[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: 8-Mbyte flash problems



One way to test the hardware might be to dump the memory at
0x8000000 and 0x80400000 and see if they match, which
would indicate that the high address space is a mirror
of the low, indicating a missing/faulty address line.

/Johan

----- Original Message -----
From: "Jonas Holmberg" <jonashg@xxxxxxx.com>
To: "David Kilp" <dkilp@xxxxxxx.net>
Cc: <dev-etrax@xxxxxxx.com>
Sent: den 8 mars 2002 09:45
Subject: RE: 8-Mbyte flash problems


> On Thu, 2002-03-07 at 19:18, David Kilp wrote:
> > Well, I had high hopes since it seems like someone else has successfully
> > used this AMD part. Unfortunately, it didn't help the programming. My
> > problem doesn't seem to be related to erasing since that seems to have
> > worked (and this part has never been fully programmed yet anyway). It
still
> > fails when it tries to program past the 4M boundary:
> >
> > PACKET_INFO
> > 0xC0004000
> > 0x00800018
> > Checksum of file is 0x73AC5E2A
> > FLASH
> > 0xC0004000
> > 0x00000000
> > 0x00800000
> > Found 1 x CFI at 0x80000000
> > 0x80000000: Erasing 0x00800000 bytes...done, verifying...OK
> > 0x80000000: Writing 0x00800000 bytes...
> > Verify error at 0x80400000
> > Verify error at 0x80400000
> >   .
> >   .
> >   .
> >
>
> What we know is that after the erase is done the entire chip contains
> only ones (0xff-bytes). So it looks like the erase is successful.
>
> Maybe it's a hardware problem. I know that on the developer board there
> aren't enough address lines to use 8MB flash (only 4MB). Make sure that
> you haven't got that limitation on your board too. Also look for and
> remove any solder between the flashchip pins.
>
> > Do I need to change something else besides the ptablespec file and
> > specifying the size when running the flashitall program??
>
> No.
>
> /Jonas
>