Hi all,
I'm new to the list. I've been working with yaffs2 for
a while and have gotten to know it well. I'm a fan of
yaffs and like the way it stores information on the
flash. I have not experienced any functional problems
that were not the fault bad integration (me!).
I've got it working on a 2GB Samsung NAND flash. This
flash supports interleaved operation (two chip
selects) and cached programming (odd and even banks on
each chip). The host CPU is a 180MHz ARM9 that has a
dedicated smart media interface with HW ECC
calculation and DMA.
I started with MTD (Linux 2.6.13) but since have
bypassed the MTD layer all together and branched out
on my own page-oriented NAND driver framework. I
copied the yaffs_mtdif code to integrate my driver.
There are two big problems I am working on solving:
1. Mount time - by ripping out all the MTD code and
really optimizing the driver I've got the mount time
down to a minute. The time budget os 10s to cold-boot
the entire system. Uh-oh.
2. Taking advantage of the chips interleaving and
cached programming features. My throughput is about
1.5 MB/s but I really want to quadruple that or better
if possible.
What is the current thinking on these two issues?
Regards,
Kent Ryhorchuk.
__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com