Re: tstile lockups - no luck

[Donald Lee <MacPPC2%c.icompute.com@localhost>]

I put the LOCKDEBUG kernel on my production web server, and the same
kernel on my test machine.  I've been seeing crashes on the server,
and have my test case reliably generating TSTILE hangs on the test
machine (with the GENERIC kernel)

Unfortunately, no luck.

With eht LOCKDEBUG kernel.  The test machine is like a rock.  It's been running
for days, and it won't fail.

The production server, on the other hand, seems less stable with the
LOCKDEBUG kernel.  The audio driver crash being evaded, it seems to crash
regularly, and more frequently than the GENERIC kernel.

UNFORTUNATELY, it is not the crash I'm looking for.  I see this "pmap" crash
sometimes, too, but that one actually crashes, and is rare.  The "TSTILE" hang
is nastier, and doesn't actually crash.  It just hangs.

My most recent traceback:

Jul  2 04:29:28 mercy syslogd[151]: restart
Jul  2 04:29:28 mercy /netbsd: panic: kernel diagnostic assertion "j < 8" 
failed: file "../../../../arch/powerpc/oea/pmap.c", line 927 
Jul  2 04:29:28 mercy /netbsd: cpu0: Begin traceback...
Jul  2 04:29:28 mercy /netbsd: 0xe1ed7e10: at kern_assert+0x68
Jul  2 04:29:28 mercy /netbsd: 0xe1ed7e50: at pmap_pte_spill+0xec
Jul  2 04:29:28 mercy /netbsd: 0xe1ed7e90: at trap+0x794 
Jul  2 04:29:28 mercy /netbsd: 0xe1ed7f20: user DSI read trap @ 0xf9ea7000 by 
0xfceaa2b4: srr1=0x200d032
Jul  2 04:29:28 mercy /netbsd:            r1=0xffffbf30 cr=0x22002088 
xer=0x20000000 ctr=0xfceaa2b0 dsisr=0x40000000
Jul  2 04:29:28 mercy /netbsd: cpu0: End traceback... 
Jul  2 04:29:28 mercy /netbsd: dumpsys: TBD      
Jul  2 04:29:28 mercy /netbsd: Skipping crash dump on recursive panic
Jul  2 04:29:28 mercy /netbsd: panic: wdc_exec_command: polled command not done
Jul  2 04:29:28 mercy /netbsd: cpu0: Begin traceback...
Jul  2 04:29:28 mercy /netbsd: 0xe1ed7c70: at panic+0x4c
Jul  2 04:29:28 mercy /netbsd: 0xe1ed7cb0: at wdc_exec_command+0x1b8
Jul  2 04:29:28 mercy /netbsd: 0xe1ed7cd0: at wd_flushcache+0x128
Jul  2 04:29:28 mercy /netbsd: 0xe1ed7d70: at wd_shutdown+0x64
Jul  2 04:29:28 mercy /netbsd: 0xe1ed7d80: at device_pmf_driver_shutdown+0x20
Jul  2 04:29:28 mercy /netbsd: 0xe1ed7d90: at pmf_system_shutdown+0xbc
Jul  2 04:29:28 mercy /netbsd: 0xe1ed7dc0: at cpu_reboot+0x19c      
Jul  2 04:29:28 mercy /netbsd: 0xe1ed7de0: at vpanic+0x1f4
Jul  2 04:29:28 mercy /netbsd: 0xe1ed7e10: at kern_assert+0x68
Jul  2 04:29:28 mercy /netbsd: 0xe1ed7e50: at pmap_pte_spill+0xec
Jul  2 04:29:28 mercy /netbsd: 0xe1ed7e90: at trap+0x794
Jul  2 04:29:28 mercy /netbsd: 0xe1ed7f20: user DSI read trap @ 0xf9ea7000 by 
0xfceaa2b4: srr1=0x200d032
Jul  2 04:29:28 mercy /netbsd:            r1=0xffffbf30 cr=0x22002088 
xer=0x20000000 ctr=0xfceaa2b0 dsisr=0x40000000
Jul  2 04:29:28 mercy /netbsd: cpu0: End traceback...
Jul  2 04:29:28 mercy /netbsd: Skipping crash dump on recursive panic
Jul  2 04:29:28 mercy /netbsd: panic: wdc_exec_command: polled command not done
Jul  2 04:29:28 mercy /netbsd: cpu0: Begin traceback...
Jul  2 04:29:28 mercy /netbsd: 0xe1ed7ad0: at panic+0x4c
Jul  2 04:29:28 mercy /netbsd: 0xe1ed7b10: at wdc_exec_command+0x1b8
Jul  2 04:29:28 mercy /netbsd: 0xe1ed7b30: at wd_flushcache+0x128
Jul  2 04:29:28 mercy /netbsd: 0xe1ed7bd0: at wd_shutdown+0x64
Jul  2 04:29:28 mercy /netbsd: 0xe1ed7be0: at device_pmf_driver_shutdown+0x20
Jul  2 04:29:28 mercy /netbsd: 0xe1ed7bf0: at pmf_system_shutdown+0xbc
Jul  2 04:29:28 mercy /netbsd: 0xe1ed7c20: at cpu_reboot+0x74
Jul  2 04:29:28 mercy /netbsd: 0xe1ed7c40: at vpanic+0x1f4
Jul  2 04:29:28 mercy /netbsd: 0xe1ed7c70: at panic+0x4c
Jul  2 04:29:28 mercy /netbsd: 0xe1ed7cb0: at wdc_exec_command+0x1b8
Jul  2 04:29:28 mercy /netbsd: 0xe1ed7cd0: at wd_flushcache+0x128
Jul  2 04:29:28 mercy /netbsd: 0xe1ed7d70: at wd_shutdown+0x64
Jul  2 04:29:28 mercy /netbsd: 0xe1ed7d80: at device_pmf_driver_shutdown+0x20
Jul  2 04:29:28 mercy /netbsd: 0xe1ed7d90: at pmf_system_shutdown+0xbc
Jul  2 04:29:28 mercy /netbsd: 0xe1ed7dc0: at cpu_reboot+0x19c
Jul  2 04:29:28 mercy /netbsd: 0xe1ed7de0: at vpanic+0x1f4
Jul  2 04:29:28 mercy /netbsd: 0xe1ed7e10: at kern_assert+0x68
Jul  2 04:29:28 mercy /netbsd: 0xe1ed7e50: at pmap_pte_spill+0xec
Jul  2 04:29:28 mercy /netbsd: 0xe1ed7e90: at trap+0x794
Jul  2 04:29:28 mercy /netbsd: 0xe1ed7f20: user DSI read trap @ 0xf9ea7000 by 
0xfceaa2b4: srr1=0x200d032
Jul  2 04:29:28 mercy /netbsd:            r1=0xffffbf30 cr=0x22002088 
xer=0x20000000 ctr=0xfceaa2b0 dsisr=0x40000000
Jul  2 04:29:28 mercy /netbsd: cpu0: End traceback...
Jul  2 04:29:28 mercy /netbsd: rebooting
J

Sigh.....

-dgl-