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

RE: Intel StrataFlash for free?

Thanks for the thread pointer; interesting discussion. Though it didn't
take up the issue of CPU-bundled flash.

In our device (and this is increasingly common) PCB space is very tight -
we have half the area of a credit card, one-sided. And power is precious,
so minimizing RAM is important, as it's always powered. (Our device is also
fairly general purpose, so we use Linux.) Thus the PXA262 looks pretty

But the K3 StrataFlash bundled on this CPU doesn't seem to support read
during write/erase, so that rules out XIP unless it's treated as ROM. And
it's NOR flash, which was branded in the referenced thread as slow-writing
and thus unsuitable for a writeable filesystem like jffs2. I do consider
this bundled flash "very inexpensive" - cheaper than RAM anyway.

This suggests a 2-chip approach, with the bundled flash used as XIP-able
ROM, and a separate NAND flash chip for the writeable filesystem. This way
we still limit use of RAM, although we sacrifice board space. Would anyone
care to critique or offer pointers on this approach?

Liam Breck, Director
PointServers.org                                  liam@xxxxxxx.org
Boston, Massachusetts                           http://pointservers.org/

At 05:14 AM 1/2/03 -0800, paulnash@xxxxxxx.com wrote:
>Yes, that is what Intel will definitely tell you is The Way.  Conveniently,
>it results in the sale of more flash.  Flash on the 261 and 262 is not
>essentially free, regardless of what Intel would have you believe.  If it
>weren't there, they wouldn't need the chip-scale package, and the processor
>itself would be cheaper alone.  So, the fact that they've priced it
>aggressively doesn't make it any more free than Internet Explorer ever was.
>This issue has been discussed quite often here.  Most recently I started a
>thread about this on the Linux-MTD mailing list which had some great
>conversation about the tradeoffs.  I recommend you check out the linux-mtd
>mailing list archives:

To unsubscribe from this list: send the line "unsubscribe jffs-dev" in
the body of a message to majordomo@xxxxxxx.com