Merge tag 'iomap-5.9-merge-5' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux

Pull iomap updates from Darrick Wong:
 "The most notable changes are:

   - iomap no longer invalidates the page cache when performing a direct
     read, since doing so is unnecessary and the old directio code
     doesn't do that either.

   - iomap embraced the use of returning ENOTBLK from a direct write to
     trigger falling back to a buffered write since ext4 already did
     this and btrfs wants it for their port.

   - iomap falls back to buffered writes if we're doing a direct write
     and the page cache invalidation after the flush fails; this was
     necessary to handle a corner case in the btrfs port.

   - Remove email virus scanner detritus that was accidentally included
     in yesterday's pull request. Clearly I need(ed) to update my git
     branch checker scripts. :("

* tag 'iomap-5.9-merge-5' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux:
  iomap: fall back to buffered writes for invalidation failures
  xfs: use ENOTBLK for direct I/O to buffered I/O fallback
  iomap: Only invalidate page cache pages on direct IO writes
  iomap: Make sure iomap_end is called after iomap_begin

7 files changed, 44 insertions, 27 deletions

diff --git a/fs/ext4/file.c b/fs/ext4/file.c
index 2a01e31a032c..129cc1dd6b79 100644
--- a/fs/ext4/file.c
+++ b/fs/ext4/file.c
@@ -544,6 +544,8 @@ static ssize_t ext4_dio_write_iter(struct kiocb *iocb, struct iov_iter *from)
 		iomap_ops = &ext4_iomap_overwrite_ops;
 	ret = iomap_dio_rw(iocb, from, iomap_ops, &ext4_dio_write_ops,
 			   is_sync_kiocb(iocb) || unaligned_io || extend);
+	if (ret == -ENOTBLK)
+		ret = 0;
 
 	if (extend)
 		ret = ext4_handle_inode_extension(inode, offset, ret, count);
diff --git a/fs/gfs2/file.c b/fs/gfs2/file.c
index bebde537ac8c..b085a3bea4f0 100644
--- a/fs/gfs2/file.c
+++ b/fs/gfs2/file.c
@@ -835,7 +835,8 @@ static ssize_t gfs2_file_direct_write(struct kiocb *iocb, struct iov_iter *from)
 
 	ret = iomap_dio_rw(iocb, from, &gfs2_iomap_ops, NULL,
 			   is_sync_kiocb(iocb));
-
+	if (ret == -ENOTBLK)
+		ret = 0;
 out:
 	gfs2_glock_dq(&gh);
 out_uninit:
diff --git a/fs/iomap/apply.c b/fs/iomap/apply.c
index 76925b40b5fd..26ab6563181f 100644
--- a/fs/iomap/apply.c
+++ b/fs/iomap/apply.c
@@ -46,10 +46,14 @@ iomap_apply(struct inode *inode, loff_t pos, loff_t length, unsigned flags,
 	ret = ops->iomap_begin(inode, pos, length, flags, &iomap, &srcmap);
 	if (ret)
 		return ret;
-	if (WARN_ON(iomap.offset > pos))
-		return -EIO;
-	if (WARN_ON(iomap.length == 0))
-		return -EIO;
+	if (WARN_ON(iomap.offset > pos)) {
+		written = -EIO;
+		goto out;
+	}
+	if (WARN_ON(iomap.length == 0)) {
+		written = -EIO;
+		goto out;
+	}
 
 	trace_iomap_apply_dstmap(inode, &iomap);
 	if (srcmap.type != IOMAP_HOLE)
@@ -80,6 +84,7 @@ iomap_apply(struct inode *inode, loff_t pos, loff_t length, unsigned flags,
 	written = actor(inode, pos, length, data, &iomap,
 			srcmap.type != IOMAP_HOLE ? &srcmap : &iomap);
 
+out:
 	/*
 	 * Now the data has been copied, commit the range we've copied.  This
 	 * should not fail unless the filesystem has had a fatal error.
diff --git a/fs/iomap/direct-io.c b/fs/iomap/direct-io.c
index ec7b78e6feca..c1aafb2ab990 100644
--- a/fs/iomap/direct-io.c
+++ b/fs/iomap/direct-io.c
@@ -10,6 +10,7 @@
 #include <linux/backing-dev.h>
 #include <linux/uio.h>
 #include <linux/task_io_accounting_ops.h>
+#include "trace.h"
 
 #include "../internal.h"
 
@@ -401,6 +402,9 @@ iomap_dio_actor(struct inode *inode, loff_t pos, loff_t length,
  * can be mapped into multiple disjoint IOs and only a subset of the IOs issued
  * may be pure data writes. In that case, we still need to do a full data sync
  * completion.
+ *
+ * Returns -ENOTBLK In case of a page invalidation invalidation failure for
+ * writes.  The callers needs to fall back to buffered I/O in this case.
  */
 ssize_t
 iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
@@ -475,23 +479,24 @@ iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
 	if (ret)
 		goto out_free_dio;
 
-	/*
-	 * Try to invalidate cache pages for the range we're direct
-	 * writing.  If this invalidation fails, tough, the write will
-	 * still work, but racing two incompatible write paths is a
-	 * pretty crazy thing to do, so we don't support it 100%.
-	 */
-	ret = invalidate_inode_pages2_range(mapping,
-			pos >> PAGE_SHIFT, end >> PAGE_SHIFT);
-	if (ret)
-		dio_warn_stale_pagecache(iocb->ki_filp);
-	ret = 0;
-
-	if (iov_iter_rw(iter) == WRITE && !wait_for_completion &&
-	    !inode->i_sb->s_dio_done_wq) {
-		ret = sb_init_dio_done_wq(inode->i_sb);
-		if (ret < 0)
+	if (iov_iter_rw(iter) == WRITE) {
+		/*
+		 * Try to invalidate cache pages for the range we are writing.
+		 * If this invalidation fails, let the caller fall back to
+		 * buffered I/O.
+		 */
+		if (invalidate_inode_pages2_range(mapping, pos >> PAGE_SHIFT,
+				end >> PAGE_SHIFT)) {
+			trace_iomap_dio_invalidate_fail(inode, pos, count);
+			ret = -ENOTBLK;
 			goto out_free_dio;
+		}
+
+		if (!wait_for_completion && !inode->i_sb->s_dio_done_wq) {
+			ret = sb_init_dio_done_wq(inode->i_sb);
+			if (ret < 0)
+				goto out_free_dio;
+		}
 	}
 
 	inode_dio_begin(inode);
diff --git a/fs/iomap/trace.h b/fs/iomap/trace.h
index 5693a39d52fb..fdc7ae388476 100644
--- a/fs/iomap/trace.h
+++ b/fs/iomap/trace.h
@@ -74,6 +74,7 @@ DEFINE_EVENT(iomap_range_class, name,	\
 DEFINE_RANGE_EVENT(iomap_writepage);
 DEFINE_RANGE_EVENT(iomap_releasepage);
 DEFINE_RANGE_EVENT(iomap_invalidatepage);
+DEFINE_RANGE_EVENT(iomap_dio_invalidate_fail);
 
 #define IOMAP_TYPE_STRINGS \
 	{ IOMAP_HOLE,		"HOLE" }, \
diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c
index fdbff4860d61..d538411c3791 100644
--- a/fs/xfs/xfs_file.c
+++ b/fs/xfs/xfs_file.c
@@ -505,7 +505,7 @@ xfs_file_dio_aio_write(
 		 */
 		if (xfs_is_cow_inode(ip)) {
 			trace_xfs_reflink_bounce_dio_write(ip, iocb->ki_pos, count);
-			return -EREMCHG;
+			return -ENOTBLK;
 		}
 		iolock = XFS_IOLOCK_EXCL;
 	} else {
@@ -553,8 +553,8 @@ out:
 	xfs_iunlock(ip, iolock);
 
 	/*
-	 * No fallback to buffered IO on errors for XFS, direct IO will either
-	 * complete fully or fail.
+	 * No fallback to buffered IO after short writes for XFS, direct I/O
+	 * will either complete fully or return an error.
 	 */
 	ASSERT(ret < 0 || ret == count);
 	return ret;
@@ -714,7 +714,7 @@ xfs_file_write_iter(
 		 * allow an operation to fall back to buffered mode.
 		 */
 		ret = xfs_file_dio_aio_write(iocb, from);
-		if (ret != -EREMCHG)
+		if (ret != -ENOTBLK)
 			return ret;
 	}
 
diff --git a/fs/zonefs/super.c b/fs/zonefs/super.c
index abfb17f88f9a..4a0bff65d3d6 100644
--- a/fs/zonefs/super.c
+++ b/fs/zonefs/super.c
@@ -786,8 +786,11 @@ static ssize_t zonefs_file_write_iter(struct kiocb *iocb, struct iov_iter *from)
 	if (iocb->ki_pos >= ZONEFS_I(inode)->i_max_size)
 		return -EFBIG;
 
-	if (iocb->ki_flags & IOCB_DIRECT)
-		return zonefs_file_dio_write(iocb, from);
+	if (iocb->ki_flags & IOCB_DIRECT) {
+		ssize_t ret = zonefs_file_dio_write(iocb, from);
+		if (ret != -ENOTBLK)
+			return ret;
+	}
 
 	return zonefs_file_buffered_write(iocb, from);
 }