misc bio allocation cleanups

@@ -678,7 +678,7 @@ static ssize_t zonefs_file_dio_append(struct kiocb *iocb, struct iov_iter *from)
 	if (!nr_pages)
 		return 0;
 
-	bio = bio_alloc_bioset(GFP_NOFS, nr_pages, &fs_bio_set);
+	bio = bio_alloc(GFP_NOFS, nr_pages);
 	if (!bio)
 		return -ENOMEM;
 

@@ -421,7 +421,7 @@ static struct btrfs_device *__alloc_device(struct btrfs_fs_info *fs_info)
 	 * Preallocate a bio that's always going to be used for flushing device
 	 * barriers and matches the device lifespan
 	 */
-	dev->flush_bio = bio_alloc_bioset(GFP_KERNEL, 0, NULL);
+	dev->flush_bio = bio_kmalloc(GFP_KERNEL, 0);
 	if (!dev->flush_bio) {
 		kfree(dev);
 		return ERR_PTR(-ENOMEM);

@@ -432,23 +432,18 @@ void blk_insert_flush(struct request *rq)
 /**
  * blkdev_issue_flush - queue a flush
  * @bdev:	blockdev to issue flush for
- * @gfp_mask:	memory allocation flags (for bio_alloc)
  *
  * Description:
  *    Issue a flush for the block device in question.
  */
-int blkdev_issue_flush(struct block_device *bdev, gfp_t gfp_mask)
+int blkdev_issue_flush(struct block_device *bdev)
 {
-	struct bio *bio;
-	int ret = 0;
+	struct bio bio;
 
-	bio = bio_alloc(gfp_mask, 0);
-	bio_set_dev(bio, bdev);
-	bio->bi_opf = REQ_OP_WRITE | REQ_PREFLUSH;
-
-	ret = submit_bio_wait(bio);
-	bio_put(bio);
-	return ret;
+	bio_init(&bio, NULL, 0);
+	bio_set_dev(&bio, bdev);
+	bio.bi_opf = REQ_OP_WRITE | REQ_PREFLUSH;
+	return submit_bio_wait(&bio);
 }
 EXPORT_SYMBOL(blkdev_issue_flush);
 

@@ -819,7 +819,7 @@ static int dmz_write_sb(struct dmz_metadata *zmd, unsigned int set)
 	ret = dmz_rdwr_block(dev, REQ_OP_WRITE, zmd->sb[set].block,
 			     mblk->page);
 	if (ret == 0)
-		ret = blkdev_issue_flush(dev->bdev, GFP_NOIO);
+		ret = blkdev_issue_flush(dev->bdev);
 
 	return ret;
 }

@@ -862,7 +862,7 @@ static int dmz_write_dirty_mblocks(struct dmz_metadata *zmd,
 
 	/* Flush drive cache (this will also sync data) */
 	if (ret == 0)
-		ret = blkdev_issue_flush(dev->bdev, GFP_NOIO);
+		ret = blkdev_issue_flush(dev->bdev);
 
 	return ret;
 }

@@ -933,7 +933,7 @@ int dmz_flush_metadata(struct dmz_metadata *zmd)
 
 	/* If there are no dirty metadata blocks, just flush the device cache */
 	if (list_empty(&write_list)) {
-		ret = blkdev_issue_flush(dev->bdev, GFP_NOIO);
+		ret = blkdev_issue_flush(dev->bdev);
 		goto err;
 	}
 

@@ -1037,7 +1037,7 @@ static int ppl_recover(struct ppl_log *log, struct ppl_header *pplhdr,
 	}
 
 	/* flush the disk cache after recovery if necessary */
-	ret = blkdev_issue_flush(rdev->bdev, GFP_KERNEL);
+	ret = blkdev_issue_flush(rdev->bdev);
 out:
 	__free_page(page);
 	return ret;

@@ -333,7 +333,7 @@ static void nvmet_bdev_execute_flush(struct nvmet_req *req)
 
 u16 nvmet_bdev_flush(struct nvmet_req *req)
 {
-	if (blkdev_issue_flush(req->ns->bdev, GFP_KERNEL))
+	if (blkdev_issue_flush(req->ns->bdev))
 		return NVME_SC_INTERNAL | NVME_SC_DNR;
 	return 0;
 }

@@ -681,7 +681,7 @@ int blkdev_fsync(struct file *filp, loff_t start, loff_t end, int datasync)
 	 * i_mutex and doing so causes performance issues with concurrent
 	 * O_SYNC writers to a block device.
 	 */
-	error = blkdev_issue_flush(bdev, GFP_KERNEL);
+	error = blkdev_issue_flush(bdev);
 	if (error == -EOPNOTSUPP)
 		error = 0;
 

@@ -361,7 +361,7 @@ int exfat_file_fsync(struct file *filp, loff_t start, loff_t end, int datasync)
 	if (err)
 		return err;
 
-	return blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL);
+	return blkdev_issue_flush(inode->i_sb->s_bdev);
 }
 
 const struct file_operations exfat_file_operations = {

@@ -1076,7 +1076,7 @@ static int ext4_fc_perform_commit(journal_t *journal)
 	 * flush before we start writing fast commit blocks.
 	 */
 	if (journal->j_fs_dev != journal->j_dev)
-		blkdev_issue_flush(journal->j_fs_dev, GFP_NOFS);
+		blkdev_issue_flush(journal->j_fs_dev);
 
 	blk_start_plug(&plug);
 	if (sbi->s_fc_bytes == 0) {

@@ -1535,7 +1535,7 @@ static int ext4_fc_replay_inode(struct super_block *sb, struct ext4_fc_tl *tl)
 out:
 	iput(inode);
 	if (!ret)
-		blkdev_issue_flush(sb->s_bdev, GFP_KERNEL);
+		blkdev_issue_flush(sb->s_bdev);
 
 	return 0;
 }

@@ -174,7 +174,7 @@ int ext4_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
 		ret = ext4_fsync_journal(inode, datasync, &needs_barrier);
 
 	if (needs_barrier) {
-		err = blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL);
+		err = blkdev_issue_flush(inode->i_sb->s_bdev);
 		if (!ret)
 			ret = err;
 	}

@@ -1583,7 +1583,7 @@ int ext4_init_inode_table(struct super_block *sb, ext4_group_t group,
 	if (ret < 0)
 		goto err_out;
 	if (barrier)
-		blkdev_issue_flush(sb->s_bdev, GFP_NOFS);
+		blkdev_issue_flush(sb->s_bdev);
 
 skip_zeroout:
 	ext4_lock_group(sb, group);

@@ -5709,7 +5709,7 @@ static int ext4_sync_fs(struct super_block *sb, int wait)
 		needs_barrier = true;
 	if (needs_barrier) {
 		int err;
-		err = blkdev_issue_flush(sb->s_bdev, GFP_KERNEL);
+		err = blkdev_issue_flush(sb->s_bdev);
 		if (!ret)
 			ret = err;
 	}

@@ -195,7 +195,7 @@ int fat_file_fsync(struct file *filp, loff_t start, loff_t end, int datasync)
 	if (err)
 		return err;
 
-	return blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL);
+	return blkdev_issue_flush(inode->i_sb->s_bdev);
 }
 
 

@@ -340,7 +340,7 @@ int hfsplus_file_fsync(struct file *file, loff_t start, loff_t end,
 	}
 
 	if (!test_bit(HFSPLUS_SB_NOBARRIER, &sbi->flags))
-		blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL);
+		blkdev_issue_flush(inode->i_sb->s_bdev);
 
 	inode_unlock(inode);
 

@@ -239,7 +239,7 @@ static int hfsplus_sync_fs(struct super_block *sb, int wait)
 	mutex_unlock(&sbi->vh_mutex);
 
 	if (!test_bit(HFSPLUS_SB_NOBARRIER, &sbi->flags))
-		blkdev_issue_flush(sb->s_bdev, GFP_KERNEL);
+		blkdev_issue_flush(sb->s_bdev);
 
 	return error;
 }

@@ -416,7 +416,7 @@ int jbd2_cleanup_journal_tail(journal_t *journal)
 	 * jbd2_cleanup_journal_tail() doesn't get called all that often.
 	 */
 	if (journal->j_flags & JBD2_BARRIER)
-		blkdev_issue_flush(journal->j_fs_dev, GFP_NOFS);
+		blkdev_issue_flush(journal->j_fs_dev);
 
 	return __jbd2_update_log_tail(journal, first_tid, blocknr);
 }

@@ -825,7 +825,7 @@ void jbd2_journal_commit_transaction(journal_t *journal)
 	if (commit_transaction->t_need_data_flush &&
 	    (journal->j_fs_dev != journal->j_dev) &&
 	    (journal->j_flags & JBD2_BARRIER))
-		blkdev_issue_flush(journal->j_fs_dev, GFP_NOFS);
+		blkdev_issue_flush(journal->j_fs_dev);
 
 	/* Done it all: now write the commit record asynchronously. */
 	if (jbd2_has_feature_async_commit(journal)) {

@@ -932,7 +932,7 @@ void jbd2_journal_commit_transaction(journal_t *journal)
 	stats.run.rs_blocks_logged++;
 	if (jbd2_has_feature_async_commit(journal) &&
 	    journal->j_flags & JBD2_BARRIER) {
-		blkdev_issue_flush(journal->j_dev, GFP_NOFS);
+		blkdev_issue_flush(journal->j_dev);
 	}
 
 	if (err)

@@ -326,7 +326,7 @@ int jbd2_journal_recover(journal_t *journal)
 		err = err2;
 	/* Make sure all replayed data is on permanent storage */
 	if (journal->j_flags & JBD2_BARRIER) {
-		err2 = blkdev_issue_flush(journal->j_fs_dev, GFP_KERNEL);
+		err2 = blkdev_issue_flush(journal->j_fs_dev);
 		if (!err)
 			err = err2;
 	}

@@ -1117,7 +1117,7 @@ int generic_file_fsync(struct file *file, loff_t start, loff_t end,
 	err = __generic_file_fsync(file, start, end, datasync);
 	if (err)
 		return err;
-	return blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL);
+	return blkdev_issue_flush(inode->i_sb->s_bdev);
 }
 EXPORT_SYMBOL(generic_file_fsync);
 

@@ -375,7 +375,7 @@ static inline int nilfs_flush_device(struct the_nilfs *nilfs)
 	 */
 	smp_wmb();
 
-	err = blkdev_issue_flush(nilfs->ns_bdev, GFP_KERNEL);
+	err = blkdev_issue_flush(nilfs->ns_bdev);
 	if (err != -EIO)
 		err = 0;
 	return err;

@@ -194,7 +194,7 @@ static int ocfs2_sync_file(struct file *file, loff_t start, loff_t end,
 		needs_barrier = true;
 	err = jbd2_complete_transaction(journal, commit_tid);
 	if (needs_barrier) {
-		ret = blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL);
+		ret = blkdev_issue_flush(inode->i_sb->s_bdev);
 		if (!err)
 			err = ret;
 	}

@@ -159,7 +159,7 @@ static int reiserfs_sync_file(struct file *filp, loff_t start, loff_t end,
 	barrier_done = reiserfs_commit_for_inode(inode);
 	reiserfs_write_unlock(inode->i_sb);
 	if (barrier_done != 1 && reiserfs_barrier_flush(inode->i_sb))
-		blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL);
+		blkdev_issue_flush(inode->i_sb->s_bdev);
 	inode_unlock(inode);
 	if (barrier_done < 0)
 		return barrier_done;

@@ -342,7 +342,7 @@ void
 xfs_blkdev_issue_flush(
 	xfs_buftarg_t		*buftarg)
 {
-	blkdev_issue_flush(buftarg->bt_bdev, GFP_NOFS);
+	blkdev_issue_flush(buftarg->bt_bdev);
 }
 
 STATIC void

@@ -541,7 +541,7 @@ static int zonefs_file_fsync(struct file *file, loff_t start, loff_t end,
 	if (ZONEFS_I(inode)->i_ztype == ZONEFS_ZTYPE_CNV)
 		ret = file_write_and_wait_range(file, start, end);
 	if (!ret)
-		ret = blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL);
+		ret = blkdev_issue_flush(inode->i_sb->s_bdev);
 
 	if (ret)
 		zonefs_io_error(inode, true);

@@ -1288,7 +1288,7 @@ static inline bool blk_needs_flush_plug(struct task_struct *tsk)
 		 !list_empty(&plug->cb_list));
 }
 
-int blkdev_issue_flush(struct block_device *, gfp_t);
+int blkdev_issue_flush(struct block_device *bdev);
 long nr_blockdev_pages(void);
 #else /* CONFIG_BLOCK */
 struct blk_plug {

@@ -1316,7 +1316,7 @@ static inline bool blk_needs_flush_plug(struct task_struct *tsk)
 	return false;
 }
 
-static inline int blkdev_issue_flush(struct block_device *bdev, gfp_t gfp_mask)
+static inline int blkdev_issue_flush(struct block_device *bdev)
 {
 	return 0;
 }

@@ -85,12 +85,6 @@ struct clone {
 
 	struct dm_clone_metadata *cmd;
 
-	/*
-	 * bio used to flush the destination device, before committing the
-	 * metadata.
-	 */
-	struct bio flush_bio;
-
 	/* Region hydration hash table */
 	struct hash_table_bucket *ht;
 

@@ -1155,11 +1149,7 @@ static int commit_metadata(struct clone *clone, bool *dest_dev_flushed)
 		goto out;
 	}
 
-	bio_reset(&clone->flush_bio);
-	bio_set_dev(&clone->flush_bio, clone->dest_dev->bdev);
-	clone->flush_bio.bi_opf = REQ_OP_WRITE | REQ_PREFLUSH;
-
-	r = submit_bio_wait(&clone->flush_bio);
+	r = blkdev_issue_flush(clone->dest_dev->bdev);
 	if (unlikely(r)) {
 		__metadata_operation_failed(clone, "flush destination device", r);
 		goto out;

@@ -1886,7 +1876,6 @@ static int clone_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 	bio_list_init(&clone->deferred_flush_completions);
 	clone->hydration_offset = 0;
 	atomic_set(&clone->hydrations_in_flight, 0);
-	bio_init(&clone->flush_bio, NULL, 0);
 
 	clone->wq = alloc_workqueue("dm-" DM_MSG_PREFIX, WQ_MEM_RECLAIM, 0);
 	if (!clone->wq) {

@@ -1958,7 +1947,6 @@ static void clone_dtr(struct dm_target *ti)
 	struct clone *clone = ti->private;
 
 	mutex_destroy(&clone->commit_lock);
-	bio_uninit(&clone->flush_bio);
 
 	for (i = 0; i < clone->nr_ctr_args; i++)
 		kfree(clone->ctr_args[i]);

@@ -56,7 +56,8 @@ static inline struct bio *__f2fs_bio_alloc(gfp_t gfp_mask,
 	return bio_alloc_bioset(gfp_mask, nr_iovecs, &f2fs_bioset);
 }
 
-struct bio *f2fs_bio_alloc(struct f2fs_sb_info *sbi, int npages, bool noio)
+static struct bio *f2fs_bio_alloc(struct f2fs_sb_info *sbi, int npages,
+		bool noio)
 {
 	if (noio) {
 		/* No failure on bio allocation */

@@ -3424,7 +3424,6 @@ void f2fs_destroy_checkpoint_caches(void);
  */
 int __init f2fs_init_bioset(void);
 void f2fs_destroy_bioset(void);
-struct bio *f2fs_bio_alloc(struct f2fs_sb_info *sbi, int npages, bool noio);
 int f2fs_init_bio_entry_cache(void);
 void f2fs_destroy_bio_entry_cache(void);
 void f2fs_submit_bio(struct f2fs_sb_info *sbi,

@@ -566,17 +566,7 @@ void f2fs_balance_fs_bg(struct f2fs_sb_info *sbi, bool from_bg)
 static int __submit_flush_wait(struct f2fs_sb_info *sbi,
 				struct block_device *bdev)
 {
-	struct bio *bio;
-	int ret;
-
-	bio = f2fs_bio_alloc(sbi, 0, false);
-	if (!bio)
-		return -ENOMEM;
-
-	bio->bi_opf = REQ_OP_WRITE | REQ_SYNC | REQ_PREFLUSH;
-	bio_set_dev(bio, bdev);
-	ret = submit_bio_wait(bio);
-	bio_put(bio);
+	int ret = blkdev_issue_flush(bdev);
 
 	trace_f2fs_issue_flush(bdev, test_opt(sbi, NOBARRIER),
 				test_opt(sbi, FLUSH_MERGE), ret);

@@ -179,7 +179,6 @@ fault_type=%d		 Support configuring fault injection type, should be
 			 FAULT_KVMALLOC		  0x000000002
 			 FAULT_PAGE_ALLOC	  0x000000004
 			 FAULT_PAGE_GET		  0x000000008
-			 FAULT_ALLOC_BIO	  0x000000010
 			 FAULT_ALLOC_NID	  0x000000020
 			 FAULT_ORPHAN		  0x000000040
 			 FAULT_BLOCK		  0x000000080

@@ -50,28 +50,6 @@ void f2fs_destroy_bioset(void)
 	bioset_exit(&f2fs_bioset);
 }
 
-static inline struct bio *__f2fs_bio_alloc(gfp_t gfp_mask,
-						unsigned int nr_iovecs)
-{
-	return bio_alloc_bioset(gfp_mask, nr_iovecs, &f2fs_bioset);
-}
-
-static struct bio *f2fs_bio_alloc(struct f2fs_sb_info *sbi, int npages,
-		bool noio)
-{
-	if (noio) {
-		/* No failure on bio allocation */
-		return __f2fs_bio_alloc(GFP_NOIO, npages);
-	}
-
-	if (time_to_inject(sbi, FAULT_ALLOC_BIO)) {
-		f2fs_show_injection_info(sbi, FAULT_ALLOC_BIO);
-		return NULL;
-	}
-
-	return __f2fs_bio_alloc(GFP_KERNEL, npages);
-}
-
 static bool __is_cp_guaranteed(struct page *page)
 {
 	struct address_space *mapping = page->mapping;

@@ -433,7 +411,7 @@ static struct bio *__bio_alloc(struct f2fs_io_info *fio, int npages)
 	struct f2fs_sb_info *sbi = fio->sbi;
 	struct bio *bio;
 
-	bio = f2fs_bio_alloc(sbi, npages, true);
+	bio = bio_alloc_bioset(GFP_NOIO, npages, &f2fs_bioset);
 
 	f2fs_target_device(sbi, fio->new_blkaddr, bio);
 	if (is_read_io(fio->op)) {

@@ -1029,8 +1007,9 @@ static struct bio *f2fs_grab_read_bio(struct inode *inode, block_t blkaddr,
 	struct bio_post_read_ctx *ctx;
 	unsigned int post_read_steps = 0;
 
-	bio = f2fs_bio_alloc(sbi, min_t(int, nr_pages, BIO_MAX_PAGES),
-								for_write);
+	bio = bio_alloc_bioset(for_write ? GFP_NOIO : GFP_KERNEL,
+			       min_t(int, nr_pages, BIO_MAX_PAGES),
+			       &f2fs_bioset);
 	if (!bio)
 		return ERR_PTR(-ENOMEM);
 

@@ -43,7 +43,6 @@ enum {
 	FAULT_KVMALLOC,
 	FAULT_PAGE_ALLOC,
 	FAULT_PAGE_GET,
-	FAULT_ALLOC_BIO,
 	FAULT_ALLOC_NID,
 	FAULT_ORPHAN,
 	FAULT_BLOCK,

@@ -45,7 +45,6 @@ const char *f2fs_fault_name[FAULT_MAX] = {
 	[FAULT_KVMALLOC]	= "kvmalloc",
 	[FAULT_PAGE_ALLOC]	= "page alloc",
 	[FAULT_PAGE_GET]	= "page get",
-	[FAULT_ALLOC_BIO]	= "alloc bio",
 	[FAULT_ALLOC_NID]	= "alloc nid",
 	[FAULT_ORPHAN]		= "orphan",
 	[FAULT_BLOCK]		= "no more block",

@@ -396,123 +396,101 @@ static void punt_bios_to_rescuer(struct bio_set *bs)
  * @nr_iovecs:	number of iovecs to pre-allocate
  * @bs:		the bio_set to allocate from.
  *
- * Description:
- *   If @bs is NULL, uses kmalloc() to allocate the bio; else the allocation is
- *   backed by the @bs's mempool.
+ * Allocate a bio from the mempools in @bs.
  *
- *   When @bs is not NULL, if %__GFP_DIRECT_RECLAIM is set then bio_alloc will
- *   always be able to allocate a bio. This is due to the mempool guarantees.
- *   To make this work, callers must never allocate more than 1 bio at a time
- *   from this pool. Callers that need to allocate more than 1 bio must always
- *   submit the previously allocated bio for IO before attempting to allocate
- *   a new one. Failure to do so can cause deadlocks under memory pressure.
+ * If %__GFP_DIRECT_RECLAIM is set then bio_alloc will always be able to
+ * allocate a bio.  This is due to the mempool guarantees.  To make this work,
+ * callers must never allocate more than 1 bio at a time from the general pool.
+ * Callers that need to allocate more than 1 bio must always submit the
+ * previously allocated bio for IO before attempting to allocate a new one.
+ * Failure to do so can cause deadlocks under memory pressure.
  *
- *   Note that when running under submit_bio_noacct() (i.e. any block
- *   driver), bios are not submitted until after you return - see the code in
- *   submit_bio_noacct() that converts recursion into iteration, to prevent
- *   stack overflows.
+ * Note that when running under submit_bio_noacct() (i.e. any block driver),
+ * bios are not submitted until after you return - see the code in
+ * submit_bio_noacct() that converts recursion into iteration, to prevent
+ * stack overflows.
  *
- *   This would normally mean allocating multiple bios under
- *   submit_bio_noacct() would be susceptible to deadlocks, but we have
- *   deadlock avoidance code that resubmits any blocked bios from a rescuer
- *   thread.
+ * This would normally mean allocating multiple bios under submit_bio_noacct()
+ * would be susceptible to deadlocks, but we have
+ * deadlock avoidance code that resubmits any blocked bios from a rescuer
+ * thread.
  *
- *   However, we do not guarantee forward progress for allocations from other
- *   mempools. Doing multiple allocations from the same mempool under
- *   submit_bio_noacct() should be avoided - instead, use bio_set's front_pad
- *   for per bio allocations.
+ * However, we do not guarantee forward progress for allocations from other
+ * mempools. Doing multiple allocations from the same mempool under
+ * submit_bio_noacct() should be avoided - instead, use bio_set's front_pad
+ * for per bio allocations.
  *
- *   RETURNS:
- *   Pointer to new bio on success, NULL on failure.
+ * Returns: Pointer to new bio on success, NULL on failure.
  */
 struct bio *bio_alloc_bioset(gfp_t gfp_mask, unsigned int nr_iovecs,
 			     struct bio_set *bs)
 {
 	gfp_t saved_gfp = gfp_mask;
-	unsigned front_pad;
-	unsigned inline_vecs;
-	struct bio_vec *bvl = NULL;
 	struct bio *bio;
 	void *p;
 
-	if (!bs) {
-		if (nr_iovecs > UIO_MAXIOV)
-			return NULL;
-
-		p = kmalloc(struct_size(bio, bi_inline_vecs, nr_iovecs), gfp_mask);
-		front_pad = 0;
-		inline_vecs = nr_iovecs;
-	} else {
-		/* should not use nobvec bioset for nr_iovecs > 0 */
-		if (WARN_ON_ONCE(!mempool_initialized(&bs->bvec_pool) &&
-				 nr_iovecs > 0))
-			return NULL;
-		/*
-		 * submit_bio_noacct() converts recursion to iteration; this
-		 * means if we're running beneath it, any bios we allocate and
-		 * submit will not be submitted (and thus freed) until after we
-		 * return.
-		 *
-		 * This exposes us to a potential deadlock if we allocate
-		 * multiple bios from the same bio_set() while running
-		 * underneath submit_bio_noacct(). If we were to allocate
-		 * multiple bios (say a stacking block driver that was splitting
-		 * bios), we would deadlock if we exhausted the mempool's
-		 * reserve.
-		 *
-		 * We solve this, and guarantee forward progress, with a rescuer
-		 * workqueue per bio_set. If we go to allocate and there are
-		 * bios on current->bio_list, we first try the allocation
-		 * without __GFP_DIRECT_RECLAIM; if that fails, we punt those
-		 * bios we would be blocking to the rescuer workqueue before
-		 * we retry with the original gfp_flags.
-		 */
-
-		if (current->bio_list &&
-		    (!bio_list_empty(&current->bio_list[0]) ||
-		     !bio_list_empty(&current->bio_list[1])) &&
-		    bs->rescue_workqueue)
-			gfp_mask &= ~__GFP_DIRECT_RECLAIM;
+	/* should not use nobvec bioset for nr_iovecs > 0 */
+	if (WARN_ON_ONCE(!mempool_initialized(&bs->bvec_pool) && nr_iovecs > 0))
+		return NULL;
 
+	/*
+	 * submit_bio_noacct() converts recursion to iteration; this means if
+	 * we're running beneath it, any bios we allocate and submit will not be
+	 * submitted (and thus freed) until after we return.
+	 *
+	 * This exposes us to a potential deadlock if we allocate multiple bios
+	 * from the same bio_set() while running underneath submit_bio_noacct().
+	 * If we were to allocate multiple bios (say a stacking block driver
+	 * that was splitting bios), we would deadlock if we exhausted the
+	 * mempool's reserve.
+	 *
+	 * We solve this, and guarantee forward progress, with a rescuer
+	 * workqueue per bio_set. If we go to allocate and there are bios on
+	 * current->bio_list, we first try the allocation without
+	 * __GFP_DIRECT_RECLAIM; if that fails, we punt those bios we would be
+	 * blocking to the rescuer workqueue before we retry with the original
+	 * gfp_flags.
+	 */
+	if (current->bio_list &&
+	    (!bio_list_empty(&current->bio_list[0]) ||
+	     !bio_list_empty(&current->bio_list[1])) &&
+	    bs->rescue_workqueue)
+		gfp_mask &= ~__GFP_DIRECT_RECLAIM;
+
+	p = mempool_alloc(&bs->bio_pool, gfp_mask);
+	if (!p && gfp_mask != saved_gfp) {
+		punt_bios_to_rescuer(bs);
+		gfp_mask = saved_gfp;
 		p = mempool_alloc(&bs->bio_pool, gfp_mask);
-		if (!p && gfp_mask != saved_gfp) {
-			punt_bios_to_rescuer(bs);
-			gfp_mask = saved_gfp;
-			p = mempool_alloc(&bs->bio_pool, gfp_mask);
-		}
-
-		front_pad = bs->front_pad;
-		inline_vecs = BIO_INLINE_VECS;
 	}
-
 	if (unlikely(!p))
 		return NULL;
 
-	bio = p + front_pad;
-	bio_init(bio, NULL, 0);
-
-	if (nr_iovecs > inline_vecs) {
+	bio = p + bs->front_pad;
+	if (nr_iovecs > BIO_INLINE_VECS) {
 		unsigned long idx = 0;
+		struct bio_vec *bvl = NULL;
 
 		bvl = bvec_alloc(gfp_mask, nr_iovecs, &idx, &bs->bvec_pool);
 		if (!bvl && gfp_mask != saved_gfp) {
 			punt_bios_to_rescuer(bs);
 			gfp_mask = saved_gfp;
-			bvl = bvec_alloc(gfp_mask, nr_iovecs, &idx, &bs->bvec_pool);
+			bvl = bvec_alloc(gfp_mask, nr_iovecs, &idx,
+					 &bs->bvec_pool);
 		}
 
 		if (unlikely(!bvl))
 			goto err_free;
 
 		bio->bi_flags |= idx << BVEC_POOL_OFFSET;
-		bio->bi_max_vecs = bvec_nr_vecs(idx);
+		bio_init(bio, bvl, bvec_nr_vecs(idx));
 	} else if (nr_iovecs) {
-		bvl = bio->bi_inline_vecs;
-		bio->bi_max_vecs = inline_vecs;
+		bio_init(bio, bio->bi_inline_vecs, BIO_INLINE_VECS);
+	} else {
+		bio_init(bio, NULL, 0);
 	}
 
 	bio->bi_pool = bs;
-	bio->bi_io_vec = bvl;
 	return bio;
 
 err_free:

@@ -521,6 +499,31 @@ struct bio *bio_alloc_bioset(gfp_t gfp_mask, unsigned int nr_iovecs,
 }
 EXPORT_SYMBOL(bio_alloc_bioset);
 
+/**
+ * bio_kmalloc - kmalloc a bio for I/O
+ * @gfp_mask:   the GFP_* mask given to the slab allocator
+ * @nr_iovecs:	number of iovecs to pre-allocate
+ *
+ * Use kmalloc to allocate and initialize a bio.
+ *
+ * Returns: Pointer to new bio on success, NULL on failure.
+ */
+struct bio *bio_kmalloc(gfp_t gfp_mask, unsigned int nr_iovecs)
+{
+	struct bio *bio;
+
+	if (nr_iovecs > UIO_MAXIOV)
+		return NULL;
+
+	bio = kmalloc(struct_size(bio, bi_inline_vecs, nr_iovecs), gfp_mask);
+	if (unlikely(!bio))
+		return NULL;
+	bio_init(bio, nr_iovecs ? bio->bi_inline_vecs : NULL, nr_iovecs);
+	bio->bi_pool = NULL;
+	return bio;
+}
+EXPORT_SYMBOL(bio_kmalloc);
+
 void zero_fill_bio_iter(struct bio *bio, struct bvec_iter start)
 {
 	unsigned long flags;

@@ -408,6 +408,7 @@ extern int biovec_init_pool(mempool_t *pool, int pool_entries);
 extern int bioset_init_from_src(struct bio_set *bs, struct bio_set *src);
 
 extern struct bio *bio_alloc_bioset(gfp_t, unsigned int, struct bio_set *);
+struct bio *bio_kmalloc(gfp_t gfp_mask, unsigned int nr_iovecs);
 extern void bio_put(struct bio *);
 
 extern void __bio_clone_fast(struct bio *, struct bio *);

@@ -420,11 +421,6 @@ static inline struct bio *bio_alloc(gfp_t gfp_mask, unsigned int nr_iovecs)
 	return bio_alloc_bioset(gfp_mask, nr_iovecs, &fs_bio_set);
 }
 
-static inline struct bio *bio_kmalloc(gfp_t gfp_mask, unsigned int nr_iovecs)
-{
-	return bio_alloc_bioset(gfp_mask, nr_iovecs, NULL);
-}
-
 extern blk_qc_t submit_bio(struct bio *);
 
 extern void bio_endio(struct bio *);

@@ -164,7 +164,7 @@ static struct bio *blk_crypto_clone_bio(struct bio *bio_src)
 	struct bio_vec bv;
 	struct bio *bio;
 
-	bio = bio_alloc_bioset(GFP_NOIO, bio_segments(bio_src), NULL);
+	bio = bio_kmalloc(GFP_NOIO, bio_segments(bio_src));
 	if (!bio)
 		return NULL;
 	bio->bi_bdev		= bio_src->bi_bdev;

@@ -138,7 +138,7 @@ static int _drbd_md_sync_page_io(struct drbd_device *device,
 		op_flags |= REQ_FUA | REQ_PREFLUSH;
 	op_flags |= REQ_SYNC;
 
-	bio = bio_alloc_drbd(GFP_NOIO);
+	bio = bio_alloc_bioset(GFP_NOIO, 1, &drbd_md_io_bio_set);
 	bio_set_dev(bio, bdev->md_bdev);
 	bio->bi_iter.bi_sector = sector;
 	err = -EIO;

@@ -976,7 +976,7 @@ static void drbd_bm_endio(struct bio *bio)
 
 static void bm_page_io_async(struct drbd_bm_aio_ctx *ctx, int page_nr) __must_hold(local)
 {
-	struct bio *bio = bio_alloc_drbd(GFP_NOIO);
+	struct bio *bio = bio_alloc_bioset(GFP_NOIO, 1, &drbd_md_io_bio_set);
 	struct drbd_device *device = ctx->device;
 	struct drbd_bitmap *b = device->bitmap;
 	struct page *page;

@@ -1422,8 +1422,6 @@ extern mempool_t drbd_md_io_page_pool;
 /* We also need to make sure we get a bio
  * when we need it for housekeeping purposes */
 extern struct bio_set drbd_md_io_bio_set;
-/* to allocate from that set */
-extern struct bio *bio_alloc_drbd(gfp_t gfp_mask);
 
 /* And a bio_set for cloning */
 extern struct bio_set drbd_io_bio_set;

@@ -138,19 +138,6 @@ static const struct block_device_operations drbd_ops = {
 	.release	= drbd_release,
 };
 
-struct bio *bio_alloc_drbd(gfp_t gfp_mask)
-{
-	struct bio *bio;
-
-	if (!bioset_initialized(&drbd_md_io_bio_set))
-		return bio_alloc(gfp_mask, 1);
-
-	bio = bio_alloc_bioset(gfp_mask, 1, &drbd_md_io_bio_set);
-	if (!bio)
-		return NULL;
-	return bio;
-}
-
 #ifdef __CHECKER__
 /* When checking with sparse, and this is an inline function, sparse will
    give tons of false positives. When this is a real functions sparse works.

@@ -30,7 +30,10 @@ static struct drbd_request *drbd_req_new(struct drbd_device *device, struct bio
 		return NULL;
 	memset(req, 0, sizeof(*req));
 
-	drbd_req_make_private_bio(req, bio_src);
+	req->private_bio = bio_clone_fast(bio_src, GFP_NOIO, &drbd_io_bio_set);
+	req->private_bio->bi_private = req;
+	req->private_bio->bi_end_io = drbd_request_endio;
+
 	req->rq_state = (bio_data_dir(bio_src) == WRITE ? RQ_WRITE : 0)
 		      | (bio_op(bio_src) == REQ_OP_WRITE_SAME ? RQ_WSAME : 0)
 		      | (bio_op(bio_src) == REQ_OP_WRITE_ZEROES ? RQ_ZEROES : 0)

@@ -256,18 +256,6 @@ enum drbd_req_state_bits {
 #define MR_WRITE       1
 #define MR_READ        2
 
-static inline void drbd_req_make_private_bio(struct drbd_request *req, struct bio *bio_src)
-{
-	struct bio *bio;
-	bio = bio_clone_fast(bio_src, GFP_NOIO, &drbd_io_bio_set);
-
-	req->private_bio = bio;
-
-	bio->bi_private  = req;
-	bio->bi_end_io   = drbd_request_endio;
-	bio->bi_next     = NULL;
-}
-
 /* Short lived temporary struct on the stack.
  * We could squirrel the error to be returned into
  * bio->bi_iter.bi_size, or similar. But that would be too ugly. */

@@ -1523,8 +1523,11 @@ int w_restart_disk_io(struct drbd_work *w, int cancel)
 	if (bio_data_dir(req->master_bio) == WRITE && req->rq_state & RQ_IN_ACT_LOG)
 		drbd_al_begin_io(device, &req->i);
 
-	drbd_req_make_private_bio(req, req->master_bio);
+	req->private_bio = bio_clone_fast(req->master_bio, GFP_NOIO,
+					  &drbd_io_bio_set);
 	bio_set_dev(req->private_bio, device->ldev->backing_bdev);
+	req->private_bio->bi_private = req;
+	req->private_bio->bi_end_io = drbd_request_endio;
 	submit_bio_noacct(req->private_bio);
 
 	return 0;