HP3000-L Archives

January 2010, Week 3

HP3000-L@RAVEN.UTC.EDU

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
"Hawkins, Jim (ESS TCE Q&W)" <[log in to unmask]>
Reply To:
Hawkins, Jim (ESS TCE Q&W)
Date:
Fri, 15 Jan 2010 05:18:39 +0000
Content-Type:
text/plain
Parts/Attachments:
text/plain (135 lines)
MPEMXQ3

Notice that PDC relocation is also done on HP-UX systems (

External Symptoms, Cause, Fix Description
=========================================
 
Intermittent small system freezes and system SW clock slips happen
on multiprocessor (MP) A- and N-class systems. The clock slips
are not gradual, they happen in steps of several seconds each.
 
The clock slips occur when a CPU that tries to load & run PDC code
from NVRAM has to block on a condition where the system bus is too busy.
 
Loading (copying) the PDC code from NVRAM (EEPROM) into normal memory
will avoid any blocking for PDC code and will also make its execution
significantly faster.  MPE code using PDC calls (mainly diagnostics)
will see increased performance.
 
This is a solution borrowed from HP-UX where similar issues were seen
with A-, L- and N-classes.  Unlike the HP-UX code though, MPE will not
'panic' if it cannot relocate the PDC.  MPE will fallback to execute
PDC from NVRAM instead (just like without this patch).
 
 
Customer Environment Changes
============================
 
This enhancement will allocate 2 MBytes of main memory for the
copy of the PDC code (image). In other words: it will reduce the
amount of memory available for the OS and applications by 2MB
(on A- and N-classes only).
 
The system bootup message will change, additional lines will
be displayed after ISL > START, showing the success or failure
to load PDC code into main memory. Should it fail to load then
it will simply be executed from NVRAM (as without the patch).
 
The most likely reason for a failure to relocate PDC will be
that the system cannot find 2MB of physically contiguous main
memory pages (free of entries in the page deallocation table,
aka PDT).
 
Success:
 
ISL > START NORECOVERY
 
Initialize_genesis - Ver bld1: <<pci  2.1601>>
Initialize_genesis - Relocating pdc...
Initialize_genesis - Relocate pdc from 0xffff800000 to 0x208ee000
completed.
WED, DEC  8, 2004,  1:44:11 PM (y/n)?
 
 
Example of a failure :
 
Initialize_genesis - Relocating pdc...
Initialize_genesis - PDC relocation failed: mm_status = 0xffdc0065,
pdc_status = 0x00, old pdc = 0xffff800000, new pdc = 0x00
 
 
Hardware Dependencies
=====================
 
Only customers using an A-class or N-class system need to
install this patch.  If this is installed on other platforms
the new code will be skipped.


-----Original Message-----
From: HP-3000 Systems Discussion [mailto:[log in to unmask]] On Behalf Of Mark Landin
Sent: Wednesday, January 13, 2010 4:52 PM
To: [log in to unmask]
Subject: Re: [HP3000-L] Relocating PDC message?

It happens on all the A-, L-, and N- class 9000 systems. It is normal
and always seems to take about 30 seconds more than past when you
really start worrying something is wrong :)

On 1/13/10, Craig Lalley <[log in to unmask]> wrote:
> John,
>
> It is just relocating the PDC from firmware to memory where access is
> supposedly faster.  Happens on all N-Classes.
>
> I am having a brain cramp... PDC stands for Processor Dependent Code????
>
> -Craig
>
>
> --- On Wed, 1/13/10, John MacLerran <[log in to unmask]> wrote:
>
> From: John MacLerran <[log in to unmask]>
> Subject: Relocating PDC message?
> To: [log in to unmask]
> Date: Wednesday, January 13, 2010, 8:03 AM
>
> Hi Everyone,
>
> We are running an HP3000 N-4000 4-way 440mhz box (MPE 7.5 PP2 + some
> reactive patches), and we've been seeing this message lately when we do a
> Start Norecovery.  It comes up just before the prompt to verify the time:
>
> Initialize_genesis - Relocating pdc...
> Initialize_genesis - Relocate pdc from 0xfff800000 to 0x245d3000
>
> Can anyone tell me what this means and whether we should be worried about
> it?  We have a patch that is staged that I haven't committed yet; I didn't
> want to commit the patch without knowing what the above message means.
>
> Thanks!
> -john maclerran
> Idaho State University
> [log in to unmask]
>
> * To join/leave the list, search archives, change list settings, *
> * etc., please visit http://raven.utc.edu/archives/hp3000-l.html *
>
> * To join/leave the list, search archives, change list settings, *
> * etc., please visit http://raven.utc.edu/archives/hp3000-l.html *
>

-- 
Sent from my mobile device

"He's old enough to know what's right and young enough not to choose
it. He's strong enough to win the world and weak enough to lose it." -
Neal Peart

* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *

* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *

ATOM RSS1 RSS2