ibm4xx_quiesce_eth?

ibm4xx_quiesce_eth?

[Sean MacLennan]

What exactly does this function do?

    /* Quiesce the MAL and EMAC(s) since PIBS/OpenBIOS don't
     * do this for us
     */

On the taco, with the 440EP, it doesn't seem to make any difference if I 
call this or not. Since I am using u-boot, is it not needed?


Cheers,
   Sean

Re: ibm4xx_quiesce_eth?

[Josh Boyer]

On Thu, 10 Jan 2008 13:38:12 -0500
Sean MacLennan <smaclennan@pikatech.com> wrote:

> What exactly does this function do?
> 
>     /* Quiesce the MAL and EMAC(s) since PIBS/OpenBIOS don't
>      * do this for us
>      */
> 
> On the taco, with the 440EP, it doesn't seem to make any difference if I 
> call this or not. Since I am using u-boot, is it not needed?

There are firmwares that do not reset the EMAC and MAL before passing
control to the client program (Linux in our case).  This can cause
weird things to happen, like spurious interrupts or DMAs from the
hardware overwriting kernel memory.  So we quiesce the hardware really
early on those.

I don't believe U-Boot has that problem.  If it does, it should be
fixed :)

josh

Re: ibm4xx_quiesce_eth?

[Stefan Roese]

On Thursday 10 January 2008, Josh Boyer wrote:
> On Thu, 10 Jan 2008 13:38:12 -0500
>
> Sean MacLennan <smaclennan@pikatech.com> wrote:
> > What exactly does this function do?
> >
> >     /* Quiesce the MAL and EMAC(s) since PIBS/OpenBIOS don't
> >      * do this for us
> >      */
> >
> > On the taco, with the 440EP, it doesn't seem to make any difference if I
> > call this or not. Since I am using u-boot, is it not needed?
>
> There are firmwares that do not reset the EMAC and MAL before passing
> control to the client program (Linux in our case).  This can cause
> weird things to happen, like spurious interrupts or DMAs from the
> hardware overwriting kernel memory.  So we quiesce the hardware really
> early on those.
>
> I don't believe U-Boot has that problem.  If it does, it should be
> fixed :)

No, U-Boot doesn't have this problem.

Best regards,
Stefan

=====================================================================
DENX Software Engineering GmbH,     MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: +49-8142-66989-0 Fax: +49-8142-66989-80  Email: office@denx.de
=====================================================================

Re: ibm4xx_quiesce_eth?

[Sean MacLennan]

Stefan Roese wrote:
>> There are firmwares that do not reset the EMAC and MAL before passing
>> control to the client program (Linux in our case).  This can cause
>> weird things to happen, like spurious interrupts or DMAs from the
>> hardware overwriting kernel memory.  So we quiesce the hardware really
>> early on those.
>>
>> I don't believe U-Boot has that problem.  If it does, it should be
>> fixed :)
>>     
>
> No, U-Boot doesn't have this problem.
>
> Best regards,
> Stefan
>   
Thanks for the confirmation! One less thing to worry about.

Cheers,
   Sean