Re: [ES40-developers] Interesting messages from Tru64

[Brian W. <bdw...@in...>]

Thanks for the info.  I'll play around with it and see what I can come
up with.  

Can you apply this patch.  It tunes the interrupt re-send code so it
doesn't trigger so often.

Brian

Index: NewIde.cpp
===================================================================
RCS file: /cvsroot/es40/es40/src/NewIde.cpp,v
retrieving revision 1.16
diff -u -r1.16 NewIde.cpp
--- NewIde.cpp  27 Feb 2008 12:34:19 -0000      1.16
+++ NewIde.cpp  27 Feb 2008 14:04:51 -0000
@@ -1881,7 +1881,7 @@
            if(CONTROLLER(index).interrupt_fired != 0) 
            {
 
-             if((SEL_COMMAND(index).command_cycle - CONTROLLER(index).interrupt_fired) > 10)
+             if((SEL_COMMAND(index).command_cycle - CONTROLLER(index).interrupt_fired) > 200)
              {
                CONTROLLER(index).interrupt_pending=1;
                theAli->pic_deassert(1,6+index); 



On Wed, 2008-02-27 at 14:35 +0100, Camiel Vanderhoeven wrote:
> Hi Brian,
> 
> The ES40 has a management processor that takes care of monitoring
> power suppies, fans, and other hardware bits and pieces. It is
> extremely difficult to find any information about this. I managed to
> find some info in the ES40 service manual, and basically played a lot
> with things to get them to show up properly in SRM.
> 
> The management processor communicates with SRM - and I presume with
> the OS - through a thing called Dual-Port-RAM. One port is connected
> to the system bus, another to the management-processor. A lot of the
> bytes are places where the OS can read device information, but in some
> locations, the OS can deposit commands for the management-processor.
> The following copmmands were found out through trial-and-error and
> partial reverse engineering of SRM:
> 
> Address locations for commands:
>   // 0xf9:	buffer size
>   // 0xfb:fa	qualifier / address
>   // 0xfc:	completion code (0 = ok, 80 = error, 81 = invalid code, 82
> = invalid qualifier)
>   // 0xfd:	rmc command id for response
>   // 0xfe:	command code
>   // 0xff:	rmc command id for command
> 
> Command codes:
>   // 01:	update EEPROM
>   // 02:	update baud rate
>   // 03:	write to OCP
>   // F0:	update RMC flash
> 
> It could very well be that there are more commands I don't know about,
> and a failure to respond to these in a satisfactory manner may well be
> the cause of the power supply failures and temperature warnings.
> 
> You can find most of what I managed to find out in DPR.cpp. Mind you,
> this is some of the oldest and least-touched code in the emulator. It
> has not changed (other than some of the standard changes that have
> been made to the entire emulator) for well over a year, and a lot of
> my comments are still in Dutch. (I'm sorry about that...) It might be
> a good idea to put some read/write tracers in DPR.cpp, and try to
> figure out what Tru64 is doing with these.
> 
> Camiel.
> 
> 
> 
> 
> 
> On Wed, Feb 27, 2008 at 2:19 PM, Brian Wheeler <bdw...@in...> wrote:
> > I did a full install from dqb0 to dqa0 and while there were some
> >  timeouts (this build was before I started on the interrupt resend code)
> >  it installed and configured correctly (I left it run unattended
> >  overnight).
> >
> >  However, once it came up, it started announcing hardware failures :)
> >
> >  -----------------
> >  Broadcast Message from ro...@tr... (???) at 23:37 ...
> >
> >  The redundant power supply has failed.
> >
> >  Broadcast Message from ro...@tr... (???) at 23:37 ...
> >
> >  The system temperature is high. Possible reasons are
> >   a clogged air filter or a high ambient room temperature.
> >   The system will shut down in 14 minutes and 0 seconds..
> >  ------------------
> >
> >  So...is there any documentation for the hardware monitoring stuff?
> >
> >  Brian
> >
> >
> >
> >  -------------------------------------------------------------------------
> >  This SF.net email is sponsored by: Microsoft
> >  Defy all challenges. Microsoft(R) Visual Studio 2008.
> >  http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> >  _______________________________________________
> >  Es40-developers mailing list
> >  Es4...@li...
> >  https://lists.sourceforge.net/lists/listinfo/es40-developers
> >
> 
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Microsoft
> Defy all challenges. Microsoft(R) Visual Studio 2008.
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> _______________________________________________
> Es40-developers mailing list
> Es4...@li...
> https://lists.sourceforge.net/lists/listinfo/es40-developers