SUMMARY: X causes panic

From: Patrick L. Nolan <pln_at_razzle.Stanford.EDU>
Date: Fri Aug 16 2002 - 21:53:54 EDT
I reported a panic in Xsun on an Ultra 5 (Solaris 7) after applying
a bunch of patches.

Thanks to Ric Anderson, Casper Dik, Steudten Thomas, Albert White,
and Mark Cain.  Some of them chimed in with anecdotal reports that
frame buffer patches or kernel patches have caused similar problems.

I took the experimental approach.  First I backed out the 60+
patches that were applied last week.  Then I put them back a
few at a time.  I was able to isolate the problem to patch 106148-14.  
Previously this machine had version -05 of that patch.  Backing
out 106148-14 cures the problem.

More information:

Casper Dik wanted to see an adb symbolic traceback.  Here it is:
$ adb -k unix.2 vmcore.2
physmem b92a
$C
complete_panic(0x0,0x10439000,0x70eab540,0x709578fa,0x0,0xc) + 24
        [savfp=0x4039b488,savpc=0x1000fe00]
do_panic(0x1,0x4039b59c,0x10410fb8,0x0,0x0,0x4) + 174
        [savfp=0x4039b4e8,savpc=0x10099dd4]
cmn_err(0x3,0x1040ce84,0x1004e800,0xf5,0xf5,0x10406c00) + 20
        [savfp=0x4039b550,savpc=0x1002c7d8]
die(0x31,0x4039b750,0x0,0x0,0x1040ce84,0x1040ce8c) + b4
        [savfp=0x4039b5b0,savpc=0x1002d450]
trap(0x4039b750,0x0,0x1,0x0,0x5,0x7107d870) + 914
        [savfp=0x4039b660,savpc=0x1001c564]

I set out to find out what frame buffer is installed.  To my
surprise, the script fbinfo, with -14 installed, also causes a 
panic with this traceback:
complete_panic(0x0,0x10439000,0x70e78940,0x70f0b95a,0x0,0xc) + 24
        [savfp=0x4032b650,savpc=0x1000fe00]
do_panic(0x1,0x4032b764,0x10410fb8,0x0,0x0,0x4) + 174
        [savfp=0x4032b6b0,savpc=0x10099dd4]
cmn_err(0x3,0x1040ce84,0x1004e800,0xf5,0xf5,0x10406c00) + 20
        [savfp=0x4032b718,savpc=0x1002c7d8]
die(0x31,0x4032b918,0x38,0x0,0x1040ce84,0x1040ce8c) + b4
        [savfp=0x4032b778,savpc=0x1002d450]
trap(0x4032b918,0x0,0x1,0x0,0x5,0x713b5c50) + 914
        [savfp=0x4032b828,savpc=0x1001c564]

That looks pretty similar.  As far as I can tell, fbinfo just runs
prtconf and parses the output.  With -14 backed out, fbinfo says:
ffb@1e  is FFB2  Vertical  Creator      1280x1024x76 Monitor type 4
(Sun 37x29cm RGB color monitor)
It doesn't have patch 106145, which would seem to be relevant to
the Creator board.

-----------------------------------------------------------------
Original question:
I rebooted my Ultra 30 (Solaris 7, many patches) today, and it 
panics when it tries to start the X server.  The log entry is
below.

It won't panic if I disable X by running "dtconfig -d".  (It's running
CDE).

I thought this might be related to a bunch of patches I installed
last week.  This was the first reboot since they were installed.
I looked through the READMEs for anything related to X or dtlogin.
This led me to back out patches 108376-38 and 107180-29, but that
didn't help.

Does this sound familiar to anyone?

------ Log entry: --------------------------------------------------

Aug 15 15:56:31 giants unix: BAD TRAP: cpu=0 type=0x31 rp=0x40391750
addr=0x0 mmu_fsr=0x0
Aug 15 15:56:31 giants unix: BAD TRAP occurred in module "genunix" due
to a NULL pointer dereference.
Aug 15 15:56:31 giants unix: Xsun:
Aug 15 15:56:31 giants unix: trap type = 0x31
Aug 15 15:56:31 giants unix: pid=303, pc=0x100cee64, sp=0x403917e0,
tstate=0x80001e05, context=0x1867
Aug 15 15:56:31 giants unix: g1-g7: 10430000, 100ce7f8, 100845d4, 0, 0,
0, 70e0b6c0
Aug 15 15:56:31 giants unix: Begin traceback... sp = 403917e0
Aug 15 15:56:31 giants unix: Called from 100a3fd8, fp=40391840, args=0
403918a4 41f66100 7004000 70fa3474 0
Aug 15 15:56:31 giants unix: Called from 101e6564, fp=403918c0,
args=70fa3420 0 0 0 41f66100 7004000
Aug 15 15:56:32 giants unix: Called from 1019efb0, fp=40391950,
args=70136000 d8 0100 70fa3420 1800 0 1045e75c
Aug 15 15:56:32 giants unix: Called from 10075284, fp=403919b8,
args=d80100 70fa 3420 0 0 7004000 40391a9c
Aug 15 15:56:32 giants unix: Called from 100a3bc8, fp=40391a18,
args=d80100 70fa 3420 0 0 7004000 40391a9c
Aug 15 15:56:32 giants unix: Called from 1010d998, fp=40391aa0, args=0 0
82 d801 00 7004000 70fa3420
Aug 15 15:56:32 giants unix: Called from 1010da3c, fp=40391b18,
args=d80100 1240 705eb3b8 40391c7c 7004000
Aug 15 15:56:32 giants unix: Called from 10084580, fp=40391ba8,
args=70dc128c 082 705eb3b8 40391c7c 7004000
Aug 15 15:56:32 giants unix: Called from 10084614, fp=40391c18, args=0
70dc128c3 7004000 70e0dbd0 0
Aug 15 15:56:32 giants unix: Called from 10035ff8, fp=40391c80, args=0
7004000 380000002 e 0
Aug 15 15:56:32 giants unix: Called from feb1a984, fp=ffbef478, args=0
7004000 380000002 e 0
Aug 15 15:56:32 giants unix: End traceback...
Aug 15 15:56:34 giants unix: panic[cpu0]/thread=70e0b6c0:
Aug 15 15:56:34 giants unix: trap
_______________________________________________
sunmanagers mailing list
sunmanagers@sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers
Received on Fri Aug 16 21:56:46 2002

This archive was generated by hypermail 2.1.8 : Thu Mar 03 2016 - 06:42:52 EST