This patch implements O_DIRECT about 80% of the way. It takes a patchset

Tor created a while ago, removes the raw I/O piece (that has cache coherency
problems), and adds a buffer cache / VM freeing piece.

Essentially this patch causes O_DIRECT I/O to not be left in the cache, but
does not prevent it from going through the cache, hence the 80%.  For
the last 20% we need a method by which the I/O can be issued directly to
buffer supplied by the user process and bypass the buffer cache entirely,
but still maintain cache coherency.

I also have the code working under -stable but the changes made to sys/file.h
may not be MFCable, so an MFC is not on the table yet.

Submitted by:	tegge, dillon

1 files changed, 9 insertions, 0 deletions

diff --git a/sys/kern/vfs_cluster.c b/sys/kern/vfs_cluster.c
index 0eb47bd..c9c09cb 100644
--- a/sys/kern/vfs_cluster.c
+++ b/sys/kern/vfs_cluster.c
@@ -505,6 +505,15 @@ cluster_callback(bp)
 			tbp->b_dirtyoff = tbp->b_dirtyend = 0;
 			tbp->b_flags &= ~B_INVAL;
 			tbp->b_ioflags &= ~BIO_ERROR;
+			/*
+			 * XXX the bdwrite()/bqrelse() issued during
+			 * cluster building clears B_RELBUF (see bqrelse()
+			 * comment).  If direct I/O was specified, we have
+			 * to restore it here to allow the buffer and VM
+			 * to be freed.
+			 */
+			if (tbp->b_flags & B_DIRECT)
+				tbp->b_flags |= B_RELBUF;
 		}
 		bufdone(tbp);
 	}