Merge tag 'block-5.17-2022-02-04' of git://git.kernel.dk/linux-block

Pull block fixes from Jens Axboe:

 - NVMe pull request
    - fix use-after-free in rdma and tcp controller reset (Sagi Grimberg)
    - fix the state check in nvmf_ctlr_matches_baseopts (Uday Shankar)

 - MD nowait null pointer fix (Song)

 - blk-integrity seed advance fix (Martin)

 - Fix a dio regression in this merge window (Ilya)

* tag 'block-5.17-2022-02-04' of git://git.kernel.dk/linux-block:
  block: bio-integrity: Advance seed correctly for larger interval sizes
  nvme-fabrics: fix state check in nvmf_ctlr_matches_baseopts()
  md: fix NULL pointer deref with nowait but no mddev->queue
  block: fix DIO handling regressions in blkdev_read_iter()
  nvme-rdma: fix possible use-after-free in transport error_recovery work
  nvme-tcp: fix possible use-after-free in transport error_recovery work
  nvme: fix a possible use-after-free in controller reset during load

block/bio-integrity.c

@@ -373,7 +373,7 @@ void bio_integrity_advance(struct bio *bio, unsigned int bytes_done)
 	struct blk_integrity *bi = blk_get_integrity(bio->bi_bdev->bd_disk);
 	unsigned bytes = bio_integrity_bytes(bi, bytes_done >> 9);
 
-	bip->bip_iter.bi_sector += bytes_done >> 9;
+	bip->bip_iter.bi_sector += bio_integrity_intervals(bi, bytes_done >> 9);
 	bvec_iter_advance(bip->bip_vec, &bip->bip_iter, bytes);
 }
 

block/fops.c

@@ -566,34 +566,37 @@ static ssize_t blkdev_read_iter(struct kiocb *iocb, struct iov_iter *to)
 {
 	struct block_device *bdev = iocb->ki_filp->private_data;
 	loff_t size = bdev_nr_bytes(bdev);
-	size_t count = iov_iter_count(to);
 	loff_t pos = iocb->ki_pos;
 	size_t shorted = 0;
 	ssize_t ret = 0;
+	size_t count;
 
-	if (unlikely(pos + count > size)) {
+	if (unlikely(pos + iov_iter_count(to) > size)) {
 		if (pos >= size)
 			return 0;
 		size -= pos;
-		if (count > size) {
-			shorted = count - size;
+		shorted = iov_iter_count(to) - size;
 		iov_iter_truncate(to, size);
 	}
-	}
+
+	count = iov_iter_count(to);
+	if (!count)
+		goto reexpand; /* skip atime */
 
 	if (iocb->ki_flags & IOCB_DIRECT) {
 		struct address_space *mapping = iocb->ki_filp->f_mapping;
 
 		if (iocb->ki_flags & IOCB_NOWAIT) {
-			if (filemap_range_needs_writeback(mapping, iocb->ki_pos,
-						iocb->ki_pos + count - 1))
-				return -EAGAIN;
+			if (filemap_range_needs_writeback(mapping, pos,
+							  pos + count - 1)) {
+				ret = -EAGAIN;
+				goto reexpand;
+			}
 		} else {
-			ret = filemap_write_and_wait_range(mapping,
-						iocb->ki_pos,
-					        iocb->ki_pos + count - 1);
+			ret = filemap_write_and_wait_range(mapping, pos,
+							   pos + count - 1);
 			if (ret < 0)
-				return ret;
+				goto reexpand;
 		}
 
 		file_accessed(iocb->ki_filp);
@@ -603,12 +606,14 @@ static ssize_t blkdev_read_iter(struct kiocb *iocb, struct iov_iter *to)
 			iocb->ki_pos += ret;
 			count -= ret;
 		}
+		iov_iter_revert(to, count - iov_iter_count(to));
 		if (ret < 0 || !count)
-			return ret;
+			goto reexpand;
 	}
 
 	ret = filemap_read(iocb, to, ret);
 
+reexpand:
 	if (unlikely(shorted))
 		iov_iter_reexpand(to, iov_iter_count(to) + shorted);
 	return ret;

drivers/md/md.c

@@ -5869,10 +5869,6 @@ int md_run(struct mddev *mddev)
 		nowait = nowait && blk_queue_nowait(bdev_get_queue(rdev->bdev));
 	}
 
-	/* Set the NOWAIT flags if all underlying devices support it */
-	if (nowait)
-		blk_queue_flag_set(QUEUE_FLAG_NOWAIT, mddev->queue);
-
 	if (!bioset_initialized(&mddev->bio_set)) {
 		err = bioset_init(&mddev->bio_set, BIO_POOL_SIZE, 0, BIOSET_NEED_BVECS);
 		if (err)
@@ -6010,6 +6006,10 @@ int md_run(struct mddev *mddev)
 		else
 			blk_queue_flag_clear(QUEUE_FLAG_NONROT, mddev->queue);
 		blk_queue_flag_set(QUEUE_FLAG_IO_STAT, mddev->queue);
+
+		/* Set the NOWAIT flags if all underlying devices support it */
+		if (nowait)
+			blk_queue_flag_set(QUEUE_FLAG_NOWAIT, mddev->queue);
 	}
 	if (pers->sync_request) {
 		if (mddev->kobj.sd &&

drivers/nvme/host/core.c

@@ -4253,6 +4253,13 @@ static void nvme_async_event_work(struct work_struct *work)
 		container_of(work, struct nvme_ctrl, async_event_work);
 
 	nvme_aen_uevent(ctrl);
+
+	/*
+	 * The transport drivers must guarantee AER submission here is safe by
+	 * flushing ctrl async_event_work after changing the controller state
+	 * from LIVE and before freeing the admin queue.
+	*/
+	if (ctrl->state == NVME_CTRL_LIVE)
 		ctrl->ops->submit_async_event(ctrl);
 }
 

drivers/nvme/host/fabrics.h

@@ -170,6 +170,7 @@ nvmf_ctlr_matches_baseopts(struct nvme_ctrl *ctrl,
 			struct nvmf_ctrl_options *opts)
 {
 	if (ctrl->state == NVME_CTRL_DELETING ||
+	    ctrl->state == NVME_CTRL_DELETING_NOIO ||
 	    ctrl->state == NVME_CTRL_DEAD ||
 	    strcmp(opts->subsysnqn, ctrl->opts->subsysnqn) ||
 	    strcmp(opts->host->nqn, ctrl->opts->host->nqn) ||

drivers/nvme/host/rdma.c

@@ -1200,6 +1200,7 @@ static void nvme_rdma_error_recovery_work(struct work_struct *work)
 			struct nvme_rdma_ctrl, err_work);
 
 	nvme_stop_keep_alive(&ctrl->ctrl);
+	flush_work(&ctrl->ctrl.async_event_work);
 	nvme_rdma_teardown_io_queues(ctrl, false);
 	nvme_start_queues(&ctrl->ctrl);
 	nvme_rdma_teardown_admin_queue(ctrl, false);

drivers/nvme/host/tcp.c

@@ -2096,6 +2096,7 @@ static void nvme_tcp_error_recovery_work(struct work_struct *work)
 	struct nvme_ctrl *ctrl = &tcp_ctrl->ctrl;
 
 	nvme_stop_keep_alive(ctrl);
+	flush_work(&ctrl->async_event_work);
 	nvme_tcp_teardown_io_queues(ctrl, false);
 	/* unquiesce to fail fast pending requests */
 	nvme_start_queues(ctrl);