The Ext2 inode-table maps nicely to the page cache with the current
page cache interface because it has uniform sized chunks that are accessed one
at a time, and likewise for the group descriptor cache.
*But* the directory code in page cache with your current approach does not
work out nicely with my directory index - it doesn't have page-sized chunks
and access to them is overlapped. This isn't an isolated problem, it's a
problem we've managed to avoid dealing with so far because much of the file
code we have does satisfy your two pre-conditions:
- Data groups naturally into pages
- Access to data items is strictly serial
We need a way of accessing the page cache that doesn't rely on either of
those two assumptions.
-- Daniel - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/