Return to BSD News archive
Path: sserve!newshost.anu.edu.au!munnari.oz.au!yarrina.connect.com.au!harbinger.cc.monash.edu.au!msuinfo!agate!howland.reston.ans.net!gatech!gt-news!prism!cc.gatech.edu!rjc From: rjc@cc.gatech.edu (Russ Clark) Newsgroups: comp.unix.bsd Subject: SunOS Kernel Debugging Date: 15 Jul 1994 14:31:47 -0400 Organization: College of Computing, Georgia Tech Lines: 55 Message-ID: <306kmj$pkg@alps.cc.gatech.edu> NNTP-Posting-Host: alps.cc.gatech.edu NNTP-Posting-User: rjc Keywords: adb SunOS kernel debug I'm deep in the morass of kernel debugging with my protocols in SunOS 4.1.3 and I'm having a heck of a time fighting with adb. Big surprise eh? Specifically, here is my current problem. I am often getting crashes where the best adb can do for me is the following. --- <rjc>adb -k vmunix.3 vmcore.3 $c _panic(0xf816ed03,0xf8140e7c,0x30,0x80,0xf8141,0xf8333d38) + 6c _trap(0x9,0xf8140e7c,0x30,0x80,0x1,0x0) + 184 --- Needless to say, this isn't very useful for finding my problem. If it helps anybody, the panic messages for the above crash were: Jul 15 13:03:03 vmunix: panic: Data fault Jul 15 13:03:03 vmunix: BAD TRAP Jul 15 13:03:03 vmunix: pid 286, `myclient': Data fault Jul 15 13:03:03 vmunix: kernel read fault at addr=0xf20110c, pme=0x1 Jul 15 13:03:03 vmunix: Sync Error Reg 80<INVALID> Jul 15 13:03:03 vmunix: pc=0xf812bd04, sp=0xf8140e60, psr=0x7c6, context= 0x6 Jul 15 13:03:03 vmunix: g1-g7: 700, 4, ffffffff, 6, f8334000, 0, f8172c00 Jul 15 13:03:03 vmunix: Begin traceback... sp = f8140e60 Jul 15 13:03:03 vmunix: Called from f812a610, fp=f8140ec8, args=ff054f70 34 f201104 ff054fa4 ff054f70 7e0 Jul 15 13:03:03 vmunix: Called from f812a198, fp=f8140f30, args=ff00715c fff00608 ff054f70 ff054f62 ff054f50 1 Jul 15 13:03:03 vmunix: Called from f8005d8c, fp=f8140fa0, args=0 ff00900 0 f8171058 fff00608 ff00715c b0 Jul 15 13:03:03 vmunix: Called from f81184d0, fp=f8333970, args=400de4 f8 17d800 fce22bb4 f817d400 ffffffff f81726d8 Jul 15 13:03:03 vmunix: End traceback... Jul 15 13:03:03 vmunix: panic: Data fault --- So, does anybody have any advice on how to find out where I'm really dying? I'm already doing the old printf/syslog debug approach. I'm looking for something a little friendlier. I'm open to any advice you can offer. Thanks in advance! -- Russ Clark College of Computing, Georgia Tech, Atlanta, GA 30332-0280 rjc@cc.gatech.edu