Re: Better Linux Kernel Patch for 4GB Inode Problem

[Paul Serice <paul@serice.net>]

> Why is it soo hard to understand that you cannot divide by 64 while
> the smallest directory entry is 34 bytes and most real life entries
> are still smaller than 64 bytes?

I'm fairly certain I'm dividing by 32.  The second patch has the
following:

    block << 6

which is the same as

    block << (11 - 5)

which (assuming arbitrary precision) is the same as

    (block << 11) >> 5

which is the byte offset divided by 32.  So, the math was correct
from the start, right?



> Code that creates st_ino == 0 or that gives away non-unique inode
> numbers is inaceptable.

If the math above is o.k., I'll be happy to go back to the approach
in the second patch.