2.6.15 failure on power4 iSeries

[will schmidt]

Michael Ellerman wrote:
> On Wed, 18 Jan 2006 01:42, will schmidt wrote:
> 
>>Michael Ellerman wrote:
>>
>>>On Tue, 17 Jan 2006 10:35, will schmidt wrote:
>>>
>>>>attempting to boot current kernels on a power4 iSeries doesnt work.  have
>>>>tried both the powerpc-git tree and the torvalds-git tree.
>>>>
>>>>OS/400 RefCode is "C200 82FF".  (which means nothing to me :-)
>>>
>>>C20082FF VSP IPL complete successfully
>>>
>>>
>>>>no console output at all.
>>>
>>>Enable early debugging for iSeries (now via Kconfig, under Kernel
>>>hacking), then enable DEBUG in platforms/iseries/* and kernel/setup_64.c,
>>>then when it dies hit "ctrl-x ctrl-x" and see if you get any output.
>>
>>  -> early_setup()
>>Probing machine type for platform 201...
>>Found, Initializing memory management...
>>  <- early_setup()
>>  -> setup_system()
>>  -> initialize_cache_info()
>>  <- initialize_cache_info()
>>  -> iSeries_init_early()
>>  <- iSeries_init_early()
>>Mapping load area - physical addr = 0000000000000000
>>                     absolute addr = 0000000018000000
>>Load area size 32768K
>>HPT absolute addr = 000000014a000000, size = 32768K
>>Mainstore_VPD: Regatta
>>ms_vpd: processing area  0  blocks=1
>>           block 0 absStart=8000000000000000 absEnd=8000000100000000
>>ms_vpd: processing area  1  blocks=1
>>           block 0 absStart=8000000100000000 absEnd=8000000200000000
>>ms_vpd: processing area  2  blocks=1
>>           block 0 absStart=8000000200000000 absEnd=8000000280000000
>>ms_vpd: 3 sorted memory blocks
>>           Bitmap range: 0000000000000000 - 0000000100000000
>>         Absolute range: 8000000000000000 - 8000000100000000
>>           Bitmap range: 0000000100000000 - 0000000200000000
>>         Absolute range: 8000000100000000 - 8000000200000000
>>           Bitmap range: 0000000200000000 - 0000000280000000
>>         Absolute range: 8000000200000000 - 8000000280000000
>>Mainstore_VPD: numMemoryBlocks = 3
>>Mainstore_VPD: block   0 logical chunks 0000000000000000 - 0000000000004000
>>                              abs chunks 0000000000000000 -
>>0000000000004000 Mainstore_VPD: block   1 logical chunks 0000000000004000 -
>>0000000000008000 abs chunks 0000000000004000 - 0000000000008000
>>Mainstore_VPD: block   2 logical chunks 0000000000008000 - 000000000000a000
>>abs chunks 0000000000008000 - 000000000000a000 Page orders: linear mapping
>>= 12, others = 12
>>mf.c: iSeries Linux LPAR Machine Facilities initialized
>>  -> smp_release_cpus()
>>  <- smp_release_cpus()
>>  <- setup_system()
>>
>>
>>This looks like the "2.6.15-mm4 failure on power5" output..  but I dont
>>see a *cpuc-to-mutexes.patch in this tree to back out.  (torvalds-git)
>>
>>Will clean my glasses and look closer in a bit.. :-)
> 
> 
> Yeah, looks similar. I can't reproduce that crash on my POWER5 box here 
> though, so I'm not sure if that patch is actually the problem. Might be worth 
> git bisecting.


I already git bisected to get it narrowed down to that one patch.   Or are you saying that
the patch is broken up into more parts in the powerpc-git tree?

Same tree builds and boots OK on power5 partition here too..   this seems to be something
unique to power4 iSeries.


> 
> You could try adding calls to udbg_printf() in start_kernel() to see if we're 
> getting in there.
> 
> cheers
>