Re: linux-2.4.10-pre5

[Andrea Arcangeli <andrea@suse.de>]

On Sun, Sep 09, 2001 at 02:14:54PM +0100, Anton Altaparmakov wrote:
> Not a pitfall, but a question: what happens with the get_block() callback 
> passed to block_read_full_page() by the readpage() address space method of 

anything related to file data uses the "real" pagecache that have
nothing to do with this getblk pagecache backed logic, this is only a
trick to share the ram memory for the physically indexed disk cache and
not to run into the alising issues. I cannot see anything obviously
wrong with this logic so I believe it worth a try to implement but I
will do that after I finished other things. Also since Daniel may have
just finished implementing that he may just send me the buffer.c diff
that I can check and integrate into the blkdev-pagecache work in the
future.

Andrea