[linux-cirrus] Re: redboot and kernel version

  • From: Cory Tusar <ctusar@xxxxxxxxxxxxxxxxxx>
  • To: linux-cirrus@xxxxxxxxxxxxx
  • Date: Fri, 15 Apr 2005 10:28:28 -0400

Henk Stegeman wrote:
> --- Adam Yergovich <ayergo@xxxxxxxxxxx> wrote:
> 
>>Still no luck getting Linux up on my protoboard. 
>>I've talked with 
>>Cirrus a bit, and they haven't yet been able to send
>>me a redboot.bin, 
>>ramdisk.gz, and zImage they know to work with the
>>ep9301 processor.  
> 
> 
> 
> That kinda sounds like there's no redboot that works
> with the ep9301 at all. Which would also mean that
> redboot wouldn't work on the edb9301 (development
> kit).
> 
> On my edb9301 redboot works for about 5% of the time.
> That 5% of the time I can actually run Linux on the
> board for about one hour.

We currently use u-boot on our EP93xx-based linux projects.  I tried to 
get the Cirrus-supplied redboot (from cirrus-arm-linux-1.4.2) to 
function on an EDB9301 with the same results you describe below.

> Most of the time redboot blinks the red LED indicating
> that the SDRAM test has failed. There was also a
> problem with my edb9301 PCB. (bad connection/track),
> took me days to figure out what's wrong. Eventhough I
> should receive custommer support from Videon
> (developer of edb9301) they haven't responded to any
> my emails about these issues.

I've kicked this posting "up the chain" internally.  You should have 
received a response, even if only a "Please contact your board vendor or 
Cirrus FAE regarding this".

> There's also another thing: I noticed a difference in
> the pinout list of the ep9301 between the previous and
> latest release of the ep9301 datasheet. The edb9301 is
> based on the previous datasheet. 

Looks like Cirrus updated their datasheet, likely in conjunction with a 
new silicon revision.  The EDB9301 rev. 2 design we did for Cirrus was 
based upon the information available from Cirrus at the time.

> This was my question to Videon:
> On the most recent EP9301 datasheet (EP9301 Product
> Data Sheet) Pin 85 is labeled gnd_ring, on the old
> EP9301 datasheet (exists on the EDB9301 CD) however it
> was labeled otherwise and was used as such in your
> EDB9301 design. Should this be modified on the
> EDB9301? Or is the new EP9301 datasheet erroneous?

Seems that both pins 85 and 86 have been changed from TREQA/TACK (on the 
CD datasheet) to gnd_ring/NC (on the latest EP9301_PP5.pdf on 
cirrus.com).  I'd venture a guess that these correspond to a chip 
"feature" which did not prove out, and was subsequently "dropped". 
Please note that this is only /my personal opinion/ from reviewing the 
datasheets and based on previous experiences.  If you need a more 
definitive answer, please contact a Cirrus FAE.

-Cory


-- 
Cory T. Tusar
Embedded Systems Engineer
Videon Central, Inc.
2171 Sandy Drive
State College, PA 16801
(814) 235-1111 x316
(814) 235-1118 fax

"Debugging is twice as hard as writing the code in the first place.
   Therefore, if you write the code as cleverly as possible, you are,
     by definition, not smart enough to debug it."  --Brian W. Kernighan


Other related posts: