[Patch] only unmap what intersects a direct_IO op

[Zach Brown]

Now that we're only invalidating the pages that intersected a direct IO write
we might as well only unmap the intersecting bytes as well.  This passed a
light fsx load with page cache, direct, and mmap IO.

Signed-off-by: Zach Brown <zach.brown@xxxxxxxxxx>

---

 filemap.c |   12 ++++++++----
 1 files changed, 8 insertions(+), 4 deletions(-)

Index: 2.6-bk-odirinv/mm/filemap.c
===================================================================
--- 2.6-bk-odirinv.orig/mm/filemap.c	2005-02-07 12:42:50.000000000 -0800
+++ 2.6-bk-odirinv/mm/filemap.c	2005-02-07 12:43:16.244253441 -0800
@@ -2285,22 +2285,26 @@
 	struct file *file = iocb->ki_filp;
 	struct address_space *mapping = file->f_mapping;
 	ssize_t retval;
+	size_t write_len = 0;
 
 	/*
 	 * If it's a write, unmap all mmappings of the file up-front.  This
 	 * will cause any pte dirty bits to be propagated into the pageframes
 	 * for the subsequent filemap_write_and_wait().
 	 */
-	if (rw == WRITE && mapping_mapped(mapping))
-		unmap_mapping_range(mapping, 0, -1, 0);
+	if (rw == WRITE) {
+		write_len = iov_length(iov, nr_segs);
+	       	if (mapping_mapped(mapping))
+			unmap_mapping_range(mapping, offset, write_len, 0);
+	}
 
 	retval = filemap_write_and_wait(mapping);
 	if (retval == 0) {
 		retval = mapping->a_ops->direct_IO(rw, iocb, iov,
 						offset, nr_segs);
 		if (rw == WRITE && mapping->nrpages) {
-			pgoff_t end = (offset + iov_length(iov, nr_segs) - 1)
-				      >> PAGE_CACHE_SHIFT;
+			pgoff_t end = (offset + write_len - 1)
+						>> PAGE_CACHE_SHIFT;
 			int err = invalidate_inode_pages2_range(mapping,
 					offset >> PAGE_CACHE_SHIFT, end);
 			if (err)
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/