1997 20:29`	485
480.6	Recompile/build on MLS+ == Better	ADISSW::FERRARA		`Fri Apr 25 1997 14:21`	26
	Another update... I re-compiled my device driver source code ON MY MLS++ MACHINE (where as before, I was taking my-driver.mod produced on my UNIX machine) and statically configured it into my MLS+ kernel. So far so good... - I've re-booted my MLS+ system several times, no panic's - I've successfully pinged another system, and vice versa, no panic's - I've successfully rlogin'ed to another system, no panic's - I've successfully ftp'ed an 8 MB file over the network, no panic's Can someone confirm that this is the "right thing to do" and panic's/crashes/etc would be expected otherwise... Thanks, BobF
480.7	It's so automatic: we didn't realize you weren't building native	SMURF::BAT	Segui la tua beatitudine	`Fri Apr 25 1997 16:38`	5
	I'll belly up to the bar: Yes, I confirm that you must build the source code on MLS+. You cannot substitute kernel modules from base OSF -- even from DIGITAL UNIX V4.0A, and expect it to work. Yes, there are modules that have no MLS+ modifications, but unless someone examines the code, you are asking for problems.