Re: PATCH: Raw device IO for 2.1.131

David S. Miller (davem@dm.cobaltmicro.com)
Mon, 14 Dec 1998 10:00:32 -0800


Date: Mon, 14 Dec 1998 18:56:07 +0100
From: Pavel Machek <pavel@atrey.karlin.mff.cuni.cz>

> For receive and writes to a disk, sendfile() is not the answer.

Why not? I see current sendfile implementation can not do it, but what
is wrong with this? I think that we should extent sendfile to do such
things in future... (Is there any problem with it?)

You are right. I meant "now, with the current architecture" it makes
no sense to use sendfile for when the disk file is the destination.

It makes sense when full blown writes directly into the page cache are
supported across the whole kernel (they sort of are right now
minimally for the sake of nfs).

Later,
David S. Miller
davem@dm.cobaltmicro.com

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/