[to-be-updated] fs-return-if-direct-write-will-trigger-writeback.patch removed from -mm tree

[to-be-updated] fs-return-if-direct-write-will-trigger-writeback.patch removed from -mm tree

[akpm]

The patch titled
     Subject: fs: return if direct write will trigger writeback
has been removed from the -mm tree.  Its filename was
     fs-return-if-direct-write-will-trigger-writeback.patch

This patch was dropped because an updated version will be merged

------------------------------------------------------
From: Goldwyn Rodrigues <rgoldwyn@suse.com>
Subject: fs: return if direct write will trigger writeback

Find out if the write will trigger a wait due to writeback. If yes,
return -EAGAIN.

Return -EINVAL for buffered AIO: there are multiple causes of delay such
as page locks, dirty throttling logic, page loading from disk etc.  which
cannot be taken care of.

Link: http://lkml.kernel.org/r/20170615160002.17233-6-rgoldwyn@suse.de
Signed-off-by: Goldwyn Rodrigues <rgoldwyn@suse.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Jan Kara <jack@suse.cz>
Cc: "Theodore Ts'o" <tytso@mit.edu>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Andreas Dilger <adilger.kernel@dilger.ca>
Cc: Chris Mason <clm@fb.com>
Cc: Darrick J. Wong <darrick.wong@oracle.com>
Cc: David Sterba <dsterba@suse.com>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Josef Bacik <jbacik@fb.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---

 mm/filemap.c |   17 ++++++++++++++---
 1 file changed, 14 insertions(+), 3 deletions(-)

diff -puN mm/filemap.c~fs-return-if-direct-write-will-trigger-writeback mm/filemap.c
--- a/mm/filemap.c~fs-return-if-direct-write-will-trigger-writeback
+++ a/mm/filemap.c
@@ -2674,6 +2674,9 @@ inline ssize_t generic_write_checks(stru
 
 	pos = iocb->ki_pos;
 
+	if ((iocb->ki_flags & IOCB_NOWAIT) && !(iocb->ki_flags & IOCB_DIRECT))
+		return -EINVAL;
+
 	if (limit != RLIM_INFINITY) {
 		if (iocb->ki_pos >= limit) {
 			send_sig(SIGXFSZ, current, 0);
@@ -2742,9 +2745,17 @@ generic_file_direct_write(struct kiocb *
 	write_len = iov_iter_count(from);
 	end = (pos + write_len - 1) >> PAGE_SHIFT;
 
-	written = filemap_write_and_wait_range(mapping, pos, pos + write_len - 1);
-	if (written)
-		goto out;
+	if (iocb->ki_flags & IOCB_NOWAIT) {
+		/* If there are pages to writeback, return */
+		if (filemap_range_has_page(inode->i_mapping, pos,
+					   pos + iov_iter_count(from)))
+			return -EAGAIN;
+	} else {
+		written = filemap_write_and_wait_range(mapping, pos,
+							pos + write_len - 1);
+		if (written)
+			goto out;
+	}
 
 	/*
 	 * After a write we want buffered reads to be sure to go to disk to get
_

Patches currently in -mm which might be from rgoldwyn@suse.com are

fs-introduce-iomap_nowait.patch
block-return-on-congested-block-device.patch
ext4-nowait-aio-support.patch
xfs-nowait-aio-support.patch
btrfs-nowait-aio-support.patch