Montecito

[Jens Kieffer-Olsen]

Our rx4640 has 4 processors 0x000000001f020204
so is their official name Montecito?
Or are they only designated Montecito, if we upgrade to dual-core?
Point is that Oracle states "The Montecito chip is certified on patch set 10.1.0.5 and higher".

[rick jones]

Well, first, an rx4640 is not a workstation :)

Second, if you have all four sockets filled, and only four cores showing, unless two of the sockets were disabled with the EFI cpuconfig command, you likely have "Madison" processors.

If you are running HP-UX I suspect that machinfo may give some clues. If running Linux then I think you would cat /proc/cpuinfo

Neither Montecito, nor Madison are "official" names, they were just codenames. Official names are much more dry.

In broad handwaving terms, dual-cores are Montecito (well, there was mx2...) however, there is also a single-core Montecito.

[Torsten.]

Hi,

do I have a montecito dual core cpu or not?

How to identify?

See this thread:

http://forums1.itrc.hp.com/service/forums/questionanswer.do?threadId=1089669

[Jens Kieffer-Olsen]

Thanks to Rick and Torsten. I managed to find an article that indicates a Montecito shows family number 32. My processors have an L3 cache of 6M just like the 9010 mono-core Montecito, but since the family number is 31, I conclude that the processors are of type Madison 9M model 2 revision id 2.
So, no cigar to the rx4640 :-)

[rick jones]

Although, if you want you _can_ upgrade an rx4640 to Montecito processors. Better still would be an upgrade from an rx4640 to an rx6600 :) Or an rx3600 if you only really need four cores (two sockets).

[Jens Kieffer-Olsen]

ok, but as it happens we have an option to upgrade the rx4640. The confusion owed to the fact that we thought we had single-core Montecito processors when in fact we had Madison 9Ms. Point here was that Oracle indicates our current 10.1.0.4 version isn't certified for Montecito, which would not make sense, if our processors were already Montecitos. - Now we will upgrade to 10.1.0.5 and then bring our upgrade option into play.

[Torsten.]

Hi,

this is the output for montecito:

/usr/contrib/bin/machinfo
CPU info:
Number of CPUs = 12
Clock speed = 1598 MHz
Bus speed = 533 MT/s
CPUID registers
vendor information = "GenuineIntel"
processor serial number = 0x0000000000000000
processor version info = 0x0000000020000704
architecture revision: 0
processor family: 32 Intel(R) Itanium 2 9000 series
...


and another:

/usr/contrib/bin/machinfo
CPU info:
32 Intel(R) Itanium 2 9000 series processors (1.6 GHz, 18 MB)
533 MT/s bus, CPU version C2
64 logical processors (2 per socket)
...

[Jens Kieffer-Olsen]

Upgrade to Montecito went ok:

# machinfo
CPU info:
Number of CPUs = 8
Clock speed = 1600 MHz
Bus speed = 400 MT/s
CPUID registers
vendor information = "GenuineIntel"
processor serial number = 0x0000000000000000
processor version info = 0x0000000020000704
architecture revision: 0
processor family: 32 Intel(R) Itanium 2 9000 series
processor model: 0 Intel(R) Itanium 2 9000 series
processor revision: 7 Stepping C2
largest CPUID reg: 4
processor capabilities = 0x0000000000000005
implements long branch: 1
implements 16-byte atomic operations: 1
Bus features
implemented = 0xbdf0000020000000
selected = 0x0000000000000000

Cache info (per core):
L1 Instruction: size = 16 KB, associativity = 4
L1 Data: size = 16 KB, associativity = 4
L2 Instruction: size = 1024 KB, associativity = 8
L2 Data: size = 256 KB, associativity = 8
L3 Unified: size = 9216 KB, associativity = 9