On Mon, Jan 26, 2004 at 07:21:58PM +0100, Ronny Buchmann wrote:
> The file system may still do smart stuff like merging requests.
It's usually used for reading from block devices (disks).
Issue a few reads and the kernel will begin readahead, it will also merge
the read and readahead into one I/O at times.
> with floppies because the 'real' blocksize of a hard
disk is now typically
Does the kernel do read-ahead in the block layer or in vfs?
Some of the file systems (especially xfs) do quite smart readahead tricks,
others use the basic facilities of the block layer and others still like
jffs and nfs don't use the block layer but their own ways of handling