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

Re: Major JFFS2 bug (?)



David Woodhouse wrote:

> I'm confused. This file was being written when power was lost. Why do you
> expect it to be the whole 448 bytes long? Were you overwriting it and it
> was already 448 bytes long, or had you completed the write and written
> something to a separate index file, or what?
>

Because, the entire write was done in a single "write()".
If the first int got written that told me "x" ints follow, then those
_should_ follow too.
The the new write() did not get complete, then the older file should have
been there as it, with all its data- including the first int- now correctly
telling my program how many ints to read- from the old file.

In other words the write() to the JFFS2 fs is not power fail
"roll_back_and_recover_to_older_data_if_newer_data_did_not_get_completely_written"
safe.

Vipin



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