Un-Setting VME3PCI Master Master Enable on MVME2431

Un-Setting VME3PCI Master Master Enable on MVME2431

[Simone Piccardi]

Hi,
today I got another "little" trouble with our board.
We finally get a dataway display that cannot work. After many attempts
to use it (it was perfectly working in other crates) we try to go back
to the default configuration of the board (I switched to VME3PCI Master
Master Enable = N according to what said in the mailing list to boot the
kernel) and everything went working.
The manual said:
N do not enable the VME Bus Interface

So is this only related to the PPCBug (and with a working kernel all
will be OK?) Or I have to worry?
Probably this is a stupid question, but as I said I'm totally new to
VME.
Thanks

-- 
Simone Piccardi
Microsoft is NOT the answer. Microsoft is the Question.
The answer is: "NO!"

[[ This message was sent via the linuxppc-dev mailing list.  Replies are ]]
[[ not  forced  back  to the list, so be sure to Cc linuxppc-dev if your ]]
[[ reply is of general interest. Please check http://lists.linuxppc.org/ ]]
[[ and http://www.linuxppc.org/ for useful information before posting.   ]]

Re: Un-Setting VME3PCI Master Master Enable on MVME2431

[Gabriel Paubert]

On Wed, 21 Apr 1999, Simone Piccardi wrote:

> 
> Hi,
> today I got another "little" trouble with our board.
> We finally get a dataway display that cannot work. After many attempts
> to use it (it was perfectly working in other crates) we try to go back
> to the default configuration of the board (I switched to VME3PCI Master
> Master Enable = N according to what said in the mailing list to boot the
> kernel) and everything went working.
> The manual said:
> N do not enable the VME Bus Interface
> 
> So is this only related to the PPCBug (and with a working kernel all
> will be OK?) Or I have to worry?

Don't worry. The settings of the Raven/Hawk are not ideal for large VME
systems and it's better to enable the Universe later in the boot process
or when loading the driver. This bridge is simply a PCI space hog,
actually if you sum the maximal size of all the programmable images you
would get something slightly over 32 Gb so it's better to disable it when
playing with the host bridge (not yet fully done). 

> Probably this is a stupid question, but as I said I'm totally new to
> VME.

No problem. We all have been newbies at some point. 

	Gabriel


[[ This message was sent via the linuxppc-dev mailing list.  Replies are ]]
[[ not  forced  back  to the list, so be sure to Cc linuxppc-dev if your ]]
[[ reply is of general interest. Please check http://lists.linuxppc.org/ ]]
[[ and http://www.linuxppc.org/ for useful information before posting.   ]]

Re: Un-Setting VME3PCI Master Master Enable on MVME2431

[David De Ridder]

On Wed, 21 Apr 1999, Gabriel Paubert wrote:
> On Wed, 21 Apr 1999, Simone Piccardi wrote:

> > We finally get a dataway display that cannot work. After many attempts
> > to use it (it was perfectly working in other crates) we try to go back
> > to the default configuration of the board (I switched to VME3PCI Master
> > Master Enable = N according to what said in the mailing list to boot the
> > kernel) and everything went working.
> > The manual said:
> > N do not enable the VME Bus Interface
> > 
> > So is this only related to the PPCBug (and with a working kernel all
> > will be OK?) Or I have to worry?
> 
> Don't worry. The settings of the Raven/Hawk are not ideal for large VME
> systems and it's better to enable the Universe later in the boot process
> or when loading the driver. This bridge is simply a PCI space hog,

 I have a report concerning VME3PCI myself : when it is set to ``N'',
 the ``SYSFAIL'' LED in our crate lights, and stays lit. I have been
 told this is bad, because it indicates that there will be no VME
 transfers possible between the MVME board and other VME devices.
 However, once the universe driver is loaded (e.g., through modprobe),
 the LED unlights. I assume this is because the Universe chipset is
 then correctly initialised, right ?

Regards,

+-----------------------------------------------------------------+
   David 'Septimus' De Ridder     <david.de.ridder@bitsmart.com>

"Blow the world apart with a press of a button
 we all say it shouldn't but we know it will happen.
 Again and again, like the rain and the ark :
 almost forever, eternally dark."
 - Anne Clark
+-----------------------------------------------------------------+


[[ This message was sent via the linuxppc-dev mailing list.  Replies are ]]
[[ not  forced  back  to the list, so be sure to Cc linuxppc-dev if your ]]
[[ reply is of general interest. Please check http://lists.linuxppc.org/ ]]
[[ and http://www.linuxppc.org/ for useful information before posting.   ]]

Re: Un-Setting VME3PCI Master Master Enable on MVME2431

[Gabriel Paubert]

On Fri, 23 Apr 1999, David De Ridder wrote:

>  I have a report concerning VME3PCI myself : when it is set to ``N'',
>  the ``SYSFAIL'' LED in our crate lights, and stays lit. I have been
>  told this is bad, because it indicates that there will be no VME
>  transfers possible between the MVME board and other VME devices.
>  However, once the universe driver is loaded (e.g., through modprobe),
>  the LED unlights. I assume this is because the Universe chipset is
>  then correctly initialised, right ?

Indeed. The driver resets the SYSFAIL signal and the associated pending
interrupts on init (Actually Universe initialization is a very complex
topic, it's easy to get conflicts in PCI memory space).

	Regards,
	Gabriel. (progressing slowly on the doc for the Universe driver)



[[ This message was sent via the linuxppc-dev mailing list.  Replies are ]]
[[ not  forced  back  to the list, so be sure to Cc linuxppc-dev if your ]]
[[ reply is of general interest. Please check http://lists.linuxppc.org/ ]]
[[ and http://www.linuxppc.org/ for useful information before posting.   ]]