On Thu, Apr 19, 2001 at 04:29:52PM -0400, Jeff Garzik wrote:
> [email protected] wrote:
> > In the long run, it probably makes sense to adjust the algorithms to
> > allow for non-power-of-two inode sizes,
>
> If you don't mind, does that imply packing inodes across block
> boundaries?
No, it means that padding the end of each block in the inode table so
that inodes don't cross block boundries. (i.e., if the inode size is
150 bytes, then there's room for 6 inodes in a 1k block, with 124
bytes left over for padding.)
- Ted