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

Re: Generalized Flash question

>> Somebody told me a week ago that flash only works well if you
>> write to it very infrequently because it, well breaks.

> It breaks after X erase-cycles, where X is usually a very large
> number today (I don't have any datasheets here, but I'd
> guess X > 100,000).

X is between 1,000 and 1,000,000 - often 10,000 based on a
failure rate of less than 1%.

> JFFS does rudimentary wear-levelling because of the log-structured
> design, we write new data at the end of the circular log instead of
> rewriting the block where the old data was. More advanced algorithms
> will come eventually..

I don't know much about JFFS. Is there a kind of allocation table in a
fixed block comparable to FAT based file systems? There were many
discussions about the flash in the Beck IPC@chip (a chip with processor,
ethernet, RAM and flash integrated). As far as I understood, the result
was that the OS of the IPC@Chip uses allocation table in a fixed block
which is updated twice during a write operation on any file. So the
allocation table will break after writing about 5,000 files, even if they
at different locations.

How does the JFFS handle this?


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