Merge branch 'for-linus' of git://git.kernel.dk/linux-block

Pull block fixes from Jens Axboe:

 - Two sets of NVMe pull requests from Christoph:
      - Fixes for the Fibre Channel host/target to fix spec compliance
      - Allow a zero keep alive timeout
      - Make the debug printk for broken SGLs work better
      - Fix queue zeroing during initialization
      - Set of RDMA and FC fixes
      - Target div-by-zero fix

 - bsg double-free fix.

 - ndb unknown ioctl fix from Josef.

 - Buffered vs O_DIRECT page cache inconsistency fix. Has been floating
   around for a long time, well reviewed. From Lukas.

 - brd overflow fix from Mikulas.

 - Fix for a loop regression in this merge window, where using a union
   for two members of the loop_cmd turned out to be a really bad idea.
   From Omar.

 - Fix for an iostat regression fix in this series, using the wrong API
   to get at the block queue. From Shaohua.

 - Fix for a potential blktrace delection deadlock. From Waiman.

* 'for-linus' of git://git.kernel.dk/linux-block: (30 commits)
  nvme-fcloop: fix port deletes and callbacks
  nvmet-fc: sync header templates with comments
  nvmet-fc: ensure target queue id within range.
  nvmet-fc: on port remove call put outside lock
  nvme-rdma: don't fully stop the controller in error recovery
  nvme-rdma: give up reconnect if state change fails
  nvme-core: Use nvme_wq to queue async events and fw activation
  nvme: fix sqhd reference when admin queue connect fails
  block: fix a crash caused by wrong API
  fs: Fix page cache inconsistency when mixing buffered and AIO DIO
  nvmet: implement valid sqhd values in completions
  nvme-fabrics: Allow 0 as KATO value
  nvme: allow timed-out ios to retry
  nvme: stop aer posting if controller state not live
  nvme-pci: Print invalid SGL only once
  nvme-pci: initialize queue memory before interrupts
  nvmet-fc: fix failing max io queue connections
  nvme-fc: use transport-specific sgl format
  nvme: add transport SGL definitions
  nvme.h: remove FC transport-specific error values
  ...

2 files changed, 59 insertions, 19 deletions

diff --git a/fs/direct-io.c b/fs/direct-io.c
index 5fa2211e49ae..62cf812ed0e5 100644
--- a/fs/direct-io.c
+++ b/fs/direct-io.c
@@ -229,6 +229,7 @@ static ssize_t dio_complete(struct dio *dio, ssize_t ret, bool is_async)
 {
 	loff_t offset = dio->iocb->ki_pos;
 	ssize_t transferred = 0;
+	int err;
 
 	/*
 	 * AIO submission can race with bio completion to get here while
@@ -258,8 +259,22 @@ static ssize_t dio_complete(struct dio *dio, ssize_t ret, bool is_async)
 	if (ret == 0)
 		ret = transferred;
 
+	/*
+	 * Try again to invalidate clean pages which might have been cached by
+	 * non-direct readahead, or faulted in by get_user_pages() if the source
+	 * of the write was an mmap'ed region of the file we're writing.  Either
+	 * one is a pretty crazy thing to do, so we don't support it 100%.  If
+	 * this invalidation fails, tough, the write still worked...
+	 */
+	if (ret > 0 && dio->op == REQ_OP_WRITE &&
+	    dio->inode->i_mapping->nrpages) {
+		err = invalidate_inode_pages2_range(dio->inode->i_mapping,
+					offset >> PAGE_SHIFT,
+					(offset + ret - 1) >> PAGE_SHIFT);
+		WARN_ON_ONCE(err);
+	}
+
 	if (dio->end_io) {
-		int err;
 
 		// XXX: ki_pos??
 		err = dio->end_io(dio->iocb, offset, ret, dio->private);
@@ -304,6 +319,7 @@ static void dio_bio_end_aio(struct bio *bio)
 	struct dio *dio = bio->bi_private;
 	unsigned long remaining;
 	unsigned long flags;
+	bool defer_completion = false;
 
 	/* cleanup the bio */
 	dio_bio_complete(dio, bio);
@@ -315,7 +331,19 @@ static void dio_bio_end_aio(struct bio *bio)
 	spin_unlock_irqrestore(&dio->bio_lock, flags);
 
 	if (remaining == 0) {
-		if (dio->result && dio->defer_completion) {
+		/*
+		 * Defer completion when defer_completion is set or
+		 * when the inode has pages mapped and this is AIO write.
+		 * We need to invalidate those pages because there is a
+		 * chance they contain stale data in the case buffered IO
+		 * went in between AIO submission and completion into the
+		 * same region.
+		 */
+		if (dio->result)
+			defer_completion = dio->defer_completion ||
+					   (dio->op == REQ_OP_WRITE &&
+					    dio->inode->i_mapping->nrpages);
+		if (defer_completion) {
 			INIT_WORK(&dio->complete_work, dio_aio_complete_work);
 			queue_work(dio->inode->i_sb->s_dio_done_wq,
 				   &dio->complete_work);
@@ -1210,10 +1238,19 @@ do_blockdev_direct_IO(struct kiocb *iocb, struct inode *inode,
 	 * For AIO O_(D)SYNC writes we need to defer completions to a workqueue
 	 * so that we can call ->fsync.
 	 */
-	if (dio->is_async && iov_iter_rw(iter) == WRITE &&
-	    ((iocb->ki_filp->f_flags & O_DSYNC) ||
-	     IS_SYNC(iocb->ki_filp->f_mapping->host))) {
-		retval = dio_set_defer_completion(dio);
+	if (dio->is_async && iov_iter_rw(iter) == WRITE) {
+		retval = 0;
+		if ((iocb->ki_filp->f_flags & O_DSYNC) ||
+		    IS_SYNC(iocb->ki_filp->f_mapping->host))
+			retval = dio_set_defer_completion(dio);
+		else if (!dio->inode->i_sb->s_dio_done_wq) {
+			/*
+			 * In case of AIO write racing with buffered read we
+			 * need to defer completion. We can't decide this now,
+			 * however the workqueue needs to be initialized here.
+			 */
+			retval = sb_init_dio_done_wq(dio->inode->i_sb);
+		}
 		if (retval) {
 			/*
 			 * We grab i_mutex only for reads so we don't have
diff --git a/fs/iomap.c b/fs/iomap.c
index 269b24a01f32..8194d30bdca0 100644
--- a/fs/iomap.c
+++ b/fs/iomap.c
@@ -713,8 +713,24 @@ struct iomap_dio {
 static ssize_t iomap_dio_complete(struct iomap_dio *dio)
 {
 	struct kiocb *iocb = dio->iocb;
+	struct inode *inode = file_inode(iocb->ki_filp);
 	ssize_t ret;
 
+	/*
+	 * Try again to invalidate clean pages which might have been cached by
+	 * non-direct readahead, or faulted in by get_user_pages() if the source
+	 * of the write was an mmap'ed region of the file we're writing.  Either
+	 * one is a pretty crazy thing to do, so we don't support it 100%.  If
+	 * this invalidation fails, tough, the write still worked...
+	 */
+	if (!dio->error &&
+	    (dio->flags & IOMAP_DIO_WRITE) && inode->i_mapping->nrpages) {
+		ret = invalidate_inode_pages2_range(inode->i_mapping,
+				iocb->ki_pos >> PAGE_SHIFT,
+				(iocb->ki_pos + dio->size - 1) >> PAGE_SHIFT);
+		WARN_ON_ONCE(ret);
+	}
+
 	if (dio->end_io) {
 		ret = dio->end_io(iocb,
 				dio->error ? dio->error : dio->size,
@@ -1042,19 +1058,6 @@ iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
 
 	ret = iomap_dio_complete(dio);
 
-	/*
-	 * Try again to invalidate clean pages which might have been cached by
-	 * non-direct readahead, or faulted in by get_user_pages() if the source
-	 * of the write was an mmap'ed region of the file we're writing.  Either
-	 * one is a pretty crazy thing to do, so we don't support it 100%.  If
-	 * this invalidation fails, tough, the write still worked...
-	 */
-	if (iov_iter_rw(iter) == WRITE) {
-		int err = invalidate_inode_pages2_range(mapping,
-				start >> PAGE_SHIFT, end >> PAGE_SHIFT);
-		WARN_ON_ONCE(err);
-	}
-
 	return ret;
 
 out_free_dio: