Re: [Openipmi-developer] FRU DIMM checksum error ( Serial Presence Detect )

[Corey M. <mi...@ac...>]

Hmm, the FRU code is not structured to handle this very well right now.  
Let me think about it a little; the code restructure will be significant.

-Corey

Pat Campbell wrote:

>Hi,
>
>On my system, running "ipmisample smi 0" results in a FRU checksum
>error.
>
>System information:
>   Intel Server Board #SE7520BD2
>   Server Memory from Wintec Inc, PC2100/1024MB
>   Intel(R) Management Module - Professional Edition
>       Board Part Number   : C46194-405
>       Board FRU File ID   : FRU Ver 0.03
>
>Results from running "ipmisample smi 0"
> .................
> Sensor added: 32.5.DIMM 3A0
> Sensor added: 32.6.DIMM 3B0
> Sensor added: 32.3.DIMM 2A0
> Sensor added: 32.4.DIMM 2B0
> Sensor added: 32.1.DIMM 1A0
> Sensor added: 32.2.DIMM 1B0
> ................
> EINF: .1.20.6.0.0.0 fru.c(process_fru_info): FRU checksum failed
> WARN: (32.1) entity.c(fru_fetched_ent_cb):Error fetching entity 32.1
>FRU: 9
> EINF: .1.20.7.0.0.0 fru.c(process_fru_info): FRU checksum failed
> WARN: (32.2) entity.c(fru_fetched_ent_cb):Error fetching entity 32.2
>FRU: 9
> EINF: .1.20.8.0.0.0 fru.c(process_fru_info): FRU checksum failed
> WARN: (32.3) entity.c(fru_fetched_ent_cb):Error fetching entity 32.3
>FRU: 9
> EINF: .1.20.9.0.0.0 fru.c(process_fru_info): FRU checksum failed
> WARN: (32.4) entity.c(fru_fetched_ent_cb):Error fetching entity 32.4
>FRU: 9
> ...........................
>
>Doing some debugging with a breakpoint at process_fru_info(), data for
>entity 32.1 
>does not contain a FRU common header, is Serial Presence Data ( SPD ).
>
>(gdb) x/128 fru->data        ( breakpoint at process_fru_info() at
>first checksum test )
>0x605920:       0x80    0x08    0x07    0x0d    0x0b    0x02    0x48   
>0x00
>0x605928:       0x04    0x75    0x75    0x02    0x82    0x04    0x04   
>0x01
>0x605930:       0x0e    0x04    0x0c    0x01    0x02    0x26    0x00   
>0xa0
>0x605938:       0x75    0x00    0x00    0x50    0x3c    0x50    0x2d   
>0x80
>0x605940:       0x90    0x90    0x50    0x50    0x00    0x00    0x00   
>0x00
>0x605948:       0x00    0x00    0x00    0x00    0x00    0x00    0x00   
>0x00
>0x605950:       0x00    0x00    0x00    0x00    0x00    0x00    0x00   
>0x00
>0x605958:       0x00    0x00    0x00    0x00    0x00    0x00    0x00   
>0x11
>0x605960:       0x00    0x00    0x00    0x00    0x00    0x00    0x00   
>0x00
>0x605968:       0x00    0x00    0x00    0x00    0x00    0x00    0x00   
>0x00
>0x605970:       0x00    0x00    0x00    0x00    0x00    0x00    0x00   
>0x00
>0x605978:       0x00    0x00    0x00    0x00    0x00    0x00    0x00   
>0x00
>0x605980:       0x00    0x00    0x00    0x00    0x00    0x00    0x00   
>0x00
>0x605988:       0x00    0x00    0x00    0x00    0x00    0x00    0x00   
>0x00
>0x605990:       0x00    0x00    0x00    0x00    0x00    0x00    0x00   
>0x00
>0x605998:       0x00    0x00    0x00    0x00    0x00    0x00    0x00   
>0x00
>
>Decode on same system using impiutil/fruconfig
> SDR[002c] FRU  20 06 20 01 DIMM 1A SPD
> Memory SPD Size     : 128
> Memory Type         : DDR
> Module Density      : 512 MB per bank
> Module Banks        : 2 banks
> Module Rows, Cols   : 13 rows, 11 cols
> DIMM Config Type    : ECC
> Manufacturer ID     :  (0x00)
> Manufacturer Part#  :
> Manufacturer Rev    : 00 00
> Manufacturer Date   : year=00 week=00
> Assembly Serial Num : 00000000
>
>Matches decode of information in the gdb dump above.  ipmitool also
>yields a matching 
>decode of the SPD data.
>
>I have looked at the IPMI FRU Information Storage Definition, really
>does not address this
>case.  Both ipmitool and ipmiutil special cased SPD data.
>
>So, are you guys working on handling a FRU that reports SPD data?  I
>have some spare cycles
>right now and with some direction can help implement this code.
>
>Looking forward to your comments.
>
>Pat Campbell  ( pl...@no... )
>
>
>
>
>
>
>-------------------------------------------------------
>SF.Net email is Sponsored by the Better Software Conference & EXPO
>September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices
>Agile & Plan-Driven Development * Managing Projects & Teams * Testing & QA
>Security * Process Improvement & Measurement * http://www.sqe.com/bsce5sf
>_______________________________________________
>Openipmi-developer mailing list
>Ope...@li...
>https://lists.sourceforge.net/lists/listinfo/openipmi-developer
>  
>