[flashrom] Olimex ARM-USB-OCD-H output shifted by one

Stefan Tauner stefan.tauner at alumni.tuwien.ac.at
Sun Nov 2 13:35:00 CET 2014

On Thu, 23 Oct 2014 20:45:09 +0200
Sanne Ausma <sanne at ausma-engineering.nl> wrote:

> On 2014-10-20 23:56, Sanne Ausma wrote:
> > Hi,
> > 
> > Using flashrom to probe for an EON 25Q32 using the ARM-USB-OCD-H I get
> > the following:
> > […]
> > RDID returned 0x0e 0x18 0x0b 0x0e. probe_spi_rdid_generic: id1 0x0e, 
> > id2 0x180b
> > ...
> > 
> > 
> > Whereas it should have read id1: 0x1c, id2: 0x3016  (0x1c = 0x0e<<1
> > and 0x3016 = 0x180b<<1)
> > 
> > Is this a known issue with the Olimex ARM-USB-OCD-H?
> > The wiki said it was currently untested...
> > 
> > Thnx  
> I looked at the signals using a scope and all looked well, the data 
> comming from the flash was ok.
> After slowing the SPI clock down from 30MHz to 15MHz, the correct data 
> also appeared from flashrom, see the output below...
> So I guess it is either a timing issue with the ARM-USB-OCD-H or USB / 
> ftdi driver related.

Hello Sanne,

thanks for your report. I am not aware of any unsolved problems with
any chip or programmer supported by the ft2232_spi driver in flashrom.
I would suspect a simple timing/sampling issue caused by the high speed
or too low signal quality. Normally that would produce way less
deterministic values though, so I am not entirely convinced.
However, since it works fine with 15 MHz and the programmer is already
marked as successfully tested, I'll leave everything as is. If you
become aware of any additional information please don't hesitate to
show it, thanks.
Kind regards/Mit freundlichen Grüßen, Stefan Tauner

More information about the flashrom mailing list