Re: tosfs, faster hash function...

[ersmith@netcom.com (Eric R. Smith)]

>Eric, the current implementation is NOT safer. Inodes have to be different
>for different files, BUT they have to be identical for identical files,
>too. The current implementation satisfies the first, but ignores the
>second point.

I disagree. The possible consequences of incorrectly identifying two
different files to be the same are much worse, I think, than any
problems that could arise from incorrectly identifying the same
file name with two different inodes. In the first case, programs such
as "cmp" or "diff" could (for example) report the files as identical
when they are not, or "tar" might not include both files. I can't
really think of a bad consequence of identifying one file with 2 different
inodes, unless you're explicitly using inodes ("find -inode") for
some reason; knowledgeable TOSFS users shouldn't make this mistake. :-)

I agree that the best solution would be to use the starting cluster
somehow. That's unlikely to happen in the near future :-(.

Regards,
Eric