Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net

Cross-merge networking fixes after downstream PR.

No conflicts or adjacent changes.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>

194 files changed, 6181 insertions, 4060 deletions

diff --git a/fs/9p/v9fs_vfs.h b/fs/9p/v9fs_vfs.h
index 731e3d14b67d..0e8418066a48 100644
--- a/fs/9p/v9fs_vfs.h
+++ b/fs/9p/v9fs_vfs.h
@@ -42,6 +42,7 @@ struct inode *v9fs_alloc_inode(struct super_block *sb);
 void v9fs_free_inode(struct inode *inode);
 struct inode *v9fs_get_inode(struct super_block *sb, umode_t mode,
 			     dev_t rdev);
+void v9fs_set_netfs_context(struct inode *inode);
 int v9fs_init_inode(struct v9fs_session_info *v9ses,
 		    struct inode *inode, umode_t mode, dev_t rdev);
 void v9fs_evict_inode(struct inode *inode);
diff --git a/fs/9p/vfs_addr.c b/fs/9p/vfs_addr.c
index 8a635999a7d6..047855033d32 100644
--- a/fs/9p/vfs_addr.c
+++ b/fs/9p/vfs_addr.c
@@ -19,12 +19,45 @@
 #include <linux/netfs.h>
 #include <net/9p/9p.h>
 #include <net/9p/client.h>
+#include <trace/events/netfs.h>
 
 #include "v9fs.h"
 #include "v9fs_vfs.h"
 #include "cache.h"
 #include "fid.h"
 
+static void v9fs_upload_to_server(struct netfs_io_subrequest *subreq)
+{
+	struct p9_fid *fid = subreq->rreq->netfs_priv;
+	int err, len;
+
+	trace_netfs_sreq(subreq, netfs_sreq_trace_submit);
+	len = p9_client_write(fid, subreq->start, &subreq->io_iter, &err);
+	netfs_write_subrequest_terminated(subreq, len ?: err, false);
+}
+
+static void v9fs_upload_to_server_worker(struct work_struct *work)
+{
+	struct netfs_io_subrequest *subreq =
+		container_of(work, struct netfs_io_subrequest, work);
+
+	v9fs_upload_to_server(subreq);
+}
+
+/*
+ * Set up write requests for a writeback slice.  We need to add a write request
+ * for each write we want to make.
+ */
+static void v9fs_create_write_requests(struct netfs_io_request *wreq, loff_t start, size_t len)
+{
+	struct netfs_io_subrequest *subreq;
+
+	subreq = netfs_create_write_request(wreq, NETFS_UPLOAD_TO_SERVER,
+					    start, len, v9fs_upload_to_server_worker);
+	if (subreq)
+		netfs_queue_write_request(subreq);
+}
+
 /**
  * v9fs_issue_read - Issue a read from 9P
  * @subreq: The read to make
@@ -33,14 +66,10 @@ static void v9fs_issue_read(struct netfs_io_subrequest *subreq)
 {
 	struct netfs_io_request *rreq = subreq->rreq;
 	struct p9_fid *fid = rreq->netfs_priv;
-	struct iov_iter to;
-	loff_t pos = subreq->start + subreq->transferred;
-	size_t len = subreq->len   - subreq->transferred;
 	int total, err;
 
-	iov_iter_xarray(&to, ITER_DEST, &rreq->mapping->i_pages, pos, len);
-
-	total = p9_client_read(fid, pos, &to, &err);
+	total = p9_client_read(fid, subreq->start + subreq->transferred,
+			       &subreq->io_iter, &err);
 
 	/* if we just extended the file size, any portion not in
 	 * cache won't be on server and is zeroes */
@@ -50,25 +79,42 @@ static void v9fs_issue_read(struct netfs_io_subrequest *subreq)
 }
 
 /**
- * v9fs_init_request - Initialise a read request
+ * v9fs_init_request - Initialise a request
  * @rreq: The read request
  * @file: The file being read from
  */
 static int v9fs_init_request(struct netfs_io_request *rreq, struct file *file)
 {
-	struct p9_fid *fid = file->private_data;
-
-	BUG_ON(!fid);
+	struct p9_fid *fid;
+	bool writing = (rreq->origin == NETFS_READ_FOR_WRITE ||
+			rreq->origin == NETFS_WRITEBACK ||
+			rreq->origin == NETFS_WRITETHROUGH ||
+			rreq->origin == NETFS_LAUNDER_WRITE ||
+			rreq->origin == NETFS_UNBUFFERED_WRITE ||
+			rreq->origin == NETFS_DIO_WRITE);
+
+	if (file) {
+		fid = file->private_data;
+		if (!fid)
+			goto no_fid;
+		p9_fid_get(fid);
+	} else {
+		fid = v9fs_fid_find_inode(rreq->inode, writing, INVALID_UID, true);
+		if (!fid)
+			goto no_fid;
+	}
 
 	/* we might need to read from a fid that was opened write-only
 	 * for read-modify-write of page cache, use the writeback fid
 	 * for that */
-	WARN_ON(rreq->origin == NETFS_READ_FOR_WRITE &&
-			!(fid->mode & P9_ORDWR));
-
-	p9_fid_get(fid);
+	WARN_ON(rreq->origin == NETFS_READ_FOR_WRITE && !(fid->mode & P9_ORDWR));
 	rreq->netfs_priv = fid;
 	return 0;
+
+no_fid:
+	WARN_ONCE(1, "folio expected an open fid inode->i_ino=%lx\n",
+		  rreq->inode->i_ino);
+	return -EINVAL;
 }
 
 /**
@@ -82,281 +128,20 @@ static void v9fs_free_request(struct netfs_io_request *rreq)
 	p9_fid_put(fid);
 }
 
-/**
- * v9fs_begin_cache_operation - Begin a cache operation for a read
- * @rreq: The read request
- */
-static int v9fs_begin_cache_operation(struct netfs_io_request *rreq)
-{
-#ifdef CONFIG_9P_FSCACHE
-	struct fscache_cookie *cookie = v9fs_inode_cookie(V9FS_I(rreq->inode));
-
-	return fscache_begin_read_operation(&rreq->cache_resources, cookie);
-#else
-	return -ENOBUFS;
-#endif
-}
-
 const struct netfs_request_ops v9fs_req_ops = {
 	.init_request		= v9fs_init_request,
 	.free_request		= v9fs_free_request,
-	.begin_cache_operation	= v9fs_begin_cache_operation,
 	.issue_read		= v9fs_issue_read,
+	.create_write_requests	= v9fs_create_write_requests,
 };
 
-/**
- * v9fs_release_folio - release the private state associated with a folio
- * @folio: The folio to be released
- * @gfp: The caller's allocation restrictions
- *
- * Returns true if the page can be released, false otherwise.
- */
-
-static bool v9fs_release_folio(struct folio *folio, gfp_t gfp)
-{
-	if (folio_test_private(folio))
-		return false;
-#ifdef CONFIG_9P_FSCACHE
-	if (folio_test_fscache(folio)) {
-		if (current_is_kswapd() || !(gfp & __GFP_FS))
-			return false;
-		folio_wait_fscache(folio);
-	}
-	fscache_note_page_release(v9fs_inode_cookie(V9FS_I(folio_inode(folio))));
-#endif
-	return true;
-}
-
-static void v9fs_invalidate_folio(struct folio *folio, size_t offset,
-				 size_t length)
-{
-	folio_wait_fscache(folio);
-}
-
-#ifdef CONFIG_9P_FSCACHE
-static void v9fs_write_to_cache_done(void *priv, ssize_t transferred_or_error,
-				     bool was_async)
-{
-	struct v9fs_inode *v9inode = priv;
-	__le32 version;
-
-	if (IS_ERR_VALUE(transferred_or_error) &&
-	    transferred_or_error != -ENOBUFS) {
-		version = cpu_to_le32(v9inode->qid.version);
-		fscache_invalidate(v9fs_inode_cookie(v9inode), &version,
-				   i_size_read(&v9inode->netfs.inode), 0);
-	}
-}
-#endif
-
-static int v9fs_vfs_write_folio_locked(struct folio *folio)
-{
-	struct inode *inode = folio_inode(folio);
-	loff_t start = folio_pos(folio);
-	loff_t i_size = i_size_read(inode);
-	struct iov_iter from;
-	size_t len = folio_size(folio);
-	struct p9_fid *writeback_fid;
-	int err;
-	struct v9fs_inode __maybe_unused *v9inode = V9FS_I(inode);
-	struct fscache_cookie __maybe_unused *cookie = v9fs_inode_cookie(v9inode);
-
-	if (start >= i_size)
-		return 0; /* Simultaneous truncation occurred */
-
-	len = min_t(loff_t, i_size - start, len);
-
-	iov_iter_xarray(&from, ITER_SOURCE, &folio_mapping(folio)->i_pages, start, len);
-
-	writeback_fid = v9fs_fid_find_inode(inode, true, INVALID_UID, true);
-	if (!writeback_fid) {
-		WARN_ONCE(1, "folio expected an open fid inode->i_private=%p\n",
-			inode->i_private);
-		return -EINVAL;
-	}
-
-	folio_wait_fscache(folio);
-	folio_start_writeback(folio);
-
-	p9_client_write(writeback_fid, start, &from, &err);
-
-#ifdef CONFIG_9P_FSCACHE
-	if (err == 0 &&
-		fscache_cookie_enabled(cookie) &&
-		test_bit(FSCACHE_COOKIE_IS_CACHING, &cookie->flags)) {
-		folio_start_fscache(folio);
-		fscache_write_to_cache(v9fs_inode_cookie(v9inode),
-					folio_mapping(folio), start, len, i_size,
-					v9fs_write_to_cache_done, v9inode,
-					true);
-	}
-#endif
-
-	folio_end_writeback(folio);
-	p9_fid_put(writeback_fid);
-
-	return err;
-}
-
-static int v9fs_vfs_writepage(struct page *page, struct writeback_control *wbc)
-{
-	struct folio *folio = page_folio(page);
-	int retval;
-
-	p9_debug(P9_DEBUG_VFS, "folio %p\n", folio);
-
-	retval = v9fs_vfs_write_folio_locked(folio);
-	if (retval < 0) {
-		if (retval == -EAGAIN) {
-			folio_redirty_for_writepage(wbc, folio);
-			retval = 0;
-		} else {
-			mapping_set_error(folio_mapping(folio), retval);
-		}
-	} else
-		retval = 0;
-
-	folio_unlock(folio);
-	return retval;
-}
-
-static int v9fs_launder_folio(struct folio *folio)
-{
-	int retval;
-
-	if (folio_clear_dirty_for_io(folio)) {
-		retval = v9fs_vfs_write_folio_locked(folio);
-		if (retval)
-			return retval;
-	}
-	folio_wait_fscache(folio);
-	return 0;
-}
-
-/**
- * v9fs_direct_IO - 9P address space operation for direct I/O
- * @iocb: target I/O control block
- * @iter: The data/buffer to use
- *
- * The presence of v9fs_direct_IO() in the address space ops vector
- * allowes open() O_DIRECT flags which would have failed otherwise.
- *
- * In the non-cached mode, we shunt off direct read and write requests before
- * the VFS gets them, so this method should never be called.
- *
- * Direct IO is not 'yet' supported in the cached mode. Hence when
- * this routine is called through generic_file_aio_read(), the read/write fails
- * with an error.
- *
- */
-static ssize_t
-v9fs_direct_IO(struct kiocb *iocb, struct iov_iter *iter)
-{
-	struct file *file = iocb->ki_filp;
-	loff_t pos = iocb->ki_pos;
-	ssize_t n;
-	int err = 0;
-
-	if (iov_iter_rw(iter) == WRITE) {
-		n = p9_client_write(file->private_data, pos, iter, &err);
-		if (n) {
-			struct inode *inode = file_inode(file);
-			loff_t i_size = i_size_read(inode);
-
-			if (pos + n > i_size)
-				inode_add_bytes(inode, pos + n - i_size);
-		}
-	} else {
-		n = p9_client_read(file->private_data, pos, iter, &err);
-	}
-	return n ? n : err;
-}
-
-static int v9fs_write_begin(struct file *filp, struct address_space *mapping,
-			    loff_t pos, unsigned int len,
-			    struct page **subpagep, void **fsdata)
-{
-	int retval;
-	struct folio *folio;
-	struct v9fs_inode *v9inode = V9FS_I(mapping->host);
-
-	p9_debug(P9_DEBUG_VFS, "filp %p, mapping %p\n", filp, mapping);
-
-	/* Prefetch area to be written into the cache if we're caching this
-	 * file.  We need to do this before we get a lock on the page in case
-	 * there's more than one writer competing for the same cache block.
-	 */
-	retval = netfs_write_begin(&v9inode->netfs, filp, mapping, pos, len, &folio, fsdata);
-	if (retval < 0)
-		return retval;
-
-	*subpagep = &folio->page;
-	return retval;
-}
-
-static int v9fs_write_end(struct file *filp, struct address_space *mapping,
-			  loff_t pos, unsigned int len, unsigned int copied,
-			  struct page *subpage, void *fsdata)
-{
-	loff_t last_pos = pos + copied;
-	struct folio *folio = page_folio(subpage);
-	struct inode *inode = mapping->host;
-
-	p9_debug(P9_DEBUG_VFS, "filp %p, mapping %p\n", filp, mapping);
-
-	if (!folio_test_uptodate(folio)) {
-		if (unlikely(copied < len)) {
-			copied = 0;
-			goto out;
-		}
-
-		folio_mark_uptodate(folio);
-	}
-
-	/*
-	 * No need to use i_size_read() here, the i_size
-	 * cannot change under us because we hold the i_mutex.
-	 */
-	if (last_pos > inode->i_size) {
-		inode_add_bytes(inode, last_pos - inode->i_size);
-		i_size_write(inode, last_pos);
-#ifdef CONFIG_9P_FSCACHE
-		fscache_update_cookie(v9fs_inode_cookie(V9FS_I(inode)), NULL,
-			&last_pos);
-#endif
-	}
-	folio_mark_dirty(folio);
-out:
-	folio_unlock(folio);
-	folio_put(folio);
-
-	return copied;
-}
-
-#ifdef CONFIG_9P_FSCACHE
-/*
- * Mark a page as having been made dirty and thus needing writeback.  We also
- * need to pin the cache object to write back to.
- */
-static bool v9fs_dirty_folio(struct address_space *mapping, struct folio *folio)
-{
-	struct v9fs_inode *v9inode = V9FS_I(mapping->host);
-
-	return fscache_dirty_folio(mapping, folio, v9fs_inode_cookie(v9inode));
-}
-#else
-#define v9fs_dirty_folio filemap_dirty_folio
-#endif
-
 const struct address_space_operations v9fs_addr_operations = {
-	.read_folio = netfs_read_folio,
-	.readahead = netfs_readahead,
-	.dirty_folio = v9fs_dirty_folio,
-	.writepage = v9fs_vfs_writepage,
-	.write_begin = v9fs_write_begin,
-	.write_end = v9fs_write_end,
-	.release_folio = v9fs_release_folio,
-	.invalidate_folio = v9fs_invalidate_folio,
-	.launder_folio = v9fs_launder_folio,
-	.direct_IO = v9fs_direct_IO,
+	.read_folio		= netfs_read_folio,
+	.readahead		= netfs_readahead,
+	.dirty_folio		= netfs_dirty_folio,
+	.release_folio		= netfs_release_folio,
+	.invalidate_folio	= netfs_invalidate_folio,
+	.launder_folio		= netfs_launder_folio,
+	.direct_IO		= noop_direct_IO,
+	.writepages		= netfs_writepages,
 };
diff --git a/fs/9p/vfs_file.c b/fs/9p/vfs_file.c
index 11cd8d23f6f2..bae330c2f0cf 100644
--- a/fs/9p/vfs_file.c
+++ b/fs/9p/vfs_file.c
@@ -353,25 +353,15 @@ static ssize_t
 v9fs_file_read_iter(struct kiocb *iocb, struct iov_iter *to)
 {
 	struct p9_fid *fid = iocb->ki_filp->private_data;
-	int ret, err = 0;
 
 	p9_debug(P9_DEBUG_VFS, "fid %d count %zu offset %lld\n",
 		 fid->fid, iov_iter_count(to), iocb->ki_pos);
 
-	if (!(fid->mode & P9L_DIRECT)) {
-		p9_debug(P9_DEBUG_VFS, "(cached)\n");
-		return generic_file_read_iter(iocb, to);
-	}
-
-	if (iocb->ki_filp->f_flags & O_NONBLOCK)
-		ret = p9_client_read_once(fid, iocb->ki_pos, to, &err);
-	else
-		ret = p9_client_read(fid, iocb->ki_pos, to, &err);
-	if (!ret)
-		return err;
+	if (fid->mode & P9L_DIRECT)
+		return netfs_unbuffered_read_iter(iocb, to);
 
-	iocb->ki_pos += ret;
-	return ret;
+	p9_debug(P9_DEBUG_VFS, "(cached)\n");
+	return netfs_file_read_iter(iocb, to);
 }
 
 /*
@@ -407,46 +397,14 @@ v9fs_file_write_iter(struct kiocb *iocb, struct iov_iter *from)
 {
 	struct file *file = iocb->ki_filp;
 	struct p9_fid *fid = file->private_data;
-	ssize_t retval;
-	loff_t origin;
-	int err = 0;
 
 	p9_debug(P9_DEBUG_VFS, "fid %d\n", fid->fid);
 
-	if (!(fid->mode & (P9L_DIRECT | P9L_NOWRITECACHE))) {
-		p9_debug(P9_DEBUG_CACHE, "(cached)\n");
-		return generic_file_write_iter(iocb, from);
-	}
+	if (fid->mode & (P9L_DIRECT | P9L_NOWRITECACHE))
+		return netfs_unbuffered_write_iter(iocb, from);
 
-	retval = generic_write_checks(iocb, from);
-	if (retval <= 0)
-		return retval;
-
-	origin = iocb->ki_pos;
-	retval = p9_client_write(file->private_data, iocb->ki_pos, from, &err);
-	if (retval > 0) {
-		struct inode *inode = file_inode(file);
-		loff_t i_size;
-		unsigned long pg_start, pg_end;
-
-		pg_start = origin >> PAGE_SHIFT;
-		pg_end = (origin + retval - 1) >> PAGE_SHIFT;
-		if (inode->i_mapping && inode->i_mapping->nrpages)
-			invalidate_inode_pages2_range(inode->i_mapping,
-						      pg_start, pg_end);
-		iocb->ki_pos += retval;
-		i_size = i_size_read(inode);
-		if (iocb->ki_pos > i_size) {
-			inode_add_bytes(inode, iocb->ki_pos - i_size);
-			/*
-			 * Need to serialize against i_size_write() in
-			 * v9fs_stat2inode()
-			 */
-			v9fs_i_size_write(inode, iocb->ki_pos);
-		}
-		return retval;
-	}
-	return err;
+	p9_debug(P9_DEBUG_CACHE, "(cached)\n");
+	return netfs_file_write_iter(iocb, from);
 }
 
 static int v9fs_file_fsync(struct file *filp, loff_t start, loff_t end,
@@ -519,36 +477,7 @@ v9fs_file_mmap(struct file *filp, struct vm_area_struct *vma)
 static vm_fault_t
 v9fs_vm_page_mkwrite(struct vm_fault *vmf)
 {
-	struct folio *folio = page_folio(vmf->page);
-	struct file *filp = vmf->vma->vm_file;
-	struct inode *inode = file_inode(filp);
-
-
-	p9_debug(P9_DEBUG_VFS, "folio %p fid %lx\n",
-		 folio, (unsigned long)filp->private_data);
-
-	/* Wait for the page to be written to the cache before we allow it to
-	 * be modified.  We then assume the entire page will need writing back.
-	 */
-#ifdef CONFIG_9P_FSCACHE
-	if (folio_test_fscache(folio) &&
-	    folio_wait_fscache_killable(folio) < 0)
-		return VM_FAULT_NOPAGE;
-#endif
-
-	/* Update file times before taking page lock */
-	file_update_time(filp);
-
-	if (folio_lock_killable(folio) < 0)
-		return VM_FAULT_RETRY;
-	if (folio_mapping(folio) != inode->i_mapping)
-		goto out_unlock;
-	folio_wait_stable(folio);
-
-	return VM_FAULT_LOCKED;
-out_unlock:
-	folio_unlock(folio);
-	return VM_FAULT_NOPAGE;
+	return netfs_page_mkwrite(vmf, NULL);
 }
 
 static void v9fs_mmap_vm_close(struct vm_area_struct *vma)
diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c
index b845ee18a80b..32572982f72e 100644
--- a/fs/9p/vfs_inode.c
+++ b/fs/9p/vfs_inode.c
@@ -246,10 +246,10 @@ void v9fs_free_inode(struct inode *inode)
 /*
  * Set parameters for the netfs library
  */
-static void v9fs_set_netfs_context(struct inode *inode)
+void v9fs_set_netfs_context(struct inode *inode)
 {
 	struct v9fs_inode *v9inode = V9FS_I(inode);
-	netfs_inode_init(&v9inode->netfs, &v9fs_req_ops);
+	netfs_inode_init(&v9inode->netfs, &v9fs_req_ops, true);
 }
 
 int v9fs_init_inode(struct v9fs_session_info *v9ses,
@@ -326,8 +326,6 @@ int v9fs_init_inode(struct v9fs_session_info *v9ses,
 		err = -EINVAL;
 		goto error;
 	}
-
-	v9fs_set_netfs_context(inode);
 error:
 	return err;
 
@@ -359,6 +357,7 @@ struct inode *v9fs_get_inode(struct super_block *sb, umode_t mode, dev_t rdev)
 		iput(inode);
 		return ERR_PTR(err);
 	}
+	v9fs_set_netfs_context(inode);
 	return inode;
 }
 
@@ -374,11 +373,8 @@ void v9fs_evict_inode(struct inode *inode)
 
 	truncate_inode_pages_final(&inode->i_data);
 
-#ifdef CONFIG_9P_FSCACHE
 	version = cpu_to_le32(v9inode->qid.version);
-	fscache_clear_inode_writeback(v9fs_inode_cookie(v9inode), inode,
-				      &version);
-#endif
+	netfs_clear_inode_writeback(inode, &version);
 
 	clear_inode(inode);
 	filemap_fdatawrite(&inode->i_data);
@@ -464,6 +460,7 @@ static struct inode *v9fs_qid_iget(struct super_block *sb,
 		goto error;
 
 	v9fs_stat2inode(st, inode, sb, 0);
+	v9fs_set_netfs_context(inode);
 	v9fs_cache_inode_get_cookie(inode);
 	unlock_new_inode(inode);
 	return inode;
@@ -1113,7 +1110,7 @@ static int v9fs_vfs_setattr(struct mnt_idmap *idmap,
 	if ((iattr->ia_valid & ATTR_SIZE) &&
 		 iattr->ia_size != i_size_read(inode)) {
 		truncate_setsize(inode, iattr->ia_size);
-		truncate_pagecache(inode, iattr->ia_size);
+		netfs_resize_file(netfs_inode(inode), iattr->ia_size, true);
 
 #ifdef CONFIG_9P_FSCACHE
 		if (v9ses->cache & CACHE_FSCACHE) {
@@ -1181,6 +1178,7 @@ v9fs_stat2inode(struct p9_wstat *stat, struct inode *inode,
 	mode |= inode->i_mode & ~S_IALLUGO;
 	inode->i_mode = mode;
 
+	v9inode->netfs.remote_i_size = stat->length;
 	if (!(flags & V9FS_STAT2INODE_KEEP_ISIZE))
 		v9fs_i_size_write(inode, stat->length);
 	/* not real number of blocks, but 512 byte ones ... */
diff --git a/fs/9p/vfs_inode_dotl.c b/fs/9p/vfs_inode_dotl.c
index c7319af2f471..3505227e1704 100644
--- a/fs/9p/vfs_inode_dotl.c
+++ b/fs/9p/vfs_inode_dotl.c
@@ -128,6 +128,7 @@ static struct inode *v9fs_qid_iget_dotl(struct super_block *sb,
 		goto error;
 
 	v9fs_stat2inode_dotl(st, inode, 0);
+	v9fs_set_netfs_context(inode);
 	v9fs_cache_inode_get_cookie(inode);
 	retval = v9fs_get_acl(inode, fid);
 	if (retval)
@@ -598,7 +599,7 @@ int v9fs_vfs_setattr_dotl(struct mnt_idmap *idmap,
 	if ((iattr->ia_valid & ATTR_SIZE) && iattr->ia_size !=
 		 i_size_read(inode)) {
 		truncate_setsize(inode, iattr->ia_size);
-		truncate_pagecache(inode, iattr->ia_size);
+		netfs_resize_file(netfs_inode(inode), iattr->ia_size, true);
 
 #ifdef CONFIG_9P_FSCACHE
 		if (v9ses->cache & CACHE_FSCACHE)
@@ -655,6 +656,7 @@ v9fs_stat2inode_dotl(struct p9_stat_dotl *stat, struct inode *inode,
 		mode |= inode->i_mode & ~S_IALLUGO;
 		inode->i_mode = mode;
 
+		v9inode->netfs.remote_i_size = stat->st_size;
 		if (!(flags & V9FS_STAT2INODE_KEEP_ISIZE))
 			v9fs_i_size_write(inode, stat->st_size);
 		inode->i_blocks = stat->st_blocks;
@@ -683,8 +685,10 @@ v9fs_stat2inode_dotl(struct p9_stat_dotl *stat, struct inode *inode,
 			inode->i_mode = mode;
 		}
 		if (!(flags & V9FS_STAT2INODE_KEEP_ISIZE) &&
-		    stat->st_result_mask & P9_STATS_SIZE)
+		    stat->st_result_mask & P9_STATS_SIZE) {
+			v9inode->netfs.remote_i_size = stat->st_size;
 			v9fs_i_size_write(inode, stat->st_size);
+		}
 		if (stat->st_result_mask & P9_STATS_BLOCKS)
 			inode->i_blocks = stat->st_blocks;
 	}
diff --git a/fs/9p/vfs_super.c b/fs/9p/vfs_super.c
index 73db55c050bf..941f7d0e0bfa 100644
--- a/fs/9p/vfs_super.c
+++ b/fs/9p/vfs_super.c
@@ -289,31 +289,21 @@ static int v9fs_drop_inode(struct inode *inode)
 static int v9fs_write_inode(struct inode *inode,
 			    struct writeback_control *wbc)
 {
-	struct v9fs_inode *v9inode;
-
 	/*
 	 * send an fsync request to server irrespective of
 	 * wbc->sync_mode.
 	 */
 	p9_debug(P9_DEBUG_VFS, "%s: inode %p\n", __func__, inode);
-
-	v9inode = V9FS_I(inode);
-	fscache_unpin_writeback(wbc, v9fs_inode_cookie(v9inode));
-
-	return 0;
+	return netfs_unpin_writeback(inode, wbc);
 }
 
 static int v9fs_write_inode_dotl(struct inode *inode,
 				 struct writeback_control *wbc)
 {
-	struct v9fs_inode *v9inode;
 
-	v9inode = V9FS_I(inode);
 	p9_debug(P9_DEBUG_VFS, "%s: inode %p\n", __func__, inode);
 
-	fscache_unpin_writeback(wbc, v9fs_inode_cookie(v9inode));
-
-	return 0;
+	return netfs_unpin_writeback(inode, wbc);
 }
 
 static const struct super_operations v9fs_super_ops = {
diff --git a/fs/Kconfig b/fs/Kconfig
index a3159831ba98..89fdbefd1075 100644
--- a/fs/Kconfig
+++ b/fs/Kconfig
@@ -144,7 +144,6 @@ source "fs/overlayfs/Kconfig"
 menu "Caches"
 
 source "fs/netfs/Kconfig"
-source "fs/fscache/Kconfig"
 source "fs/cachefiles/Kconfig"
 
 endmenu
diff --git a/fs/Makefile b/fs/Makefile
index a6962c588962..c09016257f05 100644
--- a/fs/Makefile
+++ b/fs/Makefile
@@ -61,7 +61,6 @@ obj-$(CONFIG_DLM)		+= dlm/
  
 # Do not add any filesystems before this line
 obj-$(CONFIG_NETFS_SUPPORT)	+= netfs/
-obj-$(CONFIG_FSCACHE)		+= fscache/
 obj-$(CONFIG_REISERFS_FS)	+= reiserfs/
 obj-$(CONFIG_EXT4_FS)		+= ext4/
 # We place ext4 before ext2 so that clean ext3 root fs's do NOT mount using the
diff --git a/fs/afs/dir.c b/fs/afs/dir.c
index c14533ef108f..b5b8de521f99 100644
--- a/fs/afs/dir.c
+++ b/fs/afs/dir.c
@@ -124,7 +124,7 @@ static void afs_dir_read_cleanup(struct afs_read *req)
 		if (xas_retry(&xas, folio))
 			continue;
 		BUG_ON(xa_is_value(folio));
-		ASSERTCMP(folio_file_mapping(folio), ==, mapping);
+		ASSERTCMP(folio->mapping, ==, mapping);
 
 		folio_put(folio);
 	}
@@ -202,12 +202,12 @@ static void afs_dir_dump(struct afs_vnode *dvnode, struct afs_read *req)
 		if (xas_retry(&xas, folio))
 			continue;
 
-		BUG_ON(folio_file_mapping(folio) != mapping);
+		BUG_ON(folio->mapping != mapping);
 
 		size = min_t(loff_t, folio_size(folio), req->actual_len - folio_pos(folio));
 		for (offset = 0; offset < size; offset += sizeof(*block)) {
 			block = kmap_local_folio(folio, offset);
-			pr_warn("[%02lx] %32phN\n", folio_index(folio) + offset, block);
+			pr_warn("[%02lx] %32phN\n", folio->index + offset, block);
 			kunmap_local(block);
 		}
 	}
@@ -233,7 +233,7 @@ static int afs_dir_check(struct afs_vnode *dvnode, struct afs_read *req)
 		if (xas_retry(&xas, folio))
 			continue;
 
-		BUG_ON(folio_file_mapping(folio) != mapping);
+		BUG_ON(folio->mapping != mapping);
 
 		if (!afs_dir_check_folio(dvnode, folio, req->actual_len)) {
 			afs_dir_dump(dvnode, req);
@@ -474,6 +474,14 @@ static int afs_dir_iterate_block(struct afs_vnode *dvnode,
 			continue;
 		}
 
+		/* Don't expose silly rename entries to userspace. */
+		if (nlen > 6 &&
+		    dire->u.name[0] == '.' &&
+		    ctx->actor != afs_lookup_filldir &&
+		    ctx->actor != afs_lookup_one_filldir &&
+		    memcmp(dire->u.name, ".__afs", 6) == 0)
+			continue;
+
 		/* found the next entry */
 		if (!dir_emit(ctx, dire->u.name, nlen,
 			      ntohl(dire->u.vnode),
@@ -708,6 +716,8 @@ static void afs_do_lookup_success(struct afs_operation *op)
 			break;
 		}
 
+		if (vp->scb.status.abort_code)
+			trace_afs_bulkstat_error(op, &vp->fid, i, vp->scb.status.abort_code);
 		if (!vp->scb.have_status && !vp->scb.have_error)
 			continue;
 
@@ -897,12 +907,16 @@ static struct inode *afs_do_lookup(struct inode *dir, struct dentry *dentry,
 		afs_begin_vnode_operation(op);
 		afs_wait_for_operation(op);
 	}
-	inode = ERR_PTR(afs_op_error(op));
 
 out_op:
 	if (!afs_op_error(op)) {
-		inode = &op->file[1].vnode->netfs.inode;
-		op->file[1].vnode = NULL;
+		if (op->file[1].scb.status.abort_code) {
+			afs_op_accumulate_error(op, -ECONNABORTED,
+						op->file[1].scb.status.abort_code);
+		} else {
+			inode = &op->file[1].vnode->netfs.inode;
+			op->file[1].vnode = NULL;
+		}
 	}
 
 	if (op->file[0].scb.have_status)
@@ -2022,7 +2036,7 @@ static bool afs_dir_release_folio(struct folio *folio, gfp_t gfp_flags)
 {
 	struct afs_vnode *dvnode = AFS_FS_I(folio_inode(folio));
 
-	_enter("{{%llx:%llu}[%lu]}", dvnode->fid.vid, dvnode->fid.vnode, folio_index(folio));
+	_enter("{{%llx:%llu}[%lu]}", dvnode->fid.vid, dvnode->fid.vnode, folio->index);
 
 	folio_detach_private(folio);
 
diff --git a/fs/afs/dynroot.c b/fs/afs/dynroot.c
index 2cd40ba601f1..c4d2711e20ad 100644
--- a/fs/afs/dynroot.c
+++ b/fs/afs/dynroot.c
@@ -76,7 +76,7 @@ struct inode *afs_iget_pseudo_dir(struct super_block *sb, bool root)
 	/* there shouldn't be an existing inode */
 	BUG_ON(!(inode->i_state & I_NEW));
 
-	netfs_inode_init(&vnode->netfs, NULL);
+	netfs_inode_init(&vnode->netfs, NULL, false);
 	inode->i_size		= 0;
 	inode->i_mode		= S_IFDIR | S_IRUGO | S_IXUGO;
 	if (root) {
@@ -258,16 +258,7 @@ const struct inode_operations afs_dynroot_inode_operations = {
 	.lookup		= afs_dynroot_lookup,
 };
 
-/*
- * Dirs in the dynamic root don't need revalidation.
- */
-static int afs_dynroot_d_revalidate(struct dentry *dentry, unsigned int flags)
-{
-	return 1;
-}
-
 const struct dentry_operations afs_dynroot_dentry_operations = {
-	.d_revalidate	= afs_dynroot_d_revalidate,
 	.d_delete	= always_delete_dentry,
 	.d_release	= afs_d_release,
 	.d_automount	= afs_d_automount,
diff --git a/fs/afs/file.c b/fs/afs/file.c
index 30914e0d9cb2..3d33b221d9ca 100644
--- a/fs/afs/file.c
+++ b/fs/afs/file.c
@@ -20,9 +20,6 @@
 
 static int afs_file_mmap(struct file *file, struct vm_area_struct *vma);
 static int afs_symlink_read_folio(struct file *file, struct folio *folio);
-static void afs_invalidate_folio(struct folio *folio, size_t offset,
-			       size_t length);
-static bool afs_release_folio(struct folio *folio, gfp_t gfp_flags);
 
 static ssize_t afs_file_read_iter(struct kiocb *iocb, struct iov_iter *iter);
 static ssize_t afs_file_splice_read(struct file *in, loff_t *ppos,
@@ -37,7 +34,7 @@ const struct file_operations afs_file_operations = {
 	.release	= afs_release,
 	.llseek		= generic_file_llseek,
 	.read_iter	= afs_file_read_iter,
-	.write_iter	= afs_file_write,
+	.write_iter	= netfs_file_write_iter,
 	.mmap		= afs_file_mmap,
 	.splice_read	= afs_file_splice_read,
 	.splice_write	= iter_file_splice_write,
@@ -53,22 +50,21 @@ const struct inode_operations afs_file_inode_operations = {
 };
 
 const struct address_space_operations afs_file_aops = {
+	.direct_IO	= noop_direct_IO,
 	.read_folio	= netfs_read_folio,
 	.readahead	= netfs_readahead,
-	.dirty_folio	= afs_dirty_folio,
-	.launder_folio	= afs_launder_folio,
-	.release_folio	= afs_release_folio,
-	.invalidate_folio = afs_invalidate_folio,
-	.write_begin	= afs_write_begin,
-	.write_end	= afs_write_end,
-	.writepages	= afs_writepages,
+	.dirty_folio	= netfs_dirty_folio,
+	.launder_folio	= netfs_launder_folio,
+	.release_folio	= netfs_release_folio,
+	.invalidate_folio = netfs_invalidate_folio,
 	.migrate_folio	= filemap_migrate_folio,
+	.writepages	= afs_writepages,
 };
 
 const struct address_space_operations afs_symlink_aops = {
 	.read_folio	= afs_symlink_read_folio,
-	.release_folio	= afs_release_folio,
-	.invalidate_folio = afs_invalidate_folio,
+	.release_folio	= netfs_release_folio,
+	.invalidate_folio = netfs_invalidate_folio,
 	.migrate_folio	= filemap_migrate_folio,
 };
 
@@ -323,11 +319,7 @@ static void afs_issue_read(struct netfs_io_subrequest *subreq)
 	fsreq->len	= subreq->len   - subreq->transferred;
 	fsreq->key	= key_get(subreq->rreq->netfs_priv);
 	fsreq->vnode	= vnode;
-	fsreq->iter	= &fsreq->def_iter;
-
-	iov_iter_xarray(&fsreq->def_iter, ITER_DEST,
-			&fsreq->vnode->netfs.inode.i_mapping->i_pages,
-			fsreq->pos, fsreq->len);
+	fsreq->iter	= &subreq->io_iter;
 
 	afs_fetch_data(fsreq->vnode, fsreq);
 	afs_put_read(fsreq);
@@ -359,22 +351,13 @@ static int afs_symlink_read_folio(struct file *file, struct folio *folio)
 
 static int afs_init_request(struct netfs_io_request *rreq, struct file *file)
 {
-	rreq->netfs_priv = key_get(afs_file_key(file));
+	if (file)
+		rreq->netfs_priv = key_get(afs_file_key(file));
+	rreq->rsize = 256 * 1024;
+	rreq->wsize = 256 * 1024;
 	return 0;
 }
 
-static int afs_begin_cache_operation(struct netfs_io_request *rreq)
-{
-#ifdef CONFIG_AFS_FSCACHE
-	struct afs_vnode *vnode = AFS_FS_I(rreq->inode);
-
-	return fscache_begin_read_operation(&rreq->cache_resources,
-					    afs_vnode_cache(vnode));
-#else
-	return -ENOBUFS;
-#endif
-}
-
 static int afs_check_write_begin(struct file *file, loff_t pos, unsigned len,
 				 struct folio **foliop, void **_fsdata)
 {
@@ -388,128 +371,37 @@ static void afs_free_request(struct netfs_io_request *rreq)
 	key_put(rreq->netfs_priv);
 }
 
-const struct netfs_request_ops afs_req_ops = {
-	.init_request		= afs_init_request,
-	.free_request		= afs_free_request,
-	.begin_cache_operation	= afs_begin_cache_operation,
-	.check_write_begin	= afs_check_write_begin,
-	.issue_read		= afs_issue_read,
-};
-
-int afs_write_inode(struct inode *inode, struct writeback_control *wbc)
+static void afs_update_i_size(struct inode *inode, loff_t new_i_size)
 {
-	fscache_unpin_writeback(wbc, afs_vnode_cache(AFS_FS_I(inode)));
-	return 0;
-}
-
-/*
- * Adjust the dirty region of the page on truncation or full invalidation,
- * getting rid of the markers altogether if the region is entirely invalidated.
- */
-static void afs_invalidate_dirty(struct folio *folio, size_t offset,
-				 size_t length)
-{
-	struct afs_vnode *vnode = AFS_FS_I(folio_inode(folio));
-	unsigned long priv;
-	unsigned int f, t, end = offset + length;
-
-	priv = (unsigned long)folio_get_private(folio);
-
-	/* we clean up only if the entire page is being invalidated */
-	if (offset == 0 && length == folio_size(folio))
-		goto full_invalidate;
-
-	 /* If the page was dirtied by page_mkwrite(), the PTE stays writable
-	  * and we don't get another notification to tell us to expand it
-	  * again.
-	  */
-	if (afs_is_folio_dirty_mmapped(priv))
-		return;
-
-	/* We may need to shorten the dirty region */
-	f = afs_folio_dirty_from(folio, priv);
-	t = afs_folio_dirty_to(folio, priv);
-
-	if (t <= offset || f >= end)
-		return; /* Doesn't overlap */
-
-	if (f < offset && t > end)
-		return; /* Splits the dirty region - just absorb it */
-
-	if (f >= offset && t <= end)
-		goto undirty;
+	struct afs_vnode *vnode = AFS_FS_I(inode);
+	loff_t i_size;
 
-	if (f < offset)
-		t = offset;
-	else
-		f = end;
-	if (f == t)
-		goto undirty;
-
-	priv = afs_folio_dirty(folio, f, t);
-	folio_change_private(folio, (void *)priv);
-	trace_afs_folio_dirty(vnode, tracepoint_string("trunc"), folio);
-	return;
-
-undirty:
-	trace_afs_folio_dirty(vnode, tracepoint_string("undirty"), folio);
-	folio_clear_dirty_for_io(folio);
-full_invalidate:
-	trace_afs_folio_dirty(vnode, tracepoint_string("inval"), folio);
-	folio_detach_private(folio);
+	write_seqlock(&vnode->cb_lock);
+	i_size = i_size_read(&vnode->netfs.inode);
+	if (new_i_size > i_size) {
+		i_size_write(&vnode->netfs.inode, new_i_size);
+		inode_set_bytes(&vnode->netfs.inode, new_i_size);
+	}
+	write_sequnlock(&vnode->cb_lock);
+	fscache_update_cookie(afs_vnode_cache(vnode), NULL, &new_i_size);
 }
 
-/*
- * invalidate part or all of a page
- * - release a page and clean up its private data if offset is 0 (indicating
- *   the entire page)
- */
-static void afs_invalidate_folio(struct folio *folio, size_t offset,
-			       size_t length)
+static void afs_netfs_invalidate_cache(struct netfs_io_request *wreq)
 {
-	_enter("{%lu},%zu,%zu", folio->index, offset, length);
-
-	BUG_ON(!folio_test_locked(folio));
+	struct afs_vnode *vnode = AFS_FS_I(wreq->inode);
 
-	if (folio_get_private(folio))
-		afs_invalidate_dirty(folio, offset, length);
-
-	folio_wait_fscache(folio);
-	_leave("");
+	afs_invalidate_cache(vnode, 0);
 }
 
-/*
- * release a page and clean up its private state if it's not busy
- * - return true if the page can now be released, false if not
- */
-static bool afs_release_folio(struct folio *folio, gfp_t gfp)
-{
-	struct afs_vnode *vnode = AFS_FS_I(folio_inode(folio));
-
-	_enter("{{%llx:%llu}[%lu],%lx},%x",
-	       vnode->fid.vid, vnode->fid.vnode, folio_index(folio), folio->flags,
-	       gfp);
-
-	/* deny if folio is being written to the cache and the caller hasn't
-	 * elected to wait */
-#ifdef CONFIG_AFS_FSCACHE
-	if (folio_test_fscache(folio)) {
-		if (current_is_kswapd() || !(gfp & __GFP_FS))
-			return false;
-		folio_wait_fscache(folio);
-	}
-	fscache_note_page_release(afs_vnode_cache(vnode));
-#endif
-
-	if (folio_test_private(folio)) {
-		trace_afs_folio_dirty(vnode, tracepoint_string("rel"), folio);
-		folio_detach_private(folio);
-	}
-
-	/* Indicate that the folio can be released */
-	_leave(" = T");
-	return true;
-}
+const struct netfs_request_ops afs_req_ops = {
+	.init_request		= afs_init_request,
+	.free_request		= afs_free_request,
+	.check_write_begin	= afs_check_write_begin,
+	.issue_read		= afs_issue_read,
+	.update_i_size		= afs_update_i_size,
+	.invalidate_cache	= afs_netfs_invalidate_cache,
+	.create_write_requests	= afs_create_write_requests,
+};
 
 static void afs_add_open_mmap(struct afs_vnode *vnode)
 {
@@ -576,28 +468,39 @@ static vm_fault_t afs_vm_map_pages(struct vm_fault *vmf, pgoff_t start_pgoff, pg
 
 static ssize_t afs_file_read_iter(struct kiocb *iocb, struct iov_iter *iter)
 {
-	struct afs_vnode *vnode = AFS_FS_I(file_inode(iocb->ki_filp));
+	struct inode *inode = file_inode(iocb->ki_filp);
+	struct afs_vnode *vnode = AFS_FS_I(inode);
 	struct afs_file *af = iocb->ki_filp->private_data;
-	int ret;
+	ssize_t ret;
 
-	ret = afs_validate(vnode, af->key);
+	if (iocb->ki_flags & IOCB_DIRECT)
+		return netfs_unbuffered_read_iter(iocb, iter);
+
+	ret = netfs_start_io_read(inode);
 	if (ret < 0)
 		return ret;
-
-	return generic_file_read_iter(iocb, iter);
+	ret = afs_validate(vnode, af->key);
+	if (ret == 0)
+		ret = filemap_read(iocb, iter, 0);
+	netfs_end_io_read(inode);
+	return ret;
 }
 
 static ssize_t afs_file_splice_read(struct file *in, loff_t *ppos,
 				    struct pipe_inode_info *pipe,
 				    size_t len, unsigned int flags)
 {
-	struct afs_vnode *vnode = AFS_FS_I(file_inode(in));
+	struct inode *inode = file_inode(in);
+	struct afs_vnode *vnode = AFS_FS_I(inode);
 	struct afs_file *af = in->private_data;
-	int ret;
+	ssize_t ret;
 
-	ret = afs_validate(vnode, af->key);
+	ret = netfs_start_io_read(inode);
 	if (ret < 0)
 		return ret;
-
-	return filemap_splice_read(in, ppos, pipe, len, flags);
+	ret = afs_validate(vnode, af->key);
+	if (ret == 0)
+		ret = filemap_splice_read(in, ppos, pipe, len, flags);
+	netfs_end_io_read(inode);
+	return ret;
 }
diff --git a/fs/afs/inode.c b/fs/afs/inode.c
index 4f04f6f33f46..94fc049aff58 100644
--- a/fs/afs/inode.c
+++ b/fs/afs/inode.c
@@ -58,7 +58,7 @@ static noinline void dump_vnode(struct afs_vnode *vnode, struct afs_vnode *paren
  */
 static void afs_set_netfs_context(struct afs_vnode *vnode)
 {
-	netfs_inode_init(&vnode->netfs, &afs_req_ops);
+	netfs_inode_init(&vnode->netfs, &afs_req_ops, true);
 }
 
 /*
@@ -166,6 +166,7 @@ static void afs_apply_status(struct afs_operation *op,
 	struct inode *inode = &vnode->netfs.inode;
 	struct timespec64 t;
 	umode_t mode;
+	bool unexpected_jump = false;
 	bool data_changed = false;
 	bool change_size = vp->set_size;
 
@@ -230,6 +231,7 @@ static void afs_apply_status(struct afs_operation *op,
 		}
 		change_size = true;
 		data_changed = true;
+		unexpected_jump = true;
 	} else if (vnode->status.type == AFS_FTYPE_DIR) {
 		/* Expected directory change is handled elsewhere so
 		 * that we can locally edit the directory and save on a
@@ -249,8 +251,10 @@ static void afs_apply_status(struct afs_operation *op,
 		 * what's on the server.
 		 */
 		vnode->netfs.remote_i_size = status->size;
-		if (change_size) {
+		if (change_size || status->size > i_size_read(inode)) {
 			afs_set_i_size(vnode, status->size);
+			if (unexpected_jump)
+				vnode->netfs.zero_point = status->size;
 			inode_set_ctime_to_ts(inode, t);
 			inode_set_atime_to_ts(inode, t);
 		}
@@ -647,7 +651,7 @@ void afs_evict_inode(struct inode *inode)
 	truncate_inode_pages_final(&inode->i_data);
 
 	afs_set_cache_aux(vnode, &aux);
-	fscache_clear_inode_writeback(afs_vnode_cache(vnode), inode, &aux);
+	netfs_clear_inode_writeback(inode, &aux);
 	clear_inode(inode);
 
 	while (!list_empty(&vnode->wb_keys)) {
@@ -689,17 +693,17 @@ static void afs_setattr_success(struct afs_operation *op)
 static void afs_setattr_edit_file(struct afs_operation *op)
 {
 	struct afs_vnode_param *vp = &op->file[0];
-	struct inode *inode = &vp->vnode->netfs.inode;
+	struct afs_vnode *vnode = vp->vnode;
 
 	if (op->setattr.attr->ia_valid & ATTR_SIZE) {
 		loff_t size = op->setattr.attr->ia_size;
 		loff_t i_size = op->setattr.old_i_size;
 
-		if (size < i_size)
-			truncate_pagecache(inode, size);
-		if (size != i_size)
-			fscache_resize_cookie(afs_vnode_cache(vp->vnode),
-					      vp->scb.status.size);
+		if (size != i_size) {
+			truncate_setsize(&vnode->netfs.inode, size);
+			netfs_resize_file(&vnode->netfs, size, true);
+			fscache_resize_cookie(afs_vnode_cache(vnode), size);
+		}
 	}
 }
 
@@ -767,11 +771,11 @@ int afs_setattr(struct mnt_idmap *idmap, struct dentry *dentry,
 		 */
 		if (!(attr->ia_valid & (supported & ~ATTR_SIZE & ~ATTR_MTIME)) &&
 		    attr->ia_size < i_size &&
-		    attr->ia_size > vnode->status.size) {
-			truncate_pagecache(inode, attr->ia_size);
+		    attr->ia_size > vnode->netfs.remote_i_size) {
+			truncate_setsize(inode, attr->ia_size);
+			netfs_resize_file(&vnode->netfs, size, false);
 			fscache_resize_cookie(afs_vnode_cache(vnode),
 					      attr->ia_size);
-			i_size_write(inode, attr->ia_size);
 			ret = 0;
 			goto out_unlock;
 		}
diff --git a/fs/afs/internal.h b/fs/afs/internal.h
index 55aa0679d8ce..9c03fcf7ffaa 100644
--- a/fs/afs/internal.h
+++ b/fs/afs/internal.h
@@ -985,62 +985,6 @@ static inline void afs_invalidate_cache(struct afs_vnode *vnode, unsigned int fl
 			   i_size_read(&vnode->netfs.inode), flags);
 }
 
-/*
- * We use folio->private to hold the amount of the folio that we've written to,
- * splitting the field into two parts.  However, we need to represent a range
- * 0...FOLIO_SIZE, so we reduce the resolution if the size of the folio
- * exceeds what we can encode.
- */
-#ifdef CONFIG_64BIT
-#define __AFS_FOLIO_PRIV_MASK		0x7fffffffUL
-#define __AFS_FOLIO_PRIV_SHIFT		32
-#define __AFS_FOLIO_PRIV_MMAPPED	0x80000000UL
-#else
-#define __AFS_FOLIO_PRIV_MASK		0x7fffUL
-#define __AFS_FOLIO_PRIV_SHIFT		16
-#define __AFS_FOLIO_PRIV_MMAPPED	0x8000UL
-#endif
-
-static inline unsigned int afs_folio_dirty_resolution(struct folio *folio)
-{
-	int shift = folio_shift(folio) - (__AFS_FOLIO_PRIV_SHIFT - 1);
-	return (shift > 0) ? shift : 0;
-}
-
-static inline size_t afs_folio_dirty_from(struct folio *folio, unsigned long priv)
-{
-	unsigned long x = priv & __AFS_FOLIO_PRIV_MASK;
-
-	/* The lower bound is inclusive */
-	return x << afs_folio_dirty_resolution(folio);
-}
-
-static inline size_t afs_folio_dirty_to(struct folio *folio, unsigned long priv)
-{
-	unsigned long x = (priv >> __AFS_FOLIO_PRIV_SHIFT) & __AFS_FOLIO_PRIV_MASK;
-
-	/* The upper bound is immediately beyond the region */
-	return (x + 1) << afs_folio_dirty_resolution(folio);
-}
-
-static inline unsigned long afs_folio_dirty(struct folio *folio, size_t from, size_t to)
-{
-	unsigned int res = afs_folio_dirty_resolution(folio);
-	from >>= res;
-	to = (to - 1) >> res;
-	return (to << __AFS_FOLIO_PRIV_SHIFT) | from;
-}
-
-static inline unsigned long afs_folio_dirty_mmapped(unsigned long priv)
-{
-	return priv | __AFS_FOLIO_PRIV_MMAPPED;
-}
-
-static inline bool afs_is_folio_dirty_mmapped(unsigned long priv)
-{
-	return priv & __AFS_FOLIO_PRIV_MMAPPED;
-}
-
 #include <trace/events/afs.h>
 
 /*****************************************************************************/
@@ -1167,7 +1111,6 @@ extern int afs_release(struct inode *, struct file *);
 extern int afs_fetch_data(struct afs_vnode *, struct afs_read *);
 extern struct afs_read *afs_alloc_read(gfp_t);
 extern void afs_put_read(struct afs_read *);
-extern int afs_write_inode(struct inode *, struct writeback_control *);
 
 static inline struct afs_read *afs_get_read(struct afs_read *req)
 {
@@ -1658,24 +1601,11 @@ extern int afs_check_volume_status(struct afs_volume *, struct afs_operation *);
 /*
  * write.c
  */
-#ifdef CONFIG_AFS_FSCACHE
-bool afs_dirty_folio(struct address_space *, struct folio *);
-#else
-#define afs_dirty_folio filemap_dirty_folio
-#endif
-extern int afs_write_begin(struct file *file, struct address_space *mapping,
-			loff_t pos, unsigned len,
-			struct page **pagep, void **fsdata);
-extern int afs_write_end(struct file *file, struct address_space *mapping,
-			loff_t pos, unsigned len, unsigned copied,
-			struct page *page, void *fsdata);
-extern int afs_writepage(struct page *, struct writeback_control *);
 extern int afs_writepages(struct address_space *, struct writeback_control *);
-extern ssize_t afs_file_write(struct kiocb *, struct iov_iter *);
 extern int afs_fsync(struct file *, loff_t, loff_t, int);
 extern vm_fault_t afs_page_mkwrite(struct vm_fault *vmf);
 extern void afs_prune_wb_keys(struct afs_vnode *);
-int afs_launder_folio(struct folio *);
+void afs_create_write_requests(struct netfs_io_request *wreq, loff_t start, size_t len);
 
 /*
  * xattr.c
diff --git a/fs/afs/proc.c b/fs/afs/proc.c
index 3bd02571f30d..15eab053af6d 100644
--- a/fs/afs/proc.c
+++ b/fs/afs/proc.c
@@ -166,7 +166,7 @@ static int afs_proc_addr_prefs_show(struct seq_file *m, void *v)
 
 	if (!preflist) {
 		seq_puts(m, "NO PREFS\n");
-		return 0;
+		goto out;
 	}
 
 	seq_printf(m, "PROT SUBNET                                      PRIOR (v=%u n=%u/%u/%u)\n",
@@ -191,7 +191,8 @@ static int afs_proc_addr_prefs_show(struct seq_file *m, void *v)
 		}
 	}
 
-	rcu_read_lock();
+out:
+	rcu_read_unlock();
 	return 0;
 }
 
diff --git a/fs/afs/super.c b/fs/afs/super.c
index ae2d66a52add..f3ba1c3e72f5 100644
--- a/fs/afs/super.c
+++ b/fs/afs/super.c
@@ -55,7 +55,7 @@ int afs_net_id;
 static const struct super_operations afs_super_ops = {
 	.statfs		= afs_statfs,
 	.alloc_inode	= afs_alloc_inode,
-	.write_inode	= afs_write_inode,
+	.write_inode	= netfs_unpin_writeback,
 	.drop_inode	= afs_drop_inode,
 	.destroy_inode	= afs_destroy_inode,
 	.free_inode	= afs_free_inode,
diff --git a/fs/afs/write.c b/fs/afs/write.c
index 61d34ad2ca7d..74402d95a884 100644
--- a/fs/afs/write.c
+++ b/fs/afs/write.c
@@ -12,309 +12,17 @@
 #include <linux/writeback.h>
 #include <linux/pagevec.h>
 #include <linux/netfs.h>
+#include <trace/events/netfs.h>
 #include "internal.h"
 
-static int afs_writepages_region(struct address_space *mapping,
-				 struct writeback_control *wbc,
-				 loff_t start, loff_t end, loff_t *_next,
-				 bool max_one_loop);
-
-static void afs_write_to_cache(struct afs_vnode *vnode, loff_t start, size_t len,
-			       loff_t i_size, bool caching);
-
-#ifdef CONFIG_AFS_FSCACHE
-/*
- * Mark a page as having been made dirty and thus needing writeback.  We also
- * need to pin the cache object to write back to.
- */
-bool afs_dirty_folio(struct address_space *mapping, struct folio *folio)
-{
-	return fscache_dirty_folio(mapping, folio,
-				afs_vnode_cache(AFS_FS_I(mapping->host)));
-}
-static void afs_folio_start_fscache(bool caching, struct folio *folio)
-{
-	if (caching)
-		folio_start_fscache(folio);
-}
-#else
-static void afs_folio_start_fscache(bool caching, struct folio *folio)
-{
-}
-#endif
-
-/*
- * Flush out a conflicting write.  This may extend the write to the surrounding
- * pages if also dirty and contiguous to the conflicting region..
- */
-static int afs_flush_conflicting_write(struct address_space *mapping,
-				       struct folio *folio)
-{
-	struct writeback_control wbc = {
-		.sync_mode	= WB_SYNC_ALL,
-		.nr_to_write	= LONG_MAX,
-		.range_start	= folio_pos(folio),
-		.range_end	= LLONG_MAX,
-	};
-	loff_t next;
-
-	return afs_writepages_region(mapping, &wbc, folio_pos(folio), LLONG_MAX,
-				     &next, true);
-}
-
-/*
- * prepare to perform part of a write to a page
- */
-int afs_write_begin(struct file *file, struct address_space *mapping,
-		    loff_t pos, unsigned len,
-		    struct page **_page, void **fsdata)
-{
-	struct afs_vnode *vnode = AFS_FS_I(file_inode(file));
-	struct folio *folio;
-	unsigned long priv;
-	unsigned f, from;
-	unsigned t, to;
-	pgoff_t index;
-	int ret;
-
-	_enter("{%llx:%llu},%llx,%x",
-	       vnode->fid.vid, vnode->fid.vnode, pos, len);
-
-	/* Prefetch area to be written into the cache if we're caching this
-	 * file.  We need to do this before we get a lock on the page in case
-	 * there's more than one writer competing for the same cache block.
-	 */
-	ret = netfs_write_begin(&vnode->netfs, file, mapping, pos, len, &folio, fsdata);
-	if (ret < 0)
-		return ret;
-
-	index = folio_index(folio);
-	from = pos - index * PAGE_SIZE;
-	to = from + len;
-
-try_again:
-	/* See if this page is already partially written in a way that we can
-	 * merge the new write with.
-	 */
-	if (folio_test_private(folio)) {
-		priv = (unsigned long)folio_get_private(folio);
-		f = afs_folio_dirty_from(folio, priv);
-		t = afs_folio_dirty_to(folio, priv);
-		ASSERTCMP(f, <=, t);
-
-		if (folio_test_writeback(folio)) {
-			trace_afs_folio_dirty(vnode, tracepoint_string("alrdy"), folio);
-			folio_unlock(folio);
-			goto wait_for_writeback;
-		}
-		/* If the file is being filled locally, allow inter-write
-		 * spaces to be merged into writes.  If it's not, only write
-		 * back what the user gives us.
-		 */
-		if (!test_bit(AFS_VNODE_NEW_CONTENT, &vnode->flags) &&
-		    (to < f || from > t))
-			goto flush_conflicting_write;
-	}
-
-	*_page = folio_file_page(folio, pos / PAGE_SIZE);
-	_leave(" = 0");
-	return 0;
-
-	/* The previous write and this write aren't adjacent or overlapping, so
-	 * flush the page out.
-	 */
-flush_conflicting_write:
-	trace_afs_folio_dirty(vnode, tracepoint_string("confl"), folio);
-	folio_unlock(folio);
-
-	ret = afs_flush_conflicting_write(mapping, folio);
-	if (ret < 0)
-		goto error;
-
-wait_for_writeback:
-	ret = folio_wait_writeback_killable(folio);
-	if (ret < 0)
-		goto error;
-
-	ret = folio_lock_killable(folio);
-	if (ret < 0)
-		goto error;
-	goto try_again;
-
-error:
-	folio_put(folio);
-	_leave(" = %d", ret);
-	return ret;
-}
-
-/*
- * finalise part of a write to a page
- */
-int afs_write_end(struct file *file, struct address_space *mapping,
-		  loff_t pos, unsigned len, unsigned copied,
-		  struct page *subpage, void *fsdata)
-{
-	struct folio *folio = page_folio(subpage);
-	struct afs_vnode *vnode = AFS_FS_I(file_inode(file));
-	unsigned long priv;
-	unsigned int f, from = offset_in_folio(folio, pos);
-	unsigned int t, to = from + copied;
-	loff_t i_size, write_end_pos;
-
-	_enter("{%llx:%llu},{%lx}",
-	       vnode->fid.vid, vnode->fid.vnode, folio_index(folio));
-
-	if (!folio_test_uptodate(folio)) {
-		if (copied < len) {
-			copied = 0;
-			goto out;
-		}
-
-		folio_mark_uptodate(folio);
-	}
-
-	if (copied == 0)
-		goto out;
-
-	write_end_pos = pos + copied;
-
-	i_size = i_size_read(&vnode->netfs.inode);
-	if (write_end_pos > i_size) {
-		write_seqlock(&vnode->cb_lock);
-		i_size = i_size_read(&vnode->netfs.inode);
-		if (write_end_pos > i_size)
-			afs_set_i_size(vnode, write_end_pos);
-		write_sequnlock(&vnode->cb_lock);
-		fscache_update_cookie(afs_vnode_cache(vnode), NULL, &write_end_pos);
-	}
-
-	if (folio_test_private(folio)) {
-		priv = (unsigned long)folio_get_private(folio);
-		f = afs_folio_dirty_from(folio, priv);
-		t = afs_folio_dirty_to(folio, priv);
-		if (from < f)
-			f = from;
-		if (to > t)
-			t = to;
-		priv = afs_folio_dirty(folio, f, t);
-		folio_change_private(folio, (void *)priv);
-		trace_afs_folio_dirty(vnode, tracepoint_string("dirty+"), folio);
-	} else {
-		priv = afs_folio_dirty(folio, from, to);
-		folio_attach_private(folio, (void *)priv);
-		trace_afs_folio_dirty(vnode, tracepoint_string("dirty"), folio);
-	}
-
-	if (folio_mark_dirty(folio))
-		_debug("dirtied %lx", folio_index(folio));
-
-out:
-	folio_unlock(folio);
-	folio_put(folio);
-	return copied;
-}
-
-/*
- * kill all the pages in the given range
- */
-static void afs_kill_pages(struct address_space *mapping,
-			   loff_t start, loff_t len)
-{
-	struct afs_vnode *vnode = AFS_FS_I(mapping->host);
-	struct folio *folio;
-	pgoff_t index = start / PAGE_SIZE;
-	pgoff_t last = (start + len - 1) / PAGE_SIZE, next;
-
-	_enter("{%llx:%llu},%llx @%llx",
-	       vnode->fid.vid, vnode->fid.vnode, len, start);
-
-	do {
-		_debug("kill %lx (to %lx)", index, last);
-
-		folio = filemap_get_folio(mapping, index);
-		if (IS_ERR(folio)) {
-			next = index + 1;
-			continue;
-		}
-
-		next = folio_next_index(folio);
-
-		folio_clear_uptodate(folio);
-		folio_end_writeback(folio);
-		folio_lock(folio);
-		generic_error_remove_folio(mapping, folio);
-		folio_unlock(folio);
-		folio_put(folio);
-
-	} while (index = next, index <= last);
-
-	_leave("");
-}
-
-/*
- * Redirty all the pages in a given range.
- */
-static void afs_redirty_pages(struct writeback_control *wbc,
-			      struct address_space *mapping,
-			      loff_t start, loff_t len)
-{
-	struct afs_vnode *vnode = AFS_FS_I(mapping->host);
-	struct folio *folio;
-	pgoff_t index = start / PAGE_SIZE;
-	pgoff_t last = (start + len - 1) / PAGE_SIZE, next;
-
-	_enter("{%llx:%llu},%llx @%llx",
-	       vnode->fid.vid, vnode->fid.vnode, len, start);
-
-	do {
-		_debug("redirty %llx @%llx", len, start);
-
-		folio = filemap_get_folio(mapping, index);
-		if (IS_ERR(folio)) {
-			next = index + 1;
-			continue;
-		}
-
-		next = index + folio_nr_pages(folio);
-		folio_redirty_for_writepage(wbc, folio);
-		folio_end_writeback(folio);
-		folio_put(folio);
-	} while (index = next, index <= last);
-
-	_leave("");
-}
-
 /*
  * completion of write to server
  */
 static void afs_pages_written_back(struct afs_vnode *vnode, loff_t start, unsigned int len)
 {
-	struct address_space *mapping = vnode->netfs.inode.i_mapping;
-	struct folio *folio;
-	pgoff_t end;
-
-	XA_STATE(xas, &mapping->i_pages, start / PAGE_SIZE);
-
 	_enter("{%llx:%llu},{%x @%llx}",
 	       vnode->fid.vid, vnode->fid.vnode, len, start);
 
-	rcu_read_lock();
-
-	end = (start + len - 1) / PAGE_SIZE;
-	xas_for_each(&xas, folio, end) {
-		if (!folio_test_writeback(folio)) {
-			kdebug("bad %x @%llx page %lx %lx",
-			       len, start, folio_index(folio), end);
-			ASSERT(folio_test_writeback(folio));
-		}
-
-		trace_afs_folio_dirty(vnode, tracepoint_string("clear"), folio);
-		folio_detach_private(folio);
-		folio_end_writeback(folio);
-	}
-
-	rcu_read_unlock();
-
 	afs_prune_wb_keys(vnode);
 	_leave("");
 }
@@ -451,363 +159,53 @@ try_next_key:
 	return afs_put_operation(op);
 }
 
-/*
- * Extend the region to be written back to include subsequent contiguously
- * dirty pages if possible, but don't sleep while doing so.
- *
- * If this page holds new content, then we can include filler zeros in the
- * writeback.
- */
-static void afs_extend_writeback(struct address_space *mapping,
-				 struct afs_vnode *vnode,
-				 long *_count,
-				 loff_t start,
-				 loff_t max_len,
-				 bool new_content,
-				 bool caching,
-				 unsigned int *_len)
+static void afs_upload_to_server(struct netfs_io_subrequest *subreq)
 {
-	struct folio_batch fbatch;
-	struct folio *folio;
-	unsigned long priv;
-	unsigned int psize, filler = 0;
-	unsigned int f, t;
-	loff_t len = *_len;
-	pgoff_t index = (start + len) / PAGE_SIZE;
-	bool stop = true;
-	unsigned int i;
-
-	XA_STATE(xas, &mapping->i_pages, index);
-	folio_batch_init(&fbatch);
-
-	do {
-		/* Firstly, we gather up a batch of contiguous dirty pages
-		 * under the RCU read lock - but we can't clear the dirty flags
-		 * there if any of those pages are mapped.
-		 */
-		rcu_read_lock();
-
-		xas_for_each(&xas, folio, ULONG_MAX) {
-			stop = true;
-			if (xas_retry(&xas, folio))
-				continue;
-			if (xa_is_value(folio))
-				break;
-			if (folio_index(folio) != index)
-				break;
-
-			if (!folio_try_get_rcu(folio)) {
-				xas_reset(&xas);
-				continue;
-			}
-
-			/* Has the page moved or been split? */
-			if (unlikely(folio != xas_reload(&xas))) {
-				folio_put(folio);
-				break;
-			}
-
-			if (!folio_trylock(folio)) {
-				folio_put(folio);
-				break;
-			}
-			if (!folio_test_dirty(folio) ||
-			    folio_test_writeback(folio) ||
-			    folio_test_fscache(folio)) {
-				folio_unlock(folio);
-				folio_put(folio);
-				break;
-			}
-
-			psize = folio_size(folio);
-			priv = (unsigned long)folio_get_private(folio);
-			f = afs_folio_dirty_from(folio, priv);
-			t = afs_folio_dirty_to(folio, priv);
-			if (f != 0 && !new_content) {
-				folio_unlock(folio);
-				folio_put(folio);
-				break;
-			}
-
-			len += filler + t;
-			filler = psize - t;
-			if (len >= max_len || *_count <= 0)
-				stop = true;
-			else if (t == psize || new_content)
-				stop = false;
-
-			index += folio_nr_pages(folio);
-			if (!folio_batch_add(&fbatch, folio))
-				break;
-			if (stop)
-				break;
-		}
-
-		if (!stop)
-			xas_pause(&xas);
-		rcu_read_unlock();
-
-		/* Now, if we obtained any folios, we can shift them to being
-		 * writable and mark them for caching.
-		 */
-		if (!folio_batch_count(&fbatch))
-			break;
-
-		for (i = 0; i < folio_batch_count(&fbatch); i++) {
-			folio = fbatch.folios[i];
-			trace_afs_folio_dirty(vnode, tracepoint_string("store+"), folio);
-
-			if (!folio_clear_dirty_for_io(folio))
-				BUG();
-			folio_start_writeback(folio);
-			afs_folio_start_fscache(caching, folio);
-
-			*_count -= folio_nr_pages(folio);
-			folio_unlock(folio);
-		}
+	struct afs_vnode *vnode = AFS_FS_I(subreq->rreq->inode);
+	ssize_t ret;
 
-		folio_batch_release(&fbatch);
-		cond_resched();
-	} while (!stop);
+	_enter("%x[%x],%zx",
+	       subreq->rreq->debug_id, subreq->debug_index, subreq->io_iter.count);
 
-	*_len = len;
+	trace_netfs_sreq(subreq, netfs_sreq_trace_submit);
+	ret = afs_store_data(vnode, &subreq->io_iter, subreq->start,
+			     subreq->rreq->origin == NETFS_LAUNDER_WRITE);
+	netfs_write_subrequest_terminated(subreq, ret < 0 ? ret : subreq->len,
+					  false);
 }
 
-/*
- * Synchronously write back the locked page and any subsequent non-locked dirty
- * pages.
- */
-static ssize_t afs_write_back_from_locked_folio(struct address_space *mapping,
-						struct writeback_control *wbc,
-						struct folio *folio,
-						loff_t start, loff_t end)
+static void afs_upload_to_server_worker(struct work_struct *work)
 {
-	struct afs_vnode *vnode = AFS_FS_I(mapping->host);
-	struct iov_iter iter;
-	unsigned long priv;
-	unsigned int offset, to, len, max_len;
-	loff_t i_size = i_size_read(&vnode->netfs.inode);
-	bool new_content = test_bit(AFS_VNODE_NEW_CONTENT, &vnode->flags);
-	bool caching = fscache_cookie_enabled(afs_vnode_cache(vnode));
-	long count = wbc->nr_to_write;
-	int ret;
-
-	_enter(",%lx,%llx-%llx", folio_index(folio), start, end);
-
-	folio_start_writeback(folio);
-	afs_folio_start_fscache(caching, folio);
-
-	count -= folio_nr_pages(folio);
-
-	/* Find all consecutive lockable dirty pages that have contiguous
-	 * written regions, stopping when we find a page that is not
-	 * immediately lockable, is not dirty or is missing, or we reach the
-	 * end of the range.
-	 */
-	priv = (unsigned long)folio_get_private(folio);
-	offset = afs_folio_dirty_from(folio, priv);
-	to = afs_folio_dirty_to(folio, priv);
-	trace_afs_folio_dirty(vnode, tracepoint_string("store"), folio);
-
-	len = to - offset;
-	start += offset;
-	if (start < i_size) {
-		/* Trim the write to the EOF; the extra data is ignored.  Also
-		 * put an upper limit on the size of a single storedata op.
-		 */
-		max_len = 65536 * 4096;
-		max_len = min_t(unsigned long long, max_len, end - start + 1);
-		max_len = min_t(unsigned long long, max_len, i_size - start);
-
-		if (len < max_len &&
-		    (to == folio_size(folio) || new_content))
-			afs_extend_writeback(mapping, vnode, &count,
-					     start, max_len, new_content,
-					     caching, &len);
-		len = min_t(loff_t, len, max_len);
-	}
-
-	/* We now have a contiguous set of dirty pages, each with writeback
-	 * set; the first page is still locked at this point, but all the rest
-	 * have been unlocked.
-	 */
-	folio_unlock(folio);
-
-	if (start < i_size) {
-		_debug("write back %x @%llx [%llx]", len, start, i_size);
-
-		/* Speculatively write to the cache.  We have to fix this up
-		 * later if the store fails.
-		 */
-		afs_write_to_cache(vnode, start, len, i_size, caching);
-
-		iov_iter_xarray(&iter, ITER_SOURCE, &mapping->i_pages, start, len);
-		ret = afs_store_data(vnode, &iter, start, false);
-	} else {
-		_debug("write discard %x @%llx [%llx]", len, start, i_size);
-
-		/* The dirty region was entirely beyond the EOF. */
-		fscache_clear_page_bits(mapping, start, len, caching);
-		afs_pages_written_back(vnode, start, len);
-		ret = 0;
-	}
-
-	switch (ret) {
-	case 0:
-		wbc->nr_to_write = count;
-		ret = len;
-		break;
+	struct netfs_io_subrequest *subreq =
+		container_of(work, struct netfs_io_subrequest, work);
 
-	default:
-		pr_notice("kAFS: Unexpected error from FS.StoreData %d\n", ret);
-		fallthrough;
-	case -EACCES:
-	case -EPERM:
-	case -ENOKEY:
-	case -EKEYEXPIRED:
-	case -EKEYREJECTED:
-	case -EKEYREVOKED:
-	case -ENETRESET:
-		afs_redirty_pages(wbc, mapping, start, len);
-		mapping_set_error(mapping, ret);
-		break;
-
-	case -EDQUOT:
-	case -ENOSPC:
-		afs_redirty_pages(wbc, mapping, start, len);
-		mapping_set_error(mapping, -ENOSPC);
-		break;
-
-	case -EROFS:
-	case -EIO:
-	case -EREMOTEIO:
-	case -EFBIG:
-	case -ENOENT:
-	case -ENOMEDIUM:
-	case -ENXIO:
-		trace_afs_file_error(vnode, ret, afs_file_error_writeback_fail);
-		afs_kill_pages(mapping, start, len);
-		mapping_set_error(mapping, ret);
-		break;
-	}
-
-	_leave(" = %d", ret);
-	return ret;
+	afs_upload_to_server(subreq);
 }
 
 /*
- * write a region of pages back to the server
+ * Set up write requests for a writeback slice.  We need to add a write request
+ * for each write we want to make.
  */
-static int afs_writepages_region(struct address_space *mapping,
-				 struct writeback_control *wbc,
-				 loff_t start, loff_t end, loff_t *_next,
-				 bool max_one_loop)
+void afs_create_write_requests(struct netfs_io_request *wreq, loff_t start, size_t len)
 {
-	struct folio *folio;
-	struct folio_batch fbatch;
-	ssize_t ret;
-	unsigned int i;
-	int n, skips = 0;
-
-	_enter("%llx,%llx,", start, end);
-	folio_batch_init(&fbatch);
-
-	do {
-		pgoff_t index = start / PAGE_SIZE;
-
-		n = filemap_get_folios_tag(mapping, &index, end / PAGE_SIZE,
-					PAGECACHE_TAG_DIRTY, &fbatch);
-
-		if (!n)
-			break;
-		for (i = 0; i < n; i++) {
-			folio = fbatch.folios[i];
-			start = folio_pos(folio); /* May regress with THPs */
-
-			_debug("wback %lx", folio_index(folio));
-
-			/* At this point we hold neither the i_pages lock nor the
-			 * page lock: the page may be truncated or invalidated
-			 * (changing page->mapping to NULL), or even swizzled
-			 * back from swapper_space to tmpfs file mapping
-			 */
-try_again:
-			if (wbc->sync_mode != WB_SYNC_NONE) {
-				ret = folio_lock_killable(folio);
-				if (ret < 0) {
-					folio_batch_release(&fbatch);
-					return ret;
-				}
-			} else {
-				if (!folio_trylock(folio))
-					continue;
-			}
-
-			if (folio->mapping != mapping ||
-			    !folio_test_dirty(folio)) {
-				start += folio_size(folio);
-				folio_unlock(folio);
-				continue;
-			}
-
-			if (folio_test_writeback(folio) ||
-			    folio_test_fscache(folio)) {
-				folio_unlock(folio);
-				if (wbc->sync_mode != WB_SYNC_NONE) {
-					folio_wait_writeback(folio);
-#ifdef CONFIG_AFS_FSCACHE
-					folio_wait_fscache(folio);
-#endif
-					goto try_again;
-				}
-
-				start += folio_size(folio);
-				if (wbc->sync_mode == WB_SYNC_NONE) {
-					if (skips >= 5 || need_resched()) {
-						*_next = start;
-						folio_batch_release(&fbatch);
-						_leave(" = 0 [%llx]", *_next);
-						return 0;
-					}
-					skips++;
-				}
-				continue;
-			}
-
-			if (!folio_clear_dirty_for_io(folio))
-				BUG();
-			ret = afs_write_back_from_locked_folio(mapping, wbc,
-					folio, start, end);
-			if (ret < 0) {
-				_leave(" = %zd", ret);
-				folio_batch_release(&fbatch);
-				return ret;
-			}
-
-			start += ret;
-		}
+	struct netfs_io_subrequest *subreq;
 
-		folio_batch_release(&fbatch);
-		cond_resched();
-	} while (wbc->nr_to_write > 0);
+	_enter("%x,%llx-%llx", wreq->debug_id, start, start + len);
 
-	*_next = start;
-	_leave(" = 0 [%llx]", *_next);
-	return 0;
+	subreq = netfs_create_write_request(wreq, NETFS_UPLOAD_TO_SERVER,
+					    start, len, afs_upload_to_server_worker);
+	if (subreq)
+		netfs_queue_write_request(subreq);
 }
 
 /*
  * write some of the pending data back to the server
  */
-int afs_writepages(struct address_space *mapping,
-		   struct writeback_control *wbc)
+int afs_writepages(struct address_space *mapping, struct writeback_control *wbc)
 {
 	struct afs_vnode *vnode = AFS_FS_I(mapping->host);
-	loff_t start, next;
 	int ret;
 
-	_enter("");
-
 	/* We have to be careful as we can end up racing with setattr()
 	 * truncating the pagecache since the caller doesn't take a lock here
 	 * to prevent it.
@@ -817,69 +215,12 @@ int afs_writepages(struct address_space *mapping,
 	else if (!down_read_trylock(&vnode->validate_lock))
 		return 0;
 
-	if (wbc->range_cyclic) {
-		start = mapping->writeback_index * PAGE_SIZE;
-		ret = afs_writepages_region(mapping, wbc, start, LLONG_MAX,
-					    &next, false);
-		if (ret == 0) {
-			mapping->writeback_index = next / PAGE_SIZE;
-			if (start > 0 && wbc->nr_to_write > 0) {
-				ret = afs_writepages_region(mapping, wbc, 0,
-							    start, &next, false);
-				if (ret == 0)
-					mapping->writeback_index =
-						next / PAGE_SIZE;
-			}
-		}
-	} else if (wbc->range_start == 0 && wbc->range_end == LLONG_MAX) {
-		ret = afs_writepages_region(mapping, wbc, 0, LLONG_MAX,
-					    &next, false);
-		if (wbc->nr_to_write > 0 && ret == 0)
-			mapping->writeback_index = next / PAGE_SIZE;
-	} else {
-		ret = afs_writepages_region(mapping, wbc,
-					    wbc->range_start, wbc->range_end,
-					    &next, false);
-	}
-
+	ret = netfs_writepages(mapping, wbc);
 	up_read(&vnode->validate_lock);
-	_leave(" = %d", ret);
 	return ret;
 }
 
 /*
- * write to an AFS file
- */
-ssize_t afs_file_write(struct kiocb *iocb, struct iov_iter *from)
-{
-	struct afs_vnode *vnode = AFS_FS_I(file_inode(iocb->ki_filp));
-	struct afs_file *af = iocb->ki_filp->private_data;
-	ssize_t result;
-	size_t count = iov_iter_count(from);
-
-	_enter("{%llx:%llu},{%zu},",
-	       vnode->fid.vid, vnode->fid.vnode, count);
-
-	if (IS_SWAPFILE(&vnode->netfs.inode)) {
-		printk(KERN_INFO
-		       "AFS: Attempt to write to active swap file!\n");
-		return -EBUSY;
-	}
-
-	if (!count)
-		return 0;
-
-	result = afs_validate(vnode, af->key);
-	if (result < 0)
-		return result;
-
-	result = generic_file_write_iter(iocb, from);
-
-	_leave(" = %zd", result);
-	return result;
-}
-
-/*
  * flush any dirty pages for this process, and check for write errors.
  * - the return status from this call provides a reliable indication of
  *   whether any write errors occurred for this process.
@@ -907,59 +248,11 @@ int afs_fsync(struct file *file, loff_t start, loff_t end, int datasync)
  */
 vm_fault_t afs_page_mkwrite(struct vm_fault *vmf)
 {
-	struct folio *folio = page_folio(vmf->page);
 	struct file *file = vmf->vma->vm_file;
-	struct inode *inode = file_inode(file);
-	struct afs_vnode *vnode = AFS_FS_I(inode);
-	struct afs_file *af = file->private_data;
-	unsigned long priv;
-	vm_fault_t ret = VM_FAULT_RETRY;
-
-	_enter("{{%llx:%llu}},{%lx}", vnode->fid.vid, vnode->fid.vnode, folio_index(folio));
-
-	afs_validate(vnode, af->key);
 
-	sb_start_pagefault(inode->i_sb);
-
-	/* Wait for the page to be written to the cache before we allow it to
-	 * be modified.  We then assume the entire page will need writing back.
-	 */
-#ifdef CONFIG_AFS_FSCACHE
-	if (folio_test_fscache(folio) &&
-	    folio_wait_fscache_killable(folio) < 0)
-		goto out;
-#endif
-
-	if (folio_wait_writeback_killable(folio))
-		goto out;
-
-	if (folio_lock_killable(folio) < 0)
-		goto out;
-
-	/* We mustn't change folio->private until writeback is complete as that
-	 * details the portion of the page we need to write back and we might
-	 * need to redirty the page if there's a problem.
-	 */
-	if (folio_wait_writeback_killable(folio) < 0) {
-		folio_unlock(folio);
-		goto out;
-	}
-
-	priv = afs_folio_dirty(folio, 0, folio_size(folio));
-	priv = afs_folio_dirty_mmapped(priv);
-	if (folio_test_private(folio)) {
-		folio_change_private(folio, (void *)priv);
-		trace_afs_folio_dirty(vnode, tracepoint_string("mkwrite+"), folio);
-	} else {
-		folio_attach_private(folio, (void *)priv);
-		trace_afs_folio_dirty(vnode, tracepoint_string("mkwrite"), folio);
-	}
-	file_update_time(file);
-
-	ret = VM_FAULT_LOCKED;
-out:
-	sb_end_pagefault(inode->i_sb);
-	return ret;
+	if (afs_validate(AFS_FS_I(file_inode(file)), afs_file_key(file)) < 0)
+		return VM_FAULT_SIGBUS;
+	return netfs_page_mkwrite(vmf, NULL);
 }
 
 /*
@@ -989,64 +282,3 @@ void afs_prune_wb_keys(struct afs_vnode *vnode)
 		afs_put_wb_key(wbk);
 	}
 }
-
-/*
- * Clean up a page during invalidation.
- */
-int afs_launder_folio(struct folio *folio)
-{
-	struct afs_vnode *vnode = AFS_FS_I(folio_inode(folio));
-	struct iov_iter iter;
-	struct bio_vec bv;
-	unsigned long priv;
-	unsigned int f, t;
-	int ret = 0;
-
-	_enter("{%lx}", folio->index);
-
-	priv = (unsigned long)folio_get_private(folio);
-	if (folio_clear_dirty_for_io(folio)) {
-		f = 0;
-		t = folio_size(folio);
-		if (folio_test_private(folio)) {
-			f = afs_folio_dirty_from(folio, priv);
-			t = afs_folio_dirty_to(folio, priv);
-		}
-
-		bvec_set_folio(&bv, folio, t - f, f);
-		iov_iter_bvec(&iter, ITER_SOURCE, &bv, 1, bv.bv_len);
-
-		trace_afs_folio_dirty(vnode, tracepoint_string("launder"), folio);
-		ret = afs_store_data(vnode, &iter, folio_pos(folio) + f, true);
-	}
-
-	trace_afs_folio_dirty(vnode, tracepoint_string("laundered"), folio);
-	folio_detach_private(folio);
-	folio_wait_fscache(folio);
-	return ret;
-}
-
-/*
- * Deal with the completion of writing the data to the cache.
- */
-static void afs_write_to_cache_done(void *priv, ssize_t transferred_or_error,
-				    bool was_async)
-{
-	struct afs_vnode *vnode = priv;
-
-	if (IS_ERR_VALUE(transferred_or_error) &&
-	    transferred_or_error != -ENOBUFS)
-		afs_invalidate_cache(vnode, 0);
-}
-
-/*
- * Save the write to the cache also.
- */
-static void afs_write_to_cache(struct afs_vnode *vnode,
-			       loff_t start, size_t len, loff_t i_size,
-			       bool caching)
-{
-	fscache_write_to_cache(afs_vnode_cache(vnode),
-			       vnode->netfs.inode.i_mapping, start, len, i_size,
-			       afs_write_to_cache_done, vnode, caching);
-}
diff --git a/fs/bcachefs/Makefile b/fs/bcachefs/Makefile
index 7423a3557c68..1a05cecda7cc 100644
--- a/fs/bcachefs/Makefile
+++ b/fs/bcachefs/Makefile
@@ -27,7 +27,6 @@ bcachefs-y		:=	\
 	checksum.o		\
 	clock.o			\
 	compress.o		\
-	counters.o		\
 	darray.o		\
 	debug.o			\
 	dirent.o		\
@@ -71,6 +70,7 @@ bcachefs-y		:=	\
 	reflink.o		\
 	replicas.o		\
 	sb-clean.o		\
+	sb-counters.o		\
 	sb-downgrade.o		\
 	sb-errors.o		\
 	sb-members.o		\
diff --git a/fs/bcachefs/alloc_background.c b/fs/bcachefs/alloc_background.c
index a09b9d00226a..10704f2d3af5 100644
--- a/fs/bcachefs/alloc_background.c
+++ b/fs/bcachefs/alloc_background.c
@@ -273,7 +273,7 @@ int bch2_alloc_v4_invalid(struct bch_fs *c, struct bkey_s_c k,
 		bkey_fsck_err_on(!bch2_bucket_sectors_dirty(*a.v),
 				 c, err, alloc_key_dirty_sectors_0,
 				 "data_type %s but dirty_sectors==0",
-				 bch2_data_types[a.v->data_type]);
+				 bch2_data_type_str(a.v->data_type));
 		break;
 	case BCH_DATA_cached:
 		bkey_fsck_err_on(!a.v->cached_sectors ||
@@ -321,16 +321,12 @@ void bch2_alloc_to_text(struct printbuf *out, struct bch_fs *c, struct bkey_s_c
 {
 	struct bch_alloc_v4 _a;
 	const struct bch_alloc_v4 *a = bch2_alloc_to_v4(k, &_a);
-	unsigned i;
 
 	prt_newline(out);
 	printbuf_indent_add(out, 2);
 
-	prt_printf(out, "gen %u oldest_gen %u data_type %s",
-	       a->gen, a->oldest_gen,
-	       a->data_type < BCH_DATA_NR
-	       ? bch2_data_types[a->data_type]
-	       : "(invalid data type)");
+	prt_printf(out, "gen %u oldest_gen %u data_type ", a->gen, a->oldest_gen);
+	bch2_prt_data_type(out, a->data_type);
 	prt_newline(out);
 	prt_printf(out, "journal_seq       %llu",	a->journal_seq);
 	prt_newline(out);
@@ -353,23 +349,6 @@ void bch2_alloc_to_text(struct printbuf *out, struct bch_fs *c, struct bkey_s_c
 	prt_printf(out, "fragmentation     %llu",	a->fragmentation_lru);
 	prt_newline(out);
 	prt_printf(out, "bp_start          %llu", BCH_ALLOC_V4_BACKPOINTERS_START(a));
-	prt_newline(out);
-
-	if (BCH_ALLOC_V4_NR_BACKPOINTERS(a)) {
-		struct bkey_s_c_alloc_v4 a_raw = bkey_s_c_to_alloc_v4(k);
-		const struct bch_backpointer *bps = alloc_v4_backpointers_c(a_raw.v);
-
-		prt_printf(out, "backpointers:     %llu", BCH_ALLOC_V4_NR_BACKPOINTERS(a_raw.v));
-		printbuf_indent_add(out, 2);
-
-		for (i = 0; i < BCH_ALLOC_V4_NR_BACKPOINTERS(a_raw.v); i++) {
-			prt_newline(out);
-			bch2_backpointer_to_text(out, &bps[i]);
-		}
-
-		printbuf_indent_sub(out, 2);
-	}
-
 	printbuf_indent_sub(out, 2);
 }
 
@@ -839,7 +818,7 @@ int bch2_trigger_alloc(struct btree_trans *trans,
 		}
 	}
 
-	if (!(flags & BTREE_TRIGGER_TRANSACTIONAL) && (flags & BTREE_TRIGGER_INSERT)) {
+	if ((flags & BTREE_TRIGGER_ATOMIC) && (flags & BTREE_TRIGGER_INSERT)) {
 		struct bch_alloc_v4 *new_a = bkey_s_to_alloc_v4(new).v;
 		u64 journal_seq = trans->journal_res.seq;
 		u64 bucket_journal_seq = new_a->journal_seq;
@@ -1625,13 +1604,36 @@ int bch2_check_alloc_to_lru_refs(struct bch_fs *c)
 	return ret;
 }
 
+struct discard_buckets_state {
+	u64		seen;
+	u64		open;
+	u64		need_journal_commit;
+	u64		discarded;
+	struct bch_dev	*ca;
+	u64		need_journal_commit_this_dev;
+};
+
+static void discard_buckets_next_dev(struct bch_fs *c, struct discard_buckets_state *s, struct bch_dev *ca)
+{
+	if (s->ca == ca)
+		return;
+
+	if (s->ca && s->need_journal_commit_this_dev >
+	    bch2_dev_usage_read(s->ca).d[BCH_DATA_free].buckets)
+		bch2_journal_flush_async(&c->journal, NULL);
+
+	if (s->ca)
+		percpu_ref_put(&s->ca->ref);
+	if (ca)
+		percpu_ref_get(&ca->ref);
+	s->ca = ca;
+	s->need_journal_commit_this_dev = 0;
+}
+
 static int bch2_discard_one_bucket(struct btree_trans *trans,
 				   struct btree_iter *need_discard_iter,
 				   struct bpos *discard_pos_done,
-				   u64 *seen,
-				   u64 *open,
-				   u64 *need_journal_commit,
-				   u64 *discarded)
+				   struct discard_buckets_state *s)
 {
 	struct bch_fs *c = trans->c;
 	struct bpos pos = need_discard_iter->pos;
@@ -1643,20 +1645,24 @@ static int bch2_discard_one_bucket(struct btree_trans *trans,
 	int ret = 0;
 
 	ca = bch_dev_bkey_exists(c, pos.inode);
+
 	if (!percpu_ref_tryget(&ca->io_ref)) {
 		bch2_btree_iter_set_pos(need_discard_iter, POS(pos.inode + 1, 0));
 		return 0;
 	}
 
+	discard_buckets_next_dev(c, s, ca);
+
 	if (bch2_bucket_is_open_safe(c, pos.inode, pos.offset)) {
-		(*open)++;
+		s->open++;
 		goto out;
 	}
 
 	if (bch2_bucket_needs_journal_commit(&c->buckets_waiting_for_journal,
 			c->journal.flushed_seq_ondisk,
 			pos.inode, pos.offset)) {
-		(*need_journal_commit)++;
+		s->need_journal_commit++;
+		s->need_journal_commit_this_dev++;
 		goto out;
 	}
 
@@ -1732,9 +1738,9 @@ write:
 		goto out;
 
 	count_event(c, bucket_discard);
-	(*discarded)++;
+	s->discarded++;
 out:
-	(*seen)++;
+	s->seen++;
 	bch2_trans_iter_exit(trans, &iter);
 	percpu_ref_put(&ca->io_ref);
 	printbuf_exit(&buf);
@@ -1744,7 +1750,7 @@ out:
 static void bch2_do_discards_work(struct work_struct *work)
 {
 	struct bch_fs *c = container_of(work, struct bch_fs, discard_work);
-	u64 seen = 0, open = 0, need_journal_commit = 0, discarded = 0;
+	struct discard_buckets_state s = {};
 	struct bpos discard_pos_done = POS_MAX;
 	int ret;
 
@@ -1756,19 +1762,14 @@ static void bch2_do_discards_work(struct work_struct *work)
 	ret = bch2_trans_run(c,
 		for_each_btree_key(trans, iter,
 				   BTREE_ID_need_discard, POS_MIN, 0, k,
-			bch2_discard_one_bucket(trans, &iter, &discard_pos_done,
-						&seen,
-						&open,
-						&need_journal_commit,
-						&discarded)));
-
-	if (need_journal_commit * 2 > seen)
-		bch2_journal_flush_async(&c->journal, NULL);
+			bch2_discard_one_bucket(trans, &iter, &discard_pos_done, &s)));
 
-	bch2_write_ref_put(c, BCH_WRITE_REF_discard);
+	discard_buckets_next_dev(c, &s, NULL);
 
-	trace_discard_buckets(c, seen, open, need_journal_commit, discarded,
+	trace_discard_buckets(c, s.seen, s.open, s.need_journal_commit, s.discarded,
 			      bch2_err_str(ret));
+
+	bch2_write_ref_put(c, BCH_WRITE_REF_discard);
 }
 
 void bch2_do_discards(struct bch_fs *c)
diff --git a/fs/bcachefs/alloc_background_format.h b/fs/bcachefs/alloc_background_format.h
new file mode 100644
index 000000000000..b4ec20be93b8
--- /dev/null
+++ b/fs/bcachefs/alloc_background_format.h
@@ -0,0 +1,92 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _BCACHEFS_ALLOC_BACKGROUND_FORMAT_H
+#define _BCACHEFS_ALLOC_BACKGROUND_FORMAT_H
+
+struct bch_alloc {
+	struct bch_val		v;
+	__u8			fields;
+	__u8			gen;
+	__u8			data[];
+} __packed __aligned(8);
+
+#define BCH_ALLOC_FIELDS_V1()			\
+	x(read_time,		16)		\
+	x(write_time,		16)		\
+	x(data_type,		8)		\
+	x(dirty_sectors,	16)		\
+	x(cached_sectors,	16)		\
+	x(oldest_gen,		8)		\
+	x(stripe,		32)		\
+	x(stripe_redundancy,	8)
+
+enum {
+#define x(name, _bits) BCH_ALLOC_FIELD_V1_##name,
+	BCH_ALLOC_FIELDS_V1()
+#undef x
+};
+
+struct bch_alloc_v2 {
+	struct bch_val		v;
+	__u8			nr_fields;
+	__u8			gen;
+	__u8			oldest_gen;
+	__u8			data_type;
+	__u8			data[];
+} __packed __aligned(8);
+
+#define BCH_ALLOC_FIELDS_V2()			\
+	x(read_time,		64)		\
+	x(write_time,		64)		\
+	x(dirty_sectors,	32)		\
+	x(cached_sectors,	32)		\
+	x(stripe,		32)		\
+	x(stripe_redundancy,	8)
+
+struct bch_alloc_v3 {
+	struct bch_val		v;
+	__le64			journal_seq;
+	__le32			flags;
+	__u8			nr_fields;
+	__u8			gen;
+	__u8			oldest_gen;
+	__u8			data_type;
+	__u8			data[];
+} __packed __aligned(8);
+
+LE32_BITMASK(BCH_ALLOC_V3_NEED_DISCARD,struct bch_alloc_v3, flags,  0,  1)
+LE32_BITMASK(BCH_ALLOC_V3_NEED_INC_GEN,struct bch_alloc_v3, flags,  1,  2)
+
+struct bch_alloc_v4 {
+	struct bch_val		v;
+	__u64			journal_seq;
+	__u32			flags;
+	__u8			gen;
+	__u8			oldest_gen;
+	__u8			data_type;
+	__u8			stripe_redundancy;
+	__u32			dirty_sectors;
+	__u32			cached_sectors;
+	__u64			io_time[2];
+	__u32			stripe;
+	__u32			nr_external_backpointers;
+	__u64			fragmentation_lru;
+} __packed __aligned(8);
+
+#define BCH_ALLOC_V4_U64s_V0	6
+#define BCH_ALLOC_V4_U64s	(sizeof(struct bch_alloc_v4) / sizeof(__u64))
+
+BITMASK(BCH_ALLOC_V4_NEED_DISCARD,	struct bch_alloc_v4, flags,  0,  1)
+BITMASK(BCH_ALLOC_V4_NEED_INC_GEN,	struct bch_alloc_v4, flags,  1,  2)
+BITMASK(BCH_ALLOC_V4_BACKPOINTERS_START,struct bch_alloc_v4, flags,  2,  8)
+BITMASK(BCH_ALLOC_V4_NR_BACKPOINTERS,	struct bch_alloc_v4, flags,  8,  14)
+
+#define KEY_TYPE_BUCKET_GENS_BITS	8
+#define KEY_TYPE_BUCKET_GENS_NR		(1U << KEY_TYPE_BUCKET_GENS_BITS)
+#define KEY_TYPE_BUCKET_GENS_MASK	(KEY_TYPE_BUCKET_GENS_NR - 1)
+
+struct bch_bucket_gens {
+	struct bch_val		v;
+	u8			gens[KEY_TYPE_BUCKET_GENS_NR];
+} __packed __aligned(8);
+
+#endif /* _BCACHEFS_ALLOC_BACKGROUND_FORMAT_H */
diff --git a/fs/bcachefs/alloc_foreground.c b/fs/bcachefs/alloc_foreground.c
index b0ff47998a94..633d3223b353 100644
--- a/fs/bcachefs/alloc_foreground.c
+++ b/fs/bcachefs/alloc_foreground.c
@@ -1525,10 +1525,11 @@ static void bch2_open_bucket_to_text(struct printbuf *out, struct bch_fs *c, str
 	unsigned data_type = ob->data_type;
 	barrier(); /* READ_ONCE() doesn't work on bitfields */
 
-	prt_printf(out, "%zu ref %u %s %u:%llu gen %u allocated %u/%u",
+	prt_printf(out, "%zu ref %u ",
 		   ob - c->open_buckets,
-		   atomic_read(&ob->pin),
-		   data_type < BCH_DATA_NR ? bch2_data_types[data_type] : "invalid data type",
+		   atomic_read(&ob->pin));
+	bch2_prt_data_type(out, data_type);
+	prt_printf(out, " %u:%llu gen %u allocated %u/%u",
 		   ob->dev, ob->bucket, ob->gen,
 		   ca->mi.bucket_size - ob->sectors_free, ca->mi.bucket_size);
 	if (ob->ec)
diff --git a/fs/bcachefs/backpointers.c b/fs/bcachefs/backpointers.c
index e358a2ffffde..b4dc319bcb2b 100644
--- a/fs/bcachefs/backpointers.c
+++ b/fs/bcachefs/backpointers.c
@@ -400,13 +400,24 @@ int bch2_check_btree_backpointers(struct bch_fs *c)
 	return ret;
 }
 
+static inline bool bkey_and_val_eq(struct bkey_s_c l, struct bkey_s_c r)
+{
+	return bpos_eq(l.k->p, r.k->p) &&
+		bkey_bytes(l.k) == bkey_bytes(r.k) &&
+		!memcmp(l.v, r.v, bkey_val_bytes(l.k));
+}
+
+struct extents_to_bp_state {
+	struct bpos	bucket_start;
+	struct bpos	bucket_end;
+	struct bkey_buf last_flushed;
+};
+
 static int check_bp_exists(struct btree_trans *trans,
+			   struct extents_to_bp_state *s,
 			   struct bpos bucket,
 			   struct bch_backpointer bp,
-			   struct bkey_s_c orig_k,
-			   struct bpos bucket_start,
-			   struct bpos bucket_end,
-			   struct bkey_buf *last_flushed)
+			   struct bkey_s_c orig_k)
 {
 	struct bch_fs *c = trans->c;
 	struct btree_iter bp_iter = { NULL };
@@ -417,8 +428,8 @@ static int check_bp_exists(struct btree_trans *trans,
 
 	bch2_bkey_buf_init(&tmp);
 
-	if (bpos_lt(bucket, bucket_start) ||
-	    bpos_gt(bucket, bucket_end))
+	if (bpos_lt(bucket, s->bucket_start) ||
+	    bpos_gt(bucket, s->bucket_end))
 		return 0;
 
 	if (!bch2_dev_bucket_exists(c, bucket))
@@ -433,11 +444,9 @@ static int check_bp_exists(struct btree_trans *trans,
 
 	if (bp_k.k->type != KEY_TYPE_backpointer ||
 	    memcmp(bkey_s_c_to_backpointer(bp_k).v, &bp, sizeof(bp))) {
-		if (!bpos_eq(orig_k.k->p, last_flushed->k->k.p) ||
-		    bkey_bytes(orig_k.k) != bkey_bytes(&last_flushed->k->k) ||
-		    memcmp(orig_k.v, &last_flushed->k->v, bkey_val_bytes(orig_k.k))) {
-			bch2_bkey_buf_reassemble(&tmp, c, orig_k);
+		bch2_bkey_buf_reassemble(&tmp, c, orig_k);
 
+		if (!bkey_and_val_eq(orig_k, bkey_i_to_s_c(s->last_flushed.k))) {
 			if (bp.level) {
 				bch2_trans_unlock(trans);
 				bch2_btree_interior_updates_flush(c);
@@ -447,7 +456,7 @@ static int check_bp_exists(struct btree_trans *trans,
 			if (ret)
 				goto err;
 
-			bch2_bkey_buf_copy(last_flushed, c, tmp.k);
+			bch2_bkey_buf_copy(&s->last_flushed, c, tmp.k);
 			ret = -BCH_ERR_transaction_restart_write_buffer_flush;
 			goto out;
 		}
@@ -475,10 +484,8 @@ missing:
 }
 
 static int check_extent_to_backpointers(struct btree_trans *trans,
+					struct extents_to_bp_state *s,
 					enum btree_id btree, unsigned level,
-					struct bpos bucket_start,
-					struct bpos bucket_end,
-					struct bkey_buf *last_flushed,
 					struct bkey_s_c k)
 {
 	struct bch_fs *c = trans->c;
@@ -498,9 +505,7 @@ static int check_extent_to_backpointers(struct btree_trans *trans,
 		bch2_extent_ptr_to_bp(c, btree, level,
 				      k, p, &bucket_pos, &bp);
 
-		ret = check_bp_exists(trans, bucket_pos, bp, k,
-				      bucket_start, bucket_end,
-				      last_flushed);
+		ret = check_bp_exists(trans, s, bucket_pos, bp, k);
 		if (ret)
 			return ret;
 	}
@@ -509,10 +514,8 @@ static int check_extent_to_backpointers(struct btree_trans *trans,
 }
 
 static int check_btree_root_to_backpointers(struct btree_trans *trans,
+					    struct extents_to_bp_state *s,
 					    enum btree_id btree_id,
-					    struct bpos bucket_start,
-					    struct bpos bucket_end,
-					    struct bkey_buf *last_flushed,
 					    int *level)
 {
 	struct bch_fs *c = trans->c;
@@ -536,9 +539,7 @@ retry:
 	*level = b->c.level;
 
 	k = bkey_i_to_s_c(&b->key);
-	ret = check_extent_to_backpointers(trans, btree_id, b->c.level + 1,
-				      bucket_start, bucket_end,
-				      last_flushed, k);
+	ret = check_extent_to_backpointers(trans, s, btree_id, b->c.level + 1, k);
 err:
 	bch2_trans_iter_exit(trans, &iter);
 	return ret;
@@ -559,7 +560,7 @@ static size_t btree_nodes_fit_in_ram(struct bch_fs *c)
 
 	si_meminfo(&i);
 	mem_bytes = i.totalram * i.mem_unit;
-	return div_u64(mem_bytes >> 1, btree_bytes(c));
+	return div_u64(mem_bytes >> 1, c->opts.btree_node_size);
 }
 
 static int bch2_get_btree_in_memory_pos(struct btree_trans *trans,
@@ -610,43 +611,35 @@ static int bch2_get_btree_in_memory_pos(struct btree_trans *trans,
 }
 
 static int bch2_check_extents_to_backpointers_pass(struct btree_trans *trans,
-						   struct bpos bucket_start,
-						   struct bpos bucket_end)
+						   struct extents_to_bp_state *s)
 {
 	struct bch_fs *c = trans->c;
-	struct btree_iter iter;
-	enum btree_id btree_id;
-	struct bkey_s_c k;
-	struct bkey_buf last_flushed;
 	int ret = 0;
 
-	bch2_bkey_buf_init(&last_flushed);
-	bkey_init(&last_flushed.k->k);
-
-	for (btree_id = 0; btree_id < btree_id_nr_alive(c); btree_id++) {
+	for (enum btree_id btree_id = 0;
+	     btree_id < btree_id_nr_alive(c);
+	     btree_id++) {
 		int level, depth = btree_type_has_ptrs(btree_id) ? 0 : 1;
 
 		ret = commit_do(trans, NULL, NULL,
 				BCH_TRANS_COMMIT_no_enospc,
-				check_btree_root_to_backpointers(trans, btree_id,
-							bucket_start, bucket_end,
-							&last_flushed, &level));
+				check_btree_root_to_backpointers(trans, s, btree_id, &level));
 		if (ret)
 			return ret;
 
 		while (level >= depth) {
+			struct btree_iter iter;
 			bch2_trans_node_iter_init(trans, &iter, btree_id, POS_MIN, 0,
 						  level,
 						  BTREE_ITER_PREFETCH);
 			while (1) {
 				bch2_trans_begin(trans);
-				k = bch2_btree_iter_peek(&iter);
+
+				struct bkey_s_c k = bch2_btree_iter_peek(&iter);
 				if (!k.k)
 					break;
 				ret = bkey_err(k) ?:
-					check_extent_to_backpointers(trans, btree_id, level,
-								     bucket_start, bucket_end,
-								     &last_flushed, k) ?:
+					check_extent_to_backpointers(trans, s, btree_id, level, k) ?:
 					bch2_trans_commit(trans, NULL, NULL,
 							  BCH_TRANS_COMMIT_no_enospc);
 				if (bch2_err_matches(ret, BCH_ERR_transaction_restart)) {
@@ -668,7 +661,6 @@ static int bch2_check_extents_to_backpointers_pass(struct btree_trans *trans,
 		}
 	}
 
-	bch2_bkey_buf_exit(&last_flushed, c);
 	return 0;
 }
 
@@ -731,37 +723,43 @@ static int bch2_get_alloc_in_memory_pos(struct btree_trans *trans,
 int bch2_check_extents_to_backpointers(struct bch_fs *c)
 {
 	struct btree_trans *trans = bch2_trans_get(c);
-	struct bpos start = POS_MIN, end;
+	struct extents_to_bp_state s = { .bucket_start = POS_MIN };
 	int ret;
 
+	bch2_bkey_buf_init(&s.last_flushed);
+	bkey_init(&s.last_flushed.k->k);
+
 	while (1) {
-		ret = bch2_get_alloc_in_memory_pos(trans, start, &end);
+		ret = bch2_get_alloc_in_memory_pos(trans, s.bucket_start, &s.bucket_end);
 		if (ret)
 			break;
 
-		if (bpos_eq(start, POS_MIN) && !bpos_eq(end, SPOS_MAX))
+		if ( bpos_eq(s.bucket_start, POS_MIN) &&
+		    !bpos_eq(s.bucket_end, SPOS_MAX))
 			bch_verbose(c, "%s(): alloc info does not fit in ram, running in multiple passes with %zu nodes per pass",
 				    __func__, btree_nodes_fit_in_ram(c));
 
-		if (!bpos_eq(start, POS_MIN) || !bpos_eq(end, SPOS_MAX)) {
+		if (!bpos_eq(s.bucket_start, POS_MIN) ||
+		    !bpos_eq(s.bucket_end, SPOS_MAX)) {
 			struct printbuf buf = PRINTBUF;
 
 			prt_str(&buf, "check_extents_to_backpointers(): ");
-			bch2_bpos_to_text(&buf, start);
+			bch2_bpos_to_text(&buf, s.bucket_start);
 			prt_str(&buf, "-");
-			bch2_bpos_to_text(&buf, end);
+			bch2_bpos_to_text(&buf, s.bucket_end);
 
 			bch_verbose(c, "%s", buf.buf);
 			printbuf_exit(&buf);
 		}
 
-		ret = bch2_check_extents_to_backpointers_pass(trans, start, end);
-		if (ret || bpos_eq(end, SPOS_MAX))
+		ret = bch2_check_extents_to_backpointers_pass(trans, &s);
+		if (ret || bpos_eq(s.bucket_end, SPOS_MAX))
 			break;
 
-		start = bpos_successor(end);
+		s.bucket_start = bpos_successor(s.bucket_end);
 	}
 	bch2_trans_put(trans);
+	bch2_bkey_buf_exit(&s.last_flushed, c);
 
 	bch_err_fn(c, ret);
 	return ret;
diff --git a/fs/bcachefs/backpointers.h b/fs/bcachefs/backpointers.h
index 737e2396ade7..327365a9feac 100644
--- a/fs/bcachefs/backpointers.h
+++ b/fs/bcachefs/backpointers.h
@@ -2,6 +2,7 @@
 #ifndef _BCACHEFS_BACKPOINTERS_BACKGROUND_H
 #define _BCACHEFS_BACKPOINTERS_BACKGROUND_H
 
+#include "btree_cache.h"
 #include "btree_iter.h"
 #include "btree_update.h"
 #include "buckets.h"
diff --git a/fs/bcachefs/bcachefs.h b/fs/bcachefs/bcachefs.h
index dac383e37181..b80c6c9efd8c 100644
--- a/fs/bcachefs/bcachefs.h
+++ b/fs/bcachefs/bcachefs.h
@@ -1204,11 +1204,6 @@ static inline unsigned block_sectors(const struct bch_fs *c)
 	return c->opts.block_size >> 9;
 }
 
-static inline size_t btree_sectors(const struct bch_fs *c)
-{
-	return c->opts.btree_node_size >> 9;
-}
-
 static inline bool btree_id_cached(const struct bch_fs *c, enum btree_id btree)
 {
 	return c->btree_key_cache_btrees & (1U << btree);
diff --git a/fs/bcachefs/bcachefs_format.h b/fs/bcachefs/bcachefs_format.h
index 0d5ac4184fbc..0668b682a21c 100644
--- a/fs/bcachefs/bcachefs_format.h
+++ b/fs/bcachefs/bcachefs_format.h
@@ -417,600 +417,12 @@ struct bch_set {
 	struct bch_val		v;
 };
 
-/* Extents */
-
-/*
- * In extent bkeys, the value is a list of pointers (bch_extent_ptr), optionally
- * preceded by checksum/compression information (bch_extent_crc32 or
- * bch_extent_crc64).
- *
- * One major determining factor in the format of extents is how we handle and
- * represent extents that have been partially overwritten and thus trimmed:
- *
- * If an extent is not checksummed or compressed, when the extent is trimmed we
- * don't have to remember the extent we originally allocated and wrote: we can
- * merely adjust ptr->offset to point to the start of the data that is currently
- * live. The size field in struct bkey records the current (live) size of the
- * extent, and is also used to mean "size of region on disk that we point to" in
- * this case.
- *
- * Thus an extent that is not checksummed or compressed will consist only of a
- * list of bch_extent_ptrs, with none of the fields in
- * bch_extent_crc32/bch_extent_crc64.
- *
- * When an extent is checksummed or compressed, it's not possible to read only
- * the data that is currently live: we have to read the entire extent that was
- * originally written, and then return only the part of the extent that is
- * currently live.
- *
- * Thus, in addition to the current size of the extent in struct bkey, we need
- * to store the size of the originally allocated space - this is the
- * compressed_size and uncompressed_size fields in bch_extent_crc32/64. Also,
- * when the extent is trimmed, instead of modifying the offset field of the
- * pointer, we keep a second smaller offset field - "offset into the original
- * extent of the currently live region".
- *
- * The other major determining factor is replication and data migration:
- *
- * Each pointer may have its own bch_extent_crc32/64. When doing a replicated
- * write, we will initially write all the replicas in the same format, with the
- * same checksum type and compression format - however, when copygc runs later (or
- * tiering/cache promotion, anything that moves data), it is not in general
- * going to rewrite all the pointers at once - one of the replicas may be in a
- * bucket on one device that has very little fragmentation while another lives
- * in a bucket that has become heavily fragmented, and thus is being rewritten
- * sooner than the rest.
- *
- * Thus it will only move a subset of the pointers (or in the case of
- * tiering/cache promotion perhaps add a single pointer without dropping any
- * current pointers), and if the extent has been partially overwritten it must
- * write only the currently live portion (or copygc would not be able to reduce
- * fragmentation!) - which necessitates a different bch_extent_crc format for
- * the new pointer.
- *
- * But in the interests of space efficiency, we don't want to store one
- * bch_extent_crc for each pointer if we don't have to.
- *
- * Thus, a bch_extent consists of bch_extent_crc32s, bch_extent_crc64s, and
- * bch_extent_ptrs appended arbitrarily one after the other. We determine the
- * type of a given entry with a scheme similar to utf8 (except we're encoding a
- * type, not a size), encoding the type in the position of the first set bit:
- *
- * bch_extent_crc32	- 0b1
- * bch_extent_ptr	- 0b10
- * bch_extent_crc64	- 0b100
- *
- * We do it this way because bch_extent_crc32 is _very_ constrained on bits (and
- * bch_extent_crc64 is the least constrained).
- *
- * Then, each bch_extent_crc32/64 applies to the pointers that follow after it,
- * until the next bch_extent_crc32/64.
- *
- * If there are no bch_extent_crcs preceding a bch_extent_ptr, then that pointer
- * is neither checksummed nor compressed.
- */
-
 /* 128 bits, sufficient for cryptographic MACs: */
 struct bch_csum {
 	__le64			lo;
 	__le64			hi;
 } __packed __aligned(8);
 
-#define BCH_EXTENT_ENTRY_TYPES()		\
-	x(ptr,			0)		\
-	x(crc32,		1)		\
-	x(crc64,		2)		\
-	x(crc128,		3)		\
-	x(stripe_ptr,		4)		\
-	x(rebalance,		5)
-#define BCH_EXTENT_ENTRY_MAX	6
-
-enum bch_extent_entry_type {
-#define x(f, n) BCH_EXTENT_ENTRY_##f = n,
-	BCH_EXTENT_ENTRY_TYPES()
-#undef x
-};
-
-/* Compressed/uncompressed size are stored biased by 1: */
-struct bch_extent_crc32 {
-#if defined(__LITTLE_ENDIAN_BITFIELD)
-	__u32			type:2,
-				_compressed_size:7,
-				_uncompressed_size:7,
-				offset:7,
-				_unused:1,
-				csum_type:4,
-				compression_type:4;
-	__u32			csum;
-#elif defined (__BIG_ENDIAN_BITFIELD)
-	__u32			csum;
-	__u32			compression_type:4,
-				csum_type:4,
-				_unused:1,
-				offset:7,
-				_uncompressed_size:7,
-				_compressed_size:7,
-				type:2;
-#endif
-} __packed __aligned(8);
-
-#define CRC32_SIZE_MAX		(1U << 7)
-#define CRC32_NONCE_MAX		0
-
-struct bch_extent_crc64 {
-#if defined(__LITTLE_ENDIAN_BITFIELD)
-	__u64			type:3,
-				_compressed_size:9,
-				_uncompressed_size:9,
-				offset:9,
-				nonce:10,
-				csum_type:4,
-				compression_type:4,
-				csum_hi:16;
-#elif defined (__BIG_ENDIAN_BITFIELD)
-	__u64			csum_hi:16,
-				compression_type:4,
-				csum_type:4,
-				nonce:10,
-				offset:9,
-				_uncompressed_size:9,
-				_compressed_size:9,
-				type:3;
-#endif
-	__u64			csum_lo;
-} __packed __aligned(8);
-
-#define CRC64_SIZE_MAX		(1U << 9)
-#define CRC64_NONCE_MAX		((1U << 10) - 1)
-
-struct bch_extent_crc128 {
-#if defined(__LITTLE_ENDIAN_BITFIELD)
-	__u64			type:4,
-				_compressed_size:13,
-				_uncompressed_size:13,
-				offset:13,
-				nonce:13,
-				csum_type:4,
-				compression_type:4;
-#elif defined (__BIG_ENDIAN_BITFIELD)
-	__u64			compression_type:4,
-				csum_type:4,
-				nonce:13,
-				offset:13,
-				_uncompressed_size:13,
-				_compressed_size:13,
-				type:4;
-#endif
-	struct bch_csum		csum;
-} __packed __aligned(8);
-
-#define CRC128_SIZE_MAX		(1U << 13)
-#define CRC128_NONCE_MAX	((1U << 13) - 1)
-
-/*
- * @reservation - pointer hasn't been written to, just reserved
- */
-struct bch_extent_ptr {
-#if defined(__LITTLE_ENDIAN_BITFIELD)
-	__u64			type:1,
-				cached:1,
-				unused:1,
-				unwritten:1,
-				offset:44, /* 8 petabytes */
-				dev:8,
-				gen:8;
-#elif defined (__BIG_ENDIAN_BITFIELD)
-	__u64			gen:8,
-				dev:8,
-				offset:44,
-				unwritten:1,
-				unused:1,
-				cached:1,
-				type:1;
-#endif
-} __packed __aligned(8);
-
-struct bch_extent_stripe_ptr {
-#if defined(__LITTLE_ENDIAN_BITFIELD)
-	__u64			type:5,
-				block:8,
-				redundancy:4,
-				idx:47;
-#elif defined (__BIG_ENDIAN_BITFIELD)
-	__u64			idx:47,
-				redundancy:4,
-				block:8,
-				type:5;
-#endif
-};
-
-struct bch_extent_rebalance {
-#if defined(__LITTLE_ENDIAN_BITFIELD)
-	__u64			type:6,
-				unused:34,
-				compression:8, /* enum bch_compression_opt */
-				target:16;
-#elif defined (__BIG_ENDIAN_BITFIELD)
-	__u64			target:16,
-				compression:8,
-				unused:34,
-				type:6;
-#endif
-};
-
-union bch_extent_entry {
-#if __BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__ ||  __BITS_PER_LONG == 64
-	unsigned long			type;
-#elif __BITS_PER_LONG == 32
-	struct {
-		unsigned long		pad;
-		unsigned long		type;
-	};
-#else
-#error edit for your odd byteorder.
-#endif
-
-#define x(f, n) struct bch_extent_##f	f;
-	BCH_EXTENT_ENTRY_TYPES()
-#undef x
-};
-
-struct bch_btree_ptr {
-	struct bch_val		v;
-
-	__u64			_data[0];
-	struct bch_extent_ptr	start[];
-} __packed __aligned(8);
-
-struct bch_btree_ptr_v2 {
-	struct bch_val		v;
-
-	__u64			mem_ptr;
-	__le64			seq;
-	__le16			sectors_written;
-	__le16			flags;
-	struct bpos		min_key;
-	__u64			_data[0];
-	struct bch_extent_ptr	start[];
-} __packed __aligned(8);
-
-LE16_BITMASK(BTREE_PTR_RANGE_UPDATED,	struct bch_btree_ptr_v2, flags, 0, 1);
-
-struct bch_extent {
-	struct bch_val		v;
-
-	__u64			_data[0];
-	union bch_extent_entry	start[];
-} __packed __aligned(8);
-
-struct bch_reservation {
-	struct bch_val		v;
-
-	__le32			generation;
-	__u8			nr_replicas;
-	__u8			pad[3];
-} __packed __aligned(8);
-
-/* Maximum size (in u64s) a single pointer could be: */
-#define BKEY_EXTENT_PTR_U64s_MAX\
-	((sizeof(struct bch_extent_crc128) +			\
-	  sizeof(struct bch_extent_ptr)) / sizeof(__u64))
-
-/* Maximum possible size of an entire extent value: */
-#define BKEY_EXTENT_VAL_U64s_MAX				\
-	(1 + BKEY_EXTENT_PTR_U64s_MAX * (BCH_REPLICAS_MAX + 1))
-
-/* * Maximum possible size of an entire extent, key + value: */
-#define BKEY_EXTENT_U64s_MAX		(BKEY_U64s + BKEY_EXTENT_VAL_U64s_MAX)
-
-/* Btree pointers don't carry around checksums: */
-#define BKEY_BTREE_PTR_VAL_U64s_MAX				\
-	((sizeof(struct bch_btree_ptr_v2) +			\
-	  sizeof(struct bch_extent_ptr) * BCH_REPLICAS_MAX) / sizeof(__u64))
-#define BKEY_BTREE_PTR_U64s_MAX					\
-	(BKEY_U64s + BKEY_BTREE_PTR_VAL_U64s_MAX)
-
-/* Inodes */
-
-#define BLOCKDEV_INODE_MAX	4096
-
-#define BCACHEFS_ROOT_INO	4096
-
-struct bch_inode {
-	struct bch_val		v;
-
-	__le64			bi_hash_seed;
-	__le32			bi_flags;
-	__le16			bi_mode;
-	__u8			fields[];
-} __packed __aligned(8);
-
-struct bch_inode_v2 {
-	struct bch_val		v;
-
-	__le64			bi_journal_seq;
-	__le64			bi_hash_seed;
-	__le64			bi_flags;
-	__le16			bi_mode;
-	__u8			fields[];
-} __packed __aligned(8);
-
-struct bch_inode_v3 {
-	struct bch_val		v;
-
-	__le64			bi_journal_seq;
-	__le64			bi_hash_seed;
-	__le64			bi_flags;
-	__le64			bi_sectors;
-	__le64			bi_size;
-	__le64			bi_version;
-	__u8			fields[];
-} __packed __aligned(8);
-
-#define INODEv3_FIELDS_START_INITIAL	6
-#define INODEv3_FIELDS_START_CUR	(offsetof(struct bch_inode_v3, fields) / sizeof(__u64))
-
-struct bch_inode_generation {
-	struct bch_val		v;
-
-	__le32			bi_generation;
-	__le32			pad;
-} __packed __aligned(8);
-
-/*
- * bi_subvol and bi_parent_subvol are only set for subvolume roots:
- */
-
-#define BCH_INODE_FIELDS_v2()			\
-	x(bi_atime,			96)	\
-	x(bi_ctime,			96)	\
-	x(bi_mtime,			96)	\
-	x(bi_otime,			96)	\
-	x(bi_size,			64)	\
-	x(bi_sectors,			64)	\
-	x(bi_uid,			32)	\
-	x(bi_gid,			32)	\
-	x(bi_nlink,			32)	\
-	x(bi_generation,		32)	\
-	x(bi_dev,			32)	\
-	x(bi_data_checksum,		8)	\
-	x(bi_compression,		8)	\
-	x(bi_project,			32)	\
-	x(bi_background_compression,	8)	\
-	x(bi_data_replicas,		8)	\
-	x(bi_promote_target,		16)	\
-	x(bi_foreground_target,		16)	\
-	x(bi_background_target,		16)	\
-	x(bi_erasure_code,		16)	\
-	x(bi_fields_set,		16)	\
-	x(bi_dir,			64)	\
-	x(bi_dir_offset,		64)	\
-	x(bi_subvol,			32)	\
-	x(bi_parent_subvol,		32)
-
-#define BCH_INODE_FIELDS_v3()			\
-	x(bi_atime,			96)	\
-	x(bi_ctime,			96)	\
-	x(bi_mtime,			96)	\
-	x(bi_otime,			96)	\
-	x(bi_uid,			32)	\
-	x(bi_gid,			32)	\
-	x(bi_nlink,			32)	\
-	x(bi_generation,		32)	\
-	x(bi_dev,			32)	\
-	x(bi_data_checksum,		8)	\
-	x(bi_compression,		8)	\
-	x(bi_project,			32)	\
-	x(bi_background_compression,	8)	\
-	x(bi_data_replicas,		8)	\
-	x(bi_promote_target,		16)	\
-	x(bi_foreground_target,		16)	\
-	x(bi_background_target,		16)	\
-	x(bi_erasure_code,		16)	\
-	x(bi_fields_set,		16)	\
-	x(bi_dir,			64)	\
-	x(bi_dir_offset,		64)	\
-	x(bi_subvol,			32)	\
-	x(bi_parent_subvol,		32)	\
-	x(bi_nocow,			8)
-
-/* subset of BCH_INODE_FIELDS */
-#define BCH_INODE_OPTS()			\
-	x(data_checksum,		8)	\
-	x(compression,			8)	\
-	x(project,			32)	\
-	x(background_compression,	8)	\
-	x(data_replicas,		8)	\
-	x(promote_target,		16)	\
-	x(foreground_target,		16)	\
-	x(background_target,		16)	\
-	x(erasure_code,			16)	\
-	x(nocow,			8)
-
-enum inode_opt_id {
-#define x(name, ...)				\
-	Inode_opt_##name,
-	BCH_INODE_OPTS()
-#undef  x
-	Inode_opt_nr,
-};
-
-#define BCH_INODE_FLAGS()			\
-	x(sync,				0)	\
-	x(immutable,			1)	\
-	x(append,			2)	\
-	x(nodump,			3)	\
-	x(noatime,			4)	\
-	x(i_size_dirty,			5)	\
-	x(i_sectors_dirty,		6)	\
-	x(unlinked,			7)	\
-	x(backptr_untrusted,		8)
-
-/* bits 20+ reserved for packed fields below: */
-
-enum bch_inode_flags {
-#define x(t, n)	BCH_INODE_##t = 1U << n,
-	BCH_INODE_FLAGS()
-#undef x
-};
-
-enum __bch_inode_flags {
-#define x(t, n)	__BCH_INODE_##t = n,
-	BCH_INODE_FLAGS()
-#undef x
-};
-
-LE32_BITMASK(INODE_STR_HASH,	struct bch_inode, bi_flags, 20, 24);
-LE32_BITMASK(INODE_NR_FIELDS,	struct bch_inode, bi_flags, 24, 31);
-LE32_BITMASK(INODE_NEW_VARINT,	struct bch_inode, bi_flags, 31, 32);
-
-LE64_BITMASK(INODEv2_STR_HASH,	struct bch_inode_v2, bi_flags, 20, 24);
-LE64_BITMASK(INODEv2_NR_FIELDS,	struct bch_inode_v2, bi_flags, 24, 31);
-
-LE64_BITMASK(INODEv3_STR_HASH,	struct bch_inode_v3, bi_flags, 20, 24);
-LE64_BITMASK(INODEv3_NR_FIELDS,	struct bch_inode_v3, bi_flags, 24, 31);
-
-LE64_BITMASK(INODEv3_FIELDS_START,
-				struct bch_inode_v3, bi_flags, 31, 36);
-LE64_BITMASK(INODEv3_MODE,	struct bch_inode_v3, bi_flags, 36, 52);
-
-/* Dirents */
-
-/*
- * Dirents (and xattrs) have to implement string lookups; since our b-tree
- * doesn't support arbitrary length strings for the key, we instead index by a
- * 64 bit hash (currently truncated sha1) of the string, stored in the offset
- * field of the key - using linear probing to resolve hash collisions. This also
- * provides us with the readdir cookie posix requires.
- *
- * Linear probing requires us to use whiteouts for deletions, in the event of a
- * collision:
- */
-
-struct bch_dirent {
-	struct bch_val		v;
-
-	/* Target inode number: */
-	union {
-	__le64			d_inum;
-	struct {		/* DT_SUBVOL */
-	__le32			d_child_subvol;
-	__le32			d_parent_subvol;
-	};
-	};
-
-	/*
-	 * Copy of mode bits 12-15 from the target inode - so userspace can get
-	 * the filetype without having to do a stat()
-	 */
-	__u8			d_type;
-
-	__u8			d_name[];
-} __packed __aligned(8);
-
-#define DT_SUBVOL	16
-#define BCH_DT_MAX	17
-
-#define BCH_NAME_MAX	512
-
-/* Xattrs */
-
-#define KEY_TYPE_XATTR_INDEX_USER			0
-#define KEY_TYPE_XATTR_INDEX_POSIX_ACL_ACCESS	1
-#define KEY_TYPE_XATTR_INDEX_POSIX_ACL_DEFAULT	2
-#define KEY_TYPE_XATTR_INDEX_TRUSTED			3
-#define KEY_TYPE_XATTR_INDEX_SECURITY	        4
-
-struct bch_xattr {
-	struct bch_val		v;
-	__u8			x_type;
-	__u8			x_name_len;
-	__le16			x_val_len;
-	__u8			x_name[];
-} __packed __aligned(8);
-
-/* Bucket/allocation information: */
-
-struct bch_alloc {
-	struct bch_val		v;
-	__u8			fields;
-	__u8			gen;
-	__u8			data[];
-} __packed __aligned(8);
-
-#define BCH_ALLOC_FIELDS_V1()			\
-	x(read_time,		16)		\
-	x(write_time,		16)		\
-	x(data_type,		8)		\
-	x(dirty_sectors,	16)		\
-	x(cached_sectors,	16)		\
-	x(oldest_gen,		8)		\
-	x(stripe,		32)		\
-	x(stripe_redundancy,	8)
-
-enum {
-#define x(name, _bits) BCH_ALLOC_FIELD_V1_##name,
-	BCH_ALLOC_FIELDS_V1()
-#undef x
-};
-
-struct bch_alloc_v2 {
-	struct bch_val		v;
-	__u8			nr_fields;
-	__u8			gen;
-	__u8			oldest_gen;
-	__u8			data_type;
-	__u8			data[];
-} __packed __aligned(8);
-
-#define BCH_ALLOC_FIELDS_V2()			\
-	x(read_time,		64)		\
-	x(write_time,		64)		\
-	x(dirty_sectors,	32)		\
-	x(cached_sectors,	32)		\
-	x(stripe,		32)		\
-	x(stripe_redundancy,	8)
-
-struct bch_alloc_v3 {
-	struct bch_val		v;
-	__le64			journal_seq;
-	__le32			flags;
-	__u8			nr_fields;
-	__u8			gen;
-	__u8			oldest_gen;
-	__u8			data_type;
-	__u8			data[];
-} __packed __aligned(8);
-
-LE32_BITMASK(BCH_ALLOC_V3_NEED_DISCARD,struct bch_alloc_v3, flags,  0,  1)
-LE32_BITMASK(BCH_ALLOC_V3_NEED_INC_GEN,struct bch_alloc_v3, flags,  1,  2)
-
-struct bch_alloc_v4 {
-	struct bch_val		v;
-	__u64			journal_seq;
-	__u32			flags;
-	__u8			gen;
-	__u8			oldest_gen;
-	__u8			data_type;
-	__u8			stripe_redundancy;
-	__u32			dirty_sectors;
-	__u32			cached_sectors;
-	__u64			io_time[2];
-	__u32			stripe;
-	__u32			nr_external_backpointers;
-	__u64			fragmentation_lru;
-} __packed __aligned(8);
-
-#define BCH_ALLOC_V4_U64s_V0	6
-#define BCH_ALLOC_V4_U64s	(sizeof(struct bch_alloc_v4) / sizeof(__u64))
-
-BITMASK(BCH_ALLOC_V4_NEED_DISCARD,	struct bch_alloc_v4, flags,  0,  1)
-BITMASK(BCH_ALLOC_V4_NEED_INC_GEN,	struct bch_alloc_v4, flags,  1,  2)
-BITMASK(BCH_ALLOC_V4_BACKPOINTERS_START,struct bch_alloc_v4, flags,  2,  8)
-BITMASK(BCH_ALLOC_V4_NR_BACKPOINTERS,	struct bch_alloc_v4, flags,  8,  14)
-
-#define BCH_ALLOC_V4_NR_BACKPOINTERS_MAX	40
-
 struct bch_backpointer {
 	struct bch_val		v;
 	__u8			btree_id;
@@ -1021,154 +433,6 @@ struct bch_backpointer {
 	struct bpos		pos;
 } __packed __aligned(8);
 
-#define KEY_TYPE_BUCKET_GENS_BITS	8
-#define KEY_TYPE_BUCKET_GENS_NR		(1U << KEY_TYPE_BUCKET_GENS_BITS)
-#define KEY_TYPE_BUCKET_GENS_MASK	(KEY_TYPE_BUCKET_GENS_NR - 1)
-
-struct bch_bucket_gens {
-	struct bch_val		v;
-	u8			gens[KEY_TYPE_BUCKET_GENS_NR];
-} __packed __aligned(8);
-
-/* Quotas: */
-
-enum quota_types {
-	QTYP_USR		= 0,
-	QTYP_GRP		= 1,
-	QTYP_PRJ		= 2,
-	QTYP_NR			= 3,
-};
-
-enum quota_counters {
-	Q_SPC			= 0,
-	Q_INO			= 1,
-	Q_COUNTERS		= 2,
-};
-
-struct bch_quota_counter {
-	__le64			hardlimit;
-	__le64			softlimit;
-};
-
-struct bch_quota {
-	struct bch_val		v;
-	struct bch_quota_counter c[Q_COUNTERS];
-} __packed __aligned(8);
-
-/* Erasure coding */
-
-struct bch_stripe {
-	struct bch_val		v;
-	__le16			sectors;
-	__u8			algorithm;
-	__u8			nr_blocks;
-	__u8			nr_redundant;
-
-	__u8			csum_granularity_bits;
-	__u8			csum_type;
-	__u8			pad;
-
-	struct bch_extent_ptr	ptrs[];
-} __packed __aligned(8);
-
-/* Reflink: */
-
-struct bch_reflink_p {
-	struct bch_val		v;
-	__le64			idx;
-	/*
-	 * A reflink pointer might point to an indirect extent which is then
-	 * later split (by copygc or rebalance). If we only pointed to part of
-	 * the original indirect extent, and then one of the fragments is
-	 * outside the range we point to, we'd leak a refcount: so when creating
-	 * reflink pointers, we need to store pad values to remember the full
-	 * range we were taking a reference on.
-	 */
-	__le32			front_pad;
-	__le32			back_pad;
-} __packed __aligned(8);
-
-struct bch_reflink_v {
-	struct bch_val		v;
-	__le64			refcount;
-	union bch_extent_entry	start[0];
-	__u64			_data[];
-} __packed __aligned(8);
-
-struct bch_indirect_inline_data {
-	struct bch_val		v;
-	__le64			refcount;
-	u8			data[];
-};
-
-/* Inline data */
-
-struct bch_inline_data {
-	struct bch_val		v;
-	u8			data[];
-};
-
-/* Subvolumes: */
-
-#define SUBVOL_POS_MIN		POS(0, 1)
-#define SUBVOL_POS_MAX		POS(0, S32_MAX)
-#define BCACHEFS_ROOT_SUBVOL	1
-
-struct bch_subvolume {
-	struct bch_val		v;
-	__le32			flags;
-	__le32			snapshot;
-	__le64			inode;
-	/*
-	 * Snapshot subvolumes form a tree, separate from the snapshot nodes
-	 * tree - if this subvolume is a snapshot, this is the ID of the
-	 * subvolume it was created from:
-	 */
-	__le32			parent;
-	__le32			pad;
-	bch_le128		otime;
-};
-
-LE32_BITMASK(BCH_SUBVOLUME_RO,		struct bch_subvolume, flags,  0,  1)
-/*
- * We need to know whether a subvolume is a snapshot so we can know whether we
- * can delete it (or whether it should just be rm -rf'd)
- */
-LE32_BITMASK(BCH_SUBVOLUME_SNAP,	struct bch_subvolume, flags,  1,  2)
-LE32_BITMASK(BCH_SUBVOLUME_UNLINKED,	struct bch_subvolume, flags,  2,  3)
-
-/* Snapshots */
-
-struct bch_snapshot {
-	struct bch_val		v;
-	__le32			flags;
-	__le32			parent;
-	__le32			children[2];
-	__le32			subvol;
-	/* corresponds to a bch_snapshot_tree in BTREE_ID_snapshot_trees */
-	__le32			tree;
-	__le32			depth;
-	__le32			skip[3];
-};
-
-LE32_BITMASK(BCH_SNAPSHOT_DELETED,	struct bch_snapshot, flags,  0,  1)
-
-/* True if a subvolume points to this snapshot node: */
-LE32_BITMASK(BCH_SNAPSHOT_SUBVOL,	struct bch_snapshot, flags,  1,  2)
-
-/*
- * Snapshot trees:
- *
- * The snapshot_trees btree gives us persistent indentifier for each tree of
- * bch_snapshot nodes, and allow us to record and easily find the root/master
- * subvolume that other snapshots were created from:
- */
-struct bch_snapshot_tree {
-	struct bch_val		v;
-	__le32			master_subvol;
-	__le32			root_snapshot;
-};
-
 /* LRU btree: */
 
 struct bch_lru {
@@ -1178,33 +442,6 @@ struct bch_lru {
 
 #define LRU_ID_STRIPES		(1U << 16)
 
-/* Logged operations btree: */
-
-struct bch_logged_op_truncate {
-	struct bch_val		v;
-	__le32			subvol;
-	__le32			pad;
-	__le64			inum;
-	__le64			new_i_size;
-};
-
-enum logged_op_finsert_state {
-	LOGGED_OP_FINSERT_start,
-	LOGGED_OP_FINSERT_shift_extents,
-	LOGGED_OP_FINSERT_finish,
-};
-
-struct bch_logged_op_finsert {
-	struct bch_val		v;
-	__u8			state;
-	__u8			pad[3];
-	__le32			subvol;
-	__le64			inum;
-	__le64			dst_offset;
-	__le64			src_offset;
-	__le64			pos;
-};
-
 /* Optional/variable size superblock sections: */
 
 struct bch_sb_field {
@@ -1230,6 +467,19 @@ struct bch_sb_field {
 	x(ext,				13)	\
 	x(downgrade,			14)
 
+#include "alloc_background_format.h"
+#include "extents_format.h"
+#include "reflink_format.h"
+#include "ec_format.h"
+#include "inode_format.h"
+#include "dirent_format.h"
+#include "xattr_format.h"
+#include "quota_format.h"
+#include "logged_ops_format.h"
+#include "snapshot_format.h"
+#include "subvolume_format.h"
+#include "sb-counters_format.h"
+
 enum bch_sb_field_type {
 #define x(f, nr)	BCH_SB_FIELD_##f = nr,
 	BCH_SB_FIELDS()
@@ -1465,23 +715,6 @@ struct bch_sb_field_replicas {
 	struct bch_replicas_entry_v1 entries[];
 } __packed __aligned(8);
 
-/* BCH_SB_FIELD_quota: */
-
-struct bch_sb_quota_counter {
-	__le32				timelimit;
-	__le32				warnlimit;
-};
-
-struct bch_sb_quota_type {
-	__le64				flags;
-	struct bch_sb_quota_counter	c[Q_COUNTERS];
-};
-
-struct bch_sb_field_quota {
-	struct bch_sb_field		field;
-	struct bch_sb_quota_type	q[QTYP_NR];
-} __packed __aligned(8);
-
 /* BCH_SB_FIELD_disk_groups: */
 
 #define BCH_SB_LABEL_SIZE		32
@@ -1500,101 +733,6 @@ struct bch_sb_field_disk_groups {
 	struct bch_disk_group	entries[];
 } __packed __aligned(8);
 
-/* BCH_SB_FIELD_counters */
-
-#define BCH_PERSISTENT_COUNTERS()				\
-	x(io_read,					0)	\
-	x(io_write,					1)	\
-	x(io_move,					2)	\
-	x(bucket_invalidate,				3)	\
-	x(bucket_discard,				4)	\
-	x(bucket_alloc,					5)	\
-	x(bucket_alloc_fail,				6)	\
-	x(btree_cache_scan,				7)	\
-	x(btree_cache_reap,				8)	\
-	x(btree_cache_cannibalize,			9)	\
-	x(btree_cache_cannibalize_lock,			10)	\
-	x(btree_cache_cannibalize_lock_fail,		11)	\
-	x(btree_cache_cannibalize_unlock,		12)	\
-	x(btree_node_write,				13)	\
-	x(btree_node_read,				14)	\
-	x(btree_node_compact,				15)	\
-	x(btree_node_merge,				16)	\
-	x(btree_node_split,				17)	\
-	x(btree_node_rewrite,				18)	\
-	x(btree_node_alloc,				19)	\
-	x(btree_node_free,				20)	\
-	x(btree_node_set_root,				21)	\
-	x(btree_path_relock_fail,			22)	\
-	x(btree_path_upgrade_fail,			23)	\
-	x(btree_reserve_get_fail,			24)	\
-	x(journal_entry_full,				25)	\
-	x(journal_full,					26)	\
-	x(journal_reclaim_finish,			27)	\
-	x(journal_reclaim_start,			28)	\
-	x(journal_write,				29)	\
-	x(read_promote,					30)	\
-	x(read_bounce,					31)	\
-	x(read_split,					33)	\
-	x(read_retry,					32)	\
-	x(read_reuse_race,				34)	\
-	x(move_extent_read,				35)	\
-	x(move_extent_write,				36)	\
-	x(move_extent_finish,				37)	\
-	x(move_extent_fail,				38)	\
-	x(move_extent_start_fail,			39)	\
-	x(copygc,					40)	\
-	x(copygc_wait,					41)	\
-	x(gc_gens_end,					42)	\
-	x(gc_gens_start,				43)	\
-	x(trans_blocked_journal_reclaim,		44)	\
-	x(trans_restart_btree_node_reused,		45)	\
-	x(trans_restart_btree_node_split,		46)	\
-	x(trans_restart_fault_inject,			47)	\
-	x(trans_restart_iter_upgrade,			48)	\
-	x(trans_restart_journal_preres_get,		49)	\
-	x(trans_restart_journal_reclaim,		50)	\
-	x(trans_restart_journal_res_get,		51)	\
-	x(trans_restart_key_cache_key_realloced,	52)	\
-	x(trans_restart_key_cache_raced,		53)	\
-	x(trans_restart_mark_replicas,			54)	\
-	x(trans_restart_mem_realloced,			55)	\
-	x(trans_restart_memory_allocation_failure,	56)	\
-	x(trans_restart_relock,				57)	\
-	x(trans_restart_relock_after_fill,		58)	\
-	x(trans_restart_relock_key_cache_fill,		59)	\
-	x(trans_restart_relock_next_node,		60)	\
-	x(trans_restart_relock_parent_for_fill,		61)	\
-	x(trans_restart_relock_path,			62)	\
-	x(trans_restart_relock_path_intent,		63)	\
-	x(trans_restart_too_many_iters,			64)	\
-	x(trans_restart_traverse,			65)	\
-	x(trans_restart_upgrade,			66)	\
-	x(trans_restart_would_deadlock,			67)	\
-	x(trans_restart_would_deadlock_write,		68)	\
-	x(trans_restart_injected,			69)	\
-	x(trans_restart_key_cache_upgrade,		70)	\
-	x(trans_traverse_all,				71)	\
-	x(transaction_commit,				72)	\
-	x(write_super,					73)	\
-	x(trans_restart_would_deadlock_recursion_limit,	74)	\
-	x(trans_restart_write_buffer_flush,		75)	\
-	x(trans_restart_split_race,			76)	\
-	x(write_buffer_flush_slowpath,			77)	\
-	x(write_buffer_flush_sync,			78)
-
-enum bch_persistent_counters {
-#define x(t, n, ...) BCH_COUNTER_##t,
-	BCH_PERSISTENT_COUNTERS()
-#undef x
-	BCH_COUNTER_NR
-};
-
-struct bch_sb_field_counters {
-	struct bch_sb_field	field;
-	__le64			d[];
-};
-
 /*
  * On clean shutdown, store btree roots and current journal sequence number in
  * the superblock:
diff --git a/fs/bcachefs/bkey.c b/fs/bcachefs/bkey.c
index abdb05507d16..76e79a15ba08 100644
--- a/fs/bcachefs/bkey.c
+++ b/fs/bcachefs/bkey.c
@@ -33,7 +33,7 @@ void bch2_bkey_packed_to_binary_text(struct printbuf *out,
 			next_key_bits -= 64;
 		}
 
-		bch2_prt_u64_binary(out, v, min(word_bits, nr_key_bits));
+		bch2_prt_u64_base2_nbits(out, v, min(word_bits, nr_key_bits));
 
 		if (!next_key_bits)
 			break;
diff --git a/fs/bcachefs/bkey_methods.c b/fs/bcachefs/bkey_methods.c
index 761f5e33b1e6..5e52684764eb 100644
--- a/fs/bcachefs/bkey_methods.c
+++ b/fs/bcachefs/bkey_methods.c
@@ -63,8 +63,17 @@ static int key_type_cookie_invalid(struct bch_fs *c, struct bkey_s_c k,
 	return 0;
 }
 
+static void key_type_cookie_to_text(struct printbuf *out, struct bch_fs *c,
+				    struct bkey_s_c k)
+{
+	struct bkey_s_c_cookie ck = bkey_s_c_to_cookie(k);
+
+	prt_printf(out, "%llu", le64_to_cpu(ck.v->cookie));
+}
+
 #define bch2_bkey_ops_cookie ((struct bkey_ops) {	\
 	.key_invalid	= key_type_cookie_invalid,	\
+	.val_to_text	= key_type_cookie_to_text,	\
 	.min_val_size	= 8,				\
 })
 
diff --git a/fs/bcachefs/bkey_methods.h b/fs/bcachefs/bkey_methods.h
index ee82283722b7..03efe8ee565a 100644
--- a/fs/bcachefs/bkey_methods.h
+++ b/fs/bcachefs/bkey_methods.h
@@ -83,9 +83,10 @@ enum btree_update_flags {
 
 	__BTREE_TRIGGER_NORUN,
 	__BTREE_TRIGGER_TRANSACTIONAL,
+	__BTREE_TRIGGER_ATOMIC,
+	__BTREE_TRIGGER_GC,
 	__BTREE_TRIGGER_INSERT,
 	__BTREE_TRIGGER_OVERWRITE,
-	__BTREE_TRIGGER_GC,
 	__BTREE_TRIGGER_BUCKET_INVALIDATE,
 };
 
@@ -107,6 +108,10 @@ enum btree_update_flags {
  * causing us to go emergency read-only)
  */
 #define BTREE_TRIGGER_TRANSACTIONAL	(1U << __BTREE_TRIGGER_TRANSACTIONAL)
+#define BTREE_TRIGGER_ATOMIC		(1U << __BTREE_TRIGGER_ATOMIC)
+
+/* We're in gc/fsck: running triggers to recalculate e.g. disk usage */
+#define BTREE_TRIGGER_GC		(1U << __BTREE_TRIGGER_GC)
 
 /* @new is entering the btree */
 #define BTREE_TRIGGER_INSERT		(1U << __BTREE_TRIGGER_INSERT)
@@ -114,9 +119,6 @@ enum btree_update_flags {
 /* @old is leaving the btree */
 #define BTREE_TRIGGER_OVERWRITE		(1U << __BTREE_TRIGGER_OVERWRITE)
 
-/* We're in gc/fsck: running triggers to recalculate e.g. disk usage */
-#define BTREE_TRIGGER_GC		(1U << __BTREE_TRIGGER_GC)
-
 /* signal from bucket invalidate path to alloc trigger */
 #define BTREE_TRIGGER_BUCKET_INVALIDATE	(1U << __BTREE_TRIGGER_BUCKET_INVALIDATE)
 
diff --git a/fs/bcachefs/bset.c b/fs/bcachefs/bset.c
index 74bf8eb90a4c..3fd1085b6c61 100644
--- a/fs/bcachefs/bset.c
+++ b/fs/bcachefs/bset.c
@@ -720,7 +720,7 @@ static noinline void __build_ro_aux_tree(struct btree *b, struct bset_tree *t)
 {
 	struct bkey_packed *prev = NULL, *k = btree_bkey_first(b, t);
 	struct bkey_i min_key, max_key;
-	unsigned j, cacheline = 1;
+	unsigned cacheline = 1;
 
 	t->size = min(bkey_to_cacheline(b, t, btree_bkey_last(b, t)),
 		      bset_ro_tree_capacity(b, t));
@@ -823,13 +823,12 @@ void bch2_bset_init_first(struct btree *b, struct bset *i)
 	set_btree_bset(b, t, i);
 }
 
-void bch2_bset_init_next(struct bch_fs *c, struct btree *b,
-			 struct btree_node_entry *bne)
+void bch2_bset_init_next(struct btree *b, struct btree_node_entry *bne)
 {
 	struct bset *i = &bne->keys;
 	struct bset_tree *t;
 
-	BUG_ON(bset_byte_offset(b, bne) >= btree_bytes(c));
+	BUG_ON(bset_byte_offset(b, bne) >= btree_buf_bytes(b));
 	BUG_ON((void *) bne < (void *) btree_bkey_last(b, bset_tree_last(b)));
 	BUG_ON(b->nsets >= MAX_BSETS);
 
diff --git a/fs/bcachefs/bset.h b/fs/bcachefs/bset.h
index 632c2b8c5460..79c77baaa383 100644
--- a/fs/bcachefs/bset.h
+++ b/fs/bcachefs/bset.h
@@ -264,8 +264,7 @@ static inline struct bset *bset_next_set(struct btree *b,
 void bch2_btree_keys_init(struct btree *);
 
 void bch2_bset_init_first(struct btree *, struct bset *);
-void bch2_bset_init_next(struct bch_fs *, struct btree *,
-			 struct btree_node_entry *);
+void bch2_bset_init_next(struct btree *, struct btree_node_entry *);
 void bch2_bset_build_aux_tree(struct btree *, struct bset_tree *, bool);
 
 void bch2_bset_insert(struct btree *, struct btree_node_iter *,
diff --git a/fs/bcachefs/btree_cache.c b/fs/bcachefs/btree_cache.c
index 8e2488a4b58d..d7c81beac14a 100644
--- a/fs/bcachefs/btree_cache.c
+++ b/fs/bcachefs/btree_cache.c
@@ -60,7 +60,7 @@ static void btree_node_data_free(struct bch_fs *c, struct btree *b)
 
 	clear_btree_node_just_written(b);
 
-	kvpfree(b->data, btree_bytes(c));
+	kvpfree(b->data, btree_buf_bytes(b));
 	b->data = NULL;
 #ifdef __KERNEL__
 	kvfree(b->aux_data);
@@ -94,7 +94,7 @@ static int btree_node_data_alloc(struct bch_fs *c, struct btree *b, gfp_t gfp)
 {
 	BUG_ON(b->data || b->aux_data);
 
-	b->data = kvpmalloc(btree_bytes(c), gfp);
+	b->data = kvpmalloc(btree_buf_bytes(b), gfp);
 	if (!b->data)
 		return -BCH_ERR_ENOMEM_btree_node_mem_alloc;
 #ifdef __KERNEL__
@@ -107,7 +107,7 @@ static int btree_node_data_alloc(struct bch_fs *c, struct btree *b, gfp_t gfp)
 		b->aux_data = NULL;
 #endif
 	if (!b->aux_data) {
-		kvpfree(b->data, btree_bytes(c));
+		kvpfree(b->data, btree_buf_bytes(b));
 		b->data = NULL;
 		return -BCH_ERR_ENOMEM_btree_node_mem_alloc;
 	}
@@ -126,7 +126,7 @@ static struct btree *__btree_node_mem_alloc(struct bch_fs *c, gfp_t gfp)
 	bkey_btree_ptr_init(&b->key);
 	INIT_LIST_HEAD(&b->list);
 	INIT_LIST_HEAD(&b->write_blocked);
-	b->byte_order = ilog2(btree_bytes(c));
+	b->byte_order = ilog2(c->opts.btree_node_size);
 	return b;
 }
 
@@ -408,7 +408,7 @@ void bch2_fs_btree_cache_exit(struct bch_fs *c)
 	if (c->verify_data)
 		list_move(&c->verify_data->list, &bc->live);
 
-	kvpfree(c->verify_ondisk, btree_bytes(c));
+	kvpfree(c->verify_ondisk, c->opts.btree_node_size);
 
 	for (i = 0; i < btree_id_nr_alive(c); i++) {
 		struct btree_root *r = bch2_btree_id_root(c, i);
@@ -1192,7 +1192,7 @@ void bch2_btree_node_to_text(struct printbuf *out, struct bch_fs *c, const struc
 	       "    failed unpacked %zu\n",
 	       b->unpack_fn_len,
 	       b->nr.live_u64s * sizeof(u64),
-	       btree_bytes(c) - sizeof(struct btree_node),
+	       btree_buf_bytes(b) - sizeof(struct btree_node),
 	       b->nr.live_u64s * 100 / btree_max_u64s(c),
 	       b->sib_u64s[0],
 	       b->sib_u64s[1],
diff --git a/fs/bcachefs/btree_cache.h b/fs/bcachefs/btree_cache.h
index 4e1af5882052..6d33885fdbde 100644
--- a/fs/bcachefs/btree_cache.h
+++ b/fs/bcachefs/btree_cache.h
@@ -74,22 +74,27 @@ static inline bool btree_node_hashed(struct btree *b)
 	     _iter = 0;	_iter < (_tbl)->size; _iter++)			\
 		rht_for_each_entry_rcu((_b), (_pos), _tbl, _iter, hash)
 
-static inline size_t btree_bytes(struct bch_fs *c)
+static inline size_t btree_buf_bytes(const struct btree *b)
 {
-	return c->opts.btree_node_size;
+	return 1UL << b->byte_order;
 }
 
-static inline size_t btree_max_u64s(struct bch_fs *c)
+static inline size_t btree_buf_max_u64s(const struct btree *b)
 {
-	return (btree_bytes(c) - sizeof(struct btree_node)) / sizeof(u64);
+	return (btree_buf_bytes(b) - sizeof(struct btree_node)) / sizeof(u64);
 }
 
-static inline size_t btree_pages(struct bch_fs *c)
+static inline size_t btree_max_u64s(const struct bch_fs *c)
 {
-	return btree_bytes(c) / PAGE_SIZE;
+	return (c->opts.btree_node_size - sizeof(struct btree_node)) / sizeof(u64);
 }
 
-static inline unsigned btree_blocks(struct bch_fs *c)
+static inline size_t btree_sectors(const struct bch_fs *c)
+{
+	return c->opts.btree_node_size >> SECTOR_SHIFT;
+}
+
+static inline unsigned btree_blocks(const struct bch_fs *c)
 {
 	return btree_sectors(c) >> c->block_bits;
 }
diff --git a/fs/bcachefs/btree_gc.c b/fs/bcachefs/btree_gc.c
index 49b4ade758c3..1102995643b1 100644
--- a/fs/bcachefs/btree_gc.c
+++ b/fs/bcachefs/btree_gc.c
@@ -597,7 +597,7 @@ static int bch2_check_fix_ptrs(struct btree_trans *trans, enum btree_id btree_id
 			      "bucket %u:%zu data type %s ptr gen %u missing in alloc btree\n"
 			      "while marking %s",
 			      p.ptr.dev, PTR_BUCKET_NR(ca, &p.ptr),
-			      bch2_data_types[ptr_data_type(k->k, &p.ptr)],
+			      bch2_data_type_str(ptr_data_type(k->k, &p.ptr)),
 			      p.ptr.gen,
 			      (printbuf_reset(&buf),
 			       bch2_bkey_val_to_text(&buf, c, *k), buf.buf)))) {
@@ -615,7 +615,7 @@ static int bch2_check_fix_ptrs(struct btree_trans *trans, enum btree_id btree_id
 			      "bucket %u:%zu data type %s ptr gen in the future: %u > %u\n"
 			      "while marking %s",
 			      p.ptr.dev, PTR_BUCKET_NR(ca, &p.ptr),
-			      bch2_data_types[ptr_data_type(k->k, &p.ptr)],
+			      bch2_data_type_str(ptr_data_type(k->k, &p.ptr)),
 			      p.ptr.gen, g->gen,
 			      (printbuf_reset(&buf),
 			       bch2_bkey_val_to_text(&buf, c, *k), buf.buf)))) {
@@ -637,7 +637,7 @@ static int bch2_check_fix_ptrs(struct btree_trans *trans, enum btree_id btree_id
 			      "bucket %u:%zu gen %u data type %s: ptr gen %u too stale\n"
 			      "while marking %s",
 			      p.ptr.dev, PTR_BUCKET_NR(ca, &p.ptr), g->gen,
-			      bch2_data_types[ptr_data_type(k->k, &p.ptr)],
+			      bch2_data_type_str(ptr_data_type(k->k, &p.ptr)),
 			      p.ptr.gen,
 			      (printbuf_reset(&buf),
 			       bch2_bkey_val_to_text(&buf, c, *k), buf.buf))))
@@ -649,7 +649,7 @@ static int bch2_check_fix_ptrs(struct btree_trans *trans, enum btree_id btree_id
 			      "bucket %u:%zu data type %s stale dirty ptr: %u < %u\n"
 			      "while marking %s",
 			      p.ptr.dev, PTR_BUCKET_NR(ca, &p.ptr),
-			      bch2_data_types[ptr_data_type(k->k, &p.ptr)],
+			      bch2_data_type_str(ptr_data_type(k->k, &p.ptr)),
 			      p.ptr.gen, g->gen,
 			      (printbuf_reset(&buf),
 			       bch2_bkey_val_to_text(&buf, c, *k), buf.buf))))
@@ -664,8 +664,8 @@ static int bch2_check_fix_ptrs(struct btree_trans *trans, enum btree_id btree_id
 				"bucket %u:%zu different types of data in same bucket: %s, %s\n"
 				"while marking %s",
 				p.ptr.dev, PTR_BUCKET_NR(ca, &p.ptr),
-				bch2_data_types[g->data_type],
-				bch2_data_types[data_type],
+				bch2_data_type_str(g->data_type),
+				bch2_data_type_str(data_type),
 				(printbuf_reset(&buf),
 				 bch2_bkey_val_to_text(&buf, c, *k), buf.buf))) {
 			if (data_type == BCH_DATA_btree) {
@@ -1238,11 +1238,11 @@ static int bch2_gc_done(struct bch_fs *c,
 
 		for (i = 0; i < BCH_DATA_NR; i++) {
 			copy_dev_field(dev_usage_buckets_wrong,
-				       d[i].buckets,	"%s buckets", bch2_data_types[i]);
+				       d[i].buckets,	"%s buckets", bch2_data_type_str(i));
 			copy_dev_field(dev_usage_sectors_wrong,
-				       d[i].sectors,	"%s sectors", bch2_data_types[i]);
+				       d[i].sectors,	"%s sectors", bch2_data_type_str(i));
 			copy_dev_field(dev_usage_fragmented_wrong,
-				       d[i].fragmented,	"%s fragmented", bch2_data_types[i]);
+				       d[i].fragmented,	"%s fragmented", bch2_data_type_str(i));
 		}
 	}
 
@@ -1253,19 +1253,19 @@ static int bch2_gc_done(struct bch_fs *c,
 			bch2_acc_percpu_u64s((u64 __percpu *) c->usage_gc, nr);
 
 		copy_fs_field(fs_usage_hidden_wrong,
-			      hidden,		"hidden");
+			      b.hidden,		"hidden");
 		copy_fs_field(fs_usage_btree_wrong,
-			      btree,		"btree");
+			      b.btree,		"btree");
 
 		if (!metadata_only) {
 			copy_fs_field(fs_usage_data_wrong,
-				      data,	"data");
+				      b.data,	"data");
 			copy_fs_field(fs_usage_cached_wrong,
-				      cached,	"cached");
+				      b.cached,	"cached");
 			copy_fs_field(fs_usage_reserved_wrong,
-				      reserved,	"reserved");
+				      b.reserved,	"reserved");
 			copy_fs_field(fs_usage_nr_inodes_wrong,
-				      nr_inodes,"nr_inodes");
+				      b.nr_inodes,"nr_inodes");
 
 			for (i = 0; i < BCH_REPLICAS_MAX; i++)
 				copy_fs_field(fs_usage_persistent_reserved_wrong,
@@ -1417,8 +1417,8 @@ static int bch2_alloc_write_key(struct btree_trans *trans,
 			": got %s, should be %s",
 			iter->pos.inode, iter->pos.offset,
 			gc.gen,
-			bch2_data_types[new.data_type],
-			bch2_data_types[gc.data_type]))
+			bch2_data_type_str(new.data_type),
+			bch2_data_type_str(gc.data_type)))
 		new.data_type = gc.data_type;
 
 #define copy_bucket_field(_errtype, _f)					\
@@ -1428,7 +1428,7 @@ static int bch2_alloc_write_key(struct btree_trans *trans,
 			": got %u, should be %u",			\
 			iter->pos.inode, iter->pos.offset,		\
 			gc.gen,						\
-			bch2_data_types[gc.data_type],			\
+			bch2_data_type_str(gc.data_type),		\
 			new._f, gc._f))					\
 		new._f = gc._f;						\
 
diff --git a/fs/bcachefs/btree_io.c b/fs/bcachefs/btree_io.c
index 33db48e2153f..aa9b6cbe3226 100644
--- a/fs/bcachefs/btree_io.c
+++ b/fs/bcachefs/btree_io.c
@@ -112,7 +112,7 @@ static void *btree_bounce_alloc(struct bch_fs *c, size_t size,
 	unsigned flags = memalloc_nofs_save();
 	void *p;
 
-	BUG_ON(size > btree_bytes(c));
+	BUG_ON(size > c->opts.btree_node_size);
 
 	*used_mempool = false;
 	p = vpmalloc(size, __GFP_NOWARN|GFP_NOWAIT);
@@ -174,8 +174,8 @@ static void bch2_sort_whiteouts(struct bch_fs *c, struct btree *b)
 
 	ptrs = ptrs_end = ((void *) new_whiteouts + bytes);
 
-	for (k = unwritten_whiteouts_start(c, b);
-	     k != unwritten_whiteouts_end(c, b);
+	for (k = unwritten_whiteouts_start(b);
+	     k != unwritten_whiteouts_end(b);
 	     k = bkey_p_next(k))
 		*--ptrs = k;
 
@@ -192,7 +192,7 @@ static void bch2_sort_whiteouts(struct bch_fs *c, struct btree *b)
 	verify_no_dups(b, new_whiteouts,
 		       (void *) ((u64 *) new_whiteouts + b->whiteout_u64s));
 
-	memcpy_u64s(unwritten_whiteouts_start(c, b),
+	memcpy_u64s(unwritten_whiteouts_start(b),
 		    new_whiteouts, b->whiteout_u64s);
 
 	btree_bounce_free(c, bytes, used_mempool, new_whiteouts);
@@ -313,7 +313,7 @@ static void btree_node_sort(struct bch_fs *c, struct btree *b,
 	}
 
 	bytes = sorting_entire_node
-		? btree_bytes(c)
+		? btree_buf_bytes(b)
 		: __vstruct_bytes(struct btree_node, u64s);
 
 	out = btree_bounce_alloc(c, bytes, &used_mempool);
@@ -338,7 +338,7 @@ static void btree_node_sort(struct bch_fs *c, struct btree *b,
 	if (sorting_entire_node) {
 		u64s = le16_to_cpu(out->keys.u64s);
 
-		BUG_ON(bytes != btree_bytes(c));
+		BUG_ON(bytes != btree_buf_bytes(b));
 
 		/*
 		 * Our temporary buffer is the same size as the btree node's
@@ -502,7 +502,7 @@ void bch2_btree_init_next(struct btree_trans *trans, struct btree *b)
 
 	bne = want_new_bset(c, b);
 	if (bne)
-		bch2_bset_init_next(c, b, bne);
+		bch2_bset_init_next(b, bne);
 
 	bch2_btree_build_aux_trees(b);
 
@@ -1160,7 +1160,7 @@ int bch2_btree_node_read_done(struct bch_fs *c, struct bch_dev *ca,
 			     ptr_written, b->written);
 	} else {
 		for (bne = write_block(b);
-		     bset_byte_offset(b, bne) < btree_bytes(c);
+		     bset_byte_offset(b, bne) < btree_buf_bytes(b);
 		     bne = (void *) bne + block_bytes(c))
 			btree_err_on(bne->keys.seq == b->data->keys.seq &&
 				     !bch2_journal_seq_is_blacklisted(c,
@@ -1172,7 +1172,7 @@ int bch2_btree_node_read_done(struct bch_fs *c, struct bch_dev *ca,
 				     "found bset signature after last bset");
 	}
 
-	sorted = btree_bounce_alloc(c, btree_bytes(c), &used_mempool);
+	sorted = btree_bounce_alloc(c, btree_buf_bytes(b), &used_mempool);
 	sorted->keys.u64s = 0;
 
 	set_btree_bset(b, b->set, &b->data->keys);
@@ -1188,7 +1188,7 @@ int bch2_btree_node_read_done(struct bch_fs *c, struct bch_dev *ca,
 
 	BUG_ON(b->nr.live_u64s != u64s);
 
-	btree_bounce_free(c, btree_bytes(c), used_mempool, sorted);
+	btree_bounce_free(c, btree_buf_bytes(b), used_mempool, sorted);
 
 	if (updated_range)
 		bch2_btree_node_drop_keys_outside_node(b);
@@ -1284,7 +1284,7 @@ static void btree_node_read_work(struct work_struct *work)
 		rb->have_ioref		= bch2_dev_get_ioref(ca, READ);
 		bio_reset(bio, NULL, REQ_OP_READ|REQ_SYNC|REQ_META);
 		bio->bi_iter.bi_sector	= rb->pick.ptr.offset;
-		bio->bi_iter.bi_size	= btree_bytes(c);
+		bio->bi_iter.bi_size	= btree_buf_bytes(b);
 
 		if (rb->have_ioref) {
 			bio_set_dev(bio, ca->disk_sb.bdev);
@@ -1512,7 +1512,7 @@ fsck_err:
 	}
 
 	if (best >= 0) {
-		memcpy(b->data, ra->buf[best], btree_bytes(c));
+		memcpy(b->data, ra->buf[best], btree_buf_bytes(b));
 		ret = bch2_btree_node_read_done(c, NULL, b, false, saw_error);
 	} else {
 		ret = -1;
@@ -1578,7 +1578,7 @@ static int btree_node_read_all_replicas(struct bch_fs *c, struct btree *b, bool
 	for (i = 0; i < ra->nr; i++) {
 		ra->buf[i] = mempool_alloc(&c->btree_bounce_pool, GFP_NOFS);
 		ra->bio[i] = bio_alloc_bioset(NULL,
-					      buf_pages(ra->buf[i], btree_bytes(c)),
+					      buf_pages(ra->buf[i], btree_buf_bytes(b)),
 					      REQ_OP_READ|REQ_SYNC|REQ_META,
 					      GFP_NOFS,
 					      &c->btree_bio);
@@ -1598,7 +1598,7 @@ static int btree_node_read_all_replicas(struct bch_fs *c, struct btree *b, bool
 		rb->pick		= pick;
 		rb->bio.bi_iter.bi_sector = pick.ptr.offset;
 		rb->bio.bi_end_io	= btree_node_read_all_replicas_endio;
-		bch2_bio_map(&rb->bio, ra->buf[i], btree_bytes(c));
+		bch2_bio_map(&rb->bio, ra->buf[i], btree_buf_bytes(b));
 
 		if (rb->have_ioref) {
 			this_cpu_add(ca->io_done->sectors[READ][BCH_DATA_btree],
@@ -1665,7 +1665,7 @@ void bch2_btree_node_read(struct btree_trans *trans, struct btree *b,
 	ca = bch_dev_bkey_exists(c, pick.ptr.dev);
 
 	bio = bio_alloc_bioset(NULL,
-			       buf_pages(b->data, btree_bytes(c)),
+			       buf_pages(b->data, btree_buf_bytes(b)),
 			       REQ_OP_READ|REQ_SYNC|REQ_META,
 			       GFP_NOFS,
 			       &c->btree_bio);
@@ -1679,7 +1679,7 @@ void bch2_btree_node_read(struct btree_trans *trans, struct btree *b,
 	INIT_WORK(&rb->work, btree_node_read_work);
 	bio->bi_iter.bi_sector	= pick.ptr.offset;
 	bio->bi_end_io		= btree_node_read_endio;
-	bch2_bio_map(bio, b->data, btree_bytes(c));
+	bch2_bio_map(bio, b->data, btree_buf_bytes(b));
 
 	if (rb->have_ioref) {
 		this_cpu_add(ca->io_done->sectors[READ][BCH_DATA_btree],
@@ -2074,8 +2074,8 @@ do_write:
 	i->u64s		= 0;
 
 	sort_iter_add(&sort_iter.iter,
-		      unwritten_whiteouts_start(c, b),
-		      unwritten_whiteouts_end(c, b));
+		      unwritten_whiteouts_start(b),
+		      unwritten_whiteouts_end(b));
 	SET_BSET_SEPARATE_WHITEOUTS(i, false);
 
 	b->whiteout_u64s = 0;
@@ -2251,7 +2251,7 @@ bool bch2_btree_post_write_cleanup(struct bch_fs *c, struct btree *b)
 
 	bne = want_new_bset(c, b);
 	if (bne)
-		bch2_bset_init_next(c, b, bne);
+		bch2_bset_init_next(b, bne);
 
 	bch2_btree_build_aux_trees(b);
 
diff --git a/fs/bcachefs/btree_iter.c b/fs/bcachefs/btree_iter.c
index fa298289e016..5467a8635be1 100644
--- a/fs/bcachefs/btree_iter.c
+++ b/fs/bcachefs/btree_iter.c
@@ -1337,7 +1337,7 @@ void bch2_path_put(struct btree_trans *trans, btree_path_idx_t path_idx, bool in
 
 	if (path->should_be_locked &&
 	    !trans->restarted &&
-	    (!dup || !bch2_btree_path_relock_norestart(trans, dup, _THIS_IP_)))
+	    (!dup || !bch2_btree_path_relock_norestart(trans, dup)))
 		return;
 
 	if (dup) {
diff --git a/fs/bcachefs/btree_iter.h b/fs/bcachefs/btree_iter.h
index da2b74fa63fc..24772538e4cc 100644
--- a/fs/bcachefs/btree_iter.h
+++ b/fs/bcachefs/btree_iter.h
@@ -819,6 +819,11 @@ __bch2_btree_iter_peek_and_restart(struct btree_trans *trans,
 #define for_each_btree_key_continue_norestart(_iter, _flags, _k, _ret)	\
 	for_each_btree_key_upto_continue_norestart(_iter, SPOS_MAX, _flags, _k, _ret)
 
+/*
+ * This should not be used in a fastpath, without first trying _do in
+ * nonblocking mode - it will cause excessive transaction restarts and
+ * potentially livelocking:
+ */
 #define drop_locks_do(_trans, _do)					\
 ({									\
 	bch2_trans_unlock(_trans);					\
diff --git a/fs/bcachefs/btree_locking.c b/fs/bcachefs/btree_locking.c
index 2d1c95c42f24..bed75c93c069 100644
--- a/fs/bcachefs/btree_locking.c
+++ b/fs/bcachefs/btree_locking.c
@@ -631,8 +631,7 @@ int bch2_btree_path_relock_intent(struct btree_trans *trans,
 }
 
 __flatten
-bool bch2_btree_path_relock_norestart(struct btree_trans *trans,
-			struct btree_path *path, unsigned long trace_ip)
+bool bch2_btree_path_relock_norestart(struct btree_trans *trans, struct btree_path *path)
 {
 	struct get_locks_fail f;
 
@@ -642,7 +641,7 @@ bool bch2_btree_path_relock_norestart(struct btree_trans *trans,
 int __bch2_btree_path_relock(struct btree_trans *trans,
 			struct btree_path *path, unsigned long trace_ip)
 {
-	if (!bch2_btree_path_relock_norestart(trans, path, trace_ip)) {
+	if (!bch2_btree_path_relock_norestart(trans, path)) {
 		trace_and_count(trans->c, trans_restart_relock_path, trans, trace_ip, path);
 		return btree_trans_restart(trans, BCH_ERR_transaction_restart_relock_path);
 	}
@@ -759,12 +758,39 @@ int bch2_trans_relock(struct btree_trans *trans)
 	if (unlikely(trans->restarted))
 		return -((int) trans->restarted);
 
-	trans_for_each_path(trans, path, i)
+	trans_for_each_path(trans, path, i) {
+		struct get_locks_fail f;
+
 		if (path->should_be_locked &&
-		    !bch2_btree_path_relock_norestart(trans, path, _RET_IP_)) {
-			trace_and_count(trans->c, trans_restart_relock, trans, _RET_IP_, path);
+		    !btree_path_get_locks(trans, path, false, &f)) {
+			if (trace_trans_restart_relock_enabled()) {
+				struct printbuf buf = PRINTBUF;
+
+				bch2_bpos_to_text(&buf, path->pos);
+				prt_printf(&buf, " l=%u seq=%u node seq=",
+					   f.l, path->l[f.l].lock_seq);
+				if (IS_ERR_OR_NULL(f.b)) {
+					prt_str(&buf, bch2_err_str(PTR_ERR(f.b)));
+				} else {
+					prt_printf(&buf, "%u", f.b->c.lock.seq);
+
+					struct six_lock_count c =
+						bch2_btree_node_lock_counts(trans, NULL, &f.b->c, f.l);
+					prt_printf(&buf, " self locked %u.%u.%u", c.n[0], c.n[1], c.n[2]);
+
+					c = six_lock_counts(&f.b->c.lock);
+					prt_printf(&buf, " total locked %u.%u.%u", c.n[0], c.n[1], c.n[2]);
+				}
+
+				trace_trans_restart_relock(trans, _RET_IP_, buf.buf);
+				printbuf_exit(&buf);
+			}
+
+			count_event(trans->c, trans_restart_relock);
 			return btree_trans_restart(trans, BCH_ERR_transaction_restart_relock);
 		}
+	}
+
 	return 0;
 }
 
@@ -778,7 +804,7 @@ int bch2_trans_relock_notrace(struct btree_trans *trans)
 
 	trans_for_each_path(trans, path, i)
 		if (path->should_be_locked &&
-		    !bch2_btree_path_relock_norestart(trans, path, _RET_IP_)) {
+		    !bch2_btree_path_relock_norestart(trans, path)) {
 			return btree_trans_restart(trans, BCH_ERR_transaction_restart_relock);
 		}
 	return 0;
diff --git a/fs/bcachefs/btree_locking.h b/fs/bcachefs/btree_locking.h
index cc5500a957a1..4bd72c855da1 100644
--- a/fs/bcachefs/btree_locking.h
+++ b/fs/bcachefs/btree_locking.h
@@ -312,8 +312,7 @@ void bch2_btree_node_lock_write_nofail(struct btree_trans *,
 
 /* relock: */
 
-bool bch2_btree_path_relock_norestart(struct btree_trans *,
-				      struct btree_path *, unsigned long);
+bool bch2_btree_path_relock_norestart(struct btree_trans *, struct btree_path *);
 int __bch2_btree_path_relock(struct btree_trans *,
 			     struct btree_path *, unsigned long);
 
@@ -353,12 +352,6 @@ static inline bool bch2_btree_node_relock_notrace(struct btree_trans *trans,
 
 /* upgrade */
 
-
-struct get_locks_fail {
-	unsigned	l;
-	struct btree	*b;
-};
-
 bool bch2_btree_path_upgrade_noupgrade_sibs(struct btree_trans *,
 			       struct btree_path *, unsigned,
 			       struct get_locks_fail *);
diff --git a/fs/bcachefs/btree_trans_commit.c b/fs/bcachefs/btree_trans_commit.c
index 90eb8065ff2d..30d69a6d133e 100644
--- a/fs/bcachefs/btree_trans_commit.c
+++ b/fs/bcachefs/btree_trans_commit.c
@@ -139,8 +139,7 @@ bool bch2_btree_bset_insert_key(struct btree_trans *trans,
 	EBUG_ON(bkey_deleted(&insert->k) && bkey_val_u64s(&insert->k));
 	EBUG_ON(bpos_lt(insert->k.p, b->data->min_key));
 	EBUG_ON(bpos_gt(insert->k.p, b->data->max_key));
-	EBUG_ON(insert->k.u64s >
-		bch_btree_keys_u64s_remaining(trans->c, b));
+	EBUG_ON(insert->k.u64s > bch2_btree_keys_u64s_remaining(b));
 	EBUG_ON(!b->c.level && !bpos_eq(insert->k.p, path->pos));
 
 	k = bch2_btree_node_iter_peek_all(node_iter, b);
@@ -160,7 +159,7 @@ bool bch2_btree_bset_insert_key(struct btree_trans *trans,
 		k->type = KEY_TYPE_deleted;
 
 		if (k->needs_whiteout)
-			push_whiteout(trans->c, b, insert->k.p);
+			push_whiteout(b, insert->k.p);
 		k->needs_whiteout = false;
 
 		if (k >= btree_bset_last(b)->start) {
@@ -348,9 +347,7 @@ static noinline void journal_transaction_name(struct btree_trans *trans)
 static inline int btree_key_can_insert(struct btree_trans *trans,
 				       struct btree *b, unsigned u64s)
 {
-	struct bch_fs *c = trans->c;
-
-	if (!bch2_btree_node_insert_fits(c, b, u64s))
+	if (!bch2_btree_node_insert_fits(b, u64s))
 		return -BCH_ERR_btree_insert_btree_node_full;
 
 	return 0;
@@ -418,7 +415,7 @@ static int btree_key_can_insert_cached(struct btree_trans *trans, unsigned flags
 		return 0;
 
 	new_u64s	= roundup_pow_of_two(u64s);
-	new_k		= krealloc(ck->k, new_u64s * sizeof(u64), GFP_NOWAIT);
+	new_k		= krealloc(ck->k, new_u64s * sizeof(u64), GFP_NOWAIT|__GFP_NOWARN);
 	if (unlikely(!new_k))
 		return btree_key_can_insert_cached_slowpath(trans, flags, path, new_u64s);
 
@@ -448,9 +445,6 @@ static int run_one_mem_trigger(struct btree_trans *trans,
 	if (unlikely(flags & BTREE_TRIGGER_NORUN))
 		return 0;
 
-	if (!btree_node_type_needs_gc(__btree_node_type(i->level, i->btree_id)))
-		return 0;
-
 	if (old_ops->trigger == new_ops->trigger) {
 		ret   = bch2_key_trigger(trans, i->btree_id, i->level,
 				old, bkey_i_to_s(new),
@@ -586,9 +580,6 @@ static int bch2_trans_commit_run_triggers(struct btree_trans *trans)
 
 static noinline int bch2_trans_commit_run_gc_triggers(struct btree_trans *trans)
 {
-	struct bch_fs *c = trans->c;
-	int ret = 0;
-
 	trans_for_each_update(trans, i) {
 		/*
 		 * XXX: synchronization of cached update triggers with gc
@@ -596,14 +587,15 @@ static noinline int bch2_trans_commit_run_gc_triggers(struct btree_trans *trans)
 		 */
 		BUG_ON(i->cached || i->level);
 
-		if (gc_visited(c, gc_pos_btree_node(insert_l(trans, i)->b))) {
-			ret = run_one_mem_trigger(trans, i, i->flags|BTREE_TRIGGER_GC);
+		if (btree_node_type_needs_gc(__btree_node_type(i->level, i->btree_id)) &&
+		    gc_visited(trans->c, gc_pos_btree_node(insert_l(trans, i)->b))) {
+			int ret = run_one_mem_trigger(trans, i, i->flags|BTREE_TRIGGER_GC);
 			if (ret)
-				break;
+				return ret;
 		}
 	}
 
-	return ret;
+	return 0;
 }
 
 static inline int
@@ -680,6 +672,9 @@ bch2_trans_commit_write_locked(struct btree_trans *trans, unsigned flags,
 	    bch2_trans_fs_usage_apply(trans, trans->fs_usage_deltas))
 		return -BCH_ERR_btree_insert_need_mark_replicas;
 
+	/* XXX: we only want to run this if deltas are nonzero */
+	bch2_trans_account_disk_usage_change(trans);
+
 	h = trans->hooks;
 	while (h) {
 		ret = h->fn(trans, h);
@@ -689,8 +684,8 @@ bch2_trans_commit_write_locked(struct btree_trans *trans, unsigned flags,
 	}
 
 	trans_for_each_update(trans, i)
-		if (BTREE_NODE_TYPE_HAS_MEM_TRIGGERS & (1U << i->bkey_type)) {
-			ret = run_one_mem_trigger(trans, i, i->flags);
+		if (BTREE_NODE_TYPE_HAS_ATOMIC_TRIGGERS & (1U << i->bkey_type)) {
+			ret = run_one_mem_trigger(trans, i, BTREE_TRIGGER_ATOMIC|i->flags);
 			if (ret)
 				goto fatal_err;
 		}
@@ -994,6 +989,8 @@ int __bch2_trans_commit(struct btree_trans *trans, unsigned flags)
 	    !trans->journal_entries_u64s)
 		goto out_reset;
 
+	memset(&trans->fs_usage_delta, 0, sizeof(trans->fs_usage_delta));
+
 	ret = bch2_trans_commit_run_triggers(trans);
 	if (ret)
 		goto out_reset;
diff --git a/fs/bcachefs/btree_types.h b/fs/bcachefs/btree_types.h
index d530307046f4..4a5a64499eb7 100644
--- a/fs/bcachefs/btree_types.h
+++ b/fs/bcachefs/btree_types.h
@@ -430,6 +430,9 @@ struct btree_trans {
 	struct journal_res	journal_res;
 	u64			*journal_seq;
 	struct disk_reservation *disk_res;
+
+	struct bch_fs_usage_base fs_usage_delta;
+
 	unsigned		journal_u64s;
 	unsigned		extra_disk_res; /* XXX kill */
 	struct replicas_delta_list *fs_usage_deltas;
@@ -653,7 +656,7 @@ const char *bch2_btree_node_type_str(enum btree_node_type);
 	 BIT_ULL(BKEY_TYPE_reflink)|			\
 	 BIT_ULL(BKEY_TYPE_btree))
 
-#define BTREE_NODE_TYPE_HAS_MEM_TRIGGERS		\
+#define BTREE_NODE_TYPE_HAS_ATOMIC_TRIGGERS		\
 	(BIT_ULL(BKEY_TYPE_alloc)|			\
 	 BIT_ULL(BKEY_TYPE_inodes)|			\
 	 BIT_ULL(BKEY_TYPE_stripes)|			\
@@ -661,7 +664,7 @@ const char *bch2_btree_node_type_str(enum btree_node_type);
 
 #define BTREE_NODE_TYPE_HAS_TRIGGERS			\
 	(BTREE_NODE_TYPE_HAS_TRANS_TRIGGERS|		\
-	 BTREE_NODE_TYPE_HAS_MEM_TRIGGERS)
+	 BTREE_NODE_TYPE_HAS_ATOMIC_TRIGGERS)
 
 static inline bool btree_node_type_needs_gc(enum btree_node_type type)
 {
@@ -738,4 +741,9 @@ enum btree_node_sibling {
 	btree_next_sib,
 };
 
+struct get_locks_fail {
+	unsigned	l;
+	struct btree	*b;
+};
+
 #endif /* _BCACHEFS_BTREE_TYPES_H */
diff --git a/fs/bcachefs/btree_update_interior.c b/fs/bcachefs/btree_update_interior.c
index 44f9dfa28a09..17a5938aa71a 100644
--- a/fs/bcachefs/btree_update_interior.c
+++ b/fs/bcachefs/btree_update_interior.c
@@ -159,7 +159,7 @@ static bool bch2_btree_node_format_fits(struct bch_fs *c, struct btree *b,
 {
 	size_t u64s = btree_node_u64s_with_format(nr, &b->format, new_f);
 
-	return __vstruct_bytes(struct btree_node, u64s) < btree_bytes(c);
+	return __vstruct_bytes(struct btree_node, u64s) < btree_buf_bytes(b);
 }
 
 /* Btree node freeing/allocation: */
@@ -1097,7 +1097,7 @@ bch2_btree_update_start(struct btree_trans *trans, struct btree_path *path,
 		 * Always check for space for two keys, even if we won't have to
 		 * split at prior level - it might have been a merge instead:
 		 */
-		if (bch2_btree_node_insert_fits(c, path->l[update_level].b,
+		if (bch2_btree_node_insert_fits(path->l[update_level].b,
 						BKEY_BTREE_PTR_U64s_MAX * 2))
 			break;
 
@@ -1401,7 +1401,7 @@ static void __btree_split_node(struct btree_update *as,
 
 		unsigned u64s = nr_keys[i].nr_keys * n[i]->data->format.key_u64s +
 			nr_keys[i].val_u64s;
-		if (__vstruct_bytes(struct btree_node, u64s) > btree_bytes(as->c))
+		if (__vstruct_bytes(struct btree_node, u64s) > btree_buf_bytes(b))
 			n[i]->data->format = b->format;
 
 		btree_node_set_format(n[i], n[i]->data->format);
@@ -1703,7 +1703,7 @@ static int bch2_btree_insert_node(struct btree_update *as, struct btree_trans *t
 
 	bch2_btree_node_prep_for_write(trans, path, b);
 
-	if (!bch2_btree_node_insert_fits(c, b, bch2_keylist_u64s(keys))) {
+	if (!bch2_btree_node_insert_fits(b, bch2_keylist_u64s(keys))) {
 		bch2_btree_node_unlock_write(trans, path, b);
 		goto split;
 	}
diff --git a/fs/bcachefs/btree_update_interior.h b/fs/bcachefs/btree_update_interior.h
index adfc62083844..c593c925d1e3 100644
--- a/fs/bcachefs/btree_update_interior.h
+++ b/fs/bcachefs/btree_update_interior.h
@@ -184,21 +184,19 @@ static inline void btree_node_reset_sib_u64s(struct btree *b)
 	b->sib_u64s[1] = b->nr.live_u64s;
 }
 
-static inline void *btree_data_end(struct bch_fs *c, struct btree *b)
+static inline void *btree_data_end(struct btree *b)
 {
-	return (void *) b->data + btree_bytes(c);
+	return (void *) b->data + btree_buf_bytes(b);
 }
 
-static inline struct bkey_packed *unwritten_whiteouts_start(struct bch_fs *c,
-							    struct btree *b)
+static inline struct bkey_packed *unwritten_whiteouts_start(struct btree *b)
 {
-	return (void *) ((u64 *) btree_data_end(c, b) - b->whiteout_u64s);
+	return (void *) ((u64 *) btree_data_end(b) - b->whiteout_u64s);
 }
 
-static inline struct bkey_packed *unwritten_whiteouts_end(struct bch_fs *c,
-							  struct btree *b)
+static inline struct bkey_packed *unwritten_whiteouts_end(struct btree *b)
 {
-	return btree_data_end(c, b);
+	return btree_data_end(b);
 }
 
 static inline void *write_block(struct btree *b)
@@ -221,13 +219,11 @@ static inline bool bkey_written(struct btree *b, struct bkey_packed *k)
 	return __btree_addr_written(b, k);
 }
 
-static inline ssize_t __bch_btree_u64s_remaining(struct bch_fs *c,
-						 struct btree *b,
-						 void *end)
+static inline ssize_t __bch2_btree_u64s_remaining(struct btree *b, void *end)
 {
 	ssize_t used = bset_byte_offset(b, end) / sizeof(u64) +
 		b->whiteout_u64s;
-	ssize_t total = c->opts.btree_node_size >> 3;
+	ssize_t total = btree_buf_bytes(b) >> 3;
 
 	/* Always leave one extra u64 for bch2_varint_decode: */
 	used++;
@@ -235,10 +231,9 @@ static inline ssize_t __bch_btree_u64s_remaining(struct bch_fs *c,
 	return total - used;
 }
 
-static inline size_t bch_btree_keys_u64s_remaining(struct bch_fs *c,
-						   struct btree *b)
+static inline size_t bch2_btree_keys_u64s_remaining(struct btree *b)
 {
-	ssize_t remaining = __bch_btree_u64s_remaining(c, b,
+	ssize_t remaining = __bch2_btree_u64s_remaining(b,
 				btree_bkey_last(b, bset_tree_last(b)));
 
 	BUG_ON(remaining < 0);
@@ -260,14 +255,13 @@ static inline unsigned btree_write_set_buffer(struct btree *b)
 	return 8 << BTREE_WRITE_SET_U64s_BITS;
 }
 
-static inline struct btree_node_entry *want_new_bset(struct bch_fs *c,
-						     struct btree *b)
+static inline struct btree_node_entry *want_new_bset(struct bch_fs *c, struct btree *b)
 {
 	struct bset_tree *t = bset_tree_last(b);
 	struct btree_node_entry *bne = max(write_block(b),
 			(void *) btree_bkey_last(b, bset_tree_last(b)));
 	ssize_t remaining_space =
-		__bch_btree_u64s_remaining(c, b, bne->keys.start);
+		__bch2_btree_u64s_remaining(b, bne->keys.start);
 
 	if (unlikely(bset_written(b, bset(b, t)))) {
 		if (remaining_space > (ssize_t) (block_bytes(c) >> 3))
@@ -281,12 +275,11 @@ static inline struct btree_node_entry *want_new_bset(struct bch_fs *c,
 	return NULL;
 }
 
-static inline void push_whiteout(struct bch_fs *c, struct btree *b,
-				 struct bpos pos)
+static inline void push_whiteout(struct btree *b, struct bpos pos)
 {
 	struct bkey_packed k;
 
-	BUG_ON(bch_btree_keys_u64s_remaining(c, b) < BKEY_U64s);
+	BUG_ON(bch2_btree_keys_u64s_remaining(b) < BKEY_U64s);
 	EBUG_ON(btree_node_just_written(b));
 
 	if (!bkey_pack_pos(&k, pos, b)) {
@@ -299,20 +292,19 @@ static inline void push_whiteout(struct bch_fs *c, struct btree *b,
 	k.needs_whiteout = true;
 
 	b->whiteout_u64s += k.u64s;
-	bkey_p_copy(unwritten_whiteouts_start(c, b), &k);
+	bkey_p_copy(unwritten_whiteouts_start(b), &k);
 }
 
 /*
  * write lock must be held on @b (else the dirty bset that we were going to
  * insert into could be written out from under us)
  */
-static inline bool bch2_btree_node_insert_fits(struct bch_fs *c,
-					       struct btree *b, unsigned u64s)
+static inline bool bch2_btree_node_insert_fits(struct btree *b, unsigned u64s)
 {
 	if (unlikely(btree_node_need_rewrite(b)))
 		return false;
 
-	return u64s <= bch_btree_keys_u64s_remaining(c, b);
+	return u64s <= bch2_btree_keys_u64s_remaining(b);
 }
 
 void bch2_btree_updates_to_text(struct printbuf *, struct bch_fs *);
diff --git a/fs/bcachefs/btree_write_buffer.c b/fs/bcachefs/btree_write_buffer.c
index 5c1169c78daf..ac7844861966 100644
--- a/fs/bcachefs/btree_write_buffer.c
+++ b/fs/bcachefs/btree_write_buffer.c
@@ -125,13 +125,12 @@ static inline int wb_flush_one(struct btree_trans *trans, struct btree_iter *ite
 			       struct btree_write_buffered_key *wb,
 			       bool *write_locked, size_t *fast)
 {
-	struct bch_fs *c = trans->c;
 	struct btree_path *path;
 	int ret;
 
 	EBUG_ON(!wb->journal_seq);
-	EBUG_ON(!c->btree_write_buffer.flushing.pin.seq);
-	EBUG_ON(c->btree_write_buffer.flushing.pin.seq > wb->journal_seq);
+	EBUG_ON(!trans->c->btree_write_buffer.flushing.pin.seq);
+	EBUG_ON(trans->c->btree_write_buffer.flushing.pin.seq > wb->journal_seq);
 
 	ret = bch2_btree_iter_traverse(iter);
 	if (ret)
@@ -155,7 +154,7 @@ static inline int wb_flush_one(struct btree_trans *trans, struct btree_iter *ite
 		*write_locked = true;
 	}
 
-	if (unlikely(!bch2_btree_node_insert_fits(c, path->l[0].b, wb->k.k.u64s))) {
+	if (unlikely(!bch2_btree_node_insert_fits(path->l[0].b, wb->k.k.u64s))) {
 		*write_locked = false;
 		return wb_flush_one_slowpath(trans, iter, wb);
 	}
diff --git a/fs/bcachefs/buckets.c b/fs/bcachefs/buckets.c
index d83ea0e53df3..54f7826ac498 100644
--- a/fs/bcachefs/buckets.c
+++ b/fs/bcachefs/buckets.c
@@ -25,7 +25,7 @@
 
 #include <linux/preempt.h>
 
-static inline void fs_usage_data_type_to_base(struct bch_fs_usage *fs_usage,
+static inline void fs_usage_data_type_to_base(struct bch_fs_usage_base *fs_usage,
 					      enum bch_data_type data_type,
 					      s64 sectors)
 {
@@ -54,20 +54,20 @@ void bch2_fs_usage_initialize(struct bch_fs *c)
 		bch2_fs_usage_acc_to_base(c, i);
 
 	for (unsigned i = 0; i < BCH_REPLICAS_MAX; i++)
-		usage->reserved += usage->persistent_reserved[i];
+		usage->b.reserved += usage->persistent_reserved[i];
 
 	for (unsigned i = 0; i < c->replicas.nr; i++) {
 		struct bch_replicas_entry_v1 *e =
 			cpu_replicas_entry(&c->replicas, i);
 
-		fs_usage_data_type_to_base(usage, e->data_type, usage->replicas[i]);
+		fs_usage_data_type_to_base(&usage->b, e->data_type, usage->replicas[i]);
 	}
 
 	for_each_member_device(c, ca) {
 		struct bch_dev_usage dev = bch2_dev_usage_read(ca);
 
-		usage->hidden += (dev.d[BCH_DATA_sb].buckets +
-				  dev.d[BCH_DATA_journal].buckets) *
+		usage->b.hidden += (dev.d[BCH_DATA_sb].buckets +
+				    dev.d[BCH_DATA_journal].buckets) *
 			ca->mi.bucket_size;
 	}
 
@@ -188,15 +188,15 @@ void bch2_fs_usage_to_text(struct printbuf *out,
 	prt_printf(out, "capacity:\t\t\t%llu\n", c->capacity);
 
 	prt_printf(out, "hidden:\t\t\t\t%llu\n",
-	       fs_usage->u.hidden);
+	       fs_usage->u.b.hidden);
 	prt_printf(out, "data:\t\t\t\t%llu\n",
-	       fs_usage->u.data);
+	       fs_usage->u.b.data);
 	prt_printf(out, "cached:\t\t\t\t%llu\n",
-	       fs_usage->u.cached);
+	       fs_usage->u.b.cached);
 	prt_printf(out, "reserved:\t\t\t%llu\n",
-	       fs_usage->u.reserved);
+	       fs_usage->u.b.reserved);
 	prt_printf(out, "nr_inodes:\t\t\t%llu\n",
-	       fs_usage->u.nr_inodes);
+	       fs_usage->u.b.nr_inodes);
 	prt_printf(out, "online reserved:\t\t%llu\n",
 	       fs_usage->online_reserved);
 
@@ -225,10 +225,10 @@ static u64 reserve_factor(u64 r)
 
 u64 bch2_fs_sectors_used(struct bch_fs *c, struct bch_fs_usage_online *fs_usage)
 {
-	return min(fs_usage->u.hidden +
-		   fs_usage->u.btree +
-		   fs_usage->u.data +
-		   reserve_factor(fs_usage->u.reserved +
+	return min(fs_usage->u.b.hidden +
+		   fs_usage->u.b.btree +
+		   fs_usage->u.b.data +
+		   reserve_factor(fs_usage->u.b.reserved +
 				  fs_usage->online_reserved),
 		   c->capacity);
 }
@@ -240,17 +240,17 @@ __bch2_fs_usage_read_short(struct bch_fs *c)
 	u64 data, reserved;
 
 	ret.capacity = c->capacity -
-		bch2_fs_usage_read_one(c, &c->usage_base->hidden);
+		bch2_fs_usage_read_one(c, &c->usage_base->b.hidden);
 
-	data		= bch2_fs_usage_read_one(c, &c->usage_base->data) +
-		bch2_fs_usage_read_one(c, &c->usage_base->btree);
-	reserved	= bch2_fs_usage_read_one(c, &c->usage_base->reserved) +
+	data		= bch2_fs_usage_read_one(c, &c->usage_base->b.data) +
+		bch2_fs_usage_read_one(c, &c->usage_base->b.btree);
+	reserved	= bch2_fs_usage_read_one(c, &c->usage_base->b.reserved) +
 		percpu_u64_get(c->online_reserved);
 
 	ret.used	= min(ret.capacity, data + reserve_factor(reserved));
 	ret.free	= ret.capacity - ret.used;
 
-	ret.nr_inodes	= bch2_fs_usage_read_one(c, &c->usage_base->nr_inodes);
+	ret.nr_inodes	= bch2_fs_usage_read_one(c, &c->usage_base->b.nr_inodes);
 
 	return ret;
 }
@@ -284,7 +284,7 @@ void bch2_dev_usage_to_text(struct printbuf *out, struct bch_dev_usage *usage)
 	prt_newline(out);
 
 	for (unsigned i = 0; i < BCH_DATA_NR; i++) {
-		prt_str(out, bch2_data_types[i]);
+		bch2_prt_data_type(out, i);
 		prt_tab(out);
 		prt_u64(out, usage->d[i].buckets);
 		prt_tab_rjust(out);
@@ -308,9 +308,9 @@ void bch2_dev_usage_update(struct bch_fs *c, struct bch_dev *ca,
 	fs_usage = fs_usage_ptr(c, journal_seq, gc);
 
 	if (data_type_is_hidden(old->data_type))
-		fs_usage->hidden -= ca->mi.bucket_size;
+		fs_usage->b.hidden -= ca->mi.bucket_size;
 	if (data_type_is_hidden(new->data_type))
-		fs_usage->hidden += ca->mi.bucket_size;
+		fs_usage->b.hidden += ca->mi.bucket_size;
 
 	u = dev_usage_ptr(ca, journal_seq, gc);
 
@@ -359,7 +359,7 @@ static inline int __update_replicas(struct bch_fs *c,
 	if (idx < 0)
 		return -1;
 
-	fs_usage_data_type_to_base(fs_usage, r->data_type, sectors);
+	fs_usage_data_type_to_base(&fs_usage->b, r->data_type, sectors);
 	fs_usage->replicas[idx]		+= sectors;
 	return 0;
 }
@@ -394,7 +394,7 @@ int bch2_update_replicas(struct bch_fs *c, struct bkey_s_c k,
 
 	preempt_disable();
 	fs_usage = fs_usage_ptr(c, journal_seq, gc);
-	fs_usage_data_type_to_base(fs_usage, r->data_type, sectors);
+	fs_usage_data_type_to_base(&fs_usage->b, r->data_type, sectors);
 	fs_usage->replicas[idx]		+= sectors;
 	preempt_enable();
 err:
@@ -523,8 +523,8 @@ int bch2_mark_metadata_bucket(struct bch_fs *c, struct bch_dev *ca,
 	if (bch2_fs_inconsistent_on(g->data_type &&
 			g->data_type != data_type, c,
 			"different types of data in same bucket: %s, %s",
-			bch2_data_types[g->data_type],
-			bch2_data_types[data_type])) {
+			bch2_data_type_str(g->data_type),
+			bch2_data_type_str(data_type))) {
 		ret = -EIO;
 		goto err;
 	}
@@ -532,7 +532,7 @@ int bch2_mark_metadata_bucket(struct bch_fs *c, struct bch_dev *ca,
 	if (bch2_fs_inconsistent_on((u64) g->dirty_sectors + sectors > ca->mi.bucket_size, c,
 			"bucket %u:%zu gen %u data type %s sector count overflow: %u + %u > bucket size",
 			ca->dev_idx, b, g->gen,
-			bch2_data_types[g->data_type ?: data_type],
+			bch2_data_type_str(g->data_type ?: data_type),
 			g->dirty_sectors, sectors)) {
 		ret = -EIO;
 		goto err;
@@ -575,7 +575,7 @@ int bch2_check_bucket_ref(struct btree_trans *trans,
 			"bucket %u:%zu gen %u data type %s: ptr gen %u newer than bucket gen\n"
 			"while marking %s",
 			ptr->dev, bucket_nr, b_gen,
-			bch2_data_types[bucket_data_type ?: ptr_data_type],
+			bch2_data_type_str(bucket_data_type ?: ptr_data_type),
 			ptr->gen,
 			(bch2_bkey_val_to_text(&buf, c, k), buf.buf));
 		ret = -EIO;
@@ -588,7 +588,7 @@ int bch2_check_bucket_ref(struct btree_trans *trans,
 			"bucket %u:%zu gen %u data type %s: ptr gen %u too stale\n"
 			"while marking %s",
 			ptr->dev, bucket_nr, b_gen,
-			bch2_data_types[bucket_data_type ?: ptr_data_type],
+			bch2_data_type_str(bucket_data_type ?: ptr_data_type),
 			ptr->gen,
 			(printbuf_reset(&buf),
 			 bch2_bkey_val_to_text(&buf, c, k), buf.buf));
@@ -603,7 +603,7 @@ int bch2_check_bucket_ref(struct btree_trans *trans,
 			"while marking %s",
 			ptr->dev, bucket_nr, b_gen,
 			*bucket_gen(ca, bucket_nr),
-			bch2_data_types[bucket_data_type ?: ptr_data_type],
+			bch2_data_type_str(bucket_data_type ?: ptr_data_type),
 			ptr->gen,
 			(printbuf_reset(&buf),
 			 bch2_bkey_val_to_text(&buf, c, k), buf.buf));
@@ -624,8 +624,8 @@ int bch2_check_bucket_ref(struct btree_trans *trans,
 			"bucket %u:%zu gen %u different types of data in same bucket: %s, %s\n"
 			"while marking %s",
 			ptr->dev, bucket_nr, b_gen,
-			bch2_data_types[bucket_data_type],
-			bch2_data_types[ptr_data_type],
+			bch2_data_type_str(bucket_data_type),
+			bch2_data_type_str(ptr_data_type),
 			(printbuf_reset(&buf),
 			 bch2_bkey_val_to_text(&buf, c, k), buf.buf));
 		ret = -EIO;
@@ -638,7 +638,7 @@ int bch2_check_bucket_ref(struct btree_trans *trans,
 			"bucket %u:%zu gen %u data type %s sector count overflow: %u + %lli > U32_MAX\n"
 			"while marking %s",
 			ptr->dev, bucket_nr, b_gen,
-			bch2_data_types[bucket_data_type ?: ptr_data_type],
+			bch2_data_type_str(bucket_data_type ?: ptr_data_type),
 			bucket_sectors, sectors,
 			(printbuf_reset(&buf),
 			 bch2_bkey_val_to_text(&buf, c, k), buf.buf));
@@ -677,11 +677,11 @@ void bch2_trans_fs_usage_revert(struct btree_trans *trans,
 		BUG_ON(__update_replicas(c, dst, &d->r, -d->delta));
 	}
 
-	dst->nr_inodes -= deltas->nr_inodes;
+	dst->b.nr_inodes -= deltas->nr_inodes;
 
 	for (i = 0; i < BCH_REPLICAS_MAX; i++) {
 		added				-= deltas->persistent_reserved[i];
-		dst->reserved			-= deltas->persistent_reserved[i];
+		dst->b.reserved			-= deltas->persistent_reserved[i];
 		dst->persistent_reserved[i]	-= deltas->persistent_reserved[i];
 	}
 
@@ -694,48 +694,25 @@ void bch2_trans_fs_usage_revert(struct btree_trans *trans,
 	percpu_up_read(&c->mark_lock);
 }
 
-int bch2_trans_fs_usage_apply(struct btree_trans *trans,
-			      struct replicas_delta_list *deltas)
+void bch2_trans_account_disk_usage_change(struct btree_trans *trans)
 {
 	struct bch_fs *c = trans->c;
+	u64 disk_res_sectors = trans->disk_res ? trans->disk_res->sectors : 0;
 	static int warned_disk_usage = 0;
 	bool warn = false;
-	u64 disk_res_sectors = trans->disk_res ? trans->disk_res->sectors : 0;
-	struct replicas_delta *d, *d2;
-	struct replicas_delta *top = (void *) deltas->d + deltas->used;
-	struct bch_fs_usage *dst;
-	s64 added = 0, should_not_have_added;
-	unsigned i;
 
 	percpu_down_read(&c->mark_lock);
 	preempt_disable();
-	dst = fs_usage_ptr(c, trans->journal_res.seq, false);
-
-	for (d = deltas->d; d != top; d = replicas_delta_next(d)) {
-		switch (d->r.data_type) {
-		case BCH_DATA_btree:
-		case BCH_DATA_user:
-		case BCH_DATA_parity:
-			added += d->delta;
-		}
+	struct bch_fs_usage_base *dst = &fs_usage_ptr(c, trans->journal_res.seq, false)->b;
+	struct bch_fs_usage_base *src = &trans->fs_usage_delta;
 
-		if (__update_replicas(c, dst, &d->r, d->delta))
-			goto need_mark;
-	}
-
-	dst->nr_inodes += deltas->nr_inodes;
-
-	for (i = 0; i < BCH_REPLICAS_MAX; i++) {
-		added				+= deltas->persistent_reserved[i];
-		dst->reserved			+= deltas->persistent_reserved[i];
-		dst->persistent_reserved[i]	+= deltas->persistent_reserved[i];
-	}
+	s64 added = src->btree + src->data + src->reserved;
 
 	/*
 	 * Not allowed to reduce sectors_available except by getting a
 	 * reservation:
 	 */
-	should_not_have_added = added - (s64) disk_res_sectors;
+	s64 should_not_have_added = added - (s64) disk_res_sectors;
 	if (unlikely(should_not_have_added > 0)) {
 		u64 old, new, v = atomic64_read(&c->sectors_available);
 
@@ -754,6 +731,13 @@ int bch2_trans_fs_usage_apply(struct btree_trans *trans,
 		this_cpu_sub(*c->online_reserved, added);
 	}
 
+	dst->hidden	+= src->hidden;
+	dst->btree	+= src->btree;
+	dst->data	+= src->data;
+	dst->cached	+= src->cached;
+	dst->reserved	+= src->reserved;
+	dst->nr_inodes	+= src->nr_inodes;
+
 	preempt_enable();
 	percpu_up_read(&c->mark_lock);
 
@@ -761,6 +745,34 @@ int bch2_trans_fs_usage_apply(struct btree_trans *trans,
 		bch2_trans_inconsistent(trans,
 					"disk usage increased %lli more than %llu sectors reserved)",
 					should_not_have_added, disk_res_sectors);
+}
+
+int bch2_trans_fs_usage_apply(struct btree_trans *trans,
+			      struct replicas_delta_list *deltas)
+{
+	struct bch_fs *c = trans->c;
+	struct replicas_delta *d, *d2;
+	struct replicas_delta *top = (void *) deltas->d + deltas->used;
+	struct bch_fs_usage *dst;
+	unsigned i;
+
+	percpu_down_read(&c->mark_lock);
+	preempt_disable();
+	dst = fs_usage_ptr(c, trans->journal_res.seq, false);
+
+	for (d = deltas->d; d != top; d = replicas_delta_next(d))
+		if (__update_replicas(c, dst, &d->r, d->delta))
+			goto need_mark;
+
+	dst->b.nr_inodes += deltas->nr_inodes;
+
+	for (i = 0; i < BCH_REPLICAS_MAX; i++) {
+		dst->b.reserved			+= deltas->persistent_reserved[i];
+		dst->persistent_reserved[i]	+= deltas->persistent_reserved[i];
+	}
+
+	preempt_enable();
+	percpu_up_read(&c->mark_lock);
 	return 0;
 need_mark:
 	/* revert changes: */
@@ -1084,7 +1096,7 @@ static int __trigger_reservation(struct btree_trans *trans,
 		struct bch_fs_usage *fs_usage = this_cpu_ptr(c->usage_gc);
 
 		replicas = min(replicas, ARRAY_SIZE(fs_usage->persistent_reserved));
-		fs_usage->reserved				+= sectors;
+		fs_usage->b.reserved				+= sectors;
 		fs_usage->persistent_reserved[replicas - 1]	+= sectors;
 
 		preempt_enable();
@@ -1130,9 +1142,9 @@ static int __bch2_trans_mark_metadata_bucket(struct btree_trans *trans,
 			"bucket %llu:%llu gen %u different types of data in same bucket: %s, %s\n"
 			"while marking %s",
 			iter.pos.inode, iter.pos.offset, a->v.gen,
-			bch2_data_types[a->v.data_type],
-			bch2_data_types[type],
-			bch2_data_types[type]);
+			bch2_data_type_str(a->v.data_type),
+			bch2_data_type_str(type),
+			bch2_data_type_str(type));
 		ret = -EIO;
 		goto err;
 	}
diff --git a/fs/bcachefs/buckets.h b/fs/bcachefs/buckets.h
index 2c95cc5d86be..6387e039f789 100644
--- a/fs/bcachefs/buckets.h
+++ b/fs/bcachefs/buckets.h
@@ -356,6 +356,8 @@ int bch2_trigger_reservation(struct btree_trans *, enum btree_id, unsigned,
 	ret;											\
 })
 
+void bch2_trans_account_disk_usage_change(struct btree_trans *);
+
 void bch2_trans_fs_usage_revert(struct btree_trans *, struct replicas_delta_list *);
 int bch2_trans_fs_usage_apply(struct btree_trans *, struct replicas_delta_list *);
 
@@ -385,6 +387,21 @@ static inline bool is_superblock_bucket(struct bch_dev *ca, u64 b)
 	return false;
 }
 
+static inline const char *bch2_data_type_str(enum bch_data_type type)
+{
+	return type < BCH_DATA_NR
+		? __bch2_data_types[type]
+		: "(invalid data type)";
+}
+
+static inline void bch2_prt_data_type(struct printbuf *out, enum bch_data_type type)
+{
+	if (type < BCH_DATA_NR)
+		prt_str(out, __bch2_data_types[type]);
+	else
+		prt_printf(out, "(invalid data type %u)", type);
+}
+
 /* disk reservations: */
 
 static inline void bch2_disk_reservation_put(struct bch_fs *c,
diff --git a/fs/bcachefs/buckets_types.h b/fs/bcachefs/buckets_types.h
index 783f71017204..6a31740222a7 100644
--- a/fs/bcachefs/buckets_types.h
+++ b/fs/bcachefs/buckets_types.h
@@ -45,23 +45,18 @@ struct bch_dev_usage {
 	}			d[BCH_DATA_NR];
 };
 
-struct bch_fs_usage {
-	/* all fields are in units of 512 byte sectors: */
+struct bch_fs_usage_base {
 	u64			hidden;
 	u64			btree;
 	u64			data;
 	u64			cached;
 	u64			reserved;
 	u64			nr_inodes;
+};
 
-	/* XXX: add stats for compression ratio */
-#if 0
-	u64			uncompressed;
-	u64			compressed;
-#endif
-
-	/* broken out: */
-
+struct bch_fs_usage {
+	/* all fields are in units of 512 byte sectors: */
+	struct bch_fs_usage_base b;
 	u64			persistent_reserved[BCH_REPLICAS_MAX];
 	u64			replicas[];
 };
diff --git a/fs/bcachefs/clock.c b/fs/bcachefs/clock.c
index f41889093a2c..363644451106 100644
--- a/fs/bcachefs/clock.c
+++ b/fs/bcachefs/clock.c
@@ -109,7 +109,7 @@ void bch2_kthread_io_clock_wait(struct io_clock *clock,
 	if (cpu_timeout != MAX_SCHEDULE_TIMEOUT)
 		mod_timer(&wait.cpu_timer, cpu_timeout + jiffies);
 
-	while (1) {
+	do {
 		set_current_state(TASK_INTERRUPTIBLE);
 		if (kthread && kthread_should_stop())
 			break;
@@ -119,7 +119,7 @@ void bch2_kthread_io_clock_wait(struct io_clock *clock,
 
 		schedule();
 		try_to_freeze();
-	}
+	} while (0);
 
 	__set_current_state(TASK_RUNNING);
 	del_timer_sync(&wait.cpu_timer);
diff --git a/fs/bcachefs/compress.h b/fs/bcachefs/compress.h
index 607fd5e232c9..58c2eb45570f 100644
--- a/fs/bcachefs/compress.h
+++ b/fs/bcachefs/compress.h
@@ -47,6 +47,14 @@ static inline enum bch_compression_type bch2_compression_opt_to_type(unsigned v)
 	return __bch2_compression_opt_to_type[bch2_compression_decode(v).type];
 }
 
+static inline void bch2_prt_compression_type(struct printbuf *out, enum bch_compression_type type)
+{
+	if (type < BCH_COMPRESSION_TYPE_NR)
+		prt_str(out, __bch2_compression_types[type]);
+	else
+		prt_printf(out, "(invalid compression type %u)", type);
+}
+
 int bch2_bio_uncompress_inplace(struct bch_fs *, struct bio *,
 				struct bch_extent_crc_unpacked *);
 int bch2_bio_uncompress(struct bch_fs *, struct bio *, struct bio *,
diff --git a/fs/bcachefs/data_update.c b/fs/bcachefs/data_update.c
index 6f13477ff652..4150feca42a2 100644
--- a/fs/bcachefs/data_update.c
+++ b/fs/bcachefs/data_update.c
@@ -285,9 +285,7 @@ restart_drop_extra_replicas:
 						k.k->p, bkey_start_pos(&insert->k)) ?:
 			bch2_insert_snapshot_whiteouts(trans, m->btree_id,
 						k.k->p, insert->k.p) ?:
-			bch2_bkey_set_needs_rebalance(c, insert,
-						      op->opts.background_target,
-						      op->opts.background_compression) ?:
+			bch2_bkey_set_needs_rebalance(c, insert, &op->opts) ?:
 			bch2_trans_update(trans, &iter, insert,
 				BTREE_UPDATE_INTERNAL_SNAPSHOT_NODE) ?:
 			bch2_trans_commit(trans, &op->res,
@@ -529,7 +527,7 @@ int bch2_data_update_init(struct btree_trans *trans,
 		BCH_WRITE_DATA_ENCODED|
 		BCH_WRITE_MOVE|
 		m->data_opts.write_flags;
-	m->op.compression_opt	= io_opts.background_compression ?: io_opts.compression;
+	m->op.compression_opt	= background_compression(io_opts);
 	m->op.watermark		= m->data_opts.btree_insert_flags & BCH_WATERMARK_MASK;
 
 	bkey_for_each_ptr(ptrs, ptr)
diff --git a/fs/bcachefs/debug.c b/fs/bcachefs/debug.c
index d6418948495f..cadda9bbe4a4 100644
--- a/fs/bcachefs/debug.c
+++ b/fs/bcachefs/debug.c
@@ -44,19 +44,19 @@ static bool bch2_btree_verify_replica(struct bch_fs *c, struct btree *b,
 		return false;
 
 	bio = bio_alloc_bioset(ca->disk_sb.bdev,
-			       buf_pages(n_sorted, btree_bytes(c)),
+			       buf_pages(n_sorted, btree_buf_bytes(b)),
 			       REQ_OP_READ|REQ_META,
 			       GFP_NOFS,
 			       &c->btree_bio);
 	bio->bi_iter.bi_sector	= pick.ptr.offset;
-	bch2_bio_map(bio, n_sorted, btree_bytes(c));
+	bch2_bio_map(bio, n_sorted, btree_buf_bytes(b));
 
 	submit_bio_wait(bio);
 
 	bio_put(bio);
 	percpu_ref_put(&ca->io_ref);
 
-	memcpy(n_ondisk, n_sorted, btree_bytes(c));
+	memcpy(n_ondisk, n_sorted, btree_buf_bytes(b));
 
 	v->written = 0;
 	if (bch2_btree_node_read_done(c, ca, v, false, &saw_error) || saw_error)
@@ -137,7 +137,7 @@ void __bch2_btree_verify(struct bch_fs *c, struct btree *b)
 	mutex_lock(&c->verify_lock);
 
 	if (!c->verify_ondisk) {
-		c->verify_ondisk = kvpmalloc(btree_bytes(c), GFP_KERNEL);
+		c->verify_ondisk = kvpmalloc(btree_buf_bytes(b), GFP_KERNEL);
 		if (!c->verify_ondisk)
 			goto out;
 	}
@@ -199,19 +199,19 @@ void bch2_btree_node_ondisk_to_text(struct printbuf *out, struct bch_fs *c,
 		return;
 	}
 
-	n_ondisk = kvpmalloc(btree_bytes(c), GFP_KERNEL);
+	n_ondisk = kvpmalloc(btree_buf_bytes(b), GFP_KERNEL);
 	if (!n_ondisk) {
 		prt_printf(out, "memory allocation failure\n");
 		goto out;
 	}
 
 	bio = bio_alloc_bioset(ca->disk_sb.bdev,
-			       buf_pages(n_ondisk, btree_bytes(c)),
+			       buf_pages(n_ondisk, btree_buf_bytes(b)),
 			       REQ_OP_READ|REQ_META,
 			       GFP_NOFS,
 			       &c->btree_bio);
 	bio->bi_iter.bi_sector	= pick.ptr.offset;
-	bch2_bio_map(bio, n_ondisk, btree_bytes(c));
+	bch2_bio_map(bio, n_ondisk, btree_buf_bytes(b));
 
 	ret = submit_bio_wait(bio);
 	if (ret) {
@@ -293,7 +293,7 @@ void bch2_btree_node_ondisk_to_text(struct printbuf *out, struct bch_fs *c,
 out:
 	if (bio)
 		bio_put(bio);
-	kvpfree(n_ondisk, btree_bytes(c));
+	kvpfree(n_ondisk, btree_buf_bytes(b));
 	percpu_ref_put(&ca->io_ref);
 }
 
diff --git a/fs/bcachefs/dirent_format.h b/fs/bcachefs/dirent_format.h
new file mode 100644
index 000000000000..5e116b88e814
--- /dev/null
+++ b/fs/bcachefs/dirent_format.h
@@ -0,0 +1,42 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _BCACHEFS_DIRENT_FORMAT_H
+#define _BCACHEFS_DIRENT_FORMAT_H
+
+/*
+ * Dirents (and xattrs) have to implement string lookups; since our b-tree
+ * doesn't support arbitrary length strings for the key, we instead index by a
+ * 64 bit hash (currently truncated sha1) of the string, stored in the offset
+ * field of the key - using linear probing to resolve hash collisions. This also
+ * provides us with the readdir cookie posix requires.
+ *
+ * Linear probing requires us to use whiteouts for deletions, in the event of a
+ * collision:
+ */
+
+struct bch_dirent {
+	struct bch_val		v;
+
+	/* Target inode number: */
+	union {
+	__le64			d_inum;
+	struct {		/* DT_SUBVOL */
+	__le32			d_child_subvol;
+	__le32			d_parent_subvol;
+	};
+	};
+
+	/*
+	 * Copy of mode bits 12-15 from the target inode - so userspace can get
+	 * the filetype without having to do a stat()
+	 */
+	__u8			d_type;
+
+	__u8			d_name[];
+} __packed __aligned(8);
+
+#define DT_SUBVOL	16
+#define BCH_DT_MAX	17
+
+#define BCH_NAME_MAX	512
+
+#endif /* _BCACHEFS_DIRENT_FORMAT_H */
diff --git a/fs/bcachefs/ec.c b/fs/bcachefs/ec.c
index d802bc63c8d0..d503af270024 100644
--- a/fs/bcachefs/ec.c
+++ b/fs/bcachefs/ec.c
@@ -190,7 +190,7 @@ static int bch2_trans_mark_stripe_bucket(struct btree_trans *trans,
 					       a->v.stripe_redundancy, trans,
 				"bucket %llu:%llu gen %u data type %s dirty_sectors %u: multiple stripes using same bucket (%u, %llu)",
 				iter.pos.inode, iter.pos.offset, a->v.gen,
-				bch2_data_types[a->v.data_type],
+				bch2_data_type_str(a->v.data_type),
 				a->v.dirty_sectors,
 				a->v.stripe, s.k->p.offset)) {
 			ret = -EIO;
@@ -200,7 +200,7 @@ static int bch2_trans_mark_stripe_bucket(struct btree_trans *trans,
 		if (bch2_trans_inconsistent_on(data_type && a->v.dirty_sectors, trans,
 				"bucket %llu:%llu gen %u data type %s dirty_sectors %u: data already in stripe bucket %llu",
 				iter.pos.inode, iter.pos.offset, a->v.gen,
-				bch2_data_types[a->v.data_type],
+				bch2_data_type_str(a->v.data_type),
 				a->v.dirty_sectors,
 				s.k->p.offset)) {
 			ret = -EIO;
@@ -367,7 +367,7 @@ int bch2_trigger_stripe(struct btree_trans *trans,
 		}
 	}
 
-	if (!(flags & (BTREE_TRIGGER_TRANSACTIONAL|BTREE_TRIGGER_GC))) {
+	if (flags & BTREE_TRIGGER_ATOMIC) {
 		struct stripe *m = genradix_ptr(&c->stripes, idx);
 
 		if (!m) {
diff --git a/fs/bcachefs/ec_format.h b/fs/bcachefs/ec_format.h
new file mode 100644
index 000000000000..44ce88ba08d7
--- /dev/null
+++ b/fs/bcachefs/ec_format.h
@@ -0,0 +1,19 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _BCACHEFS_EC_FORMAT_H
+#define _BCACHEFS_EC_FORMAT_H
+
+struct bch_stripe {
+	struct bch_val		v;
+	__le16			sectors;
+	__u8			algorithm;
+	__u8			nr_blocks;
+	__u8			nr_redundant;
+
+	__u8			csum_granularity_bits;
+	__u8			csum_type;
+	__u8			pad;
+
+	struct bch_extent_ptr	ptrs[];
+} __packed __aligned(8);
+
+#endif /* _BCACHEFS_EC_FORMAT_H */
diff --git a/fs/bcachefs/extents.c b/fs/bcachefs/extents.c
index 82ec056f4cdb..61395b113df9 100644
--- a/fs/bcachefs/extents.c
+++ b/fs/bcachefs/extents.c
@@ -8,6 +8,7 @@
 
 #include "bcachefs.h"
 #include "bkey_methods.h"
+#include "btree_cache.h"
 #include "btree_gc.h"
 #include "btree_io.h"
 #include "btree_iter.h"
@@ -1018,12 +1019,12 @@ void bch2_bkey_ptrs_to_text(struct printbuf *out, struct bch_fs *c,
 			struct bch_extent_crc_unpacked crc =
 				bch2_extent_crc_unpack(k.k, entry_to_crc(entry));
 
-			prt_printf(out, "crc: c_size %u size %u offset %u nonce %u csum %s compress %s",
+			prt_printf(out, "crc: c_size %u size %u offset %u nonce %u csum %s compress ",
 			       crc.compressed_size,
 			       crc.uncompressed_size,
 			       crc.offset, crc.nonce,
-			       bch2_csum_types[crc.csum_type],
-			       bch2_compression_types[crc.compression_type]);
+			       bch2_csum_types[crc.csum_type]);
+			bch2_prt_compression_type(out, crc.compression_type);
 			break;
 		}
 		case BCH_EXTENT_ENTRY_stripe_ptr: {
@@ -1334,10 +1335,12 @@ bool bch2_bkey_needs_rebalance(struct bch_fs *c, struct bkey_s_c k)
 }
 
 int bch2_bkey_set_needs_rebalance(struct bch_fs *c, struct bkey_i *_k,
-				  unsigned target, unsigned compression)
+				  struct bch_io_opts *opts)
 {
 	struct bkey_s k = bkey_i_to_s(_k);
 	struct bch_extent_rebalance *r;
+	unsigned target = opts->background_target;
+	unsigned compression = background_compression(*opts);
 	bool needs_rebalance;
 
 	if (!bkey_extent_is_direct_data(k.k))
diff --git a/fs/bcachefs/extents.h b/fs/bcachefs/extents.h
index a855c94d43dd..6bf839d69e84 100644
--- a/fs/bcachefs/extents.h
+++ b/fs/bcachefs/extents.h
@@ -708,7 +708,7 @@ unsigned bch2_bkey_ptrs_need_rebalance(struct bch_fs *, struct bkey_s_c,
 bool bch2_bkey_needs_rebalance(struct bch_fs *, struct bkey_s_c);
 
 int bch2_bkey_set_needs_rebalance(struct bch_fs *, struct bkey_i *,
-				  unsigned, unsigned);
+				  struct bch_io_opts *);
 
 /* Generic extent code: */
 
diff --git a/fs/bcachefs/extents_format.h b/fs/bcachefs/extents_format.h
new file mode 100644
index 000000000000..3bd2fdbb0817
--- /dev/null
+++ b/fs/bcachefs/extents_format.h
@@ -0,0 +1,295 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _BCACHEFS_EXTENTS_FORMAT_H
+#define _BCACHEFS_EXTENTS_FORMAT_H
+
+/*
+ * In extent bkeys, the value is a list of pointers (bch_extent_ptr), optionally
+ * preceded by checksum/compression information (bch_extent_crc32 or
+ * bch_extent_crc64).
+ *
+ * One major determining factor in the format of extents is how we handle and
+ * represent extents that have been partially overwritten and thus trimmed:
+ *
+ * If an extent is not checksummed or compressed, when the extent is trimmed we
+ * don't have to remember the extent we originally allocated and wrote: we can
+ * merely adjust ptr->offset to point to the start of the data that is currently
+ * live. The size field in struct bkey records the current (live) size of the
+ * extent, and is also used to mean "size of region on disk that we point to" in
+ * this case.
+ *
+ * Thus an extent that is not checksummed or compressed will consist only of a
+ * list of bch_extent_ptrs, with none of the fields in
+ * bch_extent_crc32/bch_extent_crc64.
+ *
+ * When an extent is checksummed or compressed, it's not possible to read only
+ * the data that is currently live: we have to read the entire extent that was
+ * originally written, and then return only the part of the extent that is
+ * currently live.
+ *
+ * Thus, in addition to the current size of the extent in struct bkey, we need
+ * to store the size of the originally allocated space - this is the
+ * compressed_size and uncompressed_size fields in bch_extent_crc32/64. Also,
+ * when the extent is trimmed, instead of modifying the offset field of the
+ * pointer, we keep a second smaller offset field - "offset into the original
+ * extent of the currently live region".
+ *
+ * The other major determining factor is replication and data migration:
+ *
+ * Each pointer may have its own bch_extent_crc32/64. When doing a replicated
+ * write, we will initially write all the replicas in the same format, with the
+ * same checksum type and compression format - however, when copygc runs later (or
+ * tiering/cache promotion, anything that moves data), it is not in general
+ * going to rewrite all the pointers at once - one of the replicas may be in a
+ * bucket on one device that has very little fragmentation while another lives
+ * in a bucket that has become heavily fragmented, and thus is being rewritten
+ * sooner than the rest.
+ *
+ * Thus it will only move a subset of the pointers (or in the case of
+ * tiering/cache promotion perhaps add a single pointer without dropping any
+ * current pointers), and if the extent has been partially overwritten it must
+ * write only the currently live portion (or copygc would not be able to reduce
+ * fragmentation!) - which necessitates a different bch_extent_crc format for
+ * the new pointer.
+ *
+ * But in the interests of space efficiency, we don't want to store one
+ * bch_extent_crc for each pointer if we don't have to.
+ *
+ * Thus, a bch_extent consists of bch_extent_crc32s, bch_extent_crc64s, and
+ * bch_extent_ptrs appended arbitrarily one after the other. We determine the
+ * type of a given entry with a scheme similar to utf8 (except we're encoding a
+ * type, not a size), encoding the type in the position of the first set bit:
+ *
+ * bch_extent_crc32	- 0b1
+ * bch_extent_ptr	- 0b10
+ * bch_extent_crc64	- 0b100
+ *
+ * We do it this way because bch_extent_crc32 is _very_ constrained on bits (and
+ * bch_extent_crc64 is the least constrained).
+ *
+ * Then, each bch_extent_crc32/64 applies to the pointers that follow after it,
+ * until the next bch_extent_crc32/64.
+ *
+ * If there are no bch_extent_crcs preceding a bch_extent_ptr, then that pointer
+ * is neither checksummed nor compressed.
+ */
+
+#define BCH_EXTENT_ENTRY_TYPES()		\
+	x(ptr,			0)		\
+	x(crc32,		1)		\
+	x(crc64,		2)		\
+	x(crc128,		3)		\
+	x(stripe_ptr,		4)		\
+	x(rebalance,		5)
+#define BCH_EXTENT_ENTRY_MAX	6
+
+enum bch_extent_entry_type {
+#define x(f, n) BCH_EXTENT_ENTRY_##f = n,
+	BCH_EXTENT_ENTRY_TYPES()
+#undef x
+};
+
+/* Compressed/uncompressed size are stored biased by 1: */
+struct bch_extent_crc32 {
+#if defined(__LITTLE_ENDIAN_BITFIELD)
+	__u32			type:2,
+				_compressed_size:7,
+				_uncompressed_size:7,
+				offset:7,
+				_unused:1,
+				csum_type:4,
+				compression_type:4;
+	__u32			csum;
+#elif defined (__BIG_ENDIAN_BITFIELD)
+	__u32			csum;
+	__u32			compression_type:4,
+				csum_type:4,
+				_unused:1,
+				offset:7,
+				_uncompressed_size:7,
+				_compressed_size:7,
+				type:2;
+#endif
+} __packed __aligned(8);
+
+#define CRC32_SIZE_MAX		(1U << 7)
+#define CRC32_NONCE_MAX		0
+
+struct bch_extent_crc64 {
+#if defined(__LITTLE_ENDIAN_BITFIELD)
+	__u64			type:3,
+				_compressed_size:9,
+				_uncompressed_size:9,
+				offset:9,
+				nonce:10,
+				csum_type:4,
+				compression_type:4,
+				csum_hi:16;
+#elif defined (__BIG_ENDIAN_BITFIELD)
+	__u64			csum_hi:16,
+				compression_type:4,
+				csum_type:4,
+				nonce:10,
+				offset:9,
+				_uncompressed_size:9,
+				_compressed_size:9,
+				type:3;
+#endif
+	__u64			csum_lo;
+} __packed __aligned(8);
+
+#define CRC64_SIZE_MAX		(1U << 9)
+#define CRC64_NONCE_MAX		((1U << 10) - 1)
+
+struct bch_extent_crc128 {
+#if defined(__LITTLE_ENDIAN_BITFIELD)
+	__u64			type:4,
+				_compressed_size:13,
+				_uncompressed_size:13,
+				offset:13,
+				nonce:13,
+				csum_type:4,
+				compression_type:4;
+#elif defined (__BIG_ENDIAN_BITFIELD)
+	__u64			compression_type:4,
+				csum_type:4,
+				nonce:13,
+				offset:13,
+				_uncompressed_size:13,
+				_compressed_size:13,
+				type:4;
+#endif
+	struct bch_csum		csum;
+} __packed __aligned(8);
+
+#define CRC128_SIZE_MAX		(1U << 13)
+#define CRC128_NONCE_MAX	((1U << 13) - 1)
+
+/*
+ * @reservation - pointer hasn't been written to, just reserved
+ */
+struct bch_extent_ptr {
+#if defined(__LITTLE_ENDIAN_BITFIELD)
+	__u64			type:1,
+				cached:1,
+				unused:1,
+				unwritten:1,
+				offset:44, /* 8 petabytes */
+				dev:8,
+				gen:8;
+#elif defined (__BIG_ENDIAN_BITFIELD)
+	__u64			gen:8,
+				dev:8,
+				offset:44,
+				unwritten:1,
+				unused:1,
+				cached:1,
+				type:1;
+#endif
+} __packed __aligned(8);
+
+struct bch_extent_stripe_ptr {
+#if defined(__LITTLE_ENDIAN_BITFIELD)
+	__u64			type:5,
+				block:8,
+				redundancy:4,
+				idx:47;
+#elif defined (__BIG_ENDIAN_BITFIELD)
+	__u64			idx:47,
+				redundancy:4,
+				block:8,
+				type:5;
+#endif
+};
+
+struct bch_extent_rebalance {
+#if defined(__LITTLE_ENDIAN_BITFIELD)
+	__u64			type:6,
+				unused:34,
+				compression:8, /* enum bch_compression_opt */
+				target:16;
+#elif defined (__BIG_ENDIAN_BITFIELD)
+	__u64			target:16,
+				compression:8,
+				unused:34,
+				type:6;
+#endif
+};
+
+union bch_extent_entry {
+#if __BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__ ||  __BITS_PER_LONG == 64
+	unsigned long			type;
+#elif __BITS_PER_LONG == 32
+	struct {
+		unsigned long		pad;
+		unsigned long		type;
+	};
+#else
+#error edit for your odd byteorder.
+#endif
+
+#define x(f, n) struct bch_extent_##f	f;
+	BCH_EXTENT_ENTRY_TYPES()
+#undef x
+};
+
+struct bch_btree_ptr {
+	struct bch_val		v;
+
+	__u64			_data[0];
+	struct bch_extent_ptr	start[];
+} __packed __aligned(8);
+
+struct bch_btree_ptr_v2 {
+	struct bch_val		v;
+
+	__u64			mem_ptr;
+	__le64			seq;
+	__le16			sectors_written;
+	__le16			flags;
+	struct bpos		min_key;
+	__u64			_data[0];
+	struct bch_extent_ptr	start[];
+} __packed __aligned(8);
+
+LE16_BITMASK(BTREE_PTR_RANGE_UPDATED,	struct bch_btree_ptr_v2, flags, 0, 1);
+
+struct bch_extent {
+	struct bch_val		v;
+
+	__u64			_data[0];
+	union bch_extent_entry	start[];
+} __packed __aligned(8);
+
+/* Maximum size (in u64s) a single pointer could be: */
+#define BKEY_EXTENT_PTR_U64s_MAX\
+	((sizeof(struct bch_extent_crc128) +			\
+	  sizeof(struct bch_extent_ptr)) / sizeof(__u64))
+
+/* Maximum possible size of an entire extent value: */
+#define BKEY_EXTENT_VAL_U64s_MAX				\
+	(1 + BKEY_EXTENT_PTR_U64s_MAX * (BCH_REPLICAS_MAX + 1))
+
+/* * Maximum possible size of an entire extent, key + value: */
+#define BKEY_EXTENT_U64s_MAX		(BKEY_U64s + BKEY_EXTENT_VAL_U64s_MAX)
+
+/* Btree pointers don't carry around checksums: */
+#define BKEY_BTREE_PTR_VAL_U64s_MAX				\
+	((sizeof(struct bch_btree_ptr_v2) +			\
+	  sizeof(struct bch_extent_ptr) * BCH_REPLICAS_MAX) / sizeof(__u64))
+#define BKEY_BTREE_PTR_U64s_MAX					\
+	(BKEY_U64s + BKEY_BTREE_PTR_VAL_U64s_MAX)
+
+struct bch_reservation {
+	struct bch_val		v;
+
+	__le32			generation;
+	__u8			nr_replicas;
+	__u8			pad[3];
+} __packed __aligned(8);
+
+struct bch_inline_data {
+	struct bch_val		v;
+	u8			data[];
+};
+
+#endif /* _BCACHEFS_EXTENTS_FORMAT_H */
diff --git a/fs/bcachefs/eytzinger.h b/fs/bcachefs/eytzinger.h
index 9637f636e32d..b04750dbf870 100644
--- a/fs/bcachefs/eytzinger.h
+++ b/fs/bcachefs/eytzinger.h
@@ -156,7 +156,7 @@ static inline unsigned inorder_to_eytzinger1(unsigned i, unsigned size)
 }
 
 #define eytzinger1_for_each(_i, _size)			\
-	for ((_i) = eytzinger1_first((_size));		\
+	for (unsigned (_i) = eytzinger1_first((_size));	\
 	     (_i) != 0;					\
 	     (_i) = eytzinger1_next((_i), (_size)))
 
@@ -227,7 +227,7 @@ static inline unsigned inorder_to_eytzinger0(unsigned i, unsigned size)
 }
 
 #define eytzinger0_for_each(_i, _size)			\
-	for ((_i) = eytzinger0_first((_size));		\
+	for (unsigned (_i) = eytzinger0_first((_size));	\
 	     (_i) != -1;				\
 	     (_i) = eytzinger0_next((_i), (_size)))
 
diff --git a/fs/bcachefs/fs-io-direct.c b/fs/bcachefs/fs-io-direct.c
index fdd57c5785c9..e3b219e19e10 100644
--- a/fs/bcachefs/fs-io-direct.c
+++ b/fs/bcachefs/fs-io-direct.c
@@ -77,6 +77,10 @@ static int bch2_direct_IO_read(struct kiocb *req, struct iov_iter *iter)
 
 	bch2_inode_opts_get(&opts, c, &inode->ei_inode);
 
+	/* bios must be 512 byte aligned: */
+	if ((offset|iter->count) & (SECTOR_SIZE - 1))
+		return -EINVAL;
+
 	ret = min_t(loff_t, iter->count,
 		    max_t(loff_t, 0, i_size_read(&inode->v) - offset));
 
diff --git a/fs/bcachefs/fs-io-pagecache.c b/fs/bcachefs/fs-io-pagecache.c
index ff664fd0d8ef..d359aa9b33b8 100644
--- a/fs/bcachefs/fs-io-pagecache.c
+++ b/fs/bcachefs/fs-io-pagecache.c
@@ -309,39 +309,49 @@ void bch2_mark_pagecache_unallocated(struct bch_inode_info *inode,
 	}
 }
 
-void bch2_mark_pagecache_reserved(struct bch_inode_info *inode,
-				  u64 start, u64 end)
+int bch2_mark_pagecache_reserved(struct bch_inode_info *inode,
+				 u64 *start, u64 end,
+				 bool nonblocking)
 {
 	struct bch_fs *c = inode->v.i_sb->s_fs_info;
-	pgoff_t index = start >> PAGE_SECTORS_SHIFT;
+	pgoff_t index = *start >> PAGE_SECTORS_SHIFT;
 	pgoff_t end_index = (end - 1) >> PAGE_SECTORS_SHIFT;
 	struct folio_batch fbatch;
 	s64 i_sectors_delta = 0;
-	unsigned i, j;
+	int ret = 0;
 
-	if (end <= start)
-		return;
+	if (end <= *start)
+		return 0;
 
 	folio_batch_init(&fbatch);
 
 	while (filemap_get_folios(inode->v.i_mapping,
 				  &index, end_index, &fbatch)) {
-		for (i = 0; i < folio_batch_count(&fbatch); i++) {
+		for (unsigned i = 0; i < folio_batch_count(&fbatch); i++) {
 			struct folio *folio = fbatch.folios[i];
+
+			if (!nonblocking)
+				folio_lock(folio);
+			else if (!folio_trylock(folio)) {
+				folio_batch_release(&fbatch);
+				ret = -EAGAIN;
+				break;
+			}
+
 			u64 folio_start = folio_sector(folio);
 			u64 folio_end = folio_end_sector(folio);
-			unsigned folio_offset = max(start, folio_start) - folio_start;
-			unsigned folio_len = min(end, folio_end) - folio_offset - folio_start;
-			struct bch_folio *s;
 
 			BUG_ON(end <= folio_start);
 
-			folio_lock(folio);
-			s = bch2_folio(folio);
+			*start = min(end, folio_end);
 
+			struct bch_folio *s = bch2_folio(folio);
 			if (s) {
+				unsigned folio_offset = max(*start, folio_start) - folio_start;
+				unsigned folio_len = min(end, folio_end) - folio_offset - folio_start;
+
 				spin_lock(&s->lock);
-				for (j = folio_offset; j < folio_offset + folio_len; j++) {
+				for (unsigned j = folio_offset; j < folio_offset + folio_len; j++) {
 					i_sectors_delta -= s->s[j].state == SECTOR_dirty;
 					bch2_folio_sector_set(folio, s, j,
 						folio_sector_reserve(s->s[j].state));
@@ -356,6 +366,7 @@ void bch2_mark_pagecache_reserved(struct bch_inode_info *inode,
 	}
 
 	bch2_i_sectors_acct(c, inode, NULL, i_sectors_delta);
+	return ret;
 }
 
 static inline unsigned sectors_to_reserve(struct bch_folio_sector *s,
diff --git a/fs/bcachefs/fs-io-pagecache.h b/fs/bcachefs/fs-io-pagecache.h
index 27f712ae37a6..8cbaba6565b4 100644
--- a/fs/bcachefs/fs-io-pagecache.h
+++ b/fs/bcachefs/fs-io-pagecache.h
@@ -143,7 +143,7 @@ int bch2_folio_set(struct bch_fs *, subvol_inum, struct folio **, unsigned);
 void bch2_bio_page_state_set(struct bio *, struct bkey_s_c);
 
 void bch2_mark_pagecache_unallocated(struct bch_inode_info *, u64, u64);
-void bch2_mark_pagecache_reserved(struct bch_inode_info *, u64, u64);
+int bch2_mark_pagecache_reserved(struct bch_inode_info *, u64 *, u64, bool);
 
 int bch2_get_folio_disk_reservation(struct bch_fs *,
 				struct bch_inode_info *,
diff --git a/fs/bcachefs/fs-io.c b/fs/bcachefs/fs-io.c
index 98bd5babab19..dc52918d06ef 100644
--- a/fs/bcachefs/fs-io.c
+++ b/fs/bcachefs/fs-io.c
@@ -675,8 +675,11 @@ static int __bchfs_fallocate(struct bch_inode_info *inode, int mode,
 
 		bch2_i_sectors_acct(c, inode, &quota_res, i_sectors_delta);
 
-		drop_locks_do(trans,
-			(bch2_mark_pagecache_reserved(inode, hole_start, iter.pos.offset), 0));
+		if (bch2_mark_pagecache_reserved(inode, &hole_start,
+						 iter.pos.offset, true))
+			drop_locks_do(trans,
+				bch2_mark_pagecache_reserved(inode, &hole_start,
+							     iter.pos.offset, false));
 bkey_err:
 		bch2_quota_reservation_put(c, inode, &quota_res);
 		if (bch2_err_matches(ret, BCH_ERR_transaction_restart))
diff --git a/fs/bcachefs/fs-ioctl.c b/fs/bcachefs/fs-ioctl.c
index 1cbc5807bc80..3a4c24c28e7f 100644
--- a/fs/bcachefs/fs-ioctl.c
+++ b/fs/bcachefs/fs-ioctl.c
@@ -337,11 +337,12 @@ static long __bch2_ioctl_subvolume_create(struct bch_fs *c, struct file *filp,
 	if (arg.flags & BCH_SUBVOL_SNAPSHOT_RO)
 		create_flags |= BCH_CREATE_SNAPSHOT_RO;
 
-	/* why do we need this lock? */
-	down_read(&c->vfs_sb->s_umount);
-
-	if (arg.flags & BCH_SUBVOL_SNAPSHOT_CREATE)
+	if (arg.flags & BCH_SUBVOL_SNAPSHOT_CREATE) {
+		/* sync_inodes_sb enforce s_umount is locked */
+		down_read(&c->vfs_sb->s_umount);
 		sync_inodes_sb(c->vfs_sb);
+		up_read(&c->vfs_sb->s_umount);
+	}
 retry:
 	if (arg.src_ptr) {
 		error = user_path_at(arg.dirfd,
@@ -425,8 +426,6 @@ err2:
 		goto retry;
 	}
 err1:
-	up_read(&c->vfs_sb->s_umount);
-
 	return error;
 }
 
diff --git a/fs/bcachefs/inode.c b/fs/bcachefs/inode.c
index 37dce96f48ac..086f0090b03a 100644
--- a/fs/bcachefs/inode.c
+++ b/fs/bcachefs/inode.c
@@ -506,22 +506,33 @@ fsck_err:
 static void __bch2_inode_unpacked_to_text(struct printbuf *out,
 					  struct bch_inode_unpacked *inode)
 {
-	prt_printf(out, "mode=%o ", inode->bi_mode);
+	printbuf_indent_add(out, 2);
+	prt_printf(out, "mode=%o", inode->bi_mode);
+	prt_newline(out);
 
 	prt_str(out, "flags=");
 	prt_bitflags(out, bch2_inode_flag_strs, inode->bi_flags & ((1U << 20) - 1));
 	prt_printf(out, " (%x)", inode->bi_flags);
+	prt_newline(out);
 
-	prt_printf(out, " journal_seq=%llu bi_size=%llu bi_sectors=%llu bi_version=%llu",
-	       inode->bi_journal_seq,
-	       inode->bi_size,
-	       inode->bi_sectors,
-	       inode->bi_version);
+	prt_printf(out, "journal_seq=%llu", inode->bi_journal_seq);
+	prt_newline(out);
+
+	prt_printf(out, "bi_size=%llu", inode->bi_size);
+	prt_newline(out);
+
+	prt_printf(out, "bi_sectors=%llu", inode->bi_sectors);
+	prt_newline(out);
+
+	prt_newline(out);
+	prt_printf(out, "bi_version=%llu", inode->bi_version);
 
 #define x(_name, _bits)						\
-	prt_printf(out, " "#_name "=%llu", (u64) inode->_name);
+	prt_printf(out, #_name "=%llu", (u64) inode->_name);	\
+	prt_newline(out);
 	BCH_INODE_FIELDS_v3()
 #undef  x
+	printbuf_indent_sub(out, 2);
 }
 
 void bch2_inode_unpacked_to_text(struct printbuf *out, struct bch_inode_unpacked *inode)
@@ -587,7 +598,7 @@ int bch2_trigger_inode(struct btree_trans *trans,
 		}
 	}
 
-	if (!(flags & BTREE_TRIGGER_TRANSACTIONAL) && (flags & BTREE_TRIGGER_INSERT)) {
+	if ((flags & BTREE_TRIGGER_ATOMIC) && (flags & BTREE_TRIGGER_INSERT)) {
 		BUG_ON(!trans->journal_res.seq);
 
 		bkey_s_to_inode_v3(new).v->bi_journal_seq = cpu_to_le64(trans->journal_res.seq);
@@ -597,7 +608,7 @@ int bch2_trigger_inode(struct btree_trans *trans,
 		struct bch_fs *c = trans->c;
 
 		percpu_down_read(&c->mark_lock);
-		this_cpu_add(c->usage_gc->nr_inodes, nr);
+		this_cpu_add(c->usage_gc->b.nr_inodes, nr);
 		percpu_up_read(&c->mark_lock);
 	}
 
diff --git a/fs/bcachefs/inode_format.h b/fs/bcachefs/inode_format.h
new file mode 100644
index 000000000000..83d107331edf
--- /dev/null
+++ b/fs/bcachefs/inode_format.h
@@ -0,0 +1,166 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _BCACHEFS_INODE_FORMAT_H
+#define _BCACHEFS_INODE_FORMAT_H
+
+#define BLOCKDEV_INODE_MAX	4096
+#define BCACHEFS_ROOT_INO	4096
+
+struct bch_inode {
+	struct bch_val		v;
+
+	__le64			bi_hash_seed;
+	__le32			bi_flags;
+	__le16			bi_mode;
+	__u8			fields[];
+} __packed __aligned(8);
+
+struct bch_inode_v2 {
+	struct bch_val		v;
+
+	__le64			bi_journal_seq;
+	__le64			bi_hash_seed;
+	__le64			bi_flags;
+	__le16			bi_mode;
+	__u8			fields[];
+} __packed __aligned(8);
+
+struct bch_inode_v3 {
+	struct bch_val		v;
+
+	__le64			bi_journal_seq;
+	__le64			bi_hash_seed;
+	__le64			bi_flags;
+	__le64			bi_sectors;
+	__le64			bi_size;
+	__le64			bi_version;
+	__u8			fields[];
+} __packed __aligned(8);
+
+#define INODEv3_FIELDS_START_INITIAL	6
+#define INODEv3_FIELDS_START_CUR	(offsetof(struct bch_inode_v3, fields) / sizeof(__u64))
+
+struct bch_inode_generation {
+	struct bch_val		v;
+
+	__le32			bi_generation;
+	__le32			pad;
+} __packed __aligned(8);
+
+/*
+ * bi_subvol and bi_parent_subvol are only set for subvolume roots:
+ */
+
+#define BCH_INODE_FIELDS_v2()			\
+	x(bi_atime,			96)	\
+	x(bi_ctime,			96)	\
+	x(bi_mtime,			96)	\
+	x(bi_otime,			96)	\
+	x(bi_size,			64)	\
+	x(bi_sectors,			64)	\
+	x(bi_uid,			32)	\
+	x(bi_gid,			32)	\
+	x(bi_nlink,			32)	\
+	x(bi_generation,		32)	\
+	x(bi_dev,			32)	\
+	x(bi_data_checksum,		8)	\
+	x(bi_compression,		8)	\
+	x(bi_project,			32)	\
+	x(bi_background_compression,	8)	\
+	x(bi_data_replicas,		8)	\
+	x(bi_promote_target,		16)	\
+	x(bi_foreground_target,		16)	\
+	x(bi_background_target,		16)	\
+	x(bi_erasure_code,		16)	\
+	x(bi_fields_set,		16)	\
+	x(bi_dir,			64)	\
+	x(bi_dir_offset,		64)	\
+	x(bi_subvol,			32)	\
+	x(bi_parent_subvol,		32)
+
+#define BCH_INODE_FIELDS_v3()			\
+	x(bi_atime,			96)	\
+	x(bi_ctime,			96)	\
+	x(bi_mtime,			96)	\
+	x(bi_otime,			96)	\
+	x(bi_uid,			32)	\
+	x(bi_gid,			32)	\
+	x(bi_nlink,			32)	\
+	x(bi_generation,		32)	\
+	x(bi_dev,			32)	\
+	x(bi_data_checksum,		8)	\
+	x(bi_compression,		8)	\
+	x(bi_project,			32)	\
+	x(bi_background_compression,	8)	\
+	x(bi_data_replicas,		8)	\
+	x(bi_promote_target,		16)	\
+	x(bi_foreground_target,		16)	\
+	x(bi_background_target,		16)	\
+	x(bi_erasure_code,		16)	\
+	x(bi_fields_set,		16)	\
+	x(bi_dir,			64)	\
+	x(bi_dir_offset,		64)	\
+	x(bi_subvol,			32)	\
+	x(bi_parent_subvol,		32)	\
+	x(bi_nocow,			8)
+
+/* subset of BCH_INODE_FIELDS */
+#define BCH_INODE_OPTS()			\
+	x(data_checksum,		8)	\
+	x(compression,			8)	\
+	x(project,			32)	\
+	x(background_compression,	8)	\
+	x(data_replicas,		8)	\
+	x(promote_target,		16)	\
+	x(foreground_target,		16)	\
+	x(background_target,		16)	\
+	x(erasure_code,			16)	\
+	x(nocow,			8)
+
+enum inode_opt_id {
+#define x(name, ...)				\
+	Inode_opt_##name,
+	BCH_INODE_OPTS()
+#undef  x
+	Inode_opt_nr,
+};
+
+#define BCH_INODE_FLAGS()			\
+	x(sync,				0)	\
+	x(immutable,			1)	\
+	x(append,			2)	\
+	x(nodump,			3)	\
+	x(noatime,			4)	\
+	x(i_size_dirty,			5)	\
+	x(i_sectors_dirty,		6)	\
+	x(unlinked,			7)	\
+	x(backptr_untrusted,		8)
+
+/* bits 20+ reserved for packed fields below: */
+
+enum bch_inode_flags {
+#define x(t, n)	BCH_INODE_##t = 1U << n,
+	BCH_INODE_FLAGS()
+#undef x
+};
+
+enum __bch_inode_flags {
+#define x(t, n)	__BCH_INODE_##t = n,
+	BCH_INODE_FLAGS()
+#undef x
+};
+
+LE32_BITMASK(INODE_STR_HASH,	struct bch_inode, bi_flags, 20, 24);
+LE32_BITMASK(INODE_NR_FIELDS,	struct bch_inode, bi_flags, 24, 31);
+LE32_BITMASK(INODE_NEW_VARINT,	struct bch_inode, bi_flags, 31, 32);
+
+LE64_BITMASK(INODEv2_STR_HASH,	struct bch_inode_v2, bi_flags, 20, 24);
+LE64_BITMASK(INODEv2_NR_FIELDS,	struct bch_inode_v2, bi_flags, 24, 31);
+
+LE64_BITMASK(INODEv3_STR_HASH,	struct bch_inode_v3, bi_flags, 20, 24);
+LE64_BITMASK(INODEv3_NR_FIELDS,	struct bch_inode_v3, bi_flags, 24, 31);
+
+LE64_BITMASK(INODEv3_FIELDS_START,
+				struct bch_inode_v3, bi_flags, 31, 36);
+LE64_BITMASK(INODEv3_MODE,	struct bch_inode_v3, bi_flags, 36, 52);
+
+#endif /* _BCACHEFS_INODE_FORMAT_H */
diff --git a/fs/bcachefs/io_misc.c b/fs/bcachefs/io_misc.c
index ca6d5f516aa2..1baf78594cca 100644
--- a/fs/bcachefs/io_misc.c
+++ b/fs/bcachefs/io_misc.c
@@ -442,9 +442,7 @@ case LOGGED_OP_FINSERT_shift_extents:
 
 		op->v.pos = cpu_to_le64(insert ? bkey_start_offset(&delete.k) : delete.k.p.offset);
 
-		ret =   bch2_bkey_set_needs_rebalance(c, copy,
-					opts.background_target,
-					opts.background_compression) ?:
+		ret =   bch2_bkey_set_needs_rebalance(c, copy, &opts) ?:
 			bch2_btree_insert_trans(trans, BTREE_ID_extents, &delete, 0) ?:
 			bch2_btree_insert_trans(trans, BTREE_ID_extents, copy, 0) ?:
 			bch2_logged_op_update(trans, &op->k_i) ?:
diff --git a/fs/bcachefs/io_write.c b/fs/bcachefs/io_write.c
index 33c0e783d546..ef3a53f9045a 100644
--- a/fs/bcachefs/io_write.c
+++ b/fs/bcachefs/io_write.c
@@ -362,9 +362,7 @@ static int bch2_write_index_default(struct bch_write_op *op)
 				     bkey_start_pos(&sk.k->k),
 				     BTREE_ITER_SLOTS|BTREE_ITER_INTENT);
 
-		ret =   bch2_bkey_set_needs_rebalance(c, sk.k,
-					op->opts.background_target,
-					op->opts.background_compression) ?:
+		ret =   bch2_bkey_set_needs_rebalance(c, sk.k, &op->opts) ?:
 			bch2_extent_update(trans, inum, &iter, sk.k,
 					&op->res,
 					op->new_i_size, &op->i_sectors_delta,
@@ -1447,10 +1445,11 @@ err:
 			op->flags |= BCH_WRITE_DONE;
 
 			if (ret < 0) {
-				bch_err_inum_offset_ratelimited(c,
-					op->pos.inode,
-					op->pos.offset << 9,
-					"%s(): error: %s", __func__, bch2_err_str(ret));
+				if (!(op->flags & BCH_WRITE_ALLOC_NOWAIT))
+					bch_err_inum_offset_ratelimited(c,
+						op->pos.inode,
+						op->pos.offset << 9,
+						"%s(): error: %s", __func__, bch2_err_str(ret));
 				op->error = ret;
 				break;
 			}
diff --git a/fs/bcachefs/journal.c b/fs/bcachefs/journal.c
index 8538ef34f62b..d71d26e39521 100644
--- a/fs/bcachefs/journal.c
+++ b/fs/bcachefs/journal.c
@@ -27,6 +27,47 @@ static const char * const bch2_journal_errors[] = {
 	NULL
 };
 
+static void bch2_journal_buf_to_text(struct printbuf *out, struct journal *j, u64 seq)
+{
+	union journal_res_state s = READ_ONCE(j->reservations);
+	unsigned i = seq & JOURNAL_BUF_MASK;
+	struct journal_buf *buf = j->buf + i;
+
+	prt_printf(out, "seq:");
+	prt_tab(out);
+	prt_printf(out, "%llu", seq);
+	prt_newline(out);
+	printbuf_indent_add(out, 2);
+
+	prt_printf(out, "refcount:");
+	prt_tab(out);
+	prt_printf(out, "%u", journal_state_count(s, i));
+	prt_newline(out);
+
+	prt_printf(out, "size:");
+	prt_tab(out);
+	prt_human_readable_u64(out, vstruct_bytes(buf->data));
+	prt_newline(out);
+
+	prt_printf(out, "expires");
+	prt_tab(out);
+	prt_printf(out, "%li jiffies", buf->expires - jiffies);
+	prt_newline(out);
+
+	printbuf_indent_sub(out, 2);
+}
+
+static void bch2_journal_bufs_to_text(struct printbuf *out, struct journal *j)
+{
+	if (!out->nr_tabstops)
+		printbuf_tabstop_push(out, 24);
+
+	for (u64 seq = journal_last_unwritten_seq(j);
+	     seq <= journal_cur_seq(j);
+	     seq++)
+		bch2_journal_buf_to_text(out, j, seq);
+}
+
 static inline bool journal_seq_unwritten(struct journal *j, u64 seq)
 {
 	return seq > j->seq_ondisk;
@@ -156,7 +197,7 @@ void bch2_journal_buf_put_final(struct journal *j, u64 seq, bool write)
  * We don't close a journal_buf until the next journal_buf is finished writing,
  * and can be opened again - this also initializes the next journal_buf:
  */
-static void __journal_entry_close(struct journal *j, unsigned closed_val)
+static void __journal_entry_close(struct journal *j, unsigned closed_val, bool trace)
 {
 	struct bch_fs *c = container_of(j, struct bch_fs, journal);
 	struct journal_buf *buf = journal_cur_buf(j);
@@ -185,7 +226,17 @@ static void __journal_entry_close(struct journal *j, unsigned closed_val)
 	/* Close out old buffer: */
 	buf->data->u64s		= cpu_to_le32(old.cur_entry_offset);
 
-	trace_journal_entry_close(c, vstruct_bytes(buf->data));
+	if (trace_journal_entry_close_enabled() && trace) {
+		struct printbuf pbuf = PRINTBUF;
+		pbuf.atomic++;
+
+		prt_str(&pbuf, "entry size: ");
+		prt_human_readable_u64(&pbuf, vstruct_bytes(buf->data));
+		prt_newline(&pbuf);
+		bch2_prt_task_backtrace(&pbuf, current, 1);
+		trace_journal_entry_close(c, pbuf.buf);
+		printbuf_exit(&pbuf);
+	}
 
 	sectors = vstruct_blocks_plus(buf->data, c->block_bits,
 				      buf->u64s_reserved) << c->block_bits;
@@ -225,7 +276,7 @@ static void __journal_entry_close(struct journal *j, unsigned closed_val)
 void bch2_journal_halt(struct journal *j)
 {
 	spin_lock(&j->lock);
-	__journal_entry_close(j, JOURNAL_ENTRY_ERROR_VAL);
+	__journal_entry_close(j, JOURNAL_ENTRY_ERROR_VAL, true);
 	if (!j->err_seq)
 		j->err_seq = journal_cur_seq(j);
 	journal_wake(j);
@@ -239,7 +290,7 @@ static bool journal_entry_want_write(struct journal *j)
 
 	/* Don't close it yet if we already have a write in flight: */
 	if (ret)
-		__journal_entry_close(j, JOURNAL_ENTRY_CLOSED_VAL);
+		__journal_entry_close(j, JOURNAL_ENTRY_CLOSED_VAL, true);
 	else if (nr_unwritten_journal_entries(j)) {
 		struct journal_buf *buf = journal_cur_buf(j);
 
@@ -406,7 +457,7 @@ static void journal_write_work(struct work_struct *work)
 	if (delta > 0)
 		mod_delayed_work(c->io_complete_wq, &j->write_work, delta);
 	else
-		__journal_entry_close(j, JOURNAL_ENTRY_CLOSED_VAL);
+		__journal_entry_close(j, JOURNAL_ENTRY_CLOSED_VAL, true);
 unlock:
 	spin_unlock(&j->lock);
 }
@@ -463,13 +514,21 @@ retry:
 	    buf->buf_size < JOURNAL_ENTRY_SIZE_MAX)
 		j->buf_size_want = max(j->buf_size_want, buf->buf_size << 1);
 
-	__journal_entry_close(j, JOURNAL_ENTRY_CLOSED_VAL);
+	__journal_entry_close(j, JOURNAL_ENTRY_CLOSED_VAL, false);
 	ret = journal_entry_open(j);
 
 	if (ret == JOURNAL_ERR_max_in_flight) {
 		track_event_change(&c->times[BCH_TIME_blocked_journal_max_in_flight],
 				   &j->max_in_flight_start, true);
-		trace_and_count(c, journal_entry_full, c);
+		if (trace_journal_entry_full_enabled()) {
+			struct printbuf buf = PRINTBUF;
+			buf.atomic++;
+
+			bch2_journal_bufs_to_text(&buf, j);
+			trace_journal_entry_full(c, buf.buf);
+			printbuf_exit(&buf);
+		}
+		count_event(c, journal_entry_full);
 	}
 unlock:
 	can_discard = j->can_discard;
@@ -549,7 +608,7 @@ void bch2_journal_entry_res_resize(struct journal *j,
 		/*
 		 * Not enough room in current journal entry, have to flush it:
 		 */
-		__journal_entry_close(j, JOURNAL_ENTRY_CLOSED_VAL);
+		__journal_entry_close(j, JOURNAL_ENTRY_CLOSED_VAL, true);
 	} else {
 		journal_cur_buf(j)->u64s_reserved += d;
 	}
@@ -606,7 +665,7 @@ recheck_need_open:
 		struct journal_res res = { 0 };
 
 		if (journal_entry_is_open(j))
-			__journal_entry_close(j, JOURNAL_ENTRY_CLOSED_VAL);
+			__journal_entry_close(j, JOURNAL_ENTRY_CLOSED_VAL, true);
 
 		spin_unlock(&j->lock);
 
@@ -786,7 +845,7 @@ static struct journal_buf *__bch2_next_write_buffer_flush_journal_buf(struct jou
 
 		if (buf->need_flush_to_write_buffer) {
 			if (seq == journal_cur_seq(j))
-				__journal_entry_close(j, JOURNAL_ENTRY_CLOSED_VAL);
+				__journal_entry_close(j, JOURNAL_ENTRY_CLOSED_VAL, true);
 
 			union journal_res_state s;
 			s.v = atomic64_read_acquire(&j->reservations.counter);
@@ -1339,35 +1398,9 @@ void __bch2_journal_debug_to_text(struct printbuf *out, struct journal *j)
 	}
 
 	prt_newline(out);
-
-	for (u64 seq = journal_cur_seq(j);
-	     seq >= journal_last_unwritten_seq(j);
-	     --seq) {
-		unsigned i = seq & JOURNAL_BUF_MASK;
-
-		prt_printf(out, "unwritten entry:");
-		prt_tab(out);
-		prt_printf(out, "%llu", seq);
-		prt_newline(out);
-		printbuf_indent_add(out, 2);
-
-		prt_printf(out, "refcount:");
-		prt_tab(out);
-		prt_printf(out, "%u", journal_state_count(s, i));
-		prt_newline(out);
-
-		prt_printf(out, "sectors:");
-		prt_tab(out);
-		prt_printf(out, "%u", j->buf[i].sectors);
-		prt_newline(out);
-
-		prt_printf(out, "expires");
-		prt_tab(out);
-		prt_printf(out, "%li jiffies", j->buf[i].expires - jiffies);
-		prt_newline(out);
-
-		printbuf_indent_sub(out, 2);
-	}
+	prt_printf(out, "unwritten entries:");
+	prt_newline(out);
+	bch2_journal_bufs_to_text(out, j);
 
 	prt_printf(out,
 	       "replay done:\t\t%i\n",
diff --git a/fs/bcachefs/journal_io.c b/fs/bcachefs/journal_io.c
index b0f4dd491e12..04a1e79a5ed3 100644
--- a/fs/bcachefs/journal_io.c
+++ b/fs/bcachefs/journal_io.c
@@ -683,10 +683,7 @@ static void journal_entry_dev_usage_to_text(struct printbuf *out, struct bch_fs
 	prt_printf(out, "dev=%u", le32_to_cpu(u->dev));
 
 	for (i = 0; i < nr_types; i++) {
-		if (i < BCH_DATA_NR)
-			prt_printf(out, " %s", bch2_data_types[i]);
-		else
-			prt_printf(out, " (unknown data type %u)", i);
+		bch2_prt_data_type(out, i);
 		prt_printf(out, ": buckets=%llu sectors=%llu fragmented=%llu",
 		       le64_to_cpu(u->d[i].buckets),
 		       le64_to_cpu(u->d[i].sectors),
diff --git a/fs/bcachefs/logged_ops_format.h b/fs/bcachefs/logged_ops_format.h
new file mode 100644
index 000000000000..6a4bf7129dba
--- /dev/null
+++ b/fs/bcachefs/logged_ops_format.h
@@ -0,0 +1,30 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _BCACHEFS_LOGGED_OPS_FORMAT_H
+#define _BCACHEFS_LOGGED_OPS_FORMAT_H
+
+struct bch_logged_op_truncate {
+	struct bch_val		v;
+	__le32			subvol;
+	__le32			pad;
+	__le64			inum;
+	__le64			new_i_size;
+};
+
+enum logged_op_finsert_state {
+	LOGGED_OP_FINSERT_start,
+	LOGGED_OP_FINSERT_shift_extents,
+	LOGGED_OP_FINSERT_finish,
+};
+
+struct bch_logged_op_finsert {
+	struct bch_val		v;
+	__u8			state;
+	__u8			pad[3];
+	__le32			subvol;
+	__le64			inum;
+	__le64			dst_offset;
+	__le64			src_offset;
+	__le64			pos;
+};
+
+#endif /* _BCACHEFS_LOGGED_OPS_FORMAT_H */
diff --git a/fs/bcachefs/move.c b/fs/bcachefs/move.c
index 7a33319dcd16..bf68ea49447b 100644
--- a/fs/bcachefs/move.c
+++ b/fs/bcachefs/move.c
@@ -6,9 +6,11 @@
 #include "backpointers.h"
 #include "bkey_buf.h"
 #include "btree_gc.h"
+#include "btree_io.h"
 #include "btree_update.h"
 #include "btree_update_interior.h"
 #include "btree_write_buffer.h"
+#include "compress.h"
 #include "disk_groups.h"
 #include "ec.h"
 #include "errcode.h"
@@ -34,12 +36,46 @@ const char * const bch2_data_ops_strs[] = {
 	NULL
 };
 
-static void trace_move_extent2(struct bch_fs *c, struct bkey_s_c k)
+static void bch2_data_update_opts_to_text(struct printbuf *out, struct bch_fs *c,
+					  struct bch_io_opts *io_opts,
+					  struct data_update_opts *data_opts)
+{
+	printbuf_tabstop_push(out, 20);
+	prt_str(out, "rewrite ptrs:");
+	prt_tab(out);
+	bch2_prt_u64_base2(out, data_opts->rewrite_ptrs);
+	prt_newline(out);
+
+	prt_str(out, "kill ptrs: ");
+	prt_tab(out);
+	bch2_prt_u64_base2(out, data_opts->kill_ptrs);
+	prt_newline(out);
+
+	prt_str(out, "target: ");
+	prt_tab(out);
+	bch2_target_to_text(out, c, data_opts->target);
+	prt_newline(out);
+
+	prt_str(out, "compression: ");
+	prt_tab(out);
+	bch2_compression_opt_to_text(out, background_compression(*io_opts));
+	prt_newline(out);
+
+	prt_str(out, "extra replicas: ");
+	prt_tab(out);
+	prt_u64(out, data_opts->extra_replicas);
+}
+
+static void trace_move_extent2(struct bch_fs *c, struct bkey_s_c k,
+			       struct bch_io_opts *io_opts,
+			       struct data_update_opts *data_opts)
 {
 	if (trace_move_extent_enabled()) {
 		struct printbuf buf = PRINTBUF;
 
 		bch2_bkey_val_to_text(&buf, c, k);
+		prt_newline(&buf);
+		bch2_data_update_opts_to_text(&buf, c, io_opts, data_opts);
 		trace_move_extent(c, buf.buf);
 		printbuf_exit(&buf);
 	}
@@ -111,6 +147,15 @@ static void move_write(struct moving_io *io)
 		return;
 	}
 
+	if (trace_move_extent_write_enabled()) {
+		struct bch_fs *c = io->write.op.c;
+		struct printbuf buf = PRINTBUF;
+
+		bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(io->write.k.k));
+		trace_move_extent_write(c, buf.buf);
+		printbuf_exit(&buf);
+	}
+
 	closure_get(&io->write.ctxt->cl);
 	atomic_add(io->write_sectors, &io->write.ctxt->write_sectors);
 	atomic_inc(&io->write.ctxt->write_ios);
@@ -241,9 +286,10 @@ int bch2_move_extent(struct moving_context *ctxt,
 	unsigned sectors = k.k->size, pages;
 	int ret = -ENOMEM;
 
+	trace_move_extent2(c, k, &io_opts, &data_opts);
+
 	if (ctxt->stats)
 		ctxt->stats->pos = BBPOS(iter->btree_id, iter->pos);
-	trace_move_extent2(c, k);
 
 	bch2_data_update_opts_normalize(k, &data_opts);
 
@@ -759,6 +805,8 @@ int bch2_evacuate_bucket(struct moving_context *ctxt,
 			if (!b)
 				goto next;
 
+			unsigned sectors = btree_ptr_sectors_written(&b->key);
+
 			ret = bch2_btree_node_rewrite(trans, &iter, b, 0);
 			bch2_trans_iter_exit(trans, &iter);
 
@@ -768,11 +816,10 @@ int bch2_evacuate_bucket(struct moving_context *ctxt,
 				goto err;
 
 			if (ctxt->rate)
-				bch2_ratelimit_increment(ctxt->rate,
-							 c->opts.btree_node_size >> 9);
+				bch2_ratelimit_increment(ctxt->rate, sectors);
 			if (ctxt->stats) {
-				atomic64_add(c->opts.btree_node_size >> 9, &ctxt->stats->sectors_seen);
-				atomic64_add(c->opts.btree_node_size >> 9, &ctxt->stats->sectors_moved);
+				atomic64_add(sectors, &ctxt->stats->sectors_seen);
+				atomic64_add(sectors, &ctxt->stats->sectors_moved);
 			}
 		}
 next:
@@ -1083,9 +1130,9 @@ int bch2_data_job(struct bch_fs *c,
 
 void bch2_move_stats_to_text(struct printbuf *out, struct bch_move_stats *stats)
 {
-	prt_printf(out, "%s: data type=%s pos=",
-		   stats->name,
-		   bch2_data_types[stats->data_type]);
+	prt_printf(out, "%s: data type==", stats->name);
+	bch2_prt_data_type(out, stats->data_type);
+	prt_str(out, " pos=");
 	bch2_bbpos_to_text(out, stats->pos);
 	prt_newline(out);
 	printbuf_indent_add(out, 2);
diff --git a/fs/bcachefs/opts.c b/fs/bcachefs/opts.c
index 8e6f230eac38..b1ed0b9a20d3 100644
--- a/fs/bcachefs/opts.c
+++ b/fs/bcachefs/opts.c
@@ -52,7 +52,7 @@ const char * const bch2_csum_opts[] = {
 	NULL
 };
 
-const char * const bch2_compression_types[] = {
+const char * const __bch2_compression_types[] = {
 	BCH_COMPRESSION_TYPES()
 	NULL
 };
@@ -72,7 +72,7 @@ const char * const bch2_str_hash_opts[] = {
 	NULL
 };
 
-const char * const bch2_data_types[] = {
+const char * const __bch2_data_types[] = {
 	BCH_DATA_TYPES()
 	NULL
 };
diff --git a/fs/bcachefs/opts.h b/fs/bcachefs/opts.h
index 93a24fef4214..9a4b7faa3765 100644
--- a/fs/bcachefs/opts.h
+++ b/fs/bcachefs/opts.h
@@ -18,11 +18,11 @@ extern const char * const bch2_sb_compat[];
 extern const char * const __bch2_btree_ids[];
 extern const char * const bch2_csum_types[];
 extern const char * const bch2_csum_opts[];
-extern const char * const bch2_compression_types[];
+extern const char * const __bch2_compression_types[];
 extern const char * const bch2_compression_opts[];
 extern const char * const bch2_str_hash_types[];
 extern const char * const bch2_str_hash_opts[];
-extern const char * const bch2_data_types[];
+extern const char * const __bch2_data_types[];
 extern const char * const bch2_member_states[];
 extern const char * const bch2_jset_entry_types[];
 extern const char * const bch2_fs_usage_types[];
@@ -564,6 +564,11 @@ struct bch_io_opts {
 #undef x
 };
 
+static inline unsigned background_compression(struct bch_io_opts opts)
+{
+	return opts.background_compression ?: opts.compression;
+}
+
 struct bch_io_opts bch2_opts_to_inode_opts(struct bch_opts);
 bool bch2_opt_is_inode_opt(enum bch_opt_id);
 
diff --git a/fs/bcachefs/quota_format.h b/fs/bcachefs/quota_format.h
new file mode 100644
index 000000000000..dc34347ef6c7
--- /dev/null
+++ b/fs/bcachefs/quota_format.h
@@ -0,0 +1,47 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _BCACHEFS_QUOTA_FORMAT_H
+#define _BCACHEFS_QUOTA_FORMAT_H
+
+/* KEY_TYPE_quota: */
+
+enum quota_types {
+	QTYP_USR		= 0,
+	QTYP_GRP		= 1,
+	QTYP_PRJ		= 2,
+	QTYP_NR			= 3,
+};
+
+enum quota_counters {
+	Q_SPC			= 0,
+	Q_INO			= 1,
+	Q_COUNTERS		= 2,
+};
+
+struct bch_quota_counter {
+	__le64			hardlimit;
+	__le64			softlimit;
+};
+
+struct bch_quota {
+	struct bch_val		v;
+	struct bch_quota_counter c[Q_COUNTERS];
+} __packed __aligned(8);
+
+/* BCH_SB_FIELD_quota: */
+
+struct bch_sb_quota_counter {
+	__le32				timelimit;
+	__le32				warnlimit;
+};
+
+struct bch_sb_quota_type {
+	__le64				flags;
+	struct bch_sb_quota_counter	c[Q_COUNTERS];
+};
+
+struct bch_sb_field_quota {
+	struct bch_sb_field		field;
+	struct bch_sb_quota_type	q[QTYP_NR];
+} __packed __aligned(8);
+
+#endif /* _BCACHEFS_QUOTA_FORMAT_H */
diff --git a/fs/bcachefs/rebalance.c b/fs/bcachefs/rebalance.c
index 95f46cb3b5bd..22d1017aa49b 100644
--- a/fs/bcachefs/rebalance.c
+++ b/fs/bcachefs/rebalance.c
@@ -177,8 +177,7 @@ static struct bkey_s_c next_rebalance_extent(struct btree_trans *trans,
 		prt_str(&buf, "target=");
 		bch2_target_to_text(&buf, c, r->target);
 		prt_str(&buf, " compression=");
-		struct bch_compression_opt opt = __bch2_compression_decode(r->compression);
-		prt_str(&buf, bch2_compression_opts[opt.type]);
+		bch2_compression_opt_to_text(&buf, r->compression);
 		prt_str(&buf, " ");
 		bch2_bkey_val_to_text(&buf, c, k);
 
@@ -254,13 +253,12 @@ static bool rebalance_pred(struct bch_fs *c, void *arg,
 
 	if (k.k->p.inode) {
 		target		= io_opts->background_target;
-		compression	= io_opts->background_compression ?: io_opts->compression;
+		compression	= background_compression(*io_opts);
 	} else {
 		const struct bch_extent_rebalance *r = bch2_bkey_rebalance_opts(k);
 
 		target		= r ? r->target : io_opts->background_target;
-		compression	= r ? r->compression :
-			(io_opts->background_compression ?: io_opts->compression);
+		compression	= r ? r->compression : background_compression(*io_opts);
 	}
 
 	data_opts->rewrite_ptrs		= bch2_bkey_ptrs_need_rebalance(c, k, target, compression);
@@ -371,6 +369,7 @@ static int do_rebalance(struct moving_context *ctxt)
 	    !kthread_should_stop() &&
 	    !atomic64_read(&r->work_stats.sectors_seen) &&
 	    !atomic64_read(&r->scan_stats.sectors_seen)) {
+		bch2_moving_ctxt_flush_all(ctxt);
 		bch2_trans_unlock_long(trans);
 		rebalance_wait(c);
 	}
@@ -385,7 +384,6 @@ static int bch2_rebalance_thread(void *arg)
 	struct bch_fs *c = arg;
 	struct bch_fs_rebalance *r = &c->rebalance;
 	struct moving_context ctxt;
-	int ret;
 
 	set_freezable();
 
@@ -393,8 +391,7 @@ static int bch2_rebalance_thread(void *arg)
 			      writepoint_ptr(&c->rebalance_write_point),
 			      true);
 
-	while (!kthread_should_stop() &&
-	       !(ret = do_rebalance(&ctxt)))
+	while (!kthread_should_stop() && !do_rebalance(&ctxt))
 		;
 
 	bch2_moving_ctxt_exit(&ctxt);
diff --git a/fs/bcachefs/recovery.c b/fs/bcachefs/recovery.c
index 725214605a05..9127d0e3ca2f 100644
--- a/fs/bcachefs/recovery.c
+++ b/fs/bcachefs/recovery.c
@@ -280,7 +280,7 @@ static int journal_replay_entry_early(struct bch_fs *c,
 					le64_to_cpu(u->v);
 			break;
 		case BCH_FS_USAGE_inodes:
-			c->usage_base->nr_inodes = le64_to_cpu(u->v);
+			c->usage_base->b.nr_inodes = le64_to_cpu(u->v);
 			break;
 		case BCH_FS_USAGE_key_version:
 			atomic64_set(&c->key_version,
diff --git a/fs/bcachefs/reflink.c b/fs/bcachefs/reflink.c
index faa5d3670058..c47c66c2b394 100644
--- a/fs/bcachefs/reflink.c
+++ b/fs/bcachefs/reflink.c
@@ -292,10 +292,10 @@ static inline void check_indirect_extent_deleting(struct bkey_s new, unsigned *f
 	}
 }
 
-int bch2_trans_mark_reflink_v(struct btree_trans *trans,
-			      enum btree_id btree_id, unsigned level,
-			      struct bkey_s_c old, struct bkey_s new,
-			      unsigned flags)
+int bch2_trigger_reflink_v(struct btree_trans *trans,
+			   enum btree_id btree_id, unsigned level,
+			   struct bkey_s_c old, struct bkey_s new,
+			   unsigned flags)
 {
 	if ((flags & BTREE_TRIGGER_TRANSACTIONAL) &&
 	    (flags & BTREE_TRIGGER_INSERT))
@@ -324,7 +324,7 @@ void bch2_indirect_inline_data_to_text(struct printbuf *out,
 	       min(datalen, 32U), d.v->data);
 }
 
-int bch2_trans_mark_indirect_inline_data(struct btree_trans *trans,
+int bch2_trigger_indirect_inline_data(struct btree_trans *trans,
 			      enum btree_id btree_id, unsigned level,
 			      struct bkey_s_c old, struct bkey_s new,
 			      unsigned flags)
@@ -486,6 +486,13 @@ s64 bch2_remap_range(struct bch_fs *c,
 
 		bch2_btree_iter_set_snapshot(&dst_iter, dst_snapshot);
 
+		if (dst_inum.inum < src_inum.inum) {
+			/* Avoid some lock cycle transaction restarts */
+			ret = bch2_btree_iter_traverse(&dst_iter);
+			if (ret)
+				continue;
+		}
+
 		dst_done = dst_iter.pos.offset - dst_start.offset;
 		src_want = POS(src_start.inode, src_start.offset + dst_done);
 		bch2_btree_iter_set_pos(&src_iter, src_want);
@@ -538,9 +545,7 @@ s64 bch2_remap_range(struct bch_fs *c,
 				min(src_k.k->p.offset - src_want.offset,
 				    dst_end.offset - dst_iter.pos.offset));
 
-		ret =   bch2_bkey_set_needs_rebalance(c, new_dst.k,
-					opts.background_target,
-					opts.background_compression) ?:
+		ret =   bch2_bkey_set_needs_rebalance(c, new_dst.k, &opts) ?:
 			bch2_extent_update(trans, dst_inum, &dst_iter,
 					new_dst.k, &disk_res,
 					new_i_size, i_sectors_delta,
diff --git a/fs/bcachefs/reflink.h b/fs/bcachefs/reflink.h
index 8ee778ec0022..4d8867289717 100644
--- a/fs/bcachefs/reflink.h
+++ b/fs/bcachefs/reflink.h
@@ -24,14 +24,14 @@ int bch2_reflink_v_invalid(struct bch_fs *, struct bkey_s_c,
 			   enum bkey_invalid_flags, struct printbuf *);
 void bch2_reflink_v_to_text(struct printbuf *, struct bch_fs *,
 			    struct bkey_s_c);
-int bch2_trans_mark_reflink_v(struct btree_trans *, enum btree_id, unsigned,
+int bch2_trigger_reflink_v(struct btree_trans *, enum btree_id, unsigned,
 			      struct bkey_s_c, struct bkey_s, unsigned);
 
 #define bch2_bkey_ops_reflink_v ((struct bkey_ops) {		\
 	.key_invalid	= bch2_reflink_v_invalid,		\
 	.val_to_text	= bch2_reflink_v_to_text,		\
 	.swab		= bch2_ptr_swab,			\
-	.trigger	= bch2_trans_mark_reflink_v,		\
+	.trigger	= bch2_trigger_reflink_v,		\
 	.min_val_size	= 8,					\
 })
 
@@ -39,7 +39,7 @@ int bch2_indirect_inline_data_invalid(struct bch_fs *, struct bkey_s_c,
 				      enum bkey_invalid_flags, struct printbuf *);
 void bch2_indirect_inline_data_to_text(struct printbuf *,
 				struct bch_fs *, struct bkey_s_c);
-int bch2_trans_mark_indirect_inline_data(struct btree_trans *,
+int bch2_trigger_indirect_inline_data(struct btree_trans *,
 					 enum btree_id, unsigned,
 			      struct bkey_s_c, struct bkey_s,
 			      unsigned);
@@ -47,7 +47,7 @@ int bch2_trans_mark_indirect_inline_data(struct btree_trans *,
 #define bch2_bkey_ops_indirect_inline_data ((struct bkey_ops) {	\
 	.key_invalid	= bch2_indirect_inline_data_invalid,	\
 	.val_to_text	= bch2_indirect_inline_data_to_text,	\
-	.trigger	= bch2_trans_mark_indirect_inline_data,	\
+	.trigger	= bch2_trigger_indirect_inline_data,	\
 	.min_val_size	= 8,					\
 })
 
diff --git a/fs/bcachefs/reflink_format.h b/fs/bcachefs/reflink_format.h
new file mode 100644
index 000000000000..6772eebb1fc6
--- /dev/null
+++ b/fs/bcachefs/reflink_format.h
@@ -0,0 +1,33 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _BCACHEFS_REFLINK_FORMAT_H
+#define _BCACHEFS_REFLINK_FORMAT_H
+
+struct bch_reflink_p {
+	struct bch_val		v;
+	__le64			idx;
+	/*
+	 * A reflink pointer might point to an indirect extent which is then
+	 * later split (by copygc or rebalance). If we only pointed to part of
+	 * the original indirect extent, and then one of the fragments is
+	 * outside the range we point to, we'd leak a refcount: so when creating
+	 * reflink pointers, we need to store pad values to remember the full
+	 * range we were taking a reference on.
+	 */
+	__le32			front_pad;
+	__le32			back_pad;
+} __packed __aligned(8);
+
+struct bch_reflink_v {
+	struct bch_val		v;
+	__le64			refcount;
+	union bch_extent_entry	start[0];
+	__u64			_data[];
+} __packed __aligned(8);
+
+struct bch_indirect_inline_data {
+	struct bch_val		v;
+	__le64			refcount;
+	u8			data[];
+};
+
+#endif /* _BCACHEFS_REFLINK_FORMAT_H */
diff --git a/fs/bcachefs/replicas.c b/fs/bcachefs/replicas.c
index 92ba56ef1fc8..cc2672c12031 100644
--- a/fs/bcachefs/replicas.c
+++ b/fs/bcachefs/replicas.c
@@ -9,6 +9,12 @@
 static int bch2_cpu_replicas_to_sb_replicas(struct bch_fs *,
 					    struct bch_replicas_cpu *);
 
+/* Some (buggy!) compilers don't allow memcmp to be passed as a pointer */
+static int bch2_memcmp(const void *l, const void *r, size_t size)
+{
+	return memcmp(l, r, size);
+}
+
 /* Replicas tracking - in memory: */
 
 static void verify_replicas_entry(struct bch_replicas_entry_v1 *e)
@@ -33,21 +39,16 @@ void bch2_replicas_entry_sort(struct bch_replicas_entry_v1 *e)
 
 static void bch2_cpu_replicas_sort(struct bch_replicas_cpu *r)
 {
-	eytzinger0_sort(r->entries, r->nr, r->entry_size, memcmp, NULL);
+	eytzinger0_sort(r->entries, r->nr, r->entry_size, bch2_memcmp, NULL);
 }
 
 static void bch2_replicas_entry_v0_to_text(struct printbuf *out,
 					   struct bch_replicas_entry_v0 *e)
 {
-	unsigned i;
-
-	if (e->data_type < BCH_DATA_NR)
-		prt_printf(out, "%s", bch2_data_types[e->data_type]);
-	else
-		prt_printf(out, "(invalid data type %u)", e->data_type);
+	bch2_prt_data_type(out, e->data_type);
 
 	prt_printf(out, ": %u [", e->nr_devs);
-	for (i = 0; i < e->nr_devs; i++)
+	for (unsigned i = 0; i < e->nr_devs; i++)
 		prt_printf(out, i ? " %u" : "%u", e->devs[i]);
 	prt_printf(out, "]");
 }
@@ -55,15 +56,10 @@ static void bch2_replicas_entry_v0_to_text(struct printbuf *out,
 void bch2_replicas_entry_to_text(struct printbuf *out,
 				 struct bch_replicas_entry_v1 *e)
 {
-	unsigned i;
-
-	if (e->data_type < BCH_DATA_NR)
-		prt_printf(out, "%s", bch2_data_types[e->data_type]);
-	else
-		prt_printf(out, "(invalid data type %u)", e->data_type);
+	bch2_prt_data_type(out, e->data_type);
 
 	prt_printf(out, ": %u/%u [", e->nr_required, e->nr_devs);
-	for (i = 0; i < e->nr_devs; i++)
+	for (unsigned i = 0; i < e->nr_devs; i++)
 		prt_printf(out, i ? " %u" : "%u", e->devs[i]);
 	prt_printf(out, "]");
 }
@@ -831,7 +827,7 @@ static int bch2_cpu_replicas_validate(struct bch_replicas_cpu *cpu_r,
 	sort_cmp_size(cpu_r->entries,
 		      cpu_r->nr,
 		      cpu_r->entry_size,
-		      memcmp, NULL);
+		      bch2_memcmp, NULL);
 
 	for (i = 0; i < cpu_r->nr; i++) {
 		struct bch_replicas_entry_v1 *e =
diff --git a/fs/bcachefs/sb-clean.c b/fs/bcachefs/sb-clean.c
index 9632f36f5f31..b6bf0ebe7e84 100644
--- a/fs/bcachefs/sb-clean.c
+++ b/fs/bcachefs/sb-clean.c
@@ -207,7 +207,7 @@ void bch2_journal_super_entries_add_common(struct bch_fs *c,
 
 		u->entry.type	= BCH_JSET_ENTRY_usage;
 		u->entry.btree_id = BCH_FS_USAGE_inodes;
-		u->v		= cpu_to_le64(c->usage_base->nr_inodes);
+		u->v		= cpu_to_le64(c->usage_base->b.nr_inodes);
 	}
 
 	{
diff --git a/fs/bcachefs/counters.c b/fs/bcachefs/sb-counters.c
index 02a996e06a64..7dc898761bb3 100644
--- a/fs/bcachefs/counters.c
+++ b/fs/bcachefs/sb-counters.c
@@ -1,7 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0
 #include "bcachefs.h"
 #include "super-io.h"
-#include "counters.h"
+#include "sb-counters.h"
 
 /* BCH_SB_FIELD_counters */
 
diff --git a/fs/bcachefs/counters.h b/fs/bcachefs/sb-counters.h
index 4778aa19bf34..81f8aec9fcb1 100644
--- a/fs/bcachefs/counters.h
+++ b/fs/bcachefs/sb-counters.h
@@ -1,11 +1,10 @@
 /* SPDX-License-Identifier: GPL-2.0 */
-#ifndef _BCACHEFS_COUNTERS_H
-#define _BCACHEFS_COUNTERS_H
+#ifndef _BCACHEFS_SB_COUNTERS_H
+#define _BCACHEFS_SB_COUNTERS_H
 
 #include "bcachefs.h"
 #include "super-io.h"
 
-
 int bch2_sb_counters_to_cpu(struct bch_fs *);
 int bch2_sb_counters_from_cpu(struct bch_fs *);
 
@@ -14,4 +13,4 @@ int bch2_fs_counters_init(struct bch_fs *);
 
 extern const struct bch_sb_field_ops bch_sb_field_ops_counters;
 
-#endif // _BCACHEFS_COUNTERS_H
+#endif // _BCACHEFS_SB_COUNTERS_H
diff --git a/fs/bcachefs/sb-counters_format.h b/fs/bcachefs/sb-counters_format.h
new file mode 100644
index 000000000000..62ea478215d0
--- /dev/null
+++ b/fs/bcachefs/sb-counters_format.h
@@ -0,0 +1,98 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _BCACHEFS_SB_COUNTERS_FORMAT_H
+#define _BCACHEFS_SB_COUNTERS_FORMAT_H
+
+#define BCH_PERSISTENT_COUNTERS()				\
+	x(io_read,					0)	\
+	x(io_write,					1)	\
+	x(io_move,					2)	\
+	x(bucket_invalidate,				3)	\
+	x(bucket_discard,				4)	\
+	x(bucket_alloc,					5)	\
+	x(bucket_alloc_fail,				6)	\
+	x(btree_cache_scan,				7)	\
+	x(btree_cache_reap,				8)	\
+	x(btree_cache_cannibalize,			9)	\
+	x(btree_cache_cannibalize_lock,			10)	\
+	x(btree_cache_cannibalize_lock_fail,		11)	\
+	x(btree_cache_cannibalize_unlock,		12)	\
+	x(btree_node_write,				13)	\
+	x(btree_node_read,				14)	\
+	x(btree_node_compact,				15)	\
+	x(btree_node_merge,				16)	\
+	x(btree_node_split,				17)	\
+	x(btree_node_rewrite,				18)	\
+	x(btree_node_alloc,				19)	\
+	x(btree_node_free,				20)	\
+	x(btree_node_set_root,				21)	\
+	x(btree_path_relock_fail,			22)	\
+	x(btree_path_upgrade_fail,			23)	\
+	x(btree_reserve_get_fail,			24)	\
+	x(journal_entry_full,				25)	\
+	x(journal_full,					26)	\
+	x(journal_reclaim_finish,			27)	\
+	x(journal_reclaim_start,			28)	\
+	x(journal_write,				29)	\
+	x(read_promote,					30)	\
+	x(read_bounce,					31)	\
+	x(read_split,					33)	\
+	x(read_retry,					32)	\
+	x(read_reuse_race,				34)	\
+	x(move_extent_read,				35)	\
+	x(move_extent_write,				36)	\
+	x(move_extent_finish,				37)	\
+	x(move_extent_fail,				38)	\
+	x(move_extent_start_fail,			39)	\
+	x(copygc,					40)	\
+	x(copygc_wait,					41)	\
+	x(gc_gens_end,					42)	\
+	x(gc_gens_start,				43)	\
+	x(trans_blocked_journal_reclaim,		44)	\
+	x(trans_restart_btree_node_reused,		45)	\
+	x(trans_restart_btree_node_split,		46)	\
+	x(trans_restart_fault_inject,			47)	\
+	x(trans_restart_iter_upgrade,			48)	\
+	x(trans_restart_journal_preres_get,		49)	\
+	x(trans_restart_journal_reclaim,		50)	\
+	x(trans_restart_journal_res_get,		51)	\
+	x(trans_restart_key_cache_key_realloced,	52)	\
+	x(trans_restart_key_cache_raced,		53)	\
+	x(trans_restart_mark_replicas,			54)	\
+	x(trans_restart_mem_realloced,			55)	\
+	x(trans_restart_memory_allocation_failure,	56)	\
+	x(trans_restart_relock,				57)	\
+	x(trans_restart_relock_after_fill,		58)	\
+	x(trans_restart_relock_key_cache_fill,		59)	\
+	x(trans_restart_relock_next_node,		60)	\
+	x(trans_restart_relock_parent_for_fill,		61)	\
+	x(trans_restart_relock_path,			62)	\
+	x(trans_restart_relock_path_intent,		63)	\
+	x(trans_restart_too_many_iters,			64)	\
+	x(trans_restart_traverse,			65)	\
+	x(trans_restart_upgrade,			66)	\
+	x(trans_restart_would_deadlock,			67)	\
+	x(trans_restart_would_deadlock_write,		68)	\
+	x(trans_restart_injected,			69)	\
+	x(trans_restart_key_cache_upgrade,		70)	\
+	x(trans_traverse_all,				71)	\
+	x(transaction_commit,				72)	\
+	x(write_super,					73)	\
+	x(trans_restart_would_deadlock_recursion_limit,	74)	\
+	x(trans_restart_write_buffer_flush,		75)	\
+	x(trans_restart_split_race,			76)	\
+	x(write_buffer_flush_slowpath,			77)	\
+	x(write_buffer_flush_sync,			78)
+
+enum bch_persistent_counters {
+#define x(t, n, ...) BCH_COUNTER_##t,
+	BCH_PERSISTENT_COUNTERS()
+#undef x
+	BCH_COUNTER_NR
+};
+
+struct bch_sb_field_counters {
+	struct bch_sb_field	field;
+	__le64			d[];
+};
+
+#endif /* _BCACHEFS_SB_COUNTERS_FORMAT_H */
diff --git a/fs/bcachefs/sb-members.c b/fs/bcachefs/sb-members.c
index a44a238bf8b5..a45354d2acde 100644
--- a/fs/bcachefs/sb-members.c
+++ b/fs/bcachefs/sb-members.c
@@ -251,7 +251,7 @@ static void member_to_text(struct printbuf *out,
 	prt_printf(out, "Data allowed:");
 	prt_tab(out);
 	if (BCH_MEMBER_DATA_ALLOWED(&m))
-		prt_bitflags(out, bch2_data_types, BCH_MEMBER_DATA_ALLOWED(&m));
+		prt_bitflags(out, __bch2_data_types, BCH_MEMBER_DATA_ALLOWED(&m));
 	else
 		prt_printf(out, "(none)");
 	prt_newline(out);
@@ -259,7 +259,7 @@ static void member_to_text(struct printbuf *out,
 	prt_printf(out, "Has data:");
 	prt_tab(out);
 	if (data_have)
-		prt_bitflags(out, bch2_data_types, data_have);
+		prt_bitflags(out, __bch2_data_types, data_have);
 	else
 		prt_printf(out, "(none)");
 	prt_newline(out);
diff --git a/fs/bcachefs/snapshot.c b/fs/bcachefs/snapshot.c
index 56af937523ff..45f67e8b29eb 100644
--- a/fs/bcachefs/snapshot.c
+++ b/fs/bcachefs/snapshot.c
@@ -1053,6 +1053,8 @@ static int create_snapids(struct btree_trans *trans, u32 parent, u32 tree,
 		n->v.subvol	= cpu_to_le32(snapshot_subvols[i]);
 		n->v.tree	= cpu_to_le32(tree);
 		n->v.depth	= cpu_to_le32(depth);
+		n->v.btime.lo	= cpu_to_le64(bch2_current_time(c));
+		n->v.btime.hi	= 0;
 
 		for (j = 0; j < ARRAY_SIZE(n->v.skip); j++)
 			n->v.skip[j] = cpu_to_le32(bch2_snapshot_skiplist_get(c, parent));
@@ -1681,5 +1683,5 @@ int bch2_snapshots_read(struct bch_fs *c)
 
 void bch2_fs_snapshots_exit(struct bch_fs *c)
 {
-	kfree(rcu_dereference_protected(c->snapshots, true));
+	kvfree(rcu_dereference_protected(c->snapshots, true));
 }
diff --git a/fs/bcachefs/snapshot_format.h b/fs/bcachefs/snapshot_format.h
new file mode 100644
index 000000000000..aabcd3a74cd9
--- /dev/null
+++ b/fs/bcachefs/snapshot_format.h
@@ -0,0 +1,36 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _BCACHEFS_SNAPSHOT_FORMAT_H
+#define _BCACHEFS_SNAPSHOT_FORMAT_H
+
+struct bch_snapshot {
+	struct bch_val		v;
+	__le32			flags;
+	__le32			parent;
+	__le32			children[2];
+	__le32			subvol;
+	/* corresponds to a bch_snapshot_tree in BTREE_ID_snapshot_trees */
+	__le32			tree;
+	__le32			depth;
+	__le32			skip[3];
+	bch_le128		btime;
+};
+
+LE32_BITMASK(BCH_SNAPSHOT_DELETED,	struct bch_snapshot, flags,  0,  1)
+
+/* True if a subvolume points to this snapshot node: */
+LE32_BITMASK(BCH_SNAPSHOT_SUBVOL,	struct bch_snapshot, flags,  1,  2)
+
+/*
+ * Snapshot trees:
+ *
+ * The snapshot_trees btree gives us persistent indentifier for each tree of
+ * bch_snapshot nodes, and allow us to record and easily find the root/master
+ * subvolume that other snapshots were created from:
+ */
+struct bch_snapshot_tree {
+	struct bch_val		v;
+	__le32			master_subvol;
+	__le32			root_snapshot;
+};
+
+#endif /* _BCACHEFS_SNAPSHOT_FORMAT_H */
diff --git a/fs/bcachefs/subvolume_format.h b/fs/bcachefs/subvolume_format.h
new file mode 100644
index 000000000000..af79134b07d6
--- /dev/null
+++ b/fs/bcachefs/subvolume_format.h
@@ -0,0 +1,35 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _BCACHEFS_SUBVOLUME_FORMAT_H
+#define _BCACHEFS_SUBVOLUME_FORMAT_H
+
+#define SUBVOL_POS_MIN		POS(0, 1)
+#define SUBVOL_POS_MAX		POS(0, S32_MAX)
+#define BCACHEFS_ROOT_SUBVOL	1
+
+struct bch_subvolume {
+	struct bch_val		v;
+	__le32			flags;
+	__le32			snapshot;
+	__le64			inode;
+	/*
+	 * Snapshot subvolumes form a tree, separate from the snapshot nodes
+	 * tree - if this subvolume is a snapshot, this is the ID of the
+	 * subvolume it was created from:
+	 *
+	 * This is _not_ necessarily the subvolume of the directory containing
+	 * this subvolume:
+	 */
+	__le32			parent;
+	__le32			pad;
+	bch_le128		otime;
+};
+
+LE32_BITMASK(BCH_SUBVOLUME_RO,		struct bch_subvolume, flags,  0,  1)
+/*
+ * We need to know whether a subvolume is a snapshot so we can know whether we
+ * can delete it (or whether it should just be rm -rf'd)
+ */
+LE32_BITMASK(BCH_SUBVOLUME_SNAP,	struct bch_subvolume, flags,  1,  2)
+LE32_BITMASK(BCH_SUBVOLUME_UNLINKED,	struct bch_subvolume, flags,  2,  3)
+
+#endif /* _BCACHEFS_SUBVOLUME_FORMAT_H */
diff --git a/fs/bcachefs/super-io.c b/fs/bcachefs/super-io.c
index 6d3db5cce5f6..d60c7d27a047 100644
--- a/fs/bcachefs/super-io.c
+++ b/fs/bcachefs/super-io.c
@@ -2,7 +2,6 @@
 
 #include "bcachefs.h"
 #include "checksum.h"
-#include "counters.h"
 #include "disk_groups.h"
 #include "ec.h"
 #include "error.h"
@@ -13,6 +12,7 @@
 #include "replicas.h"
 #include "quota.h"
 #include "sb-clean.h"
+#include "sb-counters.h"
 #include "sb-downgrade.h"
 #include "sb-errors.h"
 #include "sb-members.h"
@@ -1321,7 +1321,9 @@ void bch2_sb_to_text(struct printbuf *out, struct bch_sb *sb,
 
 	prt_printf(out, "Superblock size:");
 	prt_tab(out);
-	prt_printf(out, "%zu", vstruct_bytes(sb));
+	prt_units_u64(out, vstruct_bytes(sb));
+	prt_str(out, "/");
+	prt_units_u64(out, 512ULL << sb->layout.sb_max_size_bits);
 	prt_newline(out);
 
 	prt_printf(out, "Clean:");
diff --git a/fs/bcachefs/super.c b/fs/bcachefs/super.c
index 9dbc35940197..b9911402b175 100644
--- a/fs/bcachefs/super.c
+++ b/fs/bcachefs/super.c
@@ -23,7 +23,6 @@
 #include "checksum.h"
 #include "clock.h"
 #include "compress.h"
-#include "counters.h"
 #include "debug.h"
 #include "disk_groups.h"
 #include "ec.h"
@@ -49,6 +48,7 @@
 #include "recovery.h"
 #include "replicas.h"
 #include "sb-clean.h"
+#include "sb-counters.h"
 #include "sb-errors.h"
 #include "sb-members.h"
 #include "snapshot.h"
@@ -883,7 +883,7 @@ static struct bch_fs *bch2_fs_alloc(struct bch_sb *sb, struct bch_opts opts)
 	    !(c->pcpu = alloc_percpu(struct bch_fs_pcpu)) ||
 	    !(c->online_reserved = alloc_percpu(u64)) ||
 	    mempool_init_kvpmalloc_pool(&c->btree_bounce_pool, 1,
-					btree_bytes(c)) ||
+					c->opts.btree_node_size) ||
 	    mempool_init_kmalloc_pool(&c->large_bkey_pool, 1, 2048) ||
 	    !(c->unused_inode_hints = kcalloc(1U << c->inode_shard_bits,
 					      sizeof(u64), GFP_KERNEL))) {
@@ -1386,8 +1386,8 @@ static int bch2_dev_attach_bdev(struct bch_fs *c, struct bch_sb_handle *sb)
 	prt_bdevname(&name, ca->disk_sb.bdev);
 
 	if (c->sb.nr_devices == 1)
-		strlcpy(c->name, name.buf, sizeof(c->name));
-	strlcpy(ca->name, name.buf, sizeof(ca->name));
+		strscpy(c->name, name.buf, sizeof(c->name));
+	strscpy(ca->name, name.buf, sizeof(ca->name));
 
 	printbuf_exit(&name);
 
@@ -1625,7 +1625,7 @@ int bch2_dev_remove(struct bch_fs *c, struct bch_dev *ca, int flags)
 	if (data) {
 		struct printbuf data_has = PRINTBUF;
 
-		prt_bitflags(&data_has, bch2_data_types, data);
+		prt_bitflags(&data_has, __bch2_data_types, data);
 		bch_err(ca, "Remove failed, still has data (%s)", data_has.buf);
 		printbuf_exit(&data_has);
 		ret = -EBUSY;
diff --git a/fs/bcachefs/sysfs.c b/fs/bcachefs/sysfs.c
index 8ed52319ff68..cee80c47feea 100644
--- a/fs/bcachefs/sysfs.c
+++ b/fs/bcachefs/sysfs.c
@@ -21,6 +21,7 @@
 #include "btree_gc.h"
 #include "buckets.h"
 #include "clock.h"
+#include "compress.h"
 #include "disk_groups.h"
 #include "ec.h"
 #include "inode.h"
@@ -247,7 +248,7 @@ static size_t bch2_btree_cache_size(struct bch_fs *c)
 
 	mutex_lock(&c->btree_cache.lock);
 	list_for_each_entry(b, &c->btree_cache.live, list)
-		ret += btree_bytes(c);
+		ret += btree_buf_bytes(b);
 
 	mutex_unlock(&c->btree_cache.lock);
 	return ret;
@@ -330,7 +331,7 @@ static int bch2_compression_stats_to_text(struct printbuf *out, struct bch_fs *c
 	prt_newline(out);
 
 	for (unsigned i = 0; i < ARRAY_SIZE(s); i++) {
-		prt_str(out, bch2_compression_types[i]);
+		bch2_prt_compression_type(out, i);
 		prt_tab(out);
 
 		prt_human_readable_u64(out, s[i].sectors_compressed << 9);
@@ -725,8 +726,10 @@ STORE(bch2_fs_opts_dir)
 	bch2_opt_set_sb(c, opt, v);
 	bch2_opt_set_by_id(&c->opts, id, v);
 
-	if ((id == Opt_background_target ||
-	     id == Opt_background_compression) && v)
+	if (v &&
+	    (id == Opt_background_target ||
+	     id == Opt_background_compression ||
+	     (id == Opt_compression && !c->opts.background_compression)))
 		bch2_set_rebalance_needs_scan(c, 0);
 
 	ret = size;
@@ -883,7 +886,7 @@ static void dev_io_done_to_text(struct printbuf *out, struct bch_dev *ca)
 
 		for (i = 1; i < BCH_DATA_NR; i++)
 			prt_printf(out, "%-12s:%12llu\n",
-			       bch2_data_types[i],
+			       bch2_data_type_str(i),
 			       percpu_u64_get(&ca->io_done->sectors[rw][i]) << 9);
 	}
 }
@@ -908,7 +911,7 @@ SHOW(bch2_dev)
 	}
 
 	if (attr == &sysfs_has_data) {
-		prt_bitflags(out, bch2_data_types, bch2_dev_has_data(c, ca));
+		prt_bitflags(out, __bch2_data_types, bch2_dev_has_data(c, ca));
 		prt_char(out, '\n');
 	}
 
diff --git a/fs/bcachefs/trace.h b/fs/bcachefs/trace.h
index c94876b3bb06..293b90d704fb 100644
--- a/fs/bcachefs/trace.h
+++ b/fs/bcachefs/trace.h
@@ -46,7 +46,7 @@ DECLARE_EVENT_CLASS(fs_str,
 		__assign_str(str, str);
 	),
 
-	TP_printk("%d,%d %s", MAJOR(__entry->dev), MINOR(__entry->dev), __get_str(str))
+	TP_printk("%d,%d\n%s", MAJOR(__entry->dev), MINOR(__entry->dev), __get_str(str))
 );
 
 DECLARE_EVENT_CLASS(trans_str,
@@ -273,28 +273,14 @@ DEFINE_EVENT(bch_fs, journal_full,
 	TP_ARGS(c)
 );
 
-DEFINE_EVENT(bch_fs, journal_entry_full,
-	TP_PROTO(struct bch_fs *c),
-	TP_ARGS(c)
+DEFINE_EVENT(fs_str, journal_entry_full,
+	TP_PROTO(struct bch_fs *c, const char *str),
+	TP_ARGS(c, str)
 );
 
-TRACE_EVENT(journal_entry_close,
-	TP_PROTO(struct bch_fs *c, unsigned bytes),
-	TP_ARGS(c, bytes),
-
-	TP_STRUCT__entry(
-		__field(dev_t,		dev			)
-		__field(u32,		bytes			)
-	),
-
-	TP_fast_assign(
-		__entry->dev			= c->dev;
-		__entry->bytes			= bytes;
-	),
-
-	TP_printk("%d,%d entry bytes %u",
-		  MAJOR(__entry->dev), MINOR(__entry->dev),
-		  __entry->bytes)
+DEFINE_EVENT(fs_str, journal_entry_close,
+	TP_PROTO(struct bch_fs *c, const char *str),
+	TP_ARGS(c, str)
 );
 
 DEFINE_EVENT(bio, journal_write,
@@ -542,7 +528,7 @@ TRACE_EVENT(btree_path_relock_fail,
 		__entry->level			= path->level;
 		TRACE_BPOS_assign(pos, path->pos);
 
-		c = bch2_btree_node_lock_counts(trans, NULL, &path->l[level].b->c, level),
+		c = bch2_btree_node_lock_counts(trans, NULL, &path->l[level].b->c, level);
 		__entry->self_read_count	= c.n[SIX_LOCK_read];
 		__entry->self_intent_count	= c.n[SIX_LOCK_intent];
 
@@ -827,40 +813,28 @@ TRACE_EVENT(bucket_evacuate,
 );
 
 DEFINE_EVENT(fs_str, move_extent,
-	TP_PROTO(struct bch_fs *c, const char *k),
-	TP_ARGS(c, k)
+	TP_PROTO(struct bch_fs *c, const char *str),
+	TP_ARGS(c, str)
 );
 
 DEFINE_EVENT(fs_str, move_extent_read,
-	TP_PROTO(struct bch_fs *c, const char *k),
-	TP_ARGS(c, k)
+	TP_PROTO(struct bch_fs *c, const char *str),
+	TP_ARGS(c, str)
 );
 
 DEFINE_EVENT(fs_str, move_extent_write,
-	TP_PROTO(struct bch_fs *c, const char *k),
-	TP_ARGS(c, k)
+	TP_PROTO(struct bch_fs *c, const char *str),
+	TP_ARGS(c, str)
 );
 
 DEFINE_EVENT(fs_str, move_extent_finish,
-	TP_PROTO(struct bch_fs *c, const char *k),
-	TP_ARGS(c, k)
+	TP_PROTO(struct bch_fs *c, const char *str),
+	TP_ARGS(c, str)
 );
 
-TRACE_EVENT(move_extent_fail,
-	TP_PROTO(struct bch_fs *c, const char *msg),
-	TP_ARGS(c, msg),
-
-	TP_STRUCT__entry(
-		__field(dev_t,		dev			)
-		__string(msg,		msg			)
-	),
-
-	TP_fast_assign(
-		__entry->dev		= c->dev;
-		__assign_str(msg, msg);
-	),
-
-	TP_printk("%d:%d %s", MAJOR(__entry->dev), MINOR(__entry->dev), __get_str(msg))
+DEFINE_EVENT(fs_str, move_extent_fail,
+	TP_PROTO(struct bch_fs *c, const char *str),
+	TP_ARGS(c, str)
 );
 
 DEFINE_EVENT(fs_str, move_extent_start_fail,
@@ -1039,7 +1013,7 @@ TRACE_EVENT(trans_restart_split_race,
 		__entry->level		= b->c.level;
 		__entry->written	= b->written;
 		__entry->blocks		= btree_blocks(trans->c);
-		__entry->u64s_remaining	= bch_btree_keys_u64s_remaining(trans->c, b);
+		__entry->u64s_remaining	= bch2_btree_keys_u64s_remaining(b);
 	),
 
 	TP_printk("%s %pS l=%u written %u/%u u64s remaining %u",
@@ -1146,8 +1120,6 @@ DEFINE_EVENT(transaction_restart_iter,	trans_restart_btree_node_split,
 	TP_ARGS(trans, caller_ip, path)
 );
 
-struct get_locks_fail;
-
 TRACE_EVENT(trans_restart_upgrade,
 	TP_PROTO(struct btree_trans *trans,
 		 unsigned long caller_ip,
@@ -1195,11 +1167,9 @@ TRACE_EVENT(trans_restart_upgrade,
 		  __entry->node_seq)
 );
 
-DEFINE_EVENT(transaction_restart_iter,	trans_restart_relock,
-	TP_PROTO(struct btree_trans *trans,
-		 unsigned long caller_ip,
-		 struct btree_path *path),
-	TP_ARGS(trans, caller_ip, path)
+DEFINE_EVENT(trans_str,	trans_restart_relock,
+	TP_PROTO(struct btree_trans *trans, unsigned long caller_ip, const char *str),
+	TP_ARGS(trans, caller_ip, str)
 );
 
 DEFINE_EVENT(transaction_restart_iter,	trans_restart_relock_next_node,
diff --git a/fs/bcachefs/util.c b/fs/bcachefs/util.c
index c2ef7cddaa4f..a135136adeee 100644
--- a/fs/bcachefs/util.c
+++ b/fs/bcachefs/util.c
@@ -241,12 +241,17 @@ bool bch2_is_zero(const void *_p, size_t n)
 	return true;
 }
 
-void bch2_prt_u64_binary(struct printbuf *out, u64 v, unsigned nr_bits)
+void bch2_prt_u64_base2_nbits(struct printbuf *out, u64 v, unsigned nr_bits)
 {
 	while (nr_bits)
 		prt_char(out, '0' + ((v >> --nr_bits) & 1));
 }
 
+void bch2_prt_u64_base2(struct printbuf *out, u64 v)
+{
+	bch2_prt_u64_base2_nbits(out, v, fls64(v) ?: 1);
+}
+
 void bch2_print_string_as_lines(const char *prefix, const char *lines)
 {
 	const char *p;
@@ -1186,7 +1191,9 @@ int bch2_split_devs(const char *_dev_name, darray_str *ret)
 {
 	darray_init(ret);
 
-	char *dev_name = kstrdup(_dev_name, GFP_KERNEL), *s = dev_name;
+	char *dev_name, *s, *orig;
+
+	dev_name = orig = kstrdup(_dev_name, GFP_KERNEL);
 	if (!dev_name)
 		return -ENOMEM;
 
@@ -1201,10 +1208,10 @@ int bch2_split_devs(const char *_dev_name, darray_str *ret)
 		}
 	}
 
-	kfree(dev_name);
+	kfree(orig);
 	return 0;
 err:
 	bch2_darray_str_exit(ret);
-	kfree(dev_name);
+	kfree(orig);
 	return -ENOMEM;
 }
diff --git a/fs/bcachefs/util.h b/fs/bcachefs/util.h
index c75fc31915d3..df67bf55fe2b 100644
--- a/fs/bcachefs/util.h
+++ b/fs/bcachefs/util.h
@@ -342,7 +342,8 @@ bool bch2_is_zero(const void *, size_t);
 
 u64 bch2_read_flag_list(char *, const char * const[]);
 
-void bch2_prt_u64_binary(struct printbuf *, u64, unsigned);
+void bch2_prt_u64_base2_nbits(struct printbuf *, u64, unsigned);
+void bch2_prt_u64_base2(struct printbuf *, u64);
 
 void bch2_print_string_as_lines(const char *prefix, const char *lines);
 
diff --git a/fs/bcachefs/xattr.c b/fs/bcachefs/xattr.c
index 5a1858fb9879..9c0d2316031b 100644
--- a/fs/bcachefs/xattr.c
+++ b/fs/bcachefs/xattr.c
@@ -590,8 +590,9 @@ err:
 	mutex_unlock(&inode->ei_update_lock);
 
 	if (value &&
-	    (opt_id == Opt_background_compression ||
-	     opt_id == Opt_background_target))
+	    (opt_id == Opt_background_target ||
+	     opt_id == Opt_background_compression ||
+	     (opt_id == Opt_compression && !inode_opt_get(c, &inode->ei_inode, background_compression))))
 		bch2_set_rebalance_needs_scan(c, inode->ei_inode.bi_inum);
 
 	return bch2_err_class(ret);
diff --git a/fs/bcachefs/xattr_format.h b/fs/bcachefs/xattr_format.h
new file mode 100644
index 000000000000..e9f810539552
--- /dev/null
+++ b/fs/bcachefs/xattr_format.h
@@ -0,0 +1,19 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _BCACHEFS_XATTR_FORMAT_H
+#define _BCACHEFS_XATTR_FORMAT_H
+
+#define KEY_TYPE_XATTR_INDEX_USER		0
+#define KEY_TYPE_XATTR_INDEX_POSIX_ACL_ACCESS	1
+#define KEY_TYPE_XATTR_INDEX_POSIX_ACL_DEFAULT	2
+#define KEY_TYPE_XATTR_INDEX_TRUSTED		3
+#define KEY_TYPE_XATTR_INDEX_SECURITY	        4
+
+struct bch_xattr {
+	struct bch_val		v;
+	__u8			x_type;
+	__u8			x_name_len;
+	__le16			x_val_len;
+	__u8			x_name[];
+} __packed __aligned(8);
+
+#endif /* _BCACHEFS_XATTR_FORMAT_H */
diff --git a/fs/btrfs/compression.c b/fs/btrfs/compression.c
index 193168214eeb..68345f73d429 100644
--- a/fs/btrfs/compression.c
+++ b/fs/btrfs/compression.c
@@ -141,16 +141,16 @@ static int compression_decompress_bio(struct list_head *ws,
 }
 
 static int compression_decompress(int type, struct list_head *ws,
-               const u8 *data_in, struct page *dest_page,
-               unsigned long start_byte, size_t srclen, size_t destlen)
+		const u8 *data_in, struct page *dest_page,
+		unsigned long dest_pgoff, size_t srclen, size_t destlen)
 {
 	switch (type) {
 	case BTRFS_COMPRESS_ZLIB: return zlib_decompress(ws, data_in, dest_page,
-						start_byte, srclen, destlen);
+						dest_pgoff, srclen, destlen);
 	case BTRFS_COMPRESS_LZO:  return lzo_decompress(ws, data_in, dest_page,
-						start_byte, srclen, destlen);
+						dest_pgoff, srclen, destlen);
 	case BTRFS_COMPRESS_ZSTD: return zstd_decompress(ws, data_in, dest_page,
-						start_byte, srclen, destlen);
+						dest_pgoff, srclen, destlen);
 	case BTRFS_COMPRESS_NONE:
 	default:
 		/*
@@ -1037,14 +1037,23 @@ static int btrfs_decompress_bio(struct compressed_bio *cb)
  * start_byte tells us the offset into the compressed data we're interested in
  */
 int btrfs_decompress(int type, const u8 *data_in, struct page *dest_page,
-		     unsigned long start_byte, size_t srclen, size_t destlen)
+		     unsigned long dest_pgoff, size_t srclen, size_t destlen)
 {
+	struct btrfs_fs_info *fs_info = btrfs_sb(dest_page->mapping->host->i_sb);
 	struct list_head *workspace;
+	const u32 sectorsize = fs_info->sectorsize;
 	int ret;
 
+	/*
+	 * The full destination page range should not exceed the page size.
+	 * And the @destlen should not exceed sectorsize, as this is only called for
+	 * inline file extents, which should not exceed sectorsize.
+	 */
+	ASSERT(dest_pgoff + destlen <= PAGE_SIZE && destlen <= sectorsize);
+
 	workspace = get_workspace(type, 0);
 	ret = compression_decompress(type, workspace, data_in, dest_page,
-				     start_byte, srclen, destlen);
+				     dest_pgoff, srclen, destlen);
 	put_workspace(type, workspace);
 
 	return ret;
diff --git a/fs/btrfs/compression.h b/fs/btrfs/compression.h
index 93cc92974dee..afd7e50d073d 100644
--- a/fs/btrfs/compression.h
+++ b/fs/btrfs/compression.h
@@ -148,7 +148,7 @@ int zlib_compress_pages(struct list_head *ws, struct address_space *mapping,
 		unsigned long *total_in, unsigned long *total_out);
 int zlib_decompress_bio(struct list_head *ws, struct compressed_bio *cb);
 int zlib_decompress(struct list_head *ws, const u8 *data_in,
-		struct page *dest_page, unsigned long start_byte, size_t srclen,
+		struct page *dest_page, unsigned long dest_pgoff, size_t srclen,
 		size_t destlen);
 struct list_head *zlib_alloc_workspace(unsigned int level);
 void zlib_free_workspace(struct list_head *ws);
@@ -159,7 +159,7 @@ int lzo_compress_pages(struct list_head *ws, struct address_space *mapping,
 		unsigned long *total_in, unsigned long *total_out);
 int lzo_decompress_bio(struct list_head *ws, struct compressed_bio *cb);
 int lzo_decompress(struct list_head *ws, const u8 *data_in,
-		struct page *dest_page, unsigned long start_byte, size_t srclen,
+		struct page *dest_page, unsigned long dest_pgoff, size_t srclen,
 		size_t destlen);
 struct list_head *lzo_alloc_workspace(unsigned int level);
 void lzo_free_workspace(struct list_head *ws);
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index f396aba92c57..8e8cc1111277 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -1260,7 +1260,8 @@ static int btrfs_issue_discard(struct block_device *bdev, u64 start, u64 len,
 	u64 bytes_left, end;
 	u64 aligned_start = ALIGN(start, 1 << SECTOR_SHIFT);
 
-	if (WARN_ON(start != aligned_start)) {
+	/* Adjust the range to be aligned to 512B sectors if necessary. */
+	if (start != aligned_start) {
 		len -= aligned_start - start;
 		len = round_down(len, 1 << SECTOR_SHIFT);
 		start = aligned_start;
@@ -4298,6 +4299,42 @@ static int prepare_allocation_clustered(struct btrfs_fs_info *fs_info,
 	return 0;
 }
 
+static int prepare_allocation_zoned(struct btrfs_fs_info *fs_info,
+				    struct find_free_extent_ctl *ffe_ctl)
+{
+	if (ffe_ctl->for_treelog) {
+		spin_lock(&fs_info->treelog_bg_lock);
+		if (fs_info->treelog_bg)
+			ffe_ctl->hint_byte = fs_info->treelog_bg;
+		spin_unlock(&fs_info->treelog_bg_lock);
+	} else if (ffe_ctl->for_data_reloc) {
+		spin_lock(&fs_info->relocation_bg_lock);
+		if (fs_info->data_reloc_bg)
+			ffe_ctl->hint_byte = fs_info->data_reloc_bg;
+		spin_unlock(&fs_info->relocation_bg_lock);
+	} else if (ffe_ctl->flags & BTRFS_BLOCK_GROUP_DATA) {
+		struct btrfs_block_group *block_group;
+
+		spin_lock(&fs_info->zone_active_bgs_lock);
+		list_for_each_entry(block_group, &fs_info->zone_active_bgs, active_bg_list) {
+			/*
+			 * No lock is OK here because avail is monotinically
+			 * decreasing, and this is just a hint.
+			 */
+			u64 avail = block_group->zone_capacity - block_group->alloc_offset;
+
+			if (block_group_bits(block_group, ffe_ctl->flags) &&
+			    avail >= ffe_ctl->num_bytes) {
+				ffe_ctl->hint_byte = block_group->start;
+				break;
+			}
+		}
+		spin_unlock(&fs_info->zone_active_bgs_lock);
+	}
+
+	return 0;
+}
+
 static int prepare_allocation(struct btrfs_fs_info *fs_info,
 			      struct find_free_extent_ctl *ffe_ctl,
 			      struct btrfs_space_info *space_info,
@@ -4308,19 +4345,7 @@ static int prepare_allocation(struct btrfs_fs_info *fs_info,
 		return prepare_allocation_clustered(fs_info, ffe_ctl,
 						    space_info, ins);
 	case BTRFS_EXTENT_ALLOC_ZONED:
-		if (ffe_ctl->for_treelog) {
-			spin_lock(&fs_info->treelog_bg_lock);
-			if (fs_info->treelog_bg)
-				ffe_ctl->hint_byte = fs_info->treelog_bg;
-			spin_unlock(&fs_info->treelog_bg_lock);
-		}
-		if (ffe_ctl->for_data_reloc) {
-			spin_lock(&fs_info->relocation_bg_lock);
-			if (fs_info->data_reloc_bg)
-				ffe_ctl->hint_byte = fs_info->data_reloc_bg;
-			spin_unlock(&fs_info->relocation_bg_lock);
-		}
-		return 0;
+		return prepare_allocation_zoned(fs_info, ffe_ctl);
 	default:
 		BUG();
 	}
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 809b11472a80..1eb93d3962aa 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -4458,6 +4458,8 @@ int btrfs_delete_subvolume(struct btrfs_inode *dir, struct dentry *dentry)
 	u64 root_flags;
 	int ret;
 
+	down_write(&fs_info->subvol_sem);
+
 	/*
 	 * Don't allow to delete a subvolume with send in progress. This is
 	 * inside the inode lock so the error handling that has to drop the bit
@@ -4469,25 +4471,25 @@ int btrfs_delete_subvolume(struct btrfs_inode *dir, struct dentry *dentry)
 		btrfs_warn(fs_info,
 			   "attempt to delete subvolume %llu during send",
 			   dest->root_key.objectid);
-		return -EPERM;
+		ret = -EPERM;
+		goto out_up_write;
 	}
 	if (atomic_read(&dest->nr_swapfiles)) {
 		spin_unlock(&dest->root_item_lock);
 		btrfs_warn(fs_info,
 			   "attempt to delete subvolume %llu with active swapfile",
 			   root->root_key.objectid);
-		return -EPERM;
+		ret = -EPERM;
+		goto out_up_write;
 	}
 	root_flags = btrfs_root_flags(&dest->root_item);
 	btrfs_set_root_flags(&dest->root_item,
 			     root_flags | BTRFS_ROOT_SUBVOL_DEAD);
 	spin_unlock(&dest->root_item_lock);
 
-	down_write(&fs_info->subvol_sem);
-
 	ret = may_destroy_subvol(dest);
 	if (ret)
-		goto out_up_write;
+		goto out_undead;
 
 	btrfs_init_block_rsv(&block_rsv, BTRFS_BLOCK_RSV_TEMP);
 	/*
@@ -4497,7 +4499,7 @@ int btrfs_delete_subvolume(struct btrfs_inode *dir, struct dentry *dentry)
 	 */
 	ret = btrfs_subvolume_reserve_metadata(root, &block_rsv, 5, true);
 	if (ret)
-		goto out_up_write;
+		goto out_undead;
 
 	trans = btrfs_start_transaction(root, 0);
 	if (IS_ERR(trans)) {
@@ -4563,15 +4565,17 @@ out_end_trans:
 	inode->i_flags |= S_DEAD;
 out_release:
 	btrfs_subvolume_release_metadata(root, &block_rsv);
-out_up_write:
-	up_write(&fs_info->subvol_sem);
+out_undead:
 	if (ret) {
 		spin_lock(&dest->root_item_lock);
 		root_flags = btrfs_root_flags(&dest->root_item);
 		btrfs_set_root_flags(&dest->root_item,
 				root_flags & ~BTRFS_ROOT_SUBVOL_DEAD);
 		spin_unlock(&dest->root_item_lock);
-	} else {
+	}
+out_up_write:
+	up_write(&fs_info->subvol_sem);
+	if (!ret) {
 		d_invalidate(dentry);
 		btrfs_prune_dentries(dest);
 		ASSERT(dest->send_in_progress == 0);
diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
index 41b479861b3c..dfed9dd9c2d7 100644
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -790,6 +790,9 @@ static int create_snapshot(struct btrfs_root *root, struct inode *dir,
 		return -EOPNOTSUPP;
 	}
 
+	if (btrfs_root_refs(&root->root_item) == 0)
+		return -ENOENT;
+
 	if (!test_bit(BTRFS_ROOT_SHAREABLE, &root->state))
 		return -EINVAL;
 
@@ -2608,6 +2611,10 @@ static int btrfs_ioctl_defrag(struct file *file, void __user *argp)
 				ret = -EFAULT;
 				goto out;
 			}
+			if (range.flags & ~BTRFS_DEFRAG_RANGE_FLAGS_SUPP) {
+				ret = -EOPNOTSUPP;
+				goto out;
+			}
 			/* compression requires us to start the IO */
 			if ((range.flags & BTRFS_DEFRAG_RANGE_COMPRESS)) {
 				range.flags |= BTRFS_DEFRAG_RANGE_START_IO;
diff --git a/fs/btrfs/lzo.c b/fs/btrfs/lzo.c
index 1131d5a29d61..e43bc0fdc74e 100644
--- a/fs/btrfs/lzo.c
+++ b/fs/btrfs/lzo.c
@@ -425,16 +425,16 @@ int lzo_decompress_bio(struct list_head *ws, struct compressed_bio *cb)
 }
 
 int lzo_decompress(struct list_head *ws, const u8 *data_in,
-		struct page *dest_page, unsigned long start_byte, size_t srclen,
+		struct page *dest_page, unsigned long dest_pgoff, size_t srclen,
 		size_t destlen)
 {
 	struct workspace *workspace = list_entry(ws, struct workspace, list);
+	struct btrfs_fs_info *fs_info = btrfs_sb(dest_page->mapping->host->i_sb);
+	const u32 sectorsize = fs_info->sectorsize;
 	size_t in_len;
 	size_t out_len;
 	size_t max_segment_len = WORKSPACE_BUF_LENGTH;
 	int ret = 0;
-	char *kaddr;
-	unsigned long bytes;
 
 	if (srclen < LZO_LEN || srclen > max_segment_len + LZO_LEN * 2)
 		return -EUCLEAN;
@@ -451,7 +451,7 @@ int lzo_decompress(struct list_head *ws, const u8 *data_in,
 	}
 	data_in += LZO_LEN;
 
-	out_len = PAGE_SIZE;
+	out_len = sectorsize;
 	ret = lzo1x_decompress_safe(data_in, in_len, workspace->buf, &out_len);
 	if (ret != LZO_E_OK) {
 		pr_warn("BTRFS: decompress failed!\n");
@@ -459,29 +459,13 @@ int lzo_decompress(struct list_head *ws, const u8 *data_in,
 		goto out;
 	}
 
-	if (out_len < start_byte) {
+	ASSERT(out_len <= sectorsize);
+	memcpy_to_page(dest_page, dest_pgoff, workspace->buf, out_len);
+	/* Early end, considered as an error. */
+	if (unlikely(out_len < destlen)) {
 		ret = -EIO;
-		goto out;
+		memzero_page(dest_page, dest_pgoff + out_len, destlen - out_len);
 	}
-
-	/*
-	 * the caller is already checking against PAGE_SIZE, but lets
-	 * move this check closer to the memcpy/memset
-	 */
-	destlen = min_t(unsigned long, destlen, PAGE_SIZE);
-	bytes = min_t(unsigned long, destlen, out_len - start_byte);
-
-	kaddr = kmap_local_page(dest_page);
-	memcpy(kaddr, workspace->buf + start_byte, bytes);
-
-	/*
-	 * btrfs_getblock is doing a zero on the tail of the page too,
-	 * but this will cover anything missing from the decompressed
-	 * data.
-	 */
-	if (bytes < destlen)
-		memset(kaddr+bytes, 0, destlen-bytes);
-	kunmap_local(kaddr);
 out:
 	return ret;
 }
diff --git a/fs/btrfs/ref-verify.c b/fs/btrfs/ref-verify.c
index 6486f0d7e993..8c4fc98ca9ce 100644
--- a/fs/btrfs/ref-verify.c
+++ b/fs/btrfs/ref-verify.c
@@ -889,8 +889,10 @@ int btrfs_ref_tree_mod(struct btrfs_fs_info *fs_info,
 out_unlock:
 	spin_unlock(&fs_info->ref_verify_lock);
 out:
-	if (ret)
+	if (ret) {
+		btrfs_free_ref_cache(fs_info);
 		btrfs_clear_opt(fs_info->mount_opt, REF_VERIFY);
+	}
 	return ret;
 }
 
@@ -1021,8 +1023,8 @@ int btrfs_build_ref_tree(struct btrfs_fs_info *fs_info)
 		}
 	}
 	if (ret) {
-		btrfs_clear_opt(fs_info->mount_opt, REF_VERIFY);
 		btrfs_free_ref_cache(fs_info);
+		btrfs_clear_opt(fs_info->mount_opt, REF_VERIFY);
 	}
 	btrfs_free_path(path);
 	return ret;
diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c
index a01807cbd4d4..0123d2728923 100644
--- a/fs/btrfs/scrub.c
+++ b/fs/btrfs/scrub.c
@@ -1098,12 +1098,22 @@ out:
 static void scrub_read_endio(struct btrfs_bio *bbio)
 {
 	struct scrub_stripe *stripe = bbio->private;
+	struct bio_vec *bvec;
+	int sector_nr = calc_sector_number(stripe, bio_first_bvec_all(&bbio->bio));
+	int num_sectors;
+	u32 bio_size = 0;
+	int i;
+
+	ASSERT(sector_nr < stripe->nr_sectors);
+	bio_for_each_bvec_all(bvec, &bbio->bio, i)
+		bio_size += bvec->bv_len;
+	num_sectors = bio_size >> stripe->bg->fs_info->sectorsize_bits;
 
 	if (bbio->bio.bi_status) {
-		bitmap_set(&stripe->io_error_bitmap, 0, stripe->nr_sectors);
-		bitmap_set(&stripe->error_bitmap, 0, stripe->nr_sectors);
+		bitmap_set(&stripe->io_error_bitmap, sector_nr, num_sectors);
+		bitmap_set(&stripe->error_bitmap, sector_nr, num_sectors);
 	} else {
-		bitmap_clear(&stripe->io_error_bitmap, 0, stripe->nr_sectors);
+		bitmap_clear(&stripe->io_error_bitmap, sector_nr, num_sectors);
 	}
 	bio_put(&bbio->bio);
 	if (atomic_dec_and_test(&stripe->pending_io)) {
@@ -1636,6 +1646,9 @@ static void scrub_submit_extent_sector_read(struct scrub_ctx *sctx,
 {
 	struct btrfs_fs_info *fs_info = stripe->bg->fs_info;
 	struct btrfs_bio *bbio = NULL;
+	unsigned int nr_sectors = min(BTRFS_STRIPE_LEN, stripe->bg->start +
+				      stripe->bg->length - stripe->logical) >>
+				  fs_info->sectorsize_bits;
 	u64 stripe_len = BTRFS_STRIPE_LEN;
 	int mirror = stripe->mirror_num;
 	int i;
@@ -1646,6 +1659,10 @@ static void scrub_submit_extent_sector_read(struct scrub_ctx *sctx,
 		struct page *page = scrub_stripe_get_page(stripe, i);
 		unsigned int pgoff = scrub_stripe_get_page_offset(stripe, i);
 
+		/* We're beyond the chunk boundary, no need to read anymore. */
+		if (i >= nr_sectors)
+			break;
+
 		/* The current sector cannot be merged, submit the bio. */
 		if (bbio &&
 		    ((i > 0 &&
@@ -1701,6 +1718,9 @@ static void scrub_submit_initial_read(struct scrub_ctx *sctx,
 {
 	struct btrfs_fs_info *fs_info = sctx->fs_info;
 	struct btrfs_bio *bbio;
+	unsigned int nr_sectors = min(BTRFS_STRIPE_LEN, stripe->bg->start +
+				      stripe->bg->length - stripe->logical) >>
+				  fs_info->sectorsize_bits;
 	int mirror = stripe->mirror_num;
 
 	ASSERT(stripe->bg);
@@ -1715,14 +1735,16 @@ static void scrub_submit_initial_read(struct scrub_ctx *sctx,
 	bbio = btrfs_bio_alloc(SCRUB_STRIPE_PAGES, REQ_OP_READ, fs_info,
 			       scrub_read_endio, stripe);
 
-	/* Read the whole stripe. */
 	bbio->bio.bi_iter.bi_sector = stripe->logical >> SECTOR_SHIFT;
-	for (int i = 0; i < BTRFS_STRIPE_LEN >> PAGE_SHIFT; i++) {
+	/* Read the whole range inside the chunk boundary. */
+	for (unsigned int cur = 0; cur < nr_sectors; cur++) {
+		struct page *page = scrub_stripe_get_page(stripe, cur);
+		unsigned int pgoff = scrub_stripe_get_page_offset(stripe, cur);
 		int ret;
 
-		ret = bio_add_page(&bbio->bio, stripe->pages[i], PAGE_SIZE, 0);
+		ret = bio_add_page(&bbio->bio, page, fs_info->sectorsize, pgoff);
 		/* We should have allocated enough bio vectors. */
-		ASSERT(ret == PAGE_SIZE);
+		ASSERT(ret == fs_info->sectorsize);
 	}
 	atomic_inc(&stripe->pending_io);
 
diff --git a/fs/btrfs/send.c b/fs/btrfs/send.c
index 4e36550618e5..2d7519a6ce72 100644
--- a/fs/btrfs/send.c
+++ b/fs/btrfs/send.c
@@ -8205,8 +8205,8 @@ long btrfs_ioctl_send(struct inode *inode, struct btrfs_ioctl_send_args *arg)
 		goto out;
 	}
 
-	sctx->clone_roots = kvcalloc(sizeof(*sctx->clone_roots),
-				     arg->clone_sources_count + 1,
+	sctx->clone_roots = kvcalloc(arg->clone_sources_count + 1,
+				     sizeof(*sctx->clone_roots),
 				     GFP_KERNEL);
 	if (!sctx->clone_roots) {
 		ret = -ENOMEM;
diff --git a/fs/btrfs/subpage.c b/fs/btrfs/subpage.c
index 93511d54abf8..0e49dab8dad2 100644
--- a/fs/btrfs/subpage.c
+++ b/fs/btrfs/subpage.c
@@ -475,7 +475,8 @@ void btrfs_subpage_set_writeback(const struct btrfs_fs_info *fs_info,
 
 	spin_lock_irqsave(&subpage->lock, flags);
 	bitmap_set(subpage->bitmaps, start_bit, len >> fs_info->sectorsize_bits);
-	folio_start_writeback(folio);
+	if (!folio_test_writeback(folio))
+		folio_start_writeback(folio);
 	spin_unlock_irqrestore(&subpage->lock, flags);
 }
 
diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
index 896acfda1789..101f786963d4 100644
--- a/fs/btrfs/super.c
+++ b/fs/btrfs/super.c
@@ -1457,6 +1457,14 @@ static int btrfs_reconfigure(struct fs_context *fc)
 
 	btrfs_info_to_ctx(fs_info, &old_ctx);
 
+	/*
+	 * This is our "bind mount" trick, we don't want to allow the user to do
+	 * anything other than mount a different ro/rw and a different subvol,
+	 * all of the mount options should be maintained.
+	 */
+	if (mount_reconfigure)
+		ctx->mount_opt = old_ctx.mount_opt;
+
 	sync_filesystem(sb);
 	set_bit(BTRFS_FS_STATE_REMOUNTING, &fs_info->fs_state);
 
diff --git a/fs/btrfs/tree-checker.c b/fs/btrfs/tree-checker.c
index 50fdc69fdddf..6eccf8496486 100644
--- a/fs/btrfs/tree-checker.c
+++ b/fs/btrfs/tree-checker.c
@@ -1436,7 +1436,7 @@ static int check_extent_item(struct extent_buffer *leaf,
 		if (unlikely(ptr + btrfs_extent_inline_ref_size(inline_type) > end)) {
 			extent_err(leaf, slot,
 "inline ref item overflows extent item, ptr %lu iref size %u end %lu",
-				   ptr, inline_type, end);
+				   ptr, btrfs_extent_inline_ref_size(inline_type), end);
 			return -EUCLEAN;
 		}
 
diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
index 4c32497311d2..d67785be2c77 100644
--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -3087,7 +3087,6 @@ struct btrfs_chunk_map *btrfs_get_chunk_map(struct btrfs_fs_info *fs_info,
 	map = btrfs_find_chunk_map(fs_info, logical, length);
 
 	if (unlikely(!map)) {
-		read_unlock(&fs_info->mapping_tree_lock);
 		btrfs_crit(fs_info,
 			   "unable to find chunk map for logical %llu length %llu",
 			   logical, length);
@@ -3095,7 +3094,6 @@ struct btrfs_chunk_map *btrfs_get_chunk_map(struct btrfs_fs_info *fs_info,
 	}
 
 	if (unlikely(map->start > logical || map->start + map->chunk_len <= logical)) {
-		read_unlock(&fs_info->mapping_tree_lock);
 		btrfs_crit(fs_info,
 			   "found a bad chunk map, wanted %llu-%llu, found %llu-%llu",
 			   logical, logical + length, map->start,
diff --git a/fs/btrfs/zlib.c b/fs/btrfs/zlib.c
index 36cf1f0e338e..8da66ea699e8 100644
--- a/fs/btrfs/zlib.c
+++ b/fs/btrfs/zlib.c
@@ -354,18 +354,13 @@ done:
 }
 
 int zlib_decompress(struct list_head *ws, const u8 *data_in,
-		struct page *dest_page, unsigned long start_byte, size_t srclen,
+		struct page *dest_page, unsigned long dest_pgoff, size_t srclen,
 		size_t destlen)
 {
 	struct workspace *workspace = list_entry(ws, struct workspace, list);
 	int ret = 0;
 	int wbits = MAX_WBITS;
-	unsigned long bytes_left;
-	unsigned long total_out = 0;
-	unsigned long pg_offset = 0;
-
-	destlen = min_t(unsigned long, destlen, PAGE_SIZE);
-	bytes_left = destlen;
+	unsigned long to_copy;
 
 	workspace->strm.next_in = data_in;
 	workspace->strm.avail_in = srclen;
@@ -390,60 +385,30 @@ int zlib_decompress(struct list_head *ws, const u8 *data_in,
 		return -EIO;
 	}
 
-	while (bytes_left > 0) {
-		unsigned long buf_start;
-		unsigned long buf_offset;
-		unsigned long bytes;
-
-		ret = zlib_inflate(&workspace->strm, Z_NO_FLUSH);
-		if (ret != Z_OK && ret != Z_STREAM_END)
-			break;
-
-		buf_start = total_out;
-		total_out = workspace->strm.total_out;
-
-		if (total_out == buf_start) {
-			ret = -EIO;
-			break;
-		}
-
-		if (total_out <= start_byte)
-			goto next;
-
-		if (total_out > start_byte && buf_start < start_byte)
-			buf_offset = start_byte - buf_start;
-		else
-			buf_offset = 0;
-
-		bytes = min(PAGE_SIZE - pg_offset,
-			    PAGE_SIZE - (buf_offset % PAGE_SIZE));
-		bytes = min(bytes, bytes_left);
+	/*
+	 * Everything (in/out buf) should be at most one sector, there should
+	 * be no need to switch any input/output buffer.
+	 */
+	ret = zlib_inflate(&workspace->strm, Z_FINISH);
+	to_copy = min(workspace->strm.total_out, destlen);
+	if (ret != Z_STREAM_END)
+		goto out;
 
-		memcpy_to_page(dest_page, pg_offset,
-			       workspace->buf + buf_offset, bytes);
+	memcpy_to_page(dest_page, dest_pgoff, workspace->buf, to_copy);
 
-		pg_offset += bytes;
-		bytes_left -= bytes;
-next:
-		workspace->strm.next_out = workspace->buf;
-		workspace->strm.avail_out = workspace->buf_size;
-	}
-
-	if (ret != Z_STREAM_END && bytes_left != 0)
+out:
+	if (unlikely(to_copy != destlen)) {
+		pr_warn_ratelimited("BTRFS: infalte failed, decompressed=%lu expected=%zu\n",
+					to_copy, destlen);
 		ret = -EIO;
-	else
+	} else {
 		ret = 0;
+	}
 
 	zlib_inflateEnd(&workspace->strm);
 
-	/*
-	 * this should only happen if zlib returned fewer bytes than we
-	 * expected.  btrfs_get_block is responsible for zeroing from the
-	 * end of the inline extent (destlen) to the end of the page
-	 */
-	if (pg_offset < destlen) {
-		memzero_page(dest_page, pg_offset, destlen - pg_offset);
-	}
+	if (unlikely(to_copy < destlen))
+		memzero_page(dest_page, dest_pgoff + to_copy, destlen - to_copy);
 	return ret;
 }
 
diff --git a/fs/btrfs/zoned.c b/fs/btrfs/zoned.c
index 5bd76813b23f..168af9d000d1 100644
--- a/fs/btrfs/zoned.c
+++ b/fs/btrfs/zoned.c
@@ -2055,6 +2055,7 @@ bool btrfs_zone_activate(struct btrfs_block_group *block_group)
 
 	map = block_group->physical_map;
 
+	spin_lock(&fs_info->zone_active_bgs_lock);
 	spin_lock(&block_group->lock);
 	if (test_bit(BLOCK_GROUP_FLAG_ZONE_IS_ACTIVE, &block_group->runtime_flags)) {
 		ret = true;
@@ -2067,7 +2068,6 @@ bool btrfs_zone_activate(struct btrfs_block_group *block_group)
 		goto out_unlock;
 	}
 
-	spin_lock(&fs_info->zone_active_bgs_lock);
 	for (i = 0; i < map->num_stripes; i++) {
 		struct btrfs_zoned_device_info *zinfo;
 		int reserved = 0;
@@ -2087,20 +2087,17 @@ bool btrfs_zone_activate(struct btrfs_block_group *block_group)
 		 */
 		if (atomic_read(&zinfo->active_zones_left) <= reserved) {
 			ret = false;
-			spin_unlock(&fs_info->zone_active_bgs_lock);
 			goto out_unlock;
 		}
 
 		if (!btrfs_dev_set_active_zone(device, physical)) {
 			/* Cannot activate the zone */
 			ret = false;
-			spin_unlock(&fs_info->zone_active_bgs_lock);
 			goto out_unlock;
 		}
 		if (!is_data)
 			zinfo->reserved_active_zones--;
 	}
-	spin_unlock(&fs_info->zone_active_bgs_lock);
 
 	/* Successfully activated all the zones */
 	set_bit(BLOCK_GROUP_FLAG_ZONE_IS_ACTIVE, &block_group->runtime_flags);
@@ -2108,8 +2105,6 @@ bool btrfs_zone_activate(struct btrfs_block_group *block_group)
 
 	/* For the active block group list */
 	btrfs_get_block_group(block_group);
-
-	spin_lock(&fs_info->zone_active_bgs_lock);
 	list_add_tail(&block_group->active_bg_list, &fs_info->zone_active_bgs);
 	spin_unlock(&fs_info->zone_active_bgs_lock);
 
@@ -2117,6 +2112,7 @@ bool btrfs_zone_activate(struct btrfs_block_group *block_group)
 
 out_unlock:
 	spin_unlock(&block_group->lock);
+	spin_unlock(&fs_info->zone_active_bgs_lock);
 	return ret;
 }
 
diff --git a/fs/cachefiles/Kconfig b/fs/cachefiles/Kconfig
index 8df715640a48..c5a070550ee3 100644
--- a/fs/cachefiles/Kconfig
+++ b/fs/cachefiles/Kconfig
@@ -2,7 +2,7 @@
 
 config CACHEFILES
 	tristate "Filesystem caching on files"
-	depends on FSCACHE && BLOCK
+	depends on NETFS_SUPPORT && FSCACHE && BLOCK
 	help
 	  This permits use of a mounted filesystem as a cache for other
 	  filesystems - primarily networking filesystems - thus allowing fast
diff --git a/fs/cachefiles/internal.h b/fs/cachefiles/internal.h
index 4a87c9d714a9..d33169f0018b 100644
--- a/fs/cachefiles/internal.h
+++ b/fs/cachefiles/internal.h
@@ -246,7 +246,7 @@ extern bool cachefiles_begin_operation(struct netfs_cache_resources *cres,
 				       enum fscache_want_state want_state);
 extern int __cachefiles_prepare_write(struct cachefiles_object *object,
 				      struct file *file,
-				      loff_t *_start, size_t *_len,
+				      loff_t *_start, size_t *_len, size_t upper_len,
 				      bool no_space_allocated_yet);
 extern int __cachefiles_write(struct cachefiles_object *object,
 			      struct file *file,
diff --git a/fs/cachefiles/io.c b/fs/cachefiles/io.c
index 5857241c5918..1d685357e67f 100644
--- a/fs/cachefiles/io.c
+++ b/fs/cachefiles/io.c
@@ -517,18 +517,26 @@ cachefiles_prepare_ondemand_read(struct netfs_cache_resources *cres,
  */
 int __cachefiles_prepare_write(struct cachefiles_object *object,
 			       struct file *file,
-			       loff_t *_start, size_t *_len,
+			       loff_t *_start, size_t *_len, size_t upper_len,
 			       bool no_space_allocated_yet)
 {
 	struct cachefiles_cache *cache = object->volume->cache;
 	loff_t start = *_start, pos;
-	size_t len = *_len, down;
+	size_t len = *_len;
 	int ret;
 
 	/* Round to DIO size */
-	down = start - round_down(start, PAGE_SIZE);
-	*_start = start - down;
-	*_len = round_up(down + len, PAGE_SIZE);
+	start = round_down(*_start, PAGE_SIZE);
+	if (start != *_start || *_len > upper_len) {
+		/* Probably asked to cache a streaming write written into the
+		 * pagecache when the cookie was temporarily out of service to
+		 * culling.
+		 */
+		fscache_count_dio_misfit();
+		return -ENOBUFS;
+	}
+
+	*_len = round_up(len, PAGE_SIZE);
 
 	/* We need to work out whether there's sufficient disk space to perform
 	 * the write - but we can skip that check if we have space already
@@ -539,7 +547,7 @@ int __cachefiles_prepare_write(struct cachefiles_object *object,
 
 	pos = cachefiles_inject_read_error();
 	if (pos == 0)
-		pos = vfs_llseek(file, *_start, SEEK_DATA);
+		pos = vfs_llseek(file, start, SEEK_DATA);
 	if (pos < 0 && pos >= (loff_t)-MAX_ERRNO) {
 		if (pos == -ENXIO)
 			goto check_space; /* Unallocated tail */
@@ -547,7 +555,7 @@ int __cachefiles_prepare_write(struct cachefiles_object *object,
 					  cachefiles_trace_seek_error);
 		return pos;
 	}
-	if ((u64)pos >= (u64)*_start + *_len)
+	if ((u64)pos >= (u64)start + *_len)
 		goto check_space; /* Unallocated region */
 
 	/* We have a block that's at least partially filled - if we're low on
@@ -560,13 +568,13 @@ int __cachefiles_prepare_write(struct cachefiles_object *object,
 
 	pos = cachefiles_inject_read_error();
 	if (pos == 0)
-		pos = vfs_llseek(file, *_start, SEEK_HOLE);
+		pos = vfs_llseek(file, start, SEEK_HOLE);
 	if (pos < 0 && pos >= (loff_t)-MAX_ERRNO) {
 		trace_cachefiles_io_error(object, file_inode(file), pos,
 					  cachefiles_trace_seek_error);
 		return pos;
 	}
-	if ((u64)pos >= (u64)*_start + *_len)
+	if ((u64)pos >= (u64)start + *_len)
 		return 0; /* Fully allocated */
 
 	/* Partially allocated, but insufficient space: cull. */
@@ -574,7 +582,7 @@ int __cachefiles_prepare_write(struct cachefiles_object *object,
 	ret = cachefiles_inject_remove_error();
 	if (ret == 0)
 		ret = vfs_fallocate(file, FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE,
-				    *_start, *_len);
+				    start, *_len);
 	if (ret < 0) {
 		trace_cachefiles_io_error(object, file_inode(file), ret,
 					  cachefiles_trace_fallocate_error);
@@ -591,8 +599,8 @@ check_space:
 }
 
 static int cachefiles_prepare_write(struct netfs_cache_resources *cres,
-				    loff_t *_start, size_t *_len, loff_t i_size,
-				    bool no_space_allocated_yet)
+				    loff_t *_start, size_t *_len, size_t upper_len,
+				    loff_t i_size, bool no_space_allocated_yet)
 {
 	struct cachefiles_object *object = cachefiles_cres_object(cres);
 	struct cachefiles_cache *cache = object->volume->cache;
@@ -608,7 +616,7 @@ static int cachefiles_prepare_write(struct netfs_cache_resources *cres,
 
 	cachefiles_begin_secure(cache, &saved_cred);
 	ret = __cachefiles_prepare_write(object, cachefiles_cres_file(cres),
-					 _start, _len,
+					 _start, _len, upper_len,
 					 no_space_allocated_yet);
 	cachefiles_end_secure(cache, saved_cred);
 	return ret;
diff --git a/fs/cachefiles/ondemand.c b/fs/cachefiles/ondemand.c
index b8fbbb1961bb..4ba42f1fa3b4 100644
--- a/fs/cachefiles/ondemand.c
+++ b/fs/cachefiles/ondemand.c
@@ -50,7 +50,7 @@ static ssize_t cachefiles_ondemand_fd_write_iter(struct kiocb *kiocb,
 		return -ENOBUFS;
 
 	cachefiles_begin_secure(cache, &saved_cred);
-	ret = __cachefiles_prepare_write(object, file, &pos, &len, true);
+	ret = __cachefiles_prepare_write(object, file, &pos, &len, len, true);
 	cachefiles_end_secure(cache, saved_cred);
 	if (ret < 0)
 		return ret;
@@ -539,6 +539,9 @@ int cachefiles_ondemand_init_object(struct cachefiles_object *object)
 	struct fscache_volume *volume = object->volume->vcookie;
 	size_t volume_key_size, cookie_key_size, data_len;
 
+	if (!object->ondemand)
+		return 0;
+
 	/*
 	 * CacheFiles will firstly check the cache file under the root cache
 	 * directory. If the coherency check failed, it will fallback to
diff --git a/fs/ceph/Kconfig b/fs/ceph/Kconfig
index 94df854147d3..7249d70e1a43 100644
--- a/fs/ceph/Kconfig
+++ b/fs/ceph/Kconfig
@@ -7,6 +7,7 @@ config CEPH_FS
 	select CRYPTO_AES
 	select CRYPTO
 	select NETFS_SUPPORT
+	select FS_ENCRYPTION_ALGS if FS_ENCRYPTION
 	default n
 	help
 	  Choose Y or M here to include support for mounting the
diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c
index 13af429ab030..1340d77124ae 100644
--- a/fs/ceph/addr.c
+++ b/fs/ceph/addr.c
@@ -159,27 +159,7 @@ static void ceph_invalidate_folio(struct folio *folio, size_t offset,
 		ceph_put_snap_context(snapc);
 	}
 
-	folio_wait_fscache(folio);
-}
-
-static bool ceph_release_folio(struct folio *folio, gfp_t gfp)
-{
-	struct inode *inode = folio->mapping->host;
-	struct ceph_client *cl = ceph_inode_to_client(inode);
-
-	doutc(cl, "%llx.%llx idx %lu (%sdirty)\n", ceph_vinop(inode),
-	      folio->index, folio_test_dirty(folio) ? "" : "not ");
-
-	if (folio_test_private(folio))
-		return false;
-
-	if (folio_test_fscache(folio)) {
-		if (current_is_kswapd() || !(gfp & __GFP_FS))
-			return false;
-		folio_wait_fscache(folio);
-	}
-	ceph_fscache_note_page_release(inode);
-	return true;
+	netfs_invalidate_folio(folio, offset, length);
 }
 
 static void ceph_netfs_expand_readahead(struct netfs_io_request *rreq)
@@ -357,6 +337,7 @@ static void ceph_netfs_issue_read(struct netfs_io_subrequest *subreq)
 	u64 len = subreq->len;
 	bool sparse = IS_ENCRYPTED(inode) || ceph_test_mount_opt(fsc, SPARSEREAD);
 	u64 off = subreq->start;
+	int extent_cnt;
 
 	if (ceph_inode_is_shutdown(inode)) {
 		err = -EIO;
@@ -370,8 +351,8 @@ static void ceph_netfs_issue_read(struct netfs_io_subrequest *subreq)
 
 	req = ceph_osdc_new_request(&fsc->client->osdc, &ci->i_layout, vino,
 			off, &len, 0, 1, sparse ? CEPH_OSD_OP_SPARSE_READ : CEPH_OSD_OP_READ,
-			CEPH_OSD_FLAG_READ | fsc->client->osdc.client->options->read_from_replica,
-			NULL, ci->i_truncate_seq, ci->i_truncate_size, false);
+			CEPH_OSD_FLAG_READ, NULL, ci->i_truncate_seq,
+			ci->i_truncate_size, false);
 	if (IS_ERR(req)) {
 		err = PTR_ERR(req);
 		req = NULL;
@@ -379,7 +360,8 @@ static void ceph_netfs_issue_read(struct netfs_io_subrequest *subreq)
 	}
 
 	if (sparse) {
-		err = ceph_alloc_sparse_ext_map(&req->r_ops[0]);
+		extent_cnt = __ceph_sparse_read_ext_count(inode, len);
+		err = ceph_alloc_sparse_ext_map(&req->r_ops[0], extent_cnt);
 		if (err)
 			goto out;
 	}
@@ -509,7 +491,6 @@ static void ceph_netfs_free_request(struct netfs_io_request *rreq)
 const struct netfs_request_ops ceph_netfs_ops = {
 	.init_request		= ceph_init_request,
 	.free_request		= ceph_netfs_free_request,
-	.begin_cache_operation	= ceph_begin_cache_operation,
 	.issue_read		= ceph_netfs_issue_read,
 	.expand_readahead	= ceph_netfs_expand_readahead,
 	.clamp_length		= ceph_netfs_clamp_length,
@@ -1586,7 +1567,7 @@ const struct address_space_operations ceph_aops = {
 	.write_end = ceph_write_end,
 	.dirty_folio = ceph_dirty_folio,
 	.invalidate_folio = ceph_invalidate_folio,
-	.release_folio = ceph_release_folio,
+	.release_folio = netfs_release_folio,
 	.direct_IO = noop_direct_IO,
 };
 
diff --git a/fs/ceph/cache.h b/fs/ceph/cache.h
index dc502daac49a..20efac020394 100644
--- a/fs/ceph/cache.h
+++ b/fs/ceph/cache.h
@@ -43,38 +43,19 @@ static inline void ceph_fscache_resize(struct inode *inode, loff_t to)
 	}
 }
 
-static inline void ceph_fscache_unpin_writeback(struct inode *inode,
+static inline int ceph_fscache_unpin_writeback(struct inode *inode,
 						struct writeback_control *wbc)
 {
-	fscache_unpin_writeback(wbc, ceph_fscache_cookie(ceph_inode(inode)));
+	return netfs_unpin_writeback(inode, wbc);
 }
 
-static inline int ceph_fscache_dirty_folio(struct address_space *mapping,
-		struct folio *folio)
-{
-	struct ceph_inode_info *ci = ceph_inode(mapping->host);
-
-	return fscache_dirty_folio(mapping, folio, ceph_fscache_cookie(ci));
-}
-
-static inline int ceph_begin_cache_operation(struct netfs_io_request *rreq)
-{
-	struct fscache_cookie *cookie = ceph_fscache_cookie(ceph_inode(rreq->inode));
-
-	return fscache_begin_read_operation(&rreq->cache_resources, cookie);
-}
+#define ceph_fscache_dirty_folio netfs_dirty_folio
 
 static inline bool ceph_is_cache_enabled(struct inode *inode)
 {
 	return fscache_cookie_enabled(ceph_fscache_cookie(ceph_inode(inode)));
 }
 
-static inline void ceph_fscache_note_page_release(struct inode *inode)
-{
-	struct ceph_inode_info *ci = ceph_inode(inode);
-
-	fscache_note_page_release(ceph_fscache_cookie(ci));
-}
 #else /* CONFIG_CEPH_FSCACHE */
 static inline int ceph_fscache_register_fs(struct ceph_fs_client* fsc,
 					   struct fs_context *fc)
@@ -119,30 +100,18 @@ static inline void ceph_fscache_resize(struct inode *inode, loff_t to)
 {
 }
 
-static inline void ceph_fscache_unpin_writeback(struct inode *inode,
-						struct writeback_control *wbc)
+static inline int ceph_fscache_unpin_writeback(struct inode *inode,
+					       struct writeback_control *wbc)
 {
+	return 0;
 }
 
-static inline int ceph_fscache_dirty_folio(struct address_space *mapping,
-		struct folio *folio)
-{
-	return filemap_dirty_folio(mapping, folio);
-}
+#define ceph_fscache_dirty_folio filemap_dirty_folio
 
 static inline bool ceph_is_cache_enabled(struct inode *inode)
 {
 	return false;
 }
-
-static inline int ceph_begin_cache_operation(struct netfs_io_request *rreq)
-{
-	return -ENOBUFS;
-}
-
-static inline void ceph_fscache_note_page_release(struct inode *inode)
-{
-}
 #endif /* CONFIG_CEPH_FSCACHE */
 
 #endif
diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c
index 2c0b8dc3dd0d..9c02f328c966 100644
--- a/fs/ceph/caps.c
+++ b/fs/ceph/caps.c
@@ -4887,13 +4887,15 @@ int ceph_encode_dentry_release(void **p, struct dentry *dentry,
 			       struct inode *dir,
 			       int mds, int drop, int unless)
 {
-	struct dentry *parent = NULL;
 	struct ceph_mds_request_release *rel = *p;
 	struct ceph_dentry_info *di = ceph_dentry(dentry);
 	struct ceph_client *cl;
 	int force = 0;
 	int ret;
 
+	/* This shouldn't happen */
+	BUG_ON(!dir);
+
 	/*
 	 * force an record for the directory caps if we have a dentry lease.
 	 * this is racy (can't take i_ceph_lock and d_lock together), but it
@@ -4903,14 +4905,9 @@ int ceph_encode_dentry_release(void **p, struct dentry *dentry,
 	spin_lock(&dentry->d_lock);
 	if (di->lease_session && di->lease_session->s_mds == mds)
 		force = 1;
-	if (!dir) {
-		parent = dget(dentry->d_parent);
-		dir = d_inode(parent);
-	}
 	spin_unlock(&dentry->d_lock);
 
 	ret = ceph_encode_inode_release(p, dir, mds, drop, unless, force);
-	dput(parent);
 
 	cl = ceph_inode_to_client(dir);
 	spin_lock(&dentry->d_lock);
diff --git a/fs/ceph/dir.c b/fs/ceph/dir.c
index 678596684596..0e9f56eaba1e 100644
--- a/fs/ceph/dir.c
+++ b/fs/ceph/dir.c
@@ -1593,10 +1593,12 @@ struct ceph_lease_walk_control {
 	unsigned long dir_lease_ttl;
 };
 
+static int __dir_lease_check(const struct dentry *, struct ceph_lease_walk_control *);
+static int __dentry_lease_check(const struct dentry *);
+
 static unsigned long
 __dentry_leases_walk(struct ceph_mds_client *mdsc,
-		     struct ceph_lease_walk_control *lwc,
-		     int (*check)(struct dentry*, void*))
+		     struct ceph_lease_walk_control *lwc)
 {
 	struct ceph_dentry_info *di, *tmp;
 	struct dentry *dentry, *last = NULL;
@@ -1624,7 +1626,10 @@ __dentry_leases_walk(struct ceph_mds_client *mdsc,
 			goto next;
 		}
 
-		ret = check(dentry, lwc);
+		if (lwc->dir_lease)
+			ret = __dir_lease_check(dentry, lwc);
+		else
+			ret = __dentry_lease_check(dentry);
 		if (ret & TOUCH) {
 			/* move it into tail of dir lease list */
 			__dentry_dir_lease_touch(mdsc, di);
@@ -1681,7 +1686,7 @@ next:
 	return freed;
 }
 
-static int __dentry_lease_check(struct dentry *dentry, void *arg)
+static int __dentry_lease_check(const struct dentry *dentry)
 {
 	struct ceph_dentry_info *di = ceph_dentry(dentry);
 	int ret;
@@ -1696,9 +1701,9 @@ static int __dentry_lease_check(struct dentry *dentry, void *arg)
 	return DELETE;
 }
 
-static int __dir_lease_check(struct dentry *dentry, void *arg)
+static int __dir_lease_check(const struct dentry *dentry,
+			     struct ceph_lease_walk_control *lwc)
 {
-	struct ceph_lease_walk_control *lwc = arg;
 	struct ceph_dentry_info *di = ceph_dentry(dentry);
 
 	int ret = __dir_lease_try_check(dentry);
@@ -1737,7 +1742,7 @@ int ceph_trim_dentries(struct ceph_mds_client *mdsc)
 
 	lwc.dir_lease = false;
 	lwc.nr_to_scan  = CEPH_CAPS_PER_RELEASE * 2;
-	freed = __dentry_leases_walk(mdsc, &lwc, __dentry_lease_check);
+	freed = __dentry_leases_walk(mdsc, &lwc);
 	if (!lwc.nr_to_scan) /* more invalid leases */
 		return -EAGAIN;
 
@@ -1747,7 +1752,7 @@ int ceph_trim_dentries(struct ceph_mds_client *mdsc)
 	lwc.dir_lease = true;
 	lwc.expire_dir_lease = freed < count;
 	lwc.dir_lease_ttl = mdsc->fsc->mount_options->caps_wanted_delay_max * HZ;
-	freed +=__dentry_leases_walk(mdsc, &lwc, __dir_lease_check);
+	freed +=__dentry_leases_walk(mdsc, &lwc);
 	if (!lwc.nr_to_scan) /* more to check */
 		return -EAGAIN;
 
diff --git a/fs/ceph/export.c b/fs/ceph/export.c
index 726af69d4d62..a79f163ae4ed 100644
--- a/fs/ceph/export.c
+++ b/fs/ceph/export.c
@@ -286,8 +286,6 @@ static struct dentry *__snapfh_to_dentry(struct super_block *sb,
 		doutc(cl, "%llx.%llx parent %llx hash %x err=%d", vino.ino,
 		      vino.snap, sfh->parent_ino, sfh->hash, err);
 	}
-	if (IS_ERR(inode))
-		return ERR_CAST(inode);
 	/* see comments in ceph_get_parent() */
 	return unlinked ? d_obtain_root(inode) : d_obtain_alias(inode);
 }
diff --git a/fs/ceph/file.c b/fs/ceph/file.c
index d380d9dad0e0..abe8028d95bf 100644
--- a/fs/ceph/file.c
+++ b/fs/ceph/file.c
@@ -1029,6 +1029,7 @@ ssize_t __ceph_sync_read(struct inode *inode, loff_t *ki_pos,
 		struct ceph_osd_req_op *op;
 		u64 read_off = off;
 		u64 read_len = len;
+		int extent_cnt;
 
 		/* determine new offset/length if encrypted */
 		ceph_fscrypt_adjust_off_and_len(inode, &read_off, &read_len);
@@ -1068,7 +1069,8 @@ ssize_t __ceph_sync_read(struct inode *inode, loff_t *ki_pos,
 
 		op = &req->r_ops[0];
 		if (sparse) {
-			ret = ceph_alloc_sparse_ext_map(op);
+			extent_cnt = __ceph_sparse_read_ext_count(inode, read_len);
+			ret = ceph_alloc_sparse_ext_map(op, extent_cnt);
 			if (ret) {
 				ceph_osdc_put_request(req);
 				break;
@@ -1465,6 +1467,7 @@ ceph_direct_read_write(struct kiocb *iocb, struct iov_iter *iter,
 		ssize_t len;
 		struct ceph_osd_req_op *op;
 		int readop = sparse ? CEPH_OSD_OP_SPARSE_READ : CEPH_OSD_OP_READ;
+		int extent_cnt;
 
 		if (write)
 			size = min_t(u64, size, fsc->mount_options->wsize);
@@ -1528,7 +1531,8 @@ ceph_direct_read_write(struct kiocb *iocb, struct iov_iter *iter,
 		osd_req_op_extent_osd_data_bvecs(req, 0, bvecs, num_pages, len);
 		op = &req->r_ops[0];
 		if (sparse) {
-			ret = ceph_alloc_sparse_ext_map(op);
+			extent_cnt = __ceph_sparse_read_ext_count(inode, size);
+			ret = ceph_alloc_sparse_ext_map(op, extent_cnt);
 			if (ret) {
 				ceph_osdc_put_request(req);
 				break;
diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c
index 0679240f06db..0c25d326afc4 100644
--- a/fs/ceph/inode.c
+++ b/fs/ceph/inode.c
@@ -574,7 +574,7 @@ struct inode *ceph_alloc_inode(struct super_block *sb)
 	doutc(fsc->client, "%p\n", &ci->netfs.inode);
 
 	/* Set parameters for the netfs library */
-	netfs_inode_init(&ci->netfs, &ceph_netfs_ops);
+	netfs_inode_init(&ci->netfs, &ceph_netfs_ops, false);
 
 	spin_lock_init(&ci->i_ceph_lock);
 
@@ -694,7 +694,7 @@ void ceph_evict_inode(struct inode *inode)
 	percpu_counter_dec(&mdsc->metric.total_inodes);
 
 	truncate_inode_pages_final(&inode->i_data);
-	if (inode->i_state & I_PINNING_FSCACHE_WB)
+	if (inode->i_state & I_PINNING_NETFS_WB)
 		ceph_fscache_unuse_cookie(inode, true);
 	clear_inode(inode);
 
diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c
index 02ebfabfc8ee..548d1de379f3 100644
--- a/fs/ceph/mds_client.c
+++ b/fs/ceph/mds_client.c
@@ -1534,7 +1534,8 @@ static int encode_metric_spec(void **p, void *end)
  * session message, specialization for CEPH_SESSION_REQUEST_OPEN
  * to include additional client metadata fields.
  */
-static struct ceph_msg *create_session_open_msg(struct ceph_mds_client *mdsc, u64 seq)
+static struct ceph_msg *
+create_session_full_msg(struct ceph_mds_client *mdsc, int op, u64 seq)
 {
 	struct ceph_msg *msg;
 	struct ceph_mds_session_head *h;
@@ -1578,6 +1579,9 @@ static struct ceph_msg *create_session_open_msg(struct ceph_mds_client *mdsc, u6
 		size = METRIC_BYTES(count);
 	extra_bytes += 2 + 4 + 4 + size;
 
+	/* flags, mds auth caps and oldest_client_tid */
+	extra_bytes += 4 + 4 + 8;
+
 	/* Allocate the message */
 	msg = ceph_msg_new(CEPH_MSG_CLIENT_SESSION, sizeof(*h) + extra_bytes,
 			   GFP_NOFS, false);
@@ -1589,16 +1593,16 @@ static struct ceph_msg *create_session_open_msg(struct ceph_mds_client *mdsc, u6
 	end = p + msg->front.iov_len;
 
 	h = p;
-	h->op = cpu_to_le32(CEPH_SESSION_REQUEST_OPEN);
+	h->op = cpu_to_le32(op);
 	h->seq = cpu_to_le64(seq);
 
 	/*
 	 * Serialize client metadata into waiting buffer space, using
 	 * the format that userspace expects for map<string, string>
 	 *
-	 * ClientSession messages with metadata are v4
+	 * ClientSession messages with metadata are v7
 	 */
-	msg->hdr.version = cpu_to_le16(4);
+	msg->hdr.version = cpu_to_le16(7);
 	msg->hdr.compat_version = cpu_to_le16(1);
 
 	/* The write pointer, following the session_head structure */
@@ -1634,6 +1638,15 @@ static struct ceph_msg *create_session_open_msg(struct ceph_mds_client *mdsc, u6
 		return ERR_PTR(ret);
 	}
 
+	/* version == 5, flags */
+	ceph_encode_32(&p, 0);
+
+	/* version == 6, mds auth caps */
+	ceph_encode_32(&p, 0);
+
+	/* version == 7, oldest_client_tid */
+	ceph_encode_64(&p, mdsc->oldest_tid);
+
 	msg->front.iov_len = p - msg->front.iov_base;
 	msg->hdr.front_len = cpu_to_le32(msg->front.iov_len);
 
@@ -1663,7 +1676,8 @@ static int __open_session(struct ceph_mds_client *mdsc,
 	session->s_renew_requested = jiffies;
 
 	/* send connect message */
-	msg = create_session_open_msg(mdsc, session->s_seq);
+	msg = create_session_full_msg(mdsc, CEPH_SESSION_REQUEST_OPEN,
+				      session->s_seq);
 	if (IS_ERR(msg))
 		return PTR_ERR(msg);
 	ceph_con_send(&session->s_con, msg);
@@ -2028,10 +2042,10 @@ static int send_renew_caps(struct ceph_mds_client *mdsc,
 
 	doutc(cl, "to mds%d (%s)\n", session->s_mds,
 	      ceph_mds_state_name(state));
-	msg = ceph_create_session_msg(CEPH_SESSION_REQUEST_RENEWCAPS,
+	msg = create_session_full_msg(mdsc, CEPH_SESSION_REQUEST_RENEWCAPS,
 				      ++session->s_renew_seq);
-	if (!msg)
-		return -ENOMEM;
+	if (IS_ERR(msg))
+		return PTR_ERR(msg);
 	ceph_con_send(&session->s_con, msg);
 	return 0;
 }
@@ -4128,12 +4142,12 @@ static void handle_session(struct ceph_mds_session *session,
 			pr_info_client(cl, "mds%d reconnect success\n",
 				       session->s_mds);
 
+		session->s_features = features;
 		if (session->s_state == CEPH_MDS_SESSION_OPEN) {
 			pr_notice_client(cl, "mds%d is already opened\n",
 					 session->s_mds);
 		} else {
 			session->s_state = CEPH_MDS_SESSION_OPEN;
-			session->s_features = features;
 			renewed_caps(mdsc, session, 0);
 			if (test_bit(CEPHFS_FEATURE_METRIC_COLLECT,
 				     &session->s_features))
@@ -5870,7 +5884,8 @@ static void mds_peer_reset(struct ceph_connection *con)
 
 	pr_warn_client(mdsc->fsc->client, "mds%d closed our session\n",
 		       s->s_mds);
-	if (READ_ONCE(mdsc->fsc->mount_state) != CEPH_MOUNT_FENCE_IO)
+	if (READ_ONCE(mdsc->fsc->mount_state) != CEPH_MOUNT_FENCE_IO &&
+	    ceph_mdsmap_get_state(mdsc->mdsmap, s->s_mds) >= CEPH_MDS_STATE_RECONNECT)
 		send_mds_reconnect(mdsc, s);
 }
 
diff --git a/fs/ceph/quota.c b/fs/ceph/quota.c
index 9d36c3532de1..06ee397e0c3a 100644
--- a/fs/ceph/quota.c
+++ b/fs/ceph/quota.c
@@ -197,10 +197,10 @@ void ceph_cleanup_quotarealms_inodes(struct ceph_mds_client *mdsc)
 }
 
 /*
- * This function walks through the snaprealm for an inode and returns the
- * ceph_snap_realm for the first snaprealm that has quotas set (max_files,
+ * This function walks through the snaprealm for an inode and set the
+ * realmp with the first snaprealm that has quotas set (max_files,
  * max_bytes, or any, depending on the 'which_quota' argument).  If the root is
- * reached, return the root ceph_snap_realm instead.
+ * reached, set the realmp with the root ceph_snap_realm instead.
  *
  * Note that the caller is responsible for calling ceph_put_snap_realm() on the
  * returned realm.
@@ -211,10 +211,9 @@ void ceph_cleanup_quotarealms_inodes(struct ceph_mds_client *mdsc)
  * this function will return -EAGAIN; otherwise, the snaprealms walk-through
  * will be restarted.
  */
-static struct ceph_snap_realm *get_quota_realm(struct ceph_mds_client *mdsc,
-					       struct inode *inode,
-					       enum quota_get_realm which_quota,
-					       bool retry)
+static int get_quota_realm(struct ceph_mds_client *mdsc, struct inode *inode,
+			   enum quota_get_realm which_quota,
+			   struct ceph_snap_realm **realmp, bool retry)
 {
 	struct ceph_client *cl = mdsc->fsc->client;
 	struct ceph_inode_info *ci = NULL;
@@ -222,8 +221,10 @@ static struct ceph_snap_realm *get_quota_realm(struct ceph_mds_client *mdsc,
 	struct inode *in;
 	bool has_quota;
 
+	if (realmp)
+		*realmp = NULL;
 	if (ceph_snap(inode) != CEPH_NOSNAP)
-		return NULL;
+		return 0;
 
 restart:
 	realm = ceph_inode(inode)->i_snap_realm;
@@ -250,7 +251,7 @@ restart:
 				break;
 			ceph_put_snap_realm(mdsc, realm);
 			if (!retry)
-				return ERR_PTR(-EAGAIN);
+				return -EAGAIN;
 			goto restart;
 		}
 
@@ -259,8 +260,11 @@ restart:
 		iput(in);
 
 		next = realm->parent;
-		if (has_quota || !next)
-		       return realm;
+		if (has_quota || !next) {
+			if (realmp)
+				*realmp = realm;
+			return 0;
+		}
 
 		ceph_get_snap_realm(mdsc, next);
 		ceph_put_snap_realm(mdsc, realm);
@@ -269,7 +273,7 @@ restart:
 	if (realm)
 		ceph_put_snap_realm(mdsc, realm);
 
-	return NULL;
+	return 0;
 }
 
 bool ceph_quota_is_same_realm(struct inode *old, struct inode *new)
@@ -277,6 +281,7 @@ bool ceph_quota_is_same_realm(struct inode *old, struct inode *new)
 	struct ceph_mds_client *mdsc = ceph_sb_to_mdsc(old->i_sb);
 	struct ceph_snap_realm *old_realm, *new_realm;
 	bool is_same;
+	int ret;
 
 restart:
 	/*
@@ -286,9 +291,9 @@ restart:
 	 * dropped and we can then restart the whole operation.
 	 */
 	down_read(&mdsc->snap_rwsem);
-	old_realm = get_quota_realm(mdsc, old, QUOTA_GET_ANY, true);
-	new_realm = get_quota_realm(mdsc, new, QUOTA_GET_ANY, false);
-	if (PTR_ERR(new_realm) == -EAGAIN) {
+	get_quota_realm(mdsc, old, QUOTA_GET_ANY, &old_realm, true);
+	ret = get_quota_realm(mdsc, new, QUOTA_GET_ANY, &new_realm, false);
+	if (ret == -EAGAIN) {
 		up_read(&mdsc->snap_rwsem);
 		if (old_realm)
 			ceph_put_snap_realm(mdsc, old_realm);
@@ -492,8 +497,8 @@ bool ceph_quota_update_statfs(struct ceph_fs_client *fsc, struct kstatfs *buf)
 	bool is_updated = false;
 
 	down_read(&mdsc->snap_rwsem);
-	realm = get_quota_realm(mdsc, d_inode(fsc->sb->s_root),
-				QUOTA_GET_MAX_BYTES, true);
+	get_quota_realm(mdsc, d_inode(fsc->sb->s_root), QUOTA_GET_MAX_BYTES,
+			&realm, true);
 	up_read(&mdsc->snap_rwsem);
 	if (!realm)
 		return false;
diff --git a/fs/ceph/super.h b/fs/ceph/super.h
index fe0f64a0acb2..b06e2bc86221 100644
--- a/fs/ceph/super.h
+++ b/fs/ceph/super.h
@@ -3,6 +3,7 @@
 #define _FS_CEPH_SUPER_H
 
 #include <linux/ceph/ceph_debug.h>
+#include <linux/ceph/osd_client.h>
 
 #include <asm/unaligned.h>
 #include <linux/backing-dev.h>
@@ -1407,6 +1408,19 @@ static inline void __ceph_update_quota(struct ceph_inode_info *ci,
 		ceph_adjust_quota_realms_count(&ci->netfs.inode, has_quota);
 }
 
+static inline int __ceph_sparse_read_ext_count(struct inode *inode, u64 len)
+{
+	int cnt = 0;
+
+	if (IS_ENCRYPTED(inode)) {
+		cnt = len >> CEPH_FSCRYPT_BLOCK_SHIFT;
+		if (cnt > CEPH_SPARSE_EXT_ARRAY_INITIAL)
+			cnt = 0;
+	}
+
+	return cnt;
+}
+
 extern void ceph_handle_quota(struct ceph_mds_client *mdsc,
 			      struct ceph_mds_session *session,
 			      struct ceph_msg *msg);
diff --git a/fs/erofs/Kconfig b/fs/erofs/Kconfig
index 1d318f85232d..fffd3919343e 100644
--- a/fs/erofs/Kconfig
+++ b/fs/erofs/Kconfig
@@ -114,8 +114,11 @@ config EROFS_FS_ZIP_DEFLATE
 
 config EROFS_FS_ONDEMAND
 	bool "EROFS fscache-based on-demand read support"
-	depends on CACHEFILES_ONDEMAND && (EROFS_FS=m && FSCACHE || EROFS_FS=y && FSCACHE=y)
-	default n
+	depends on EROFS_FS
+	select NETFS_SUPPORT
+	select FSCACHE
+	select CACHEFILES
+	select CACHEFILES_ONDEMAND
 	help
 	  This permits EROFS to use fscache-backed data blobs with on-demand
 	  read support.
diff --git a/fs/erofs/decompressor.c b/fs/erofs/decompressor.c
index 1d65b9f60a39..072ef6a66823 100644
--- a/fs/erofs/decompressor.c
+++ b/fs/erofs/decompressor.c
@@ -408,7 +408,7 @@ int z_erofs_parse_cfgs(struct super_block *sb, struct erofs_super_block *dsb)
 	int size, ret = 0;
 
 	if (!erofs_sb_has_compr_cfgs(sbi)) {
-		sbi->available_compr_algs = Z_EROFS_COMPRESSION_LZ4;
+		sbi->available_compr_algs = 1 << Z_EROFS_COMPRESSION_LZ4;
 		return z_erofs_load_lz4_config(sb, dsb, NULL, 0);
 	}
 
diff --git a/fs/erofs/fscache.c b/fs/erofs/fscache.c
index 87ff35bff8d5..bc12030393b2 100644
--- a/fs/erofs/fscache.c
+++ b/fs/erofs/fscache.c
@@ -165,10 +165,10 @@ static int erofs_fscache_read_folios_async(struct fscache_cookie *cookie,
 static int erofs_fscache_meta_read_folio(struct file *data, struct folio *folio)
 {
 	int ret;
-	struct erofs_fscache *ctx = folio_mapping(folio)->host->i_private;
+	struct erofs_fscache *ctx = folio->mapping->host->i_private;
 	struct erofs_fscache_request *req;
 
-	req = erofs_fscache_req_alloc(folio_mapping(folio),
+	req = erofs_fscache_req_alloc(folio->mapping,
 				folio_pos(folio), folio_size(folio));
 	if (IS_ERR(req)) {
 		folio_unlock(folio);
@@ -276,7 +276,7 @@ static int erofs_fscache_read_folio(struct file *file, struct folio *folio)
 	struct erofs_fscache_request *req;
 	int ret;
 
-	req = erofs_fscache_req_alloc(folio_mapping(folio),
+	req = erofs_fscache_req_alloc(folio->mapping,
 			folio_pos(folio), folio_size(folio));
 	if (IS_ERR(req)) {
 		folio_unlock(folio);
diff --git a/fs/erofs/zmap.c b/fs/erofs/zmap.c
index 9753875e41cb..e313c936351d 100644
--- a/fs/erofs/zmap.c
+++ b/fs/erofs/zmap.c
@@ -454,7 +454,7 @@ static int z_erofs_do_map_blocks(struct inode *inode,
 		.map = map,
 	};
 	int err = 0;
-	unsigned int lclusterbits, endoff;
+	unsigned int lclusterbits, endoff, afmt;
 	unsigned long initial_lcn;
 	unsigned long long ofs, end;
 
@@ -543,17 +543,20 @@ static int z_erofs_do_map_blocks(struct inode *inode,
 			err = -EFSCORRUPTED;
 			goto unmap_out;
 		}
-		if (vi->z_advise & Z_EROFS_ADVISE_INTERLACED_PCLUSTER)
-			map->m_algorithmformat =
-				Z_EROFS_COMPRESSION_INTERLACED;
-		else
-			map->m_algorithmformat =
-				Z_EROFS_COMPRESSION_SHIFTED;
-	} else if (m.headtype == Z_EROFS_LCLUSTER_TYPE_HEAD2) {
-		map->m_algorithmformat = vi->z_algorithmtype[1];
+		afmt = vi->z_advise & Z_EROFS_ADVISE_INTERLACED_PCLUSTER ?
+			Z_EROFS_COMPRESSION_INTERLACED :
+			Z_EROFS_COMPRESSION_SHIFTED;
 	} else {
-		map->m_algorithmformat = vi->z_algorithmtype[0];
+		afmt = m.headtype == Z_EROFS_LCLUSTER_TYPE_HEAD2 ?
+			vi->z_algorithmtype[1] : vi->z_algorithmtype[0];
+		if (!(EROFS_I_SB(inode)->available_compr_algs & (1 << afmt))) {
+			erofs_err(inode->i_sb, "inconsistent algorithmtype %u for nid %llu",
+				  afmt, vi->nid);
+			err = -EFSCORRUPTED;
+			goto unmap_out;
+		}
 	}
+	map->m_algorithmformat = afmt;
 
 	if ((flags & EROFS_GET_BLOCKS_FIEMAP) ||
 	    ((flags & EROFS_GET_BLOCKS_READMORE) &&
diff --git a/fs/exec.c b/fs/exec.c
index 73e4045df271..af4fbb61cd53 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -128,7 +128,7 @@ SYSCALL_DEFINE1(uselib, const char __user *, library)
 	struct filename *tmp = getname(library);
 	int error = PTR_ERR(tmp);
 	static const struct open_flags uselib_flags = {
-		.open_flag = O_LARGEFILE | O_RDONLY | __FMODE_EXEC,
+		.open_flag = O_LARGEFILE | O_RDONLY,
 		.acc_mode = MAY_READ | MAY_EXEC,
 		.intent = LOOKUP_OPEN,
 		.lookup_flags = LOOKUP_FOLLOW,
@@ -904,6 +904,10 @@ EXPORT_SYMBOL(transfer_args_to_stack);
 
 #endif /* CONFIG_MMU */
 
+/*
+ * On success, caller must call do_close_execat() on the returned
+ * struct file to close it.
+ */
 static struct file *do_open_execat(int fd, struct filename *name, int flags)
 {
 	struct file *file;
@@ -948,6 +952,17 @@ exit:
 	return ERR_PTR(err);
 }
 
+/**
+ * open_exec - Open a path name for execution
+ *
+ * @name: path name to open with the intent of executing it.
+ *
+ * Returns ERR_PTR on failure or allocated struct file on success.
+ *
+ * As this is a wrapper for the internal do_open_execat(), callers
+ * must call allow_write_access() before fput() on release. Also see
+ * do_close_execat().
+ */
 struct file *open_exec(const char *name)
 {
 	struct filename *filename = getname_kernel(name);
@@ -1409,6 +1424,9 @@ int begin_new_exec(struct linux_binprm * bprm)
 
 out_unlock:
 	up_write(&me->signal->exec_update_lock);
+	if (!bprm->cred)
+		mutex_unlock(&me->signal->cred_guard_mutex);
+
 out:
 	return retval;
 }
@@ -1484,6 +1502,15 @@ static int prepare_bprm_creds(struct linux_binprm *bprm)
 	return -ENOMEM;
 }
 
+/* Matches do_open_execat() */
+static void do_close_execat(struct file *file)
+{
+	if (!file)
+		return;
+	allow_write_access(file);
+	fput(file);
+}
+
 static void free_bprm(struct linux_binprm *bprm)
 {
 	if (bprm->mm) {
@@ -1495,10 +1522,7 @@ static void free_bprm(struct linux_binprm *bprm)
 		mutex_unlock(&current->signal->cred_guard_mutex);
 		abort_creds(bprm->cred);
 	}
-	if (bprm->file) {
-		allow_write_access(bprm->file);
-		fput(bprm->file);
-	}
+	do_close_execat(bprm->file);
 	if (bprm->executable)
 		fput(bprm->executable);
 	/* If a binfmt changed the interp, free it. */
@@ -1508,12 +1532,23 @@ static void free_bprm(struct linux_binprm *bprm)
 	kfree(bprm);
 }
 
-static struct linux_binprm *alloc_bprm(int fd, struct filename *filename)
+static struct linux_binprm *alloc_bprm(int fd, struct filename *filename, int flags)
 {
-	struct linux_binprm *bprm = kzalloc(sizeof(*bprm), GFP_KERNEL);
+	struct linux_binprm *bprm;
+	struct file *file;
 	int retval = -ENOMEM;
-	if (!bprm)
-		goto out;
+
+	file = do_open_execat(fd, filename, flags);
+	if (IS_ERR(file))
+		return ERR_CAST(file);
+
+	bprm = kzalloc(sizeof(*bprm), GFP_KERNEL);
+	if (!bprm) {
+		do_close_execat(file);
+		return ERR_PTR(-ENOMEM);
+	}
+
+	bprm->file = file;
 
 	if (fd == AT_FDCWD || filename->name[0] == '/') {
 		bprm->filename = filename->name;
@@ -1526,18 +1561,28 @@ static struct linux_binprm *alloc_bprm(int fd, struct filename *filename)
 		if (!bprm->fdpath)
 			goto out_free;
 
+		/*
+		 * Record that a name derived from an O_CLOEXEC fd will be
+		 * inaccessible after exec.  This allows the code in exec to
+		 * choose to fail when the executable is not mmaped into the
+		 * interpreter and an open file descriptor is not passed to
+		 * the interpreter.  This makes for a better user experience
+		 * than having the interpreter start and then immediately fail
+		 * when it finds the executable is inaccessible.
+		 */
+		if (get_close_on_exec(fd))
+			bprm->interp_flags |= BINPRM_FLAGS_PATH_INACCESSIBLE;
+
 		bprm->filename = bprm->fdpath;
 	}
 	bprm->interp = bprm->filename;
 
 	retval = bprm_mm_init(bprm);
-	if (retval)
-		goto out_free;
-	return bprm;
+	if (!retval)
+		return bprm;
 
 out_free:
 	free_bprm(bprm);
-out:
 	return ERR_PTR(retval);
 }
 
@@ -1588,6 +1633,7 @@ static void check_unsafe_exec(struct linux_binprm *bprm)
 	}
 	rcu_read_unlock();
 
+	/* "users" and "in_exec" locked for copy_fs() */
 	if (p->fs->users > n_fs)
 		bprm->unsafe |= LSM_UNSAFE_SHARE;
 	else
@@ -1804,13 +1850,8 @@ static int exec_binprm(struct linux_binprm *bprm)
 	return 0;
 }
 
-/*
- * sys_execve() executes a new program.
- */
-static int bprm_execve(struct linux_binprm *bprm,
-		       int fd, struct filename *filename, int flags)
+static int bprm_execve(struct linux_binprm *bprm)
 {
-	struct file *file;
 	int retval;
 
 	retval = prepare_bprm_creds(bprm);
@@ -1826,26 +1867,8 @@ static int bprm_execve(struct linux_binprm *bprm,
 	current->in_execve = 1;
 	sched_mm_cid_before_execve(current);
 
-	file = do_open_execat(fd, filename, flags);
-	retval = PTR_ERR(file);
-	if (IS_ERR(file))
-		goto out_unmark;
-
 	sched_exec();
 
-	bprm->file = file;
-	/*
-	 * Record that a name derived from an O_CLOEXEC fd will be
-	 * inaccessible after exec.  This allows the code in exec to
-	 * choose to fail when the executable is not mmaped into the
-	 * interpreter and an open file descriptor is not passed to
-	 * the interpreter.  This makes for a better user experience
-	 * than having the interpreter start and then immediately fail
-	 * when it finds the executable is inaccessible.
-	 */
-	if (bprm->fdpath && get_close_on_exec(fd))
-		bprm->interp_flags |= BINPRM_FLAGS_PATH_INACCESSIBLE;
-
 	/* Set the unchanging part of bprm->cred */
 	retval = security_bprm_creds_for_exec(bprm);
 	if (retval)
@@ -1875,7 +1898,6 @@ out:
 	if (bprm->point_of_no_return && !fatal_signal_pending(current))
 		force_fatal_sig(SIGSEGV);
 
-out_unmark:
 	sched_mm_cid_after_execve(current);
 	current->fs->in_exec = 0;
 	current->in_execve = 0;
@@ -1910,7 +1932,7 @@ static int do_execveat_common(int fd, struct filename *filename,
 	 * further execve() calls fail. */
 	current->flags &= ~PF_NPROC_EXCEEDED;
 
-	bprm = alloc_bprm(fd, filename);
+	bprm = alloc_bprm(fd, filename, flags);
 	if (IS_ERR(bprm)) {
 		retval = PTR_ERR(bprm);
 		goto out_ret;
@@ -1959,7 +1981,7 @@ static int do_execveat_common(int fd, struct filename *filename,
 		bprm->argc = 1;
 	}
 
-	retval = bprm_execve(bprm, fd, filename, flags);
+	retval = bprm_execve(bprm);
 out_free:
 	free_bprm(bprm);
 
@@ -1984,7 +2006,7 @@ int kernel_execve(const char *kernel_filename,
 	if (IS_ERR(filename))
 		return PTR_ERR(filename);
 
-	bprm = alloc_bprm(fd, filename);
+	bprm = alloc_bprm(fd, filename, 0);
 	if (IS_ERR(bprm)) {
 		retval = PTR_ERR(bprm);
 		goto out_ret;
@@ -2019,7 +2041,7 @@ int kernel_execve(const char *kernel_filename,
 	if (retval < 0)
 		goto out_free;
 
-	retval = bprm_execve(bprm, fd, filename, 0);
+	retval = bprm_execve(bprm);
 out_free:
 	free_bprm(bprm);
 out_ret:
diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c
index 1767493dffda..3d84fcc471c6 100644
--- a/fs/fs-writeback.c
+++ b/fs/fs-writeback.c
@@ -1675,11 +1675,11 @@ __writeback_single_inode(struct inode *inode, struct writeback_control *wbc)
 
 	if (mapping_tagged(mapping, PAGECACHE_TAG_DIRTY))
 		inode->i_state |= I_DIRTY_PAGES;
-	else if (unlikely(inode->i_state & I_PINNING_FSCACHE_WB)) {
+	else if (unlikely(inode->i_state & I_PINNING_NETFS_WB)) {
 		if (!(inode->i_state & I_DIRTY_PAGES)) {
-			inode->i_state &= ~I_PINNING_FSCACHE_WB;
-			wbc->unpinned_fscache_wb = true;
-			dirty |= I_PINNING_FSCACHE_WB; /* Cause write_inode */
+			inode->i_state &= ~I_PINNING_NETFS_WB;
+			wbc->unpinned_netfs_wb = true;
+			dirty |= I_PINNING_NETFS_WB; /* Cause write_inode */
 		}
 	}
 
@@ -1691,7 +1691,7 @@ __writeback_single_inode(struct inode *inode, struct writeback_control *wbc)
 		if (ret == 0)
 			ret = err;
 	}
-	wbc->unpinned_fscache_wb = false;
+	wbc->unpinned_netfs_wb = false;
 	trace_writeback_single_inode(inode, wbc, nr_to_write);
 	return ret;
 }
diff --git a/fs/fscache/Kconfig b/fs/fscache/Kconfig
deleted file mode 100644
index b313a978ae0a..000000000000
--- a/fs/fscache/Kconfig
+++ /dev/null
@@ -1,40 +0,0 @@
-# SPDX-License-Identifier: GPL-2.0-only
-
-config FSCACHE
-	tristate "General filesystem local caching manager"
-	select NETFS_SUPPORT
-	help
-	  This option enables a generic filesystem caching manager that can be
-	  used by various network and other filesystems to cache data locally.
-	  Different sorts of caches can be plugged in, depending on the
-	  resources available.
-
-	  See Documentation/filesystems/caching/fscache.rst for more information.
-
-config FSCACHE_STATS
-	bool "Gather statistical information on local caching"
-	depends on FSCACHE && PROC_FS
-	select NETFS_STATS
-	help
-	  This option causes statistical information to be gathered on local
-	  caching and exported through file:
-
-		/proc/fs/fscache/stats
-
-	  The gathering of statistics adds a certain amount of overhead to
-	  execution as there are a quite a few stats gathered, and on a
-	  multi-CPU system these may be on cachelines that keep bouncing
-	  between CPUs.  On the other hand, the stats are very useful for
-	  debugging purposes.  Saying 'Y' here is recommended.
-
-	  See Documentation/filesystems/caching/fscache.rst for more information.
-
-config FSCACHE_DEBUG
-	bool "Debug FS-Cache"
-	depends on FSCACHE
-	help
-	  This permits debugging to be dynamically enabled in the local caching
-	  management module.  If this is set, the debugging output may be
-	  enabled by setting bits in /sys/modules/fscache/parameter/debug.
-
-	  See Documentation/filesystems/caching/fscache.rst for more information.
diff --git a/fs/fscache/Makefile b/fs/fscache/Makefile
deleted file mode 100644
index afb090ea16c4..000000000000
--- a/fs/fscache/Makefile
+++ /dev/null
@@ -1,16 +0,0 @@
-# SPDX-License-Identifier: GPL-2.0
-#
-# Makefile for general filesystem caching code
-#
-
-fscache-y := \
-	cache.o \
-	cookie.o \
-	io.o \
-	main.o \
-	volume.o
-
-fscache-$(CONFIG_PROC_FS) += proc.o
-fscache-$(CONFIG_FSCACHE_STATS) += stats.o
-
-obj-$(CONFIG_FSCACHE) := fscache.o
diff --git a/fs/fscache/internal.h b/fs/fscache/internal.h
deleted file mode 100644
index 1336f517e9b1..000000000000
--- a/fs/fscache/internal.h
+++ /dev/null
@@ -1,277 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0-or-later */
-/* Internal definitions for FS-Cache
- *
- * Copyright (C) 2021 Red Hat, Inc. All Rights Reserved.
- * Written by David Howells (dhowells@redhat.com)
- */
-
-#ifdef pr_fmt
-#undef pr_fmt
-#endif
-
-#define pr_fmt(fmt) "FS-Cache: " fmt
-
-#include <linux/slab.h>
-#include <linux/fscache-cache.h>
-#include <trace/events/fscache.h>
-#include <linux/sched.h>
-#include <linux/seq_file.h>
-
-/*
- * cache.c
- */
-#ifdef CONFIG_PROC_FS
-extern const struct seq_operations fscache_caches_seq_ops;
-#endif
-bool fscache_begin_cache_access(struct fscache_cache *cache, enum fscache_access_trace why);
-void fscache_end_cache_access(struct fscache_cache *cache, enum fscache_access_trace why);
-struct fscache_cache *fscache_lookup_cache(const char *name, bool is_cache);
-void fscache_put_cache(struct fscache_cache *cache, enum fscache_cache_trace where);
-
-static inline enum fscache_cache_state fscache_cache_state(const struct fscache_cache *cache)
-{
-	return smp_load_acquire(&cache->state);
-}
-
-static inline bool fscache_cache_is_live(const struct fscache_cache *cache)
-{
-	return fscache_cache_state(cache) == FSCACHE_CACHE_IS_ACTIVE;
-}
-
-static inline void fscache_set_cache_state(struct fscache_cache *cache,
-					   enum fscache_cache_state new_state)
-{
-	smp_store_release(&cache->state, new_state);
-
-}
-
-static inline bool fscache_set_cache_state_maybe(struct fscache_cache *cache,
-						 enum fscache_cache_state old_state,
-						 enum fscache_cache_state new_state)
-{
-	return try_cmpxchg_release(&cache->state, &old_state, new_state);
-}
-
-/*
- * cookie.c
- */
-extern struct kmem_cache *fscache_cookie_jar;
-#ifdef CONFIG_PROC_FS
-extern const struct seq_operations fscache_cookies_seq_ops;
-#endif
-extern struct timer_list fscache_cookie_lru_timer;
-
-extern void fscache_print_cookie(struct fscache_cookie *cookie, char prefix);
-extern bool fscache_begin_cookie_access(struct fscache_cookie *cookie,
-					enum fscache_access_trace why);
-
-static inline void fscache_see_cookie(struct fscache_cookie *cookie,
-				      enum fscache_cookie_trace where)
-{
-	trace_fscache_cookie(cookie->debug_id, refcount_read(&cookie->ref),
-			     where);
-}
-
-/*
- * main.c
- */
-extern unsigned fscache_debug;
-
-extern unsigned int fscache_hash(unsigned int salt, const void *data, size_t len);
-
-/*
- * proc.c
- */
-#ifdef CONFIG_PROC_FS
-extern int __init fscache_proc_init(void);
-extern void fscache_proc_cleanup(void);
-#else
-#define fscache_proc_init()	(0)
-#define fscache_proc_cleanup()	do {} while (0)
-#endif
-
-/*
- * stats.c
- */
-#ifdef CONFIG_FSCACHE_STATS
-extern atomic_t fscache_n_volumes;
-extern atomic_t fscache_n_volumes_collision;
-extern atomic_t fscache_n_volumes_nomem;
-extern atomic_t fscache_n_cookies;
-extern atomic_t fscache_n_cookies_lru;
-extern atomic_t fscache_n_cookies_lru_expired;
-extern atomic_t fscache_n_cookies_lru_removed;
-extern atomic_t fscache_n_cookies_lru_dropped;
-
-extern atomic_t fscache_n_acquires;
-extern atomic_t fscache_n_acquires_ok;
-extern atomic_t fscache_n_acquires_oom;
-
-extern atomic_t fscache_n_invalidates;
-
-extern atomic_t fscache_n_relinquishes;
-extern atomic_t fscache_n_relinquishes_retire;
-extern atomic_t fscache_n_relinquishes_dropped;
-
-extern atomic_t fscache_n_resizes;
-extern atomic_t fscache_n_resizes_null;
-
-static inline void fscache_stat(atomic_t *stat)
-{
-	atomic_inc(stat);
-}
-
-static inline void fscache_stat_d(atomic_t *stat)
-{
-	atomic_dec(stat);
-}
-
-#define __fscache_stat(stat) (stat)
-
-int fscache_stats_show(struct seq_file *m, void *v);
-#else
-
-#define __fscache_stat(stat) (NULL)
-#define fscache_stat(stat) do {} while (0)
-#define fscache_stat_d(stat) do {} while (0)
-#endif
-
-/*
- * volume.c
- */
-#ifdef CONFIG_PROC_FS
-extern const struct seq_operations fscache_volumes_seq_ops;
-#endif
-
-struct fscache_volume *fscache_get_volume(struct fscache_volume *volume,
-					  enum fscache_volume_trace where);
-void fscache_put_volume(struct fscache_volume *volume,
-			enum fscache_volume_trace where);
-bool fscache_begin_volume_access(struct fscache_volume *volume,
-				 struct fscache_cookie *cookie,
-				 enum fscache_access_trace why);
-void fscache_create_volume(struct fscache_volume *volume, bool wait);
-
-
-/*****************************************************************************/
-/*
- * debug tracing
- */
-#define dbgprintk(FMT, ...) \
-	printk("[%-6.6s] "FMT"\n", current->comm, ##__VA_ARGS__)
-
-#define kenter(FMT, ...) dbgprintk("==> %s("FMT")", __func__, ##__VA_ARGS__)
-#define kleave(FMT, ...) dbgprintk("<== %s()"FMT"", __func__, ##__VA_ARGS__)
-#define kdebug(FMT, ...) dbgprintk(FMT, ##__VA_ARGS__)
-
-#define kjournal(FMT, ...) no_printk(FMT, ##__VA_ARGS__)
-
-#ifdef __KDEBUG
-#define _enter(FMT, ...) kenter(FMT, ##__VA_ARGS__)
-#define _leave(FMT, ...) kleave(FMT, ##__VA_ARGS__)
-#define _debug(FMT, ...) kdebug(FMT, ##__VA_ARGS__)
-
-#elif defined(CONFIG_FSCACHE_DEBUG)
-#define _enter(FMT, ...)			\
-do {						\
-	if (__do_kdebug(ENTER))			\
-		kenter(FMT, ##__VA_ARGS__);	\
-} while (0)
-
-#define _leave(FMT, ...)			\
-do {						\
-	if (__do_kdebug(LEAVE))			\
-		kleave(FMT, ##__VA_ARGS__);	\
-} while (0)
-
-#define _debug(FMT, ...)			\
-do {						\
-	if (__do_kdebug(DEBUG))			\
-		kdebug(FMT, ##__VA_ARGS__);	\
-} while (0)
-
-#else
-#define _enter(FMT, ...) no_printk("==> %s("FMT")", __func__, ##__VA_ARGS__)
-#define _leave(FMT, ...) no_printk("<== %s()"FMT"", __func__, ##__VA_ARGS__)
-#define _debug(FMT, ...) no_printk(FMT, ##__VA_ARGS__)
-#endif
-
-/*
- * determine whether a particular optional debugging point should be logged
- * - we need to go through three steps to persuade cpp to correctly join the
- *   shorthand in FSCACHE_DEBUG_LEVEL with its prefix
- */
-#define ____do_kdebug(LEVEL, POINT) \
-	unlikely((fscache_debug & \
-		  (FSCACHE_POINT_##POINT << (FSCACHE_DEBUG_ ## LEVEL * 3))))
-#define ___do_kdebug(LEVEL, POINT) \
-	____do_kdebug(LEVEL, POINT)
-#define __do_kdebug(POINT) \
-	___do_kdebug(FSCACHE_DEBUG_LEVEL, POINT)
-
-#define FSCACHE_DEBUG_CACHE	0
-#define FSCACHE_DEBUG_COOKIE	1
-#define FSCACHE_DEBUG_OBJECT	2
-#define FSCACHE_DEBUG_OPERATION	3
-
-#define FSCACHE_POINT_ENTER	1
-#define FSCACHE_POINT_LEAVE	2
-#define FSCACHE_POINT_DEBUG	4
-
-#ifndef FSCACHE_DEBUG_LEVEL
-#define FSCACHE_DEBUG_LEVEL CACHE
-#endif
-
-/*
- * assertions
- */
-#if 1 /* defined(__KDEBUGALL) */
-
-#define ASSERT(X)							\
-do {									\
-	if (unlikely(!(X))) {						\
-		pr_err("\n");					\
-		pr_err("Assertion failed\n");	\
-		BUG();							\
-	}								\
-} while (0)
-
-#define ASSERTCMP(X, OP, Y)						\
-do {									\
-	if (unlikely(!((X) OP (Y)))) {					\
-		pr_err("\n");					\
-		pr_err("Assertion failed\n");	\
-		pr_err("%lx " #OP " %lx is false\n",		\
-		       (unsigned long)(X), (unsigned long)(Y));		\
-		BUG();							\
-	}								\
-} while (0)
-
-#define ASSERTIF(C, X)							\
-do {									\
-	if (unlikely((C) && !(X))) {					\
-		pr_err("\n");					\
-		pr_err("Assertion failed\n");	\
-		BUG();							\
-	}								\
-} while (0)
-
-#define ASSERTIFCMP(C, X, OP, Y)					\
-do {									\
-	if (unlikely((C) && !((X) OP (Y)))) {				\
-		pr_err("\n");					\
-		pr_err("Assertion failed\n");	\
-		pr_err("%lx " #OP " %lx is false\n",		\
-		       (unsigned long)(X), (unsigned long)(Y));		\
-		BUG();							\
-	}								\
-} while (0)
-
-#else
-
-#define ASSERT(X)			do {} while (0)
-#define ASSERTCMP(X, OP, Y)		do {} while (0)
-#define ASSERTIF(C, X)			do {} while (0)
-#define ASSERTIFCMP(C, X, OP, Y)	do {} while (0)
-
-#endif /* assert or not */
diff --git a/fs/netfs/Kconfig b/fs/netfs/Kconfig
index b4db21022cb4..bec805e0c44c 100644
--- a/fs/netfs/Kconfig
+++ b/fs/netfs/Kconfig
@@ -21,3 +21,42 @@ config NETFS_STATS
 	  multi-CPU system these may be on cachelines that keep bouncing
 	  between CPUs.  On the other hand, the stats are very useful for
 	  debugging purposes.  Saying 'Y' here is recommended.
+
+config FSCACHE
+	bool "General filesystem local caching manager"
+	depends on NETFS_SUPPORT
+	help
+	  This option enables a generic filesystem caching manager that can be
+	  used by various network and other filesystems to cache data locally.
+	  Different sorts of caches can be plugged in, depending on the
+	  resources available.
+
+	  See Documentation/filesystems/caching/fscache.rst for more information.
+
+config FSCACHE_STATS
+	bool "Gather statistical information on local caching"
+	depends on FSCACHE && PROC_FS
+	select NETFS_STATS
+	help
+	  This option causes statistical information to be gathered on local
+	  caching and exported through file:
+
+		/proc/fs/fscache/stats
+
+	  The gathering of statistics adds a certain amount of overhead to
+	  execution as there are a quite a few stats gathered, and on a
+	  multi-CPU system these may be on cachelines that keep bouncing
+	  between CPUs.  On the other hand, the stats are very useful for
+	  debugging purposes.  Saying 'Y' here is recommended.
+
+	  See Documentation/filesystems/caching/fscache.rst for more information.
+
+config FSCACHE_DEBUG
+	bool "Debug FS-Cache"
+	depends on FSCACHE
+	help
+	  This permits debugging to be dynamically enabled in the local caching
+	  management module.  If this is set, the debugging output may be
+	  enabled by setting bits in /sys/modules/fscache/parameter/debug.
+
+	  See Documentation/filesystems/caching/fscache.rst for more information.
diff --git a/fs/netfs/Makefile b/fs/netfs/Makefile
index 386d6fb92793..d4d1d799819e 100644
--- a/fs/netfs/Makefile
+++ b/fs/netfs/Makefile
@@ -2,11 +2,29 @@
 
 netfs-y := \
 	buffered_read.o \
+	buffered_write.o \
+	direct_read.o \
+	direct_write.o \
 	io.o \
 	iterator.o \
+	locking.o \
 	main.o \
-	objects.o
+	misc.o \
+	objects.o \
+	output.o
 
 netfs-$(CONFIG_NETFS_STATS) += stats.o
 
-obj-$(CONFIG_NETFS_SUPPORT) := netfs.o
+netfs-$(CONFIG_FSCACHE) += \
+	fscache_cache.o \
+	fscache_cookie.o \
+	fscache_io.o \
+	fscache_main.o \
+	fscache_volume.o
+
+ifeq ($(CONFIG_PROC_FS),y)
+netfs-$(CONFIG_FSCACHE) += fscache_proc.o
+endif
+netfs-$(CONFIG_FSCACHE_STATS) += fscache_stats.o
+
+obj-$(CONFIG_NETFS_SUPPORT) += netfs.o
diff --git a/fs/netfs/buffered_read.c b/fs/netfs/buffered_read.c
index 2cd3ccf4c439..3298c29b5548 100644
--- a/fs/netfs/buffered_read.c
+++ b/fs/netfs/buffered_read.c
@@ -16,6 +16,7 @@
 void netfs_rreq_unlock_folios(struct netfs_io_request *rreq)
 {
 	struct netfs_io_subrequest *subreq;
+	struct netfs_folio *finfo;
 	struct folio *folio;
 	pgoff_t start_page = rreq->start / PAGE_SIZE;
 	pgoff_t last_page = ((rreq->start + rreq->len) / PAGE_SIZE) - 1;
@@ -63,6 +64,7 @@ void netfs_rreq_unlock_folios(struct netfs_io_request *rreq)
 				break;
 			}
 			if (!folio_started && test_bit(NETFS_SREQ_COPY_TO_CACHE, &subreq->flags)) {
+				trace_netfs_folio(folio, netfs_folio_trace_copy_to_cache);
 				folio_start_fscache(folio);
 				folio_started = true;
 			}
@@ -86,11 +88,20 @@ void netfs_rreq_unlock_folios(struct netfs_io_request *rreq)
 
 		if (!pg_failed) {
 			flush_dcache_folio(folio);
+			finfo = netfs_folio_info(folio);
+			if (finfo) {
+				trace_netfs_folio(folio, netfs_folio_trace_filled_gaps);
+				if (finfo->netfs_group)
+					folio_change_private(folio, finfo->netfs_group);
+				else
+					folio_detach_private(folio);
+				kfree(finfo);
+			}
 			folio_mark_uptodate(folio);
 		}
 
 		if (!test_bit(NETFS_RREQ_DONT_UNLOCK_FOLIOS, &rreq->flags)) {
-			if (folio_index(folio) == rreq->no_unlock_folio &&
+			if (folio->index == rreq->no_unlock_folio &&
 			    test_bit(NETFS_RREQ_NO_UNLOCK_FOLIO, &rreq->flags))
 				_debug("no unlock");
 			else
@@ -147,6 +158,15 @@ static void netfs_rreq_expand(struct netfs_io_request *rreq,
 	}
 }
 
+/*
+ * Begin an operation, and fetch the stored zero point value from the cookie if
+ * available.
+ */
+static int netfs_begin_cache_read(struct netfs_io_request *rreq, struct netfs_inode *ctx)
+{
+	return fscache_begin_read_operation(&rreq->cache_resources, netfs_i_cookie(ctx));
+}
+
 /**
  * netfs_readahead - Helper to manage a read request
  * @ractl: The description of the readahead request
@@ -180,11 +200,9 @@ void netfs_readahead(struct readahead_control *ractl)
 	if (IS_ERR(rreq))
 		return;
 
-	if (ctx->ops->begin_cache_operation) {
-		ret = ctx->ops->begin_cache_operation(rreq);
-		if (ret == -ENOMEM || ret == -EINTR || ret == -ERESTARTSYS)
-			goto cleanup_free;
-	}
+	ret = netfs_begin_cache_read(rreq, ctx);
+	if (ret == -ENOMEM || ret == -EINTR || ret == -ERESTARTSYS)
+		goto cleanup_free;
 
 	netfs_stat(&netfs_n_rh_readahead);
 	trace_netfs_read(rreq, readahead_pos(ractl), readahead_length(ractl),
@@ -192,6 +210,10 @@ void netfs_readahead(struct readahead_control *ractl)
 
 	netfs_rreq_expand(rreq, ractl);
 
+	/* Set up the output buffer */
+	iov_iter_xarray(&rreq->iter, ITER_DEST, &ractl->mapping->i_pages,
+			rreq->start, rreq->len);
+
 	/* Drop the refs on the folios here rather than in the cache or
 	 * filesystem.  The locks will be dropped in netfs_rreq_unlock().
 	 */
@@ -199,6 +221,7 @@ void netfs_readahead(struct readahead_control *ractl)
 		;
 
 	netfs_begin_read(rreq, false);
+	netfs_put_request(rreq, false, netfs_rreq_trace_put_return);
 	return;
 
 cleanup_free:
@@ -223,12 +246,13 @@ EXPORT_SYMBOL(netfs_readahead);
  */
 int netfs_read_folio(struct file *file, struct folio *folio)
 {
-	struct address_space *mapping = folio_file_mapping(folio);
+	struct address_space *mapping = folio->mapping;
 	struct netfs_io_request *rreq;
 	struct netfs_inode *ctx = netfs_inode(mapping->host);
+	struct folio *sink = NULL;
 	int ret;
 
-	_enter("%lx", folio_index(folio));
+	_enter("%lx", folio->index);
 
 	rreq = netfs_alloc_request(mapping, file,
 				   folio_file_pos(folio), folio_size(folio),
@@ -238,15 +262,64 @@ int netfs_read_folio(struct file *file, struct folio *folio)
 		goto alloc_error;
 	}
 
-	if (ctx->ops->begin_cache_operation) {
-		ret = ctx->ops->begin_cache_operation(rreq);
-		if (ret == -ENOMEM || ret == -EINTR || ret == -ERESTARTSYS)
-			goto discard;
-	}
+	ret = netfs_begin_cache_read(rreq, ctx);
+	if (ret == -ENOMEM || ret == -EINTR || ret == -ERESTARTSYS)
+		goto discard;
 
 	netfs_stat(&netfs_n_rh_readpage);
 	trace_netfs_read(rreq, rreq->start, rreq->len, netfs_read_trace_readpage);
-	return netfs_begin_read(rreq, true);
+
+	/* Set up the output buffer */
+	if (folio_test_dirty(folio)) {
+		/* Handle someone trying to read from an unflushed streaming
+		 * write.  We fiddle the buffer so that a gap at the beginning
+		 * and/or a gap at the end get copied to, but the middle is
+		 * discarded.
+		 */
+		struct netfs_folio *finfo = netfs_folio_info(folio);
+		struct bio_vec *bvec;
+		unsigned int from = finfo->dirty_offset;
+		unsigned int to = from + finfo->dirty_len;
+		unsigned int off = 0, i = 0;
+		size_t flen = folio_size(folio);
+		size_t nr_bvec = flen / PAGE_SIZE + 2;
+		size_t part;
+
+		ret = -ENOMEM;
+		bvec = kmalloc_array(nr_bvec, sizeof(*bvec), GFP_KERNEL);
+		if (!bvec)
+			goto discard;
+
+		sink = folio_alloc(GFP_KERNEL, 0);
+		if (!sink)
+			goto discard;
+
+		trace_netfs_folio(folio, netfs_folio_trace_read_gaps);
+
+		rreq->direct_bv = bvec;
+		rreq->direct_bv_count = nr_bvec;
+		if (from > 0) {
+			bvec_set_folio(&bvec[i++], folio, from, 0);
+			off = from;
+		}
+		while (off < to) {
+			part = min_t(size_t, to - off, PAGE_SIZE);
+			bvec_set_folio(&bvec[i++], sink, part, 0);
+			off += part;
+		}
+		if (to < flen)
+			bvec_set_folio(&bvec[i++], folio, flen - to, to);
+		iov_iter_bvec(&rreq->iter, ITER_DEST, bvec, i, rreq->len);
+	} else {
+		iov_iter_xarray(&rreq->iter, ITER_DEST, &mapping->i_pages,
+				rreq->start, rreq->len);
+	}
+
+	ret = netfs_begin_read(rreq, true);
+	if (sink)
+		folio_put(sink);
+	netfs_put_request(rreq, false, netfs_rreq_trace_put_return);
+	return ret < 0 ? ret : 0;
 
 discard:
 	netfs_put_request(rreq, false, netfs_rreq_trace_put_discard);
@@ -387,14 +460,12 @@ retry:
 		ret = PTR_ERR(rreq);
 		goto error;
 	}
-	rreq->no_unlock_folio	= folio_index(folio);
+	rreq->no_unlock_folio	= folio->index;
 	__set_bit(NETFS_RREQ_NO_UNLOCK_FOLIO, &rreq->flags);
 
-	if (ctx->ops->begin_cache_operation) {
-		ret = ctx->ops->begin_cache_operation(rreq);
-		if (ret == -ENOMEM || ret == -EINTR || ret == -ERESTARTSYS)
-			goto error_put;
-	}
+	ret = netfs_begin_cache_read(rreq, ctx);
+	if (ret == -ENOMEM || ret == -EINTR || ret == -ERESTARTSYS)
+		goto error_put;
 
 	netfs_stat(&netfs_n_rh_write_begin);
 	trace_netfs_read(rreq, pos, len, netfs_read_trace_write_begin);
@@ -405,6 +476,10 @@ retry:
 	ractl._nr_pages = folio_nr_pages(folio);
 	netfs_rreq_expand(rreq, &ractl);
 
+	/* Set up the output buffer */
+	iov_iter_xarray(&rreq->iter, ITER_DEST, &mapping->i_pages,
+			rreq->start, rreq->len);
+
 	/* We hold the folio locks, so we can drop the references */
 	folio_get(folio);
 	while (readahead_folio(&ractl))
@@ -413,6 +488,7 @@ retry:
 	ret = netfs_begin_read(rreq, true);
 	if (ret < 0)
 		goto error;
+	netfs_put_request(rreq, false, netfs_rreq_trace_put_return);
 
 have_folio:
 	ret = folio_wait_fscache_killable(folio);
@@ -434,3 +510,124 @@ error:
 	return ret;
 }
 EXPORT_SYMBOL(netfs_write_begin);
+
+/*
+ * Preload the data into a page we're proposing to write into.
+ */
+int netfs_prefetch_for_write(struct file *file, struct folio *folio,
+			     size_t offset, size_t len)
+{
+	struct netfs_io_request *rreq;
+	struct address_space *mapping = folio->mapping;
+	struct netfs_inode *ctx = netfs_inode(mapping->host);
+	unsigned long long start = folio_pos(folio);
+	size_t flen = folio_size(folio);
+	int ret;
+
+	_enter("%zx @%llx", flen, start);
+
+	ret = -ENOMEM;
+
+	rreq = netfs_alloc_request(mapping, file, start, flen,
+				   NETFS_READ_FOR_WRITE);
+	if (IS_ERR(rreq)) {
+		ret = PTR_ERR(rreq);
+		goto error;
+	}
+
+	rreq->no_unlock_folio = folio->index;
+	__set_bit(NETFS_RREQ_NO_UNLOCK_FOLIO, &rreq->flags);
+	ret = netfs_begin_cache_read(rreq, ctx);
+	if (ret == -ENOMEM || ret == -EINTR || ret == -ERESTARTSYS)
+		goto error_put;
+
+	netfs_stat(&netfs_n_rh_write_begin);
+	trace_netfs_read(rreq, start, flen, netfs_read_trace_prefetch_for_write);
+
+	/* Set up the output buffer */
+	iov_iter_xarray(&rreq->iter, ITER_DEST, &mapping->i_pages,
+			rreq->start, rreq->len);
+
+	ret = netfs_begin_read(rreq, true);
+	netfs_put_request(rreq, false, netfs_rreq_trace_put_return);
+	return ret;
+
+error_put:
+	netfs_put_request(rreq, false, netfs_rreq_trace_put_discard);
+error:
+	_leave(" = %d", ret);
+	return ret;
+}
+
+/**
+ * netfs_buffered_read_iter - Filesystem buffered I/O read routine
+ * @iocb: kernel I/O control block
+ * @iter: destination for the data read
+ *
+ * This is the ->read_iter() routine for all filesystems that can use the page
+ * cache directly.
+ *
+ * The IOCB_NOWAIT flag in iocb->ki_flags indicates that -EAGAIN shall be
+ * returned when no data can be read without waiting for I/O requests to
+ * complete; it doesn't prevent readahead.
+ *
+ * The IOCB_NOIO flag in iocb->ki_flags indicates that no new I/O requests
+ * shall be made for the read or for readahead.  When no data can be read,
+ * -EAGAIN shall be returned.  When readahead would be triggered, a partial,
+ * possibly empty read shall be returned.
+ *
+ * Return:
+ * * number of bytes copied, even for partial reads
+ * * negative error code (or 0 if IOCB_NOIO) if nothing was read
+ */
+ssize_t netfs_buffered_read_iter(struct kiocb *iocb, struct iov_iter *iter)
+{
+	struct inode *inode = file_inode(iocb->ki_filp);
+	struct netfs_inode *ictx = netfs_inode(inode);
+	ssize_t ret;
+
+	if (WARN_ON_ONCE((iocb->ki_flags & IOCB_DIRECT) ||
+			 test_bit(NETFS_ICTX_UNBUFFERED, &ictx->flags)))
+		return -EINVAL;
+
+	ret = netfs_start_io_read(inode);
+	if (ret == 0) {
+		ret = filemap_read(iocb, iter, 0);
+		netfs_end_io_read(inode);
+	}
+	return ret;
+}
+EXPORT_SYMBOL(netfs_buffered_read_iter);
+
+/**
+ * netfs_file_read_iter - Generic filesystem read routine
+ * @iocb: kernel I/O control block
+ * @iter: destination for the data read
+ *
+ * This is the ->read_iter() routine for all filesystems that can use the page
+ * cache directly.
+ *
+ * The IOCB_NOWAIT flag in iocb->ki_flags indicates that -EAGAIN shall be
+ * returned when no data can be read without waiting for I/O requests to
+ * complete; it doesn't prevent readahead.
+ *
+ * The IOCB_NOIO flag in iocb->ki_flags indicates that no new I/O requests
+ * shall be made for the read or for readahead.  When no data can be read,
+ * -EAGAIN shall be returned.  When readahead would be triggered, a partial,
+ * possibly empty read shall be returned.
+ *
+ * Return:
+ * * number of bytes copied, even for partial reads
+ * * negative error code (or 0 if IOCB_NOIO) if nothing was read
+ */
+ssize_t netfs_file_read_iter(struct kiocb *iocb, struct iov_iter *iter)
+{
+	struct netfs_inode *ictx = netfs_inode(iocb->ki_filp->f_mapping->host);
+
+	if ((iocb->ki_flags & IOCB_DIRECT) ||
+	    test_bit(NETFS_ICTX_UNBUFFERED, &ictx->flags))
+		return netfs_unbuffered_read_iter(iocb, iter);
+
+	return netfs_buffered_read_iter(iocb, iter);
+}
+EXPORT_SYMBOL(netfs_file_read_iter);
diff --git a/fs/netfs/buffered_write.c b/fs/netfs/buffered_write.c
new file mode 100644
index 000000000000..a3059b3168fd
--- /dev/null
+++ b/fs/netfs/buffered_write.c
@@ -0,0 +1,1254 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Network filesystem high-level write support.
+ *
+ * Copyright (C) 2023 Red Hat, Inc. All Rights Reserved.
+ * Written by David Howells (dhowells@redhat.com)
+ */
+
+#include <linux/export.h>
+#include <linux/fs.h>
+#include <linux/mm.h>
+#include <linux/pagemap.h>
+#include <linux/slab.h>
+#include <linux/pagevec.h>
+#include "internal.h"
+
+/*
+ * Determined write method.  Adjust netfs_folio_traces if this is changed.
+ */
+enum netfs_how_to_modify {
+	NETFS_FOLIO_IS_UPTODATE,	/* Folio is uptodate already */
+	NETFS_JUST_PREFETCH,		/* We have to read the folio anyway */
+	NETFS_WHOLE_FOLIO_MODIFY,	/* We're going to overwrite the whole folio */
+	NETFS_MODIFY_AND_CLEAR,		/* We can assume there is no data to be downloaded. */
+	NETFS_STREAMING_WRITE,		/* Store incomplete data in non-uptodate page. */
+	NETFS_STREAMING_WRITE_CONT,	/* Continue streaming write. */
+	NETFS_FLUSH_CONTENT,		/* Flush incompatible content. */
+};
+
+static void netfs_cleanup_buffered_write(struct netfs_io_request *wreq);
+
+static void netfs_set_group(struct folio *folio, struct netfs_group *netfs_group)
+{
+	if (netfs_group && !folio_get_private(folio))
+		folio_attach_private(folio, netfs_get_group(netfs_group));
+}
+
+#if IS_ENABLED(CONFIG_FSCACHE)
+static void netfs_folio_start_fscache(bool caching, struct folio *folio)
+{
+	if (caching)
+		folio_start_fscache(folio);
+}
+#else
+static void netfs_folio_start_fscache(bool caching, struct folio *folio)
+{
+}
+#endif
+
+/*
+ * Decide how we should modify a folio.  We might be attempting to do
+ * write-streaming, in which case we don't want to a local RMW cycle if we can
+ * avoid it.  If we're doing local caching or content crypto, we award that
+ * priority over avoiding RMW.  If the file is open readably, then we also
+ * assume that we may want to read what we wrote.
+ */
+static enum netfs_how_to_modify netfs_how_to_modify(struct netfs_inode *ctx,
+						    struct file *file,
+						    struct folio *folio,
+						    void *netfs_group,
+						    size_t flen,
+						    size_t offset,
+						    size_t len,
+						    bool maybe_trouble)
+{
+	struct netfs_folio *finfo = netfs_folio_info(folio);
+	loff_t pos = folio_file_pos(folio);
+
+	_enter("");
+
+	if (netfs_folio_group(folio) != netfs_group)
+		return NETFS_FLUSH_CONTENT;
+
+	if (folio_test_uptodate(folio))
+		return NETFS_FOLIO_IS_UPTODATE;
+
+	if (pos >= ctx->zero_point)
+		return NETFS_MODIFY_AND_CLEAR;
+
+	if (!maybe_trouble && offset == 0 && len >= flen)
+		return NETFS_WHOLE_FOLIO_MODIFY;
+
+	if (file->f_mode & FMODE_READ)
+		goto no_write_streaming;
+	if (test_bit(NETFS_ICTX_NO_WRITE_STREAMING, &ctx->flags))
+		goto no_write_streaming;
+
+	if (netfs_is_cache_enabled(ctx)) {
+		/* We don't want to get a streaming write on a file that loses
+		 * caching service temporarily because the backing store got
+		 * culled.
+		 */
+		if (!test_bit(NETFS_ICTX_NO_WRITE_STREAMING, &ctx->flags))
+			set_bit(NETFS_ICTX_NO_WRITE_STREAMING, &ctx->flags);
+		goto no_write_streaming;
+	}
+
+	if (!finfo)
+		return NETFS_STREAMING_WRITE;
+
+	/* We can continue a streaming write only if it continues on from the
+	 * previous.  If it overlaps, we must flush lest we suffer a partial
+	 * copy and disjoint dirty regions.
+	 */
+	if (offset == finfo->dirty_offset + finfo->dirty_len)
+		return NETFS_STREAMING_WRITE_CONT;
+	return NETFS_FLUSH_CONTENT;
+
+no_write_streaming:
+	if (finfo) {
+		netfs_stat(&netfs_n_wh_wstream_conflict);
+		return NETFS_FLUSH_CONTENT;
+	}
+	return NETFS_JUST_PREFETCH;
+}
+
+/*
+ * Grab a folio for writing and lock it.  Attempt to allocate as large a folio
+ * as possible to hold as much of the remaining length as possible in one go.
+ */
+static struct folio *netfs_grab_folio_for_write(struct address_space *mapping,
+						loff_t pos, size_t part)
+{
+	pgoff_t index = pos / PAGE_SIZE;
+	fgf_t fgp_flags = FGP_WRITEBEGIN;
+
+	if (mapping_large_folio_support(mapping))
+		fgp_flags |= fgf_set_order(pos % PAGE_SIZE + part);
+
+	return __filemap_get_folio(mapping, index, fgp_flags,
+				   mapping_gfp_mask(mapping));
+}
+
+/**
+ * netfs_perform_write - Copy data into the pagecache.
+ * @iocb: The operation parameters
+ * @iter: The source buffer
+ * @netfs_group: Grouping for dirty pages (eg. ceph snaps).
+ *
+ * Copy data into pagecache pages attached to the inode specified by @iocb.
+ * The caller must hold appropriate inode locks.
+ *
+ * Dirty pages are tagged with a netfs_folio struct if they're not up to date
+ * to indicate the range modified.  Dirty pages may also be tagged with a
+ * netfs-specific grouping such that data from an old group gets flushed before
+ * a new one is started.
+ */
+ssize_t netfs_perform_write(struct kiocb *iocb, struct iov_iter *iter,
+			    struct netfs_group *netfs_group)
+{
+	struct file *file = iocb->ki_filp;
+	struct inode *inode = file_inode(file);
+	struct address_space *mapping = inode->i_mapping;
+	struct netfs_inode *ctx = netfs_inode(inode);
+	struct writeback_control wbc = {
+		.sync_mode	= WB_SYNC_NONE,
+		.for_sync	= true,
+		.nr_to_write	= LONG_MAX,
+		.range_start	= iocb->ki_pos,
+		.range_end	= iocb->ki_pos + iter->count,
+	};
+	struct netfs_io_request *wreq = NULL;
+	struct netfs_folio *finfo;
+	struct folio *folio;
+	enum netfs_how_to_modify howto;
+	enum netfs_folio_trace trace;
+	unsigned int bdp_flags = (iocb->ki_flags & IOCB_SYNC) ? 0: BDP_ASYNC;
+	ssize_t written = 0, ret;
+	loff_t i_size, pos = iocb->ki_pos, from, to;
+	size_t max_chunk = PAGE_SIZE << MAX_PAGECACHE_ORDER;
+	bool maybe_trouble = false;
+
+	if (unlikely(test_bit(NETFS_ICTX_WRITETHROUGH, &ctx->flags) ||
+		     iocb->ki_flags & (IOCB_DSYNC | IOCB_SYNC))
+	    ) {
+		if (pos < i_size_read(inode)) {
+			ret = filemap_write_and_wait_range(mapping, pos, pos + iter->count);
+			if (ret < 0) {
+				goto out;
+			}
+		}
+
+		wbc_attach_fdatawrite_inode(&wbc, mapping->host);
+
+		wreq = netfs_begin_writethrough(iocb, iter->count);
+		if (IS_ERR(wreq)) {
+			wbc_detach_inode(&wbc);
+			ret = PTR_ERR(wreq);
+			wreq = NULL;
+			goto out;
+		}
+		if (!is_sync_kiocb(iocb))
+			wreq->iocb = iocb;
+		wreq->cleanup = netfs_cleanup_buffered_write;
+	}
+
+	do {
+		size_t flen;
+		size_t offset;	/* Offset into pagecache folio */
+		size_t part;	/* Bytes to write to folio */
+		size_t copied;	/* Bytes copied from user */
+
+		ret = balance_dirty_pages_ratelimited_flags(mapping, bdp_flags);
+		if (unlikely(ret < 0))
+			break;
+
+		offset = pos & (max_chunk - 1);
+		part = min(max_chunk - offset, iov_iter_count(iter));
+
+		/* Bring in the user pages that we will copy from _first_ lest
+		 * we hit a nasty deadlock on copying from the same page as
+		 * we're writing to, without it being marked uptodate.
+		 *
+		 * Not only is this an optimisation, but it is also required to
+		 * check that the address is actually valid, when atomic
+		 * usercopies are used below.
+		 *
+		 * We rely on the page being held onto long enough by the LRU
+		 * that we can grab it below if this causes it to be read.
+		 */
+		ret = -EFAULT;
+		if (unlikely(fault_in_iov_iter_readable(iter, part) == part))
+			break;
+
+		folio = netfs_grab_folio_for_write(mapping, pos, part);
+		if (IS_ERR(folio)) {
+			ret = PTR_ERR(folio);
+			break;
+		}
+
+		flen = folio_size(folio);
+		offset = pos & (flen - 1);
+		part = min_t(size_t, flen - offset, part);
+
+		if (signal_pending(current)) {
+			ret = written ? -EINTR : -ERESTARTSYS;
+			goto error_folio_unlock;
+		}
+
+		/* See if we need to prefetch the area we're going to modify.
+		 * We need to do this before we get a lock on the folio in case
+		 * there's more than one writer competing for the same cache
+		 * block.
+		 */
+		howto = netfs_how_to_modify(ctx, file, folio, netfs_group,
+					    flen, offset, part, maybe_trouble);
+		_debug("howto %u", howto);
+		switch (howto) {
+		case NETFS_JUST_PREFETCH:
+			ret = netfs_prefetch_for_write(file, folio, offset, part);
+			if (ret < 0) {
+				_debug("prefetch = %zd", ret);
+				goto error_folio_unlock;
+			}
+			break;
+		case NETFS_FOLIO_IS_UPTODATE:
+		case NETFS_WHOLE_FOLIO_MODIFY:
+		case NETFS_STREAMING_WRITE_CONT:
+			break;
+		case NETFS_MODIFY_AND_CLEAR:
+			zero_user_segment(&folio->page, 0, offset);
+			break;
+		case NETFS_STREAMING_WRITE:
+			ret = -EIO;
+			if (WARN_ON(folio_get_private(folio)))
+				goto error_folio_unlock;
+			break;
+		case NETFS_FLUSH_CONTENT:
+			trace_netfs_folio(folio, netfs_flush_content);
+			from = folio_pos(folio);
+			to = from + folio_size(folio) - 1;
+			folio_unlock(folio);
+			folio_put(folio);
+			ret = filemap_write_and_wait_range(mapping, from, to);
+			if (ret < 0)
+				goto error_folio_unlock;
+			continue;
+		}
+
+		if (mapping_writably_mapped(mapping))
+			flush_dcache_folio(folio);
+
+		copied = copy_folio_from_iter_atomic(folio, offset, part, iter);
+
+		flush_dcache_folio(folio);
+
+		/* Deal with a (partially) failed copy */
+		if (copied == 0) {
+			ret = -EFAULT;
+			goto error_folio_unlock;
+		}
+
+		trace = (enum netfs_folio_trace)howto;
+		switch (howto) {
+		case NETFS_FOLIO_IS_UPTODATE:
+		case NETFS_JUST_PREFETCH:
+			netfs_set_group(folio, netfs_group);
+			break;
+		case NETFS_MODIFY_AND_CLEAR:
+			zero_user_segment(&folio->page, offset + copied, flen);
+			netfs_set_group(folio, netfs_group);
+			folio_mark_uptodate(folio);
+			break;
+		case NETFS_WHOLE_FOLIO_MODIFY:
+			if (unlikely(copied < part)) {
+				maybe_trouble = true;
+				iov_iter_revert(iter, copied);
+				copied = 0;
+				goto retry;
+			}
+			netfs_set_group(folio, netfs_group);
+			folio_mark_uptodate(folio);
+			break;
+		case NETFS_STREAMING_WRITE:
+			if (offset == 0 && copied == flen) {
+				netfs_set_group(folio, netfs_group);
+				folio_mark_uptodate(folio);
+				trace = netfs_streaming_filled_page;
+				break;
+			}
+			finfo = kzalloc(sizeof(*finfo), GFP_KERNEL);
+			if (!finfo) {
+				iov_iter_revert(iter, copied);
+				ret = -ENOMEM;
+				goto error_folio_unlock;
+			}
+			finfo->netfs_group = netfs_get_group(netfs_group);
+			finfo->dirty_offset = offset;
+			finfo->dirty_len = copied;
+			folio_attach_private(folio, (void *)((unsigned long)finfo |
+							     NETFS_FOLIO_INFO));
+			break;
+		case NETFS_STREAMING_WRITE_CONT:
+			finfo = netfs_folio_info(folio);
+			finfo->dirty_len += copied;
+			if (finfo->dirty_offset == 0 && finfo->dirty_len == flen) {
+				if (finfo->netfs_group)
+					folio_change_private(folio, finfo->netfs_group);
+				else
+					folio_detach_private(folio);
+				folio_mark_uptodate(folio);
+				kfree(finfo);
+				trace = netfs_streaming_cont_filled_page;
+			}
+			break;
+		default:
+			WARN(true, "Unexpected modify type %u ix=%lx\n",
+			     howto, folio->index);
+			ret = -EIO;
+			goto error_folio_unlock;
+		}
+
+		trace_netfs_folio(folio, trace);
+
+		/* Update the inode size if we moved the EOF marker */
+		i_size = i_size_read(inode);
+		pos += copied;
+		if (pos > i_size) {
+			if (ctx->ops->update_i_size) {
+				ctx->ops->update_i_size(inode, pos);
+			} else {
+				i_size_write(inode, pos);
+#if IS_ENABLED(CONFIG_FSCACHE)
+				fscache_update_cookie(ctx->cache, NULL, &pos);
+#endif
+			}
+		}
+		written += copied;
+
+		if (likely(!wreq)) {
+			folio_mark_dirty(folio);
+		} else {
+			if (folio_test_dirty(folio))
+				/* Sigh.  mmap. */
+				folio_clear_dirty_for_io(folio);
+			/* We make multiple writes to the folio... */
+			if (!folio_test_writeback(folio)) {
+				folio_wait_fscache(folio);
+				folio_start_writeback(folio);
+				folio_start_fscache(folio);
+				if (wreq->iter.count == 0)
+					trace_netfs_folio(folio, netfs_folio_trace_wthru);
+				else
+					trace_netfs_folio(folio, netfs_folio_trace_wthru_plus);
+			}
+			netfs_advance_writethrough(wreq, copied,
+						   offset + copied == flen);
+		}
+	retry:
+		folio_unlock(folio);
+		folio_put(folio);
+		folio = NULL;
+
+		cond_resched();
+	} while (iov_iter_count(iter));
+
+out:
+	if (unlikely(wreq)) {
+		ret = netfs_end_writethrough(wreq, iocb);
+		wbc_detach_inode(&wbc);
+		if (ret == -EIOCBQUEUED)
+			return ret;
+	}
+
+	iocb->ki_pos += written;
+	_leave(" = %zd [%zd]", written, ret);
+	return written ? written : ret;
+
+error_folio_unlock:
+	folio_unlock(folio);
+	folio_put(folio);
+	goto out;
+}
+EXPORT_SYMBOL(netfs_perform_write);
+
+/**
+ * netfs_buffered_write_iter_locked - write data to a file
+ * @iocb:	IO state structure (file, offset, etc.)
+ * @from:	iov_iter with data to write
+ * @netfs_group: Grouping for dirty pages (eg. ceph snaps).
+ *
+ * This function does all the work needed for actually writing data to a
+ * file. It does all basic checks, removes SUID from the file, updates
+ * modification times and calls proper subroutines depending on whether we
+ * do direct IO or a standard buffered write.
+ *
+ * The caller must hold appropriate locks around this function and have called
+ * generic_write_checks() already.  The caller is also responsible for doing
+ * any necessary syncing afterwards.
+ *
+ * This function does *not* take care of syncing data in case of O_SYNC write.
+ * A caller has to handle it. This is mainly due to the fact that we want to
+ * avoid syncing under i_rwsem.
+ *
+ * Return:
+ * * number of bytes written, even for truncated writes
+ * * negative error code if no data has been written at all
+ */
+ssize_t netfs_buffered_write_iter_locked(struct kiocb *iocb, struct iov_iter *from,
+					 struct netfs_group *netfs_group)
+{
+	struct file *file = iocb->ki_filp;
+	ssize_t ret;
+
+	trace_netfs_write_iter(iocb, from);
+
+	ret = file_remove_privs(file);
+	if (ret)
+		return ret;
+
+	ret = file_update_time(file);
+	if (ret)
+		return ret;
+
+	return netfs_perform_write(iocb, from, netfs_group);
+}
+EXPORT_SYMBOL(netfs_buffered_write_iter_locked);
+
+/**
+ * netfs_file_write_iter - write data to a file
+ * @iocb: IO state structure
+ * @from: iov_iter with data to write
+ *
+ * Perform a write to a file, writing into the pagecache if possible and doing
+ * an unbuffered write instead if not.
+ *
+ * Return:
+ * * Negative error code if no data has been written at all of
+ *   vfs_fsync_range() failed for a synchronous write
+ * * Number of bytes written, even for truncated writes
+ */
+ssize_t netfs_file_write_iter(struct kiocb *iocb, struct iov_iter *from)
+{
+	struct file *file = iocb->ki_filp;
+	struct inode *inode = file->f_mapping->host;
+	struct netfs_inode *ictx = netfs_inode(inode);
+	ssize_t ret;
+
+	_enter("%llx,%zx,%llx", iocb->ki_pos, iov_iter_count(from), i_size_read(inode));
+
+	if ((iocb->ki_flags & IOCB_DIRECT) ||
+	    test_bit(NETFS_ICTX_UNBUFFERED, &ictx->flags))
+		return netfs_unbuffered_write_iter(iocb, from);
+
+	ret = netfs_start_io_write(inode);
+	if (ret < 0)
+		return ret;
+
+	ret = generic_write_checks(iocb, from);
+	if (ret > 0)
+		ret = netfs_buffered_write_iter_locked(iocb, from, NULL);
+	netfs_end_io_write(inode);
+	if (ret > 0)
+		ret = generic_write_sync(iocb, ret);
+	return ret;
+}
+EXPORT_SYMBOL(netfs_file_write_iter);
+
+/*
+ * Notification that a previously read-only page is about to become writable.
+ * Note that the caller indicates a single page of a multipage folio.
+ */
+vm_fault_t netfs_page_mkwrite(struct vm_fault *vmf, struct netfs_group *netfs_group)
+{
+	struct folio *folio = page_folio(vmf->page);
+	struct file *file = vmf->vma->vm_file;
+	struct inode *inode = file_inode(file);
+	vm_fault_t ret = VM_FAULT_RETRY;
+	int err;
+
+	_enter("%lx", folio->index);
+
+	sb_start_pagefault(inode->i_sb);
+
+	if (folio_wait_writeback_killable(folio))
+		goto out;
+
+	if (folio_lock_killable(folio) < 0)
+		goto out;
+
+	/* Can we see a streaming write here? */
+	if (WARN_ON(!folio_test_uptodate(folio))) {
+		ret = VM_FAULT_SIGBUS | VM_FAULT_LOCKED;
+		goto out;
+	}
+
+	if (netfs_folio_group(folio) != netfs_group) {
+		folio_unlock(folio);
+		err = filemap_fdatawait_range(inode->i_mapping,
+					      folio_pos(folio),
+					      folio_pos(folio) + folio_size(folio));
+		switch (err) {
+		case 0:
+			ret = VM_FAULT_RETRY;
+			goto out;
+		case -ENOMEM:
+			ret = VM_FAULT_OOM;
+			goto out;
+		default:
+			ret = VM_FAULT_SIGBUS;
+			goto out;
+		}
+	}
+
+	if (folio_test_dirty(folio))
+		trace_netfs_folio(folio, netfs_folio_trace_mkwrite_plus);
+	else
+		trace_netfs_folio(folio, netfs_folio_trace_mkwrite);
+	netfs_set_group(folio, netfs_group);
+	file_update_time(file);
+	ret = VM_FAULT_LOCKED;
+out:
+	sb_end_pagefault(inode->i_sb);
+	return ret;
+}
+EXPORT_SYMBOL(netfs_page_mkwrite);
+
+/*
+ * Kill all the pages in the given range
+ */
+static void netfs_kill_pages(struct address_space *mapping,
+			     loff_t start, loff_t len)
+{
+	struct folio *folio;
+	pgoff_t index = start / PAGE_SIZE;
+	pgoff_t last = (start + len - 1) / PAGE_SIZE, next;
+
+	_enter("%llx-%llx", start, start + len - 1);
+
+	do {
+		_debug("kill %lx (to %lx)", index, last);
+
+		folio = filemap_get_folio(mapping, index);
+		if (IS_ERR(folio)) {
+			next = index + 1;
+			continue;
+		}
+
+		next = folio_next_index(folio);
+
+		trace_netfs_folio(folio, netfs_folio_trace_kill);
+		folio_clear_uptodate(folio);
+		if (folio_test_fscache(folio))
+			folio_end_fscache(folio);
+		folio_end_writeback(folio);
+		folio_lock(folio);
+		generic_error_remove_folio(mapping, folio);
+		folio_unlock(folio);
+		folio_put(folio);
+
+	} while (index = next, index <= last);
+
+	_leave("");
+}
+
+/*
+ * Redirty all the pages in a given range.
+ */
+static void netfs_redirty_pages(struct address_space *mapping,
+				loff_t start, loff_t len)
+{
+	struct folio *folio;
+	pgoff_t index = start / PAGE_SIZE;
+	pgoff_t last = (start + len - 1) / PAGE_SIZE, next;
+
+	_enter("%llx-%llx", start, start + len - 1);
+
+	do {
+		_debug("redirty %llx @%llx", len, start);
+
+		folio = filemap_get_folio(mapping, index);
+		if (IS_ERR(folio)) {
+			next = index + 1;
+			continue;
+		}
+
+		next = folio_next_index(folio);
+		trace_netfs_folio(folio, netfs_folio_trace_redirty);
+		filemap_dirty_folio(mapping, folio);
+		if (folio_test_fscache(folio))
+			folio_end_fscache(folio);
+		folio_end_writeback(folio);
+		folio_put(folio);
+	} while (index = next, index <= last);
+
+	balance_dirty_pages_ratelimited(mapping);
+
+	_leave("");
+}
+
+/*
+ * Completion of write to server
+ */
+static void netfs_pages_written_back(struct netfs_io_request *wreq)
+{
+	struct address_space *mapping = wreq->mapping;
+	struct netfs_folio *finfo;
+	struct netfs_group *group = NULL;
+	struct folio *folio;
+	pgoff_t last;
+	int gcount = 0;
+
+	XA_STATE(xas, &mapping->i_pages, wreq->start / PAGE_SIZE);
+
+	_enter("%llx-%llx", wreq->start, wreq->start + wreq->len);
+
+	rcu_read_lock();
+
+	last = (wreq->start + wreq->len - 1) / PAGE_SIZE;
+	xas_for_each(&xas, folio, last) {
+		WARN(!folio_test_writeback(folio),
+		     "bad %zx @%llx page %lx %lx\n",
+		     wreq->len, wreq->start, folio->index, last);
+
+		if ((finfo = netfs_folio_info(folio))) {
+			/* Streaming writes cannot be redirtied whilst under
+			 * writeback, so discard the streaming record.
+			 */
+			folio_detach_private(folio);
+			group = finfo->netfs_group;
+			gcount++;
+			trace_netfs_folio(folio, netfs_folio_trace_clear_s);
+			kfree(finfo);
+		} else if ((group = netfs_folio_group(folio))) {
+			/* Need to detach the group pointer if the page didn't
+			 * get redirtied.  If it has been redirtied, then it
+			 * must be within the same group.
+			 */
+			if (folio_test_dirty(folio)) {
+				trace_netfs_folio(folio, netfs_folio_trace_redirtied);
+				goto end_wb;
+			}
+			if (folio_trylock(folio)) {
+				if (!folio_test_dirty(folio)) {
+					folio_detach_private(folio);
+					gcount++;
+					trace_netfs_folio(folio, netfs_folio_trace_clear_g);
+				} else {
+					trace_netfs_folio(folio, netfs_folio_trace_redirtied);
+				}
+				folio_unlock(folio);
+				goto end_wb;
+			}
+
+			xas_pause(&xas);
+			rcu_read_unlock();
+			folio_lock(folio);
+			if (!folio_test_dirty(folio)) {
+				folio_detach_private(folio);
+				gcount++;
+				trace_netfs_folio(folio, netfs_folio_trace_clear_g);
+			} else {
+				trace_netfs_folio(folio, netfs_folio_trace_redirtied);
+			}
+			folio_unlock(folio);
+			rcu_read_lock();
+		} else {
+			trace_netfs_folio(folio, netfs_folio_trace_clear);
+		}
+	end_wb:
+		if (folio_test_fscache(folio))
+			folio_end_fscache(folio);
+		xas_advance(&xas, folio_next_index(folio) - 1);
+		folio_end_writeback(folio);
+	}
+
+	rcu_read_unlock();
+	netfs_put_group_many(group, gcount);
+	_leave("");
+}
+
+/*
+ * Deal with the disposition of the folios that are under writeback to close
+ * out the operation.
+ */
+static void netfs_cleanup_buffered_write(struct netfs_io_request *wreq)
+{
+	struct address_space *mapping = wreq->mapping;
+
+	_enter("");
+
+	switch (wreq->error) {
+	case 0:
+		netfs_pages_written_back(wreq);
+		break;
+
+	default:
+		pr_notice("R=%08x Unexpected error %d\n", wreq->debug_id, wreq->error);
+		fallthrough;
+	case -EACCES:
+	case -EPERM:
+	case -ENOKEY:
+	case -EKEYEXPIRED:
+	case -EKEYREJECTED:
+	case -EKEYREVOKED:
+	case -ENETRESET:
+	case -EDQUOT:
+	case -ENOSPC:
+		netfs_redirty_pages(mapping, wreq->start, wreq->len);
+		break;
+
+	case -EROFS:
+	case -EIO:
+	case -EREMOTEIO:
+	case -EFBIG:
+	case -ENOENT:
+	case -ENOMEDIUM:
+	case -ENXIO:
+		netfs_kill_pages(mapping, wreq->start, wreq->len);
+		break;
+	}
+
+	if (wreq->error)
+		mapping_set_error(mapping, wreq->error);
+	if (wreq->netfs_ops->done)
+		wreq->netfs_ops->done(wreq);
+}
+
+/*
+ * Extend the region to be written back to include subsequent contiguously
+ * dirty pages if possible, but don't sleep while doing so.
+ *
+ * If this page holds new content, then we can include filler zeros in the
+ * writeback.
+ */
+static void netfs_extend_writeback(struct address_space *mapping,
+				   struct netfs_group *group,
+				   struct xa_state *xas,
+				   long *_count,
+				   loff_t start,
+				   loff_t max_len,
+				   bool caching,
+				   size_t *_len,
+				   size_t *_top)
+{
+	struct netfs_folio *finfo;
+	struct folio_batch fbatch;
+	struct folio *folio;
+	unsigned int i;
+	pgoff_t index = (start + *_len) / PAGE_SIZE;
+	size_t len;
+	void *priv;
+	bool stop = true;
+
+	folio_batch_init(&fbatch);
+
+	do {
+		/* Firstly, we gather up a batch of contiguous dirty pages
+		 * under the RCU read lock - but we can't clear the dirty flags
+		 * there if any of those pages are mapped.
+		 */
+		rcu_read_lock();
+
+		xas_for_each(xas, folio, ULONG_MAX) {
+			stop = true;
+			if (xas_retry(xas, folio))
+				continue;
+			if (xa_is_value(folio))
+				break;
+			if (folio->index != index) {
+				xas_reset(xas);
+				break;
+			}
+
+			if (!folio_try_get_rcu(folio)) {
+				xas_reset(xas);
+				continue;
+			}
+
+			/* Has the folio moved or been split? */
+			if (unlikely(folio != xas_reload(xas))) {
+				folio_put(folio);
+				xas_reset(xas);
+				break;
+			}
+
+			if (!folio_trylock(folio)) {
+				folio_put(folio);
+				xas_reset(xas);
+				break;
+			}
+			if (!folio_test_dirty(folio) ||
+			    folio_test_writeback(folio) ||
+			    folio_test_fscache(folio)) {
+				folio_unlock(folio);
+				folio_put(folio);
+				xas_reset(xas);
+				break;
+			}
+
+			stop = false;
+			len = folio_size(folio);
+			priv = folio_get_private(folio);
+			if ((const struct netfs_group *)priv != group) {
+				stop = true;
+				finfo = netfs_folio_info(folio);
+				if (finfo->netfs_group != group ||
+				    finfo->dirty_offset > 0) {
+					folio_unlock(folio);
+					folio_put(folio);
+					xas_reset(xas);
+					break;
+				}
+				len = finfo->dirty_len;
+			}
+
+			*_top += folio_size(folio);
+			index += folio_nr_pages(folio);
+			*_count -= folio_nr_pages(folio);
+			*_len += len;
+			if (*_len >= max_len || *_count <= 0)
+				stop = true;
+
+			if (!folio_batch_add(&fbatch, folio))
+				break;
+			if (stop)
+				break;
+		}
+
+		xas_pause(xas);
+		rcu_read_unlock();
+
+		/* Now, if we obtained any folios, we can shift them to being
+		 * writable and mark them for caching.
+		 */
+		if (!folio_batch_count(&fbatch))
+			break;
+
+		for (i = 0; i < folio_batch_count(&fbatch); i++) {
+			folio = fbatch.folios[i];
+			trace_netfs_folio(folio, netfs_folio_trace_store_plus);
+
+			if (!folio_clear_dirty_for_io(folio))
+				BUG();
+			folio_start_writeback(folio);
+			netfs_folio_start_fscache(caching, folio);
+			folio_unlock(folio);
+		}
+
+		folio_batch_release(&fbatch);
+		cond_resched();
+	} while (!stop);
+}
+
+/*
+ * Synchronously write back the locked page and any subsequent non-locked dirty
+ * pages.
+ */
+static ssize_t netfs_write_back_from_locked_folio(struct address_space *mapping,
+						  struct writeback_control *wbc,
+						  struct netfs_group *group,
+						  struct xa_state *xas,
+						  struct folio *folio,
+						  unsigned long long start,
+						  unsigned long long end)
+{
+	struct netfs_io_request *wreq;
+	struct netfs_folio *finfo;
+	struct netfs_inode *ctx = netfs_inode(mapping->host);
+	unsigned long long i_size = i_size_read(&ctx->inode);
+	size_t len, max_len;
+	bool caching = netfs_is_cache_enabled(ctx);
+	long count = wbc->nr_to_write;
+	int ret;
+
+	_enter(",%lx,%llx-%llx,%u", folio->index, start, end, caching);
+
+	wreq = netfs_alloc_request(mapping, NULL, start, folio_size(folio),
+				   NETFS_WRITEBACK);
+	if (IS_ERR(wreq)) {
+		folio_unlock(folio);
+		return PTR_ERR(wreq);
+	}
+
+	if (!folio_clear_dirty_for_io(folio))
+		BUG();
+	folio_start_writeback(folio);
+	netfs_folio_start_fscache(caching, folio);
+
+	count -= folio_nr_pages(folio);
+
+	/* Find all consecutive lockable dirty pages that have contiguous
+	 * written regions, stopping when we find a page that is not
+	 * immediately lockable, is not dirty or is missing, or we reach the
+	 * end of the range.
+	 */
+	trace_netfs_folio(folio, netfs_folio_trace_store);
+
+	len = wreq->len;
+	finfo = netfs_folio_info(folio);
+	if (finfo) {
+		start += finfo->dirty_offset;
+		if (finfo->dirty_offset + finfo->dirty_len != len) {
+			len = finfo->dirty_len;
+			goto cant_expand;
+		}
+		len = finfo->dirty_len;
+	}
+
+	if (start < i_size) {
+		/* Trim the write to the EOF; the extra data is ignored.  Also
+		 * put an upper limit on the size of a single storedata op.
+		 */
+		max_len = 65536 * 4096;
+		max_len = min_t(unsigned long long, max_len, end - start + 1);
+		max_len = min_t(unsigned long long, max_len, i_size - start);
+
+		if (len < max_len)
+			netfs_extend_writeback(mapping, group, xas, &count, start,
+					       max_len, caching, &len, &wreq->upper_len);
+	}
+
+cant_expand:
+	len = min_t(unsigned long long, len, i_size - start);
+
+	/* We now have a contiguous set of dirty pages, each with writeback
+	 * set; the first page is still locked at this point, but all the rest
+	 * have been unlocked.
+	 */
+	folio_unlock(folio);
+	wreq->start = start;
+	wreq->len = len;
+
+	if (start < i_size) {
+		_debug("write back %zx @%llx [%llx]", len, start, i_size);
+
+		/* Speculatively write to the cache.  We have to fix this up
+		 * later if the store fails.
+		 */
+		wreq->cleanup = netfs_cleanup_buffered_write;
+
+		iov_iter_xarray(&wreq->iter, ITER_SOURCE, &mapping->i_pages, start,
+				wreq->upper_len);
+		__set_bit(NETFS_RREQ_UPLOAD_TO_SERVER, &wreq->flags);
+		ret = netfs_begin_write(wreq, true, netfs_write_trace_writeback);
+		if (ret == 0 || ret == -EIOCBQUEUED)
+			wbc->nr_to_write -= len / PAGE_SIZE;
+	} else {
+		_debug("write discard %zx @%llx [%llx]", len, start, i_size);
+
+		/* The dirty region was entirely beyond the EOF. */
+		fscache_clear_page_bits(mapping, start, len, caching);
+		netfs_pages_written_back(wreq);
+		ret = 0;
+	}
+
+	netfs_put_request(wreq, false, netfs_rreq_trace_put_return);
+	_leave(" = 1");
+	return 1;
+}
+
+/*
+ * Write a region of pages back to the server
+ */
+static ssize_t netfs_writepages_begin(struct address_space *mapping,
+				      struct writeback_control *wbc,
+				      struct netfs_group *group,
+				      struct xa_state *xas,
+				      unsigned long long *_start,
+				      unsigned long long end)
+{
+	const struct netfs_folio *finfo;
+	struct folio *folio;
+	unsigned long long start = *_start;
+	ssize_t ret;
+	void *priv;
+	int skips = 0;
+
+	_enter("%llx,%llx,", start, end);
+
+search_again:
+	/* Find the first dirty page in the group. */
+	rcu_read_lock();
+
+	for (;;) {
+		folio = xas_find_marked(xas, end / PAGE_SIZE, PAGECACHE_TAG_DIRTY);
+		if (xas_retry(xas, folio) || xa_is_value(folio))
+			continue;
+		if (!folio)
+			break;
+
+		if (!folio_try_get_rcu(folio)) {
+			xas_reset(xas);
+			continue;
+		}
+
+		if (unlikely(folio != xas_reload(xas))) {
+			folio_put(folio);
+			xas_reset(xas);
+			continue;
+		}
+
+		/* Skip any dirty folio that's not in the group of interest. */
+		priv = folio_get_private(folio);
+		if ((const struct netfs_group *)priv != group) {
+			finfo = netfs_folio_info(folio);
+			if (finfo->netfs_group != group) {
+				folio_put(folio);
+				continue;
+			}
+		}
+
+		xas_pause(xas);
+		break;
+	}
+	rcu_read_unlock();
+	if (!folio)
+		return 0;
+
+	start = folio_pos(folio); /* May regress with THPs */
+
+	_debug("wback %lx", folio->index);
+
+	/* At this point we hold neither the i_pages lock nor the page lock:
+	 * the page may be truncated or invalidated (changing page->mapping to
+	 * NULL), or even swizzled back from swapper_space to tmpfs file
+	 * mapping
+	 */
+lock_again:
+	if (wbc->sync_mode != WB_SYNC_NONE) {
+		ret = folio_lock_killable(folio);
+		if (ret < 0)
+			return ret;
+	} else {
+		if (!folio_trylock(folio))
+			goto search_again;
+	}
+
+	if (folio->mapping != mapping ||
+	    !folio_test_dirty(folio)) {
+		start += folio_size(folio);
+		folio_unlock(folio);
+		goto search_again;
+	}
+
+	if (folio_test_writeback(folio) ||
+	    folio_test_fscache(folio)) {
+		folio_unlock(folio);
+		if (wbc->sync_mode != WB_SYNC_NONE) {
+			folio_wait_writeback(folio);
+#ifdef CONFIG_FSCACHE
+			folio_wait_fscache(folio);
+#endif
+			goto lock_again;
+		}
+
+		start += folio_size(folio);
+		if (wbc->sync_mode == WB_SYNC_NONE) {
+			if (skips >= 5 || need_resched()) {
+				ret = 0;
+				goto out;
+			}
+			skips++;
+		}
+		goto search_again;
+	}
+
+	ret = netfs_write_back_from_locked_folio(mapping, wbc, group, xas,
+						 folio, start, end);
+out:
+	if (ret > 0)
+		*_start = start + ret;
+	_leave(" = %zd [%llx]", ret, *_start);
+	return ret;
+}
+
+/*
+ * Write a region of pages back to the server
+ */
+static int netfs_writepages_region(struct address_space *mapping,
+				   struct writeback_control *wbc,
+				   struct netfs_group *group,
+				   unsigned long long *_start,
+				   unsigned long long end)
+{
+	ssize_t ret;
+
+	XA_STATE(xas, &mapping->i_pages, *_start / PAGE_SIZE);
+
+	do {
+		ret = netfs_writepages_begin(mapping, wbc, group, &xas,
+					     _start, end);
+		if (ret > 0 && wbc->nr_to_write > 0)
+			cond_resched();
+	} while (ret > 0 && wbc->nr_to_write > 0);
+
+	return ret > 0 ? 0 : ret;
+}
+
+/*
+ * write some of the pending data back to the server
+ */
+int netfs_writepages(struct address_space *mapping,
+		     struct writeback_control *wbc)
+{
+	struct netfs_group *group = NULL;
+	loff_t start, end;
+	int ret;
+
+	_enter("");
+
+	/* We have to be careful as we can end up racing with setattr()
+	 * truncating the pagecache since the caller doesn't take a lock here
+	 * to prevent it.
+	 */
+
+	if (wbc->range_cyclic && mapping->writeback_index) {
+		start = mapping->writeback_index * PAGE_SIZE;
+		ret = netfs_writepages_region(mapping, wbc, group,
+					      &start, LLONG_MAX);
+		if (ret < 0)
+			goto out;
+
+		if (wbc->nr_to_write <= 0) {
+			mapping->writeback_index = start / PAGE_SIZE;
+			goto out;
+		}
+
+		start = 0;
+		end = mapping->writeback_index * PAGE_SIZE;
+		mapping->writeback_index = 0;
+		ret = netfs_writepages_region(mapping, wbc, group, &start, end);
+		if (ret == 0)
+			mapping->writeback_index = start / PAGE_SIZE;
+	} else if (wbc->range_start == 0 && wbc->range_end == LLONG_MAX) {
+		start = 0;
+		ret = netfs_writepages_region(mapping, wbc, group,
+					      &start, LLONG_MAX);
+		if (wbc->nr_to_write > 0 && ret == 0)
+			mapping->writeback_index = start / PAGE_SIZE;
+	} else {
+		start = wbc->range_start;
+		ret = netfs_writepages_region(mapping, wbc, group,
+					      &start, wbc->range_end);
+	}
+
+out:
+	_leave(" = %d", ret);
+	return ret;
+}
+EXPORT_SYMBOL(netfs_writepages);
+
+/*
+ * Deal with the disposition of a laundered folio.
+ */
+static void netfs_cleanup_launder_folio(struct netfs_io_request *wreq)
+{
+	if (wreq->error) {
+		pr_notice("R=%08x Laundering error %d\n", wreq->debug_id, wreq->error);
+		mapping_set_error(wreq->mapping, wreq->error);
+	}
+}
+
+/**
+ * netfs_launder_folio - Clean up a dirty folio that's being invalidated
+ * @folio: The folio to clean
+ *
+ * This is called to write back a folio that's being invalidated when an inode
+ * is getting torn down.  Ideally, writepages would be used instead.
+ */
+int netfs_launder_folio(struct folio *folio)
+{
+	struct netfs_io_request *wreq;
+	struct address_space *mapping = folio->mapping;
+	struct netfs_folio *finfo = netfs_folio_info(folio);
+	struct netfs_group *group = netfs_folio_group(folio);
+	struct bio_vec bvec;
+	unsigned long long i_size = i_size_read(mapping->host);
+	unsigned long long start = folio_pos(folio);
+	size_t offset = 0, len;
+	int ret = 0;
+
+	if (finfo) {
+		offset = finfo->dirty_offset;
+		start += offset;
+		len = finfo->dirty_len;
+	} else {
+		len = folio_size(folio);
+	}
+	len = min_t(unsigned long long, len, i_size - start);
+
+	wreq = netfs_alloc_request(mapping, NULL, start, len, NETFS_LAUNDER_WRITE);
+	if (IS_ERR(wreq)) {
+		ret = PTR_ERR(wreq);
+		goto out;
+	}
+
+	if (!folio_clear_dirty_for_io(folio))
+		goto out_put;
+
+	trace_netfs_folio(folio, netfs_folio_trace_launder);
+
+	_debug("launder %llx-%llx", start, start + len - 1);
+
+	/* Speculatively write to the cache.  We have to fix this up later if
+	 * the store fails.
+	 */
+	wreq->cleanup = netfs_cleanup_launder_folio;
+
+	bvec_set_folio(&bvec, folio, len, offset);
+	iov_iter_bvec(&wreq->iter, ITER_SOURCE, &bvec, 1, len);
+	__set_bit(NETFS_RREQ_UPLOAD_TO_SERVER, &wreq->flags);
+	ret = netfs_begin_write(wreq, true, netfs_write_trace_launder);
+
+out_put:
+	folio_detach_private(folio);
+	netfs_put_group(group);
+	kfree(finfo);
+	netfs_put_request(wreq, false, netfs_rreq_trace_put_return);
+out:
+	folio_wait_fscache(folio);
+	_leave(" = %d", ret);
+	return ret;
+}
+EXPORT_SYMBOL(netfs_launder_folio);
diff --git a/fs/netfs/direct_read.c b/fs/netfs/direct_read.c
new file mode 100644
index 000000000000..ad4370b3935d
--- /dev/null
+++ b/fs/netfs/direct_read.c
@@ -0,0 +1,125 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/* Direct I/O support.
+ *
+ * Copyright (C) 2023 Red Hat, Inc. All Rights Reserved.
+ * Written by David Howells (dhowells@redhat.com)
+ */
+
+#include <linux/export.h>
+#include <linux/fs.h>
+#include <linux/mm.h>
+#include <linux/pagemap.h>
+#include <linux/slab.h>
+#include <linux/uio.h>
+#include <linux/sched/mm.h>
+#include <linux/task_io_accounting_ops.h>
+#include <linux/netfs.h>
+#include "internal.h"
+
+/**
+ * netfs_unbuffered_read_iter_locked - Perform an unbuffered or direct I/O read
+ * @iocb: The I/O control descriptor describing the read
+ * @iter: The output buffer (also specifies read length)
+ *
+ * Perform an unbuffered I/O or direct I/O from the file in @iocb to the
+ * output buffer.  No use is made of the pagecache.
+ *
+ * The caller must hold any appropriate locks.
+ */
+static ssize_t netfs_unbuffered_read_iter_locked(struct kiocb *iocb, struct iov_iter *iter)
+{
+	struct netfs_io_request *rreq;
+	ssize_t ret;
+	size_t orig_count = iov_iter_count(iter);
+	bool async = !is_sync_kiocb(iocb);
+
+	_enter("");
+
+	if (!orig_count)
+		return 0; /* Don't update atime */
+
+	ret = kiocb_write_and_wait(iocb, orig_count);
+	if (ret < 0)
+		return ret;
+	file_accessed(iocb->ki_filp);
+
+	rreq = netfs_alloc_request(iocb->ki_filp->f_mapping, iocb->ki_filp,
+				   iocb->ki_pos, orig_count,
+				   NETFS_DIO_READ);
+	if (IS_ERR(rreq))
+		return PTR_ERR(rreq);
+
+	netfs_stat(&netfs_n_rh_dio_read);
+	trace_netfs_read(rreq, rreq->start, rreq->len, netfs_read_trace_dio_read);
+
+	/* If this is an async op, we have to keep track of the destination
+	 * buffer for ourselves as the caller's iterator will be trashed when
+	 * we return.
+	 *
+	 * In such a case, extract an iterator to represent as much of the the
+	 * output buffer as we can manage.  Note that the extraction might not
+	 * be able to allocate a sufficiently large bvec array and may shorten
+	 * the request.
+	 */
+	if (user_backed_iter(iter)) {
+		ret = netfs_extract_user_iter(iter, rreq->len, &rreq->iter, 0);
+		if (ret < 0)
+			goto out;
+		rreq->direct_bv = (struct bio_vec *)rreq->iter.bvec;
+		rreq->direct_bv_count = ret;
+		rreq->direct_bv_unpin = iov_iter_extract_will_pin(iter);
+		rreq->len = iov_iter_count(&rreq->iter);
+	} else {
+		rreq->iter = *iter;
+		rreq->len = orig_count;
+		rreq->direct_bv_unpin = false;
+		iov_iter_advance(iter, orig_count);
+	}
+
+	// TODO: Set up bounce buffer if needed
+
+	if (async)
+		rreq->iocb = iocb;
+
+	ret = netfs_begin_read(rreq, is_sync_kiocb(iocb));
+	if (ret < 0)
+		goto out; /* May be -EIOCBQUEUED */
+	if (!async) {
+		// TODO: Copy from bounce buffer
+		iocb->ki_pos += rreq->transferred;
+		ret = rreq->transferred;
+	}
+
+out:
+	netfs_put_request(rreq, false, netfs_rreq_trace_put_return);
+	if (ret > 0)
+		orig_count -= ret;
+	if (ret != -EIOCBQUEUED)
+		iov_iter_revert(iter, orig_count - iov_iter_count(iter));
+	return ret;
+}
+
+/**
+ * netfs_unbuffered_read_iter - Perform an unbuffered or direct I/O read
+ * @iocb: The I/O control descriptor describing the read
+ * @iter: The output buffer (also specifies read length)
+ *
+ * Perform an unbuffered I/O or direct I/O from the file in @iocb to the
+ * output buffer.  No use is made of the pagecache.
+ */
+ssize_t netfs_unbuffered_read_iter(struct kiocb *iocb, struct iov_iter *iter)
+{
+	struct inode *inode = file_inode(iocb->ki_filp);
+	ssize_t ret;
+
+	if (!iter->count)
+		return 0; /* Don't update atime */
+
+	ret = netfs_start_io_direct(inode);
+	if (ret == 0) {
+		ret = netfs_unbuffered_read_iter_locked(iocb, iter);
+		netfs_end_io_direct(inode);
+	}
+	return ret;
+}
+EXPORT_SYMBOL(netfs_unbuffered_read_iter);
diff --git a/fs/netfs/direct_write.c b/fs/netfs/direct_write.c
new file mode 100644
index 000000000000..60a40d293c87
--- /dev/null
+++ b/fs/netfs/direct_write.c
@@ -0,0 +1,171 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/* Unbuffered and direct write support.
+ *
+ * Copyright (C) 2023 Red Hat, Inc. All Rights Reserved.
+ * Written by David Howells (dhowells@redhat.com)
+ */
+
+#include <linux/export.h>
+#include <linux/uio.h>
+#include "internal.h"
+
+static void netfs_cleanup_dio_write(struct netfs_io_request *wreq)
+{
+	struct inode *inode = wreq->inode;
+	unsigned long long end = wreq->start + wreq->len;
+
+	if (!wreq->error &&
+	    i_size_read(inode) < end) {
+		if (wreq->netfs_ops->update_i_size)
+			wreq->netfs_ops->update_i_size(inode, end);
+		else
+			i_size_write(inode, end);
+	}
+}
+
+/*
+ * Perform an unbuffered write where we may have to do an RMW operation on an
+ * encrypted file.  This can also be used for direct I/O writes.
+ */
+static ssize_t netfs_unbuffered_write_iter_locked(struct kiocb *iocb, struct iov_iter *iter,
+						  struct netfs_group *netfs_group)
+{
+	struct netfs_io_request *wreq;
+	unsigned long long start = iocb->ki_pos;
+	unsigned long long end = start + iov_iter_count(iter);
+	ssize_t ret, n;
+	bool async = !is_sync_kiocb(iocb);
+
+	_enter("");
+
+	/* We're going to need a bounce buffer if what we transmit is going to
+	 * be different in some way to the source buffer, e.g. because it gets
+	 * encrypted/compressed or because it needs expanding to a block size.
+	 */
+	// TODO
+
+	_debug("uw %llx-%llx", start, end);
+
+	wreq = netfs_alloc_request(iocb->ki_filp->f_mapping, iocb->ki_filp,
+				   start, end - start,
+				   iocb->ki_flags & IOCB_DIRECT ?
+				   NETFS_DIO_WRITE : NETFS_UNBUFFERED_WRITE);
+	if (IS_ERR(wreq))
+		return PTR_ERR(wreq);
+
+	{
+		/* If this is an async op and we're not using a bounce buffer,
+		 * we have to save the source buffer as the iterator is only
+		 * good until we return.  In such a case, extract an iterator
+		 * to represent as much of the the output buffer as we can
+		 * manage.  Note that the extraction might not be able to
+		 * allocate a sufficiently large bvec array and may shorten the
+		 * request.
+		 */
+		if (async || user_backed_iter(iter)) {
+			n = netfs_extract_user_iter(iter, wreq->len, &wreq->iter, 0);
+			if (n < 0) {
+				ret = n;
+				goto out;
+			}
+			wreq->direct_bv = (struct bio_vec *)wreq->iter.bvec;
+			wreq->direct_bv_count = n;
+			wreq->direct_bv_unpin = iov_iter_extract_will_pin(iter);
+			wreq->len = iov_iter_count(&wreq->iter);
+		} else {
+			wreq->iter = *iter;
+		}
+
+		wreq->io_iter = wreq->iter;
+	}
+
+	/* Copy the data into the bounce buffer and encrypt it. */
+	// TODO
+
+	/* Dispatch the write. */
+	__set_bit(NETFS_RREQ_UPLOAD_TO_SERVER, &wreq->flags);
+	if (async)
+		wreq->iocb = iocb;
+	wreq->cleanup = netfs_cleanup_dio_write;
+	ret = netfs_begin_write(wreq, is_sync_kiocb(iocb),
+				iocb->ki_flags & IOCB_DIRECT ?
+				netfs_write_trace_dio_write :
+				netfs_write_trace_unbuffered_write);
+	if (ret < 0) {
+		_debug("begin = %zd", ret);
+		goto out;
+	}
+
+	if (!async) {
+		trace_netfs_rreq(wreq, netfs_rreq_trace_wait_ip);
+		wait_on_bit(&wreq->flags, NETFS_RREQ_IN_PROGRESS,
+			    TASK_UNINTERRUPTIBLE);
+
+		ret = wreq->error;
+		_debug("waited = %zd", ret);
+		if (ret == 0) {
+			ret = wreq->transferred;
+			iocb->ki_pos += ret;
+		}
+	} else {
+		ret = -EIOCBQUEUED;
+	}
+
+out:
+	netfs_put_request(wreq, false, netfs_rreq_trace_put_return);
+	return ret;
+}
+
+/**
+ * netfs_unbuffered_write_iter - Unbuffered write to a file
+ * @iocb: IO state structure
+ * @from: iov_iter with data to write
+ *
+ * Do an unbuffered write to a file, writing the data directly to the server
+ * and not lodging the data in the pagecache.
+ *
+ * Return:
+ * * Negative error code if no data has been written at all of
+ *   vfs_fsync_range() failed for a synchronous write
+ * * Number of bytes written, even for truncated writes
+ */
+ssize_t netfs_unbuffered_write_iter(struct kiocb *iocb, struct iov_iter *from)
+{
+	struct file *file = iocb->ki_filp;
+	struct inode *inode = file->f_mapping->host;
+	struct netfs_inode *ictx = netfs_inode(inode);
+	unsigned long long end;
+	ssize_t ret;
+
+	_enter("%llx,%zx,%llx", iocb->ki_pos, iov_iter_count(from), i_size_read(inode));
+
+	trace_netfs_write_iter(iocb, from);
+	netfs_stat(&netfs_n_rh_dio_write);
+
+	ret = netfs_start_io_direct(inode);
+	if (ret < 0)
+		return ret;
+	ret = generic_write_checks(iocb, from);
+	if (ret < 0)
+		goto out;
+	ret = file_remove_privs(file);
+	if (ret < 0)
+		goto out;
+	ret = file_update_time(file);
+	if (ret < 0)
+		goto out;
+	ret = kiocb_invalidate_pages(iocb, iov_iter_count(from));
+	if (ret < 0)
+		goto out;
+	end = iocb->ki_pos + iov_iter_count(from);
+	if (end > ictx->zero_point)
+		ictx->zero_point = end;
+
+	fscache_invalidate(netfs_i_cookie(ictx), NULL, i_size_read(inode),
+			   FSCACHE_INVAL_DIO_WRITE);
+	ret = netfs_unbuffered_write_iter_locked(iocb, from, NULL);
+out:
+	netfs_end_io_direct(inode);
+	return ret;
+}
+EXPORT_SYMBOL(netfs_unbuffered_write_iter);
diff --git a/fs/fscache/cache.c b/fs/netfs/fscache_cache.c
index d645f8b302a2..9397ed39b0b4 100644
--- a/fs/fscache/cache.c
+++ b/fs/netfs/fscache_cache.c
@@ -179,13 +179,14 @@ EXPORT_SYMBOL(fscache_acquire_cache);
 void fscache_put_cache(struct fscache_cache *cache,
 		       enum fscache_cache_trace where)
 {
-	unsigned int debug_id = cache->debug_id;
+	unsigned int debug_id;
 	bool zero;
 	int ref;
 
 	if (IS_ERR_OR_NULL(cache))
 		return;
 
+	debug_id = cache->debug_id;
 	zero = __refcount_dec_and_test(&cache->ref, &ref);
 	trace_fscache_cache(debug_id, ref - 1, where);
 
diff --git a/fs/fscache/cookie.c b/fs/netfs/fscache_cookie.c
index bce2492186d0..bce2492186d0 100644
--- a/fs/fscache/cookie.c
+++ b/fs/netfs/fscache_cookie.c
diff --git a/fs/netfs/fscache_internal.h b/fs/netfs/fscache_internal.h
new file mode 100644
index 000000000000..a09b948fcef2
--- /dev/null
+++ b/fs/netfs/fscache_internal.h
@@ -0,0 +1,14 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/* Internal definitions for FS-Cache
+ *
+ * Copyright (C) 2021 Red Hat, Inc. All Rights Reserved.
+ * Written by David Howells (dhowells@redhat.com)
+ */
+
+#include "internal.h"
+
+#ifdef pr_fmt
+#undef pr_fmt
+#endif
+
+#define pr_fmt(fmt) "FS-Cache: " fmt
diff --git a/fs/fscache/io.c b/fs/netfs/fscache_io.c
index 0d2b8dec8f82..ad572f7ee897 100644
--- a/fs/fscache/io.c
+++ b/fs/netfs/fscache_io.c
@@ -158,46 +158,6 @@ int __fscache_begin_write_operation(struct netfs_cache_resources *cres,
 }
 EXPORT_SYMBOL(__fscache_begin_write_operation);
 
-/**
- * fscache_dirty_folio - Mark folio dirty and pin a cache object for writeback
- * @mapping: The mapping the folio belongs to.
- * @folio: The folio being dirtied.
- * @cookie: The cookie referring to the cache object
- *
- * Set the dirty flag on a folio and pin an in-use cache object in memory
- * so that writeback can later write to it.  This is intended
- * to be called from the filesystem's ->dirty_folio() method.
- *
- * Return: true if the dirty flag was set on the folio, false otherwise.
- */
-bool fscache_dirty_folio(struct address_space *mapping, struct folio *folio,
-				struct fscache_cookie *cookie)
-{
-	struct inode *inode = mapping->host;
-	bool need_use = false;
-
-	_enter("");
-
-	if (!filemap_dirty_folio(mapping, folio))
-		return false;
-	if (!fscache_cookie_valid(cookie))
-		return true;
-
-	if (!(inode->i_state & I_PINNING_FSCACHE_WB)) {
-		spin_lock(&inode->i_lock);
-		if (!(inode->i_state & I_PINNING_FSCACHE_WB)) {
-			inode->i_state |= I_PINNING_FSCACHE_WB;
-			need_use = true;
-		}
-		spin_unlock(&inode->i_lock);
-
-		if (need_use)
-			fscache_use_cookie(cookie, true);
-	}
-	return true;
-}
-EXPORT_SYMBOL(fscache_dirty_folio);
-
 struct fscache_write_request {
 	struct netfs_cache_resources cache_resources;
 	struct address_space	*mapping;
@@ -277,7 +237,7 @@ void __fscache_write_to_cache(struct fscache_cookie *cookie,
 				    fscache_access_io_write) < 0)
 		goto abandon_free;
 
-	ret = cres->ops->prepare_write(cres, &start, &len, i_size, false);
+	ret = cres->ops->prepare_write(cres, &start, &len, len, i_size, false);
 	if (ret < 0)
 		goto abandon_end;
 
diff --git a/fs/fscache/main.c b/fs/netfs/fscache_main.c
index dad85fd84f6f..42e98bb523e3 100644
--- a/fs/fscache/main.c
+++ b/fs/netfs/fscache_main.c
@@ -8,18 +8,9 @@
 #define FSCACHE_DEBUG_LEVEL CACHE
 #include <linux/module.h>
 #include <linux/init.h>
-#define CREATE_TRACE_POINTS
 #include "internal.h"
-
-MODULE_DESCRIPTION("FS Cache Manager");
-MODULE_AUTHOR("Red Hat, Inc.");
-MODULE_LICENSE("GPL");
-
-unsigned fscache_debug;
-module_param_named(debug, fscache_debug, uint,
-		   S_IWUSR | S_IRUGO);
-MODULE_PARM_DESC(fscache_debug,
-		 "FS-Cache debugging mask");
+#define CREATE_TRACE_POINTS
+#include <trace/events/fscache.h>
 
 EXPORT_TRACEPOINT_SYMBOL(fscache_access_cache);
 EXPORT_TRACEPOINT_SYMBOL(fscache_access_volume);
@@ -71,7 +62,7 @@ unsigned int fscache_hash(unsigned int salt, const void *data, size_t len)
 /*
  * initialise the fs caching module
  */
-static int __init fscache_init(void)
+int __init fscache_init(void)
 {
 	int ret = -ENOMEM;
 
@@ -92,7 +83,7 @@ static int __init fscache_init(void)
 		goto error_cookie_jar;
 	}
 
-	pr_notice("Loaded\n");
+	pr_notice("FS-Cache loaded\n");
 	return 0;
 
 error_cookie_jar:
@@ -103,19 +94,15 @@ error_wq:
 	return ret;
 }
 
-fs_initcall(fscache_init);
-
 /*
  * clean up on module removal
  */
-static void __exit fscache_exit(void)
+void __exit fscache_exit(void)
 {
 	_enter("");
 
 	kmem_cache_destroy(fscache_cookie_jar);
 	fscache_proc_cleanup();
 	destroy_workqueue(fscache_wq);
-	pr_notice("Unloaded\n");
+	pr_notice("FS-Cache unloaded\n");
 }
-
-module_exit(fscache_exit);
diff --git a/fs/fscache/proc.c b/fs/netfs/fscache_proc.c
index dc3b0e9c8cce..874d951bc390 100644
--- a/fs/fscache/proc.c
+++ b/fs/netfs/fscache_proc.c
@@ -12,41 +12,34 @@
 #include "internal.h"
 
 /*
- * initialise the /proc/fs/fscache/ directory
+ * Add files to /proc/fs/netfs/.
  */
 int __init fscache_proc_init(void)
 {
-	if (!proc_mkdir("fs/fscache", NULL))
-		goto error_dir;
+	if (!proc_symlink("fs/fscache", NULL, "netfs"))
+		goto error_sym;
 
-	if (!proc_create_seq("fs/fscache/caches", S_IFREG | 0444, NULL,
+	if (!proc_create_seq("fs/netfs/caches", S_IFREG | 0444, NULL,
 			     &fscache_caches_seq_ops))
 		goto error;
 
-	if (!proc_create_seq("fs/fscache/volumes", S_IFREG | 0444, NULL,
+	if (!proc_create_seq("fs/netfs/volumes", S_IFREG | 0444, NULL,
 			     &fscache_volumes_seq_ops))
 		goto error;
 
-	if (!proc_create_seq("fs/fscache/cookies", S_IFREG | 0444, NULL,
+	if (!proc_create_seq("fs/netfs/cookies", S_IFREG | 0444, NULL,
 			     &fscache_cookies_seq_ops))
 		goto error;
-
-#ifdef CONFIG_FSCACHE_STATS
-	if (!proc_create_single("fs/fscache/stats", S_IFREG | 0444, NULL,
-				fscache_stats_show))
-		goto error;
-#endif
-
 	return 0;
 
 error:
 	remove_proc_entry("fs/fscache", NULL);
-error_dir:
+error_sym:
 	return -ENOMEM;
 }
 
 /*
- * clean up the /proc/fs/fscache/ directory
+ * Clean up the /proc/fs/fscache symlink.
  */
 void fscache_proc_cleanup(void)
 {
diff --git a/fs/fscache/stats.c b/fs/netfs/fscache_stats.c
index fc94e5e79f1c..add21abdf713 100644
--- a/fs/fscache/stats.c
+++ b/fs/netfs/fscache_stats.c
@@ -48,13 +48,15 @@ atomic_t fscache_n_no_create_space;
 EXPORT_SYMBOL(fscache_n_no_create_space);
 atomic_t fscache_n_culled;
 EXPORT_SYMBOL(fscache_n_culled);
+atomic_t fscache_n_dio_misfit;
+EXPORT_SYMBOL(fscache_n_dio_misfit);
 
 /*
  * display the general statistics
  */
-int fscache_stats_show(struct seq_file *m, void *v)
+int fscache_stats_show(struct seq_file *m)
 {
-	seq_puts(m, "FS-Cache statistics\n");
+	seq_puts(m, "-- FS-Cache statistics --\n");
 	seq_printf(m, "Cookies: n=%d v=%d vcol=%u voom=%u\n",
 		   atomic_read(&fscache_n_cookies),
 		   atomic_read(&fscache_n_volumes),
@@ -93,10 +95,9 @@ int fscache_stats_show(struct seq_file *m, void *v)
 		   atomic_read(&fscache_n_no_create_space),
 		   atomic_read(&fscache_n_culled));
 
-	seq_printf(m, "IO     : rd=%u wr=%u\n",
+	seq_printf(m, "IO     : rd=%u wr=%u mis=%u\n",
 		   atomic_read(&fscache_n_read),
-		   atomic_read(&fscache_n_write));
-
-	netfs_stats_show(m);
+		   atomic_read(&fscache_n_write),
+		   atomic_read(&fscache_n_dio_misfit));
 	return 0;
 }
diff --git a/fs/fscache/volume.c b/fs/netfs/fscache_volume.c
index cdf991bdd9de..cdf991bdd9de 100644
--- a/fs/fscache/volume.c
+++ b/fs/netfs/fscache_volume.c
diff --git a/fs/netfs/internal.h b/fs/netfs/internal.h
index 43fac1b14e40..ec7045d24400 100644
--- a/fs/netfs/internal.h
+++ b/fs/netfs/internal.h
@@ -5,9 +5,13 @@
  * Written by David Howells (dhowells@redhat.com)
  */
 
+#include <linux/slab.h>
+#include <linux/seq_file.h>
 #include <linux/netfs.h>
 #include <linux/fscache.h>
+#include <linux/fscache-cache.h>
 #include <trace/events/netfs.h>
+#include <trace/events/fscache.h>
 
 #ifdef pr_fmt
 #undef pr_fmt
@@ -19,6 +23,8 @@
  * buffered_read.c
  */
 void netfs_rreq_unlock_folios(struct netfs_io_request *rreq);
+int netfs_prefetch_for_write(struct file *file, struct folio *folio,
+			     size_t offset, size_t len);
 
 /*
  * io.c
@@ -29,6 +35,41 @@ int netfs_begin_read(struct netfs_io_request *rreq, bool sync);
  * main.c
  */
 extern unsigned int netfs_debug;
+extern struct list_head netfs_io_requests;
+extern spinlock_t netfs_proc_lock;
+
+#ifdef CONFIG_PROC_FS
+static inline void netfs_proc_add_rreq(struct netfs_io_request *rreq)
+{
+	spin_lock(&netfs_proc_lock);
+	list_add_tail_rcu(&rreq->proc_link, &netfs_io_requests);
+	spin_unlock(&netfs_proc_lock);
+}
+static inline void netfs_proc_del_rreq(struct netfs_io_request *rreq)
+{
+	if (!list_empty(&rreq->proc_link)) {
+		spin_lock(&netfs_proc_lock);
+		list_del_rcu(&rreq->proc_link);
+		spin_unlock(&netfs_proc_lock);
+	}
+}
+#else
+static inline void netfs_proc_add_rreq(struct netfs_io_request *rreq) {}
+static inline void netfs_proc_del_rreq(struct netfs_io_request *rreq) {}
+#endif
+
+/*
+ * misc.c
+ */
+#define NETFS_FLAG_PUT_MARK		BIT(0)
+#define NETFS_FLAG_PAGECACHE_MARK	BIT(1)
+int netfs_xa_store_and_mark(struct xarray *xa, unsigned long index,
+			    struct folio *folio, unsigned int flags,
+			    gfp_t gfp_mask);
+int netfs_add_folios_to_buffer(struct xarray *buffer,
+			       struct address_space *mapping,
+			       pgoff_t index, pgoff_t to, gfp_t gfp_mask);
+void netfs_clear_buffer(struct xarray *buffer);
 
 /*
  * objects.c
@@ -50,9 +91,20 @@ static inline void netfs_see_request(struct netfs_io_request *rreq,
 }
 
 /*
+ * output.c
+ */
+int netfs_begin_write(struct netfs_io_request *wreq, bool may_wait,
+		      enum netfs_write_trace what);
+struct netfs_io_request *netfs_begin_writethrough(struct kiocb *iocb, size_t len);
+int netfs_advance_writethrough(struct netfs_io_request *wreq, size_t copied, bool to_page_end);
+int netfs_end_writethrough(struct netfs_io_request *wreq, struct kiocb *iocb);
+
+/*
  * stats.c
  */
 #ifdef CONFIG_NETFS_STATS
+extern atomic_t netfs_n_rh_dio_read;
+extern atomic_t netfs_n_rh_dio_write;
 extern atomic_t netfs_n_rh_readahead;
 extern atomic_t netfs_n_rh_readpage;
 extern atomic_t netfs_n_rh_rreq;
@@ -71,7 +123,15 @@ extern atomic_t netfs_n_rh_write_begin;
 extern atomic_t netfs_n_rh_write_done;
 extern atomic_t netfs_n_rh_write_failed;
 extern atomic_t netfs_n_rh_write_zskip;
+extern atomic_t netfs_n_wh_wstream_conflict;
+extern atomic_t netfs_n_wh_upload;
+extern atomic_t netfs_n_wh_upload_done;
+extern atomic_t netfs_n_wh_upload_failed;
+extern atomic_t netfs_n_wh_write;
+extern atomic_t netfs_n_wh_write_done;
+extern atomic_t netfs_n_wh_write_failed;
 
+int netfs_stats_show(struct seq_file *m, void *v);
 
 static inline void netfs_stat(atomic_t *stat)
 {
@@ -103,6 +163,176 @@ static inline bool netfs_is_cache_enabled(struct netfs_inode *ctx)
 #endif
 }
 
+/*
+ * Get a ref on a netfs group attached to a dirty page (e.g. a ceph snap).
+ */
+static inline struct netfs_group *netfs_get_group(struct netfs_group *netfs_group)
+{
+	if (netfs_group)
+		refcount_inc(&netfs_group->ref);
+	return netfs_group;
+}
+
+/*
+ * Dispose of a netfs group attached to a dirty page (e.g. a ceph snap).
+ */
+static inline void netfs_put_group(struct netfs_group *netfs_group)
+{
+	if (netfs_group && refcount_dec_and_test(&netfs_group->ref))
+		netfs_group->free(netfs_group);
+}
+
+/*
+ * Dispose of a netfs group attached to a dirty page (e.g. a ceph snap).
+ */
+static inline void netfs_put_group_many(struct netfs_group *netfs_group, int nr)
+{
+	if (netfs_group && refcount_sub_and_test(nr, &netfs_group->ref))
+		netfs_group->free(netfs_group);
+}
+
+/*
+ * fscache-cache.c
+ */
+#ifdef CONFIG_PROC_FS
+extern const struct seq_operations fscache_caches_seq_ops;
+#endif
+bool fscache_begin_cache_access(struct fscache_cache *cache, enum fscache_access_trace why);
+void fscache_end_cache_access(struct fscache_cache *cache, enum fscache_access_trace why);
+struct fscache_cache *fscache_lookup_cache(const char *name, bool is_cache);
+void fscache_put_cache(struct fscache_cache *cache, enum fscache_cache_trace where);
+
+static inline enum fscache_cache_state fscache_cache_state(const struct fscache_cache *cache)
+{
+	return smp_load_acquire(&cache->state);
+}
+
+static inline bool fscache_cache_is_live(const struct fscache_cache *cache)
+{
+	return fscache_cache_state(cache) == FSCACHE_CACHE_IS_ACTIVE;
+}
+
+static inline void fscache_set_cache_state(struct fscache_cache *cache,
+					   enum fscache_cache_state new_state)
+{
+	smp_store_release(&cache->state, new_state);
+
+}
+
+static inline bool fscache_set_cache_state_maybe(struct fscache_cache *cache,
+						 enum fscache_cache_state old_state,
+						 enum fscache_cache_state new_state)
+{
+	return try_cmpxchg_release(&cache->state, &old_state, new_state);
+}
+
+/*
+ * fscache-cookie.c
+ */
+extern struct kmem_cache *fscache_cookie_jar;
+#ifdef CONFIG_PROC_FS
+extern const struct seq_operations fscache_cookies_seq_ops;
+#endif
+extern struct timer_list fscache_cookie_lru_timer;
+
+extern void fscache_print_cookie(struct fscache_cookie *cookie, char prefix);
+extern bool fscache_begin_cookie_access(struct fscache_cookie *cookie,
+					enum fscache_access_trace why);
+
+static inline void fscache_see_cookie(struct fscache_cookie *cookie,
+				      enum fscache_cookie_trace where)
+{
+	trace_fscache_cookie(cookie->debug_id, refcount_read(&cookie->ref),
+			     where);
+}
+
+/*
+ * fscache-main.c
+ */
+extern unsigned int fscache_hash(unsigned int salt, const void *data, size_t len);
+#ifdef CONFIG_FSCACHE
+int __init fscache_init(void);
+void __exit fscache_exit(void);
+#else
+static inline int fscache_init(void) { return 0; }
+static inline void fscache_exit(void) {}
+#endif
+
+/*
+ * fscache-proc.c
+ */
+#ifdef CONFIG_PROC_FS
+extern int __init fscache_proc_init(void);
+extern void fscache_proc_cleanup(void);
+#else
+#define fscache_proc_init()	(0)
+#define fscache_proc_cleanup()	do {} while (0)
+#endif
+
+/*
+ * fscache-stats.c
+ */
+#ifdef CONFIG_FSCACHE_STATS
+extern atomic_t fscache_n_volumes;
+extern atomic_t fscache_n_volumes_collision;
+extern atomic_t fscache_n_volumes_nomem;
+extern atomic_t fscache_n_cookies;
+extern atomic_t fscache_n_cookies_lru;
+extern atomic_t fscache_n_cookies_lru_expired;
+extern atomic_t fscache_n_cookies_lru_removed;
+extern atomic_t fscache_n_cookies_lru_dropped;
+
+extern atomic_t fscache_n_acquires;
+extern atomic_t fscache_n_acquires_ok;
+extern atomic_t fscache_n_acquires_oom;
+
+extern atomic_t fscache_n_invalidates;
+
+extern atomic_t fscache_n_relinquishes;
+extern atomic_t fscache_n_relinquishes_retire;
+extern atomic_t fscache_n_relinquishes_dropped;
+
+extern atomic_t fscache_n_resizes;
+extern atomic_t fscache_n_resizes_null;
+
+static inline void fscache_stat(atomic_t *stat)
+{
+	atomic_inc(stat);
+}
+
+static inline void fscache_stat_d(atomic_t *stat)
+{
+	atomic_dec(stat);
+}
+
+#define __fscache_stat(stat) (stat)
+
+int fscache_stats_show(struct seq_file *m);
+#else
+
+#define __fscache_stat(stat) (NULL)
+#define fscache_stat(stat) do {} while (0)
+#define fscache_stat_d(stat) do {} while (0)
+
+static inline int fscache_stats_show(struct seq_file *m) { return 0; }
+#endif
+
+/*
+ * fscache-volume.c
+ */
+#ifdef CONFIG_PROC_FS
+extern const struct seq_operations fscache_volumes_seq_ops;
+#endif
+
+struct fscache_volume *fscache_get_volume(struct fscache_volume *volume,
+					  enum fscache_volume_trace where);
+void fscache_put_volume(struct fscache_volume *volume,
+			enum fscache_volume_trace where);
+bool fscache_begin_volume_access(struct fscache_volume *volume,
+				 struct fscache_cookie *cookie,
+				 enum fscache_access_trace why);
+void fscache_create_volume(struct fscache_volume *volume, bool wait);
+
 /*****************************************************************************/
 /*
  * debug tracing
@@ -143,3 +373,57 @@ do {						\
 #define _leave(FMT, ...) no_printk("<== %s()"FMT"", __func__, ##__VA_ARGS__)
 #define _debug(FMT, ...) no_printk(FMT, ##__VA_ARGS__)
 #endif
+
+/*
+ * assertions
+ */
+#if 1 /* defined(__KDEBUGALL) */
+
+#define ASSERT(X)							\
+do {									\
+	if (unlikely(!(X))) {						\
+		pr_err("\n");					\
+		pr_err("Assertion failed\n");	\
+		BUG();							\
+	}								\
+} while (0)
+
+#define ASSERTCMP(X, OP, Y)						\
+do {									\
+	if (unlikely(!((X) OP (Y)))) {					\
+		pr_err("\n");					\
+		pr_err("Assertion failed\n");	\
+		pr_err("%lx " #OP " %lx is false\n",		\
+		       (unsigned long)(X), (unsigned long)(Y));		\
+		BUG();							\
+	}								\
+} while (0)
+
+#define ASSERTIF(C, X)							\
+do {									\
+	if (unlikely((C) && !(X))) {					\
+		pr_err("\n");					\
+		pr_err("Assertion failed\n");	\
+		BUG();							\
+	}								\
+} while (0)
+
+#define ASSERTIFCMP(C, X, OP, Y)					\
+do {									\
+	if (unlikely((C) && !((X) OP (Y)))) {				\
+		pr_err("\n");					\
+		pr_err("Assertion failed\n");	\
+		pr_err("%lx " #OP " %lx is false\n",		\
+		       (unsigned long)(X), (unsigned long)(Y));		\
+		BUG();							\
+	}								\
+} while (0)
+
+#else
+
+#define ASSERT(X)			do {} while (0)
+#define ASSERTCMP(X, OP, Y)		do {} while (0)
+#define ASSERTIF(C, X)			do {} while (0)
+#define ASSERTIFCMP(C, X, OP, Y)	do {} while (0)
+
+#endif /* assert or not */
diff --git a/fs/netfs/io.c b/fs/netfs/io.c
index 7f753380e047..e8ff1e61ce79 100644
--- a/fs/netfs/io.c
+++ b/fs/netfs/io.c
@@ -21,12 +21,7 @@
  */
 static void netfs_clear_unread(struct netfs_io_subrequest *subreq)
 {
-	struct iov_iter iter;
-
-	iov_iter_xarray(&iter, ITER_DEST, &subreq->rreq->mapping->i_pages,
-			subreq->start + subreq->transferred,
-			subreq->len   - subreq->transferred);
-	iov_iter_zero(iov_iter_count(&iter), &iter);
+	iov_iter_zero(iov_iter_count(&subreq->io_iter), &subreq->io_iter);
 }
 
 static void netfs_cache_read_terminated(void *priv, ssize_t transferred_or_error,
@@ -46,14 +41,9 @@ static void netfs_read_from_cache(struct netfs_io_request *rreq,
 				  enum netfs_read_from_hole read_hole)
 {
 	struct netfs_cache_resources *cres = &rreq->cache_resources;
-	struct iov_iter iter;
 
 	netfs_stat(&netfs_n_rh_read);
-	iov_iter_xarray(&iter, ITER_DEST, &rreq->mapping->i_pages,
-			subreq->start + subreq->transferred,
-			subreq->len   - subreq->transferred);
-
-	cres->ops->read(cres, subreq->start, &iter, read_hole,
+	cres->ops->read(cres, subreq->start, &subreq->io_iter, read_hole,
 			netfs_cache_read_terminated, subreq);
 }
 
@@ -88,6 +78,13 @@ static void netfs_read_from_server(struct netfs_io_request *rreq,
 				   struct netfs_io_subrequest *subreq)
 {
 	netfs_stat(&netfs_n_rh_download);
+
+	if (rreq->origin != NETFS_DIO_READ &&
+	    iov_iter_count(&subreq->io_iter) != subreq->len - subreq->transferred)
+		pr_warn("R=%08x[%u] ITER PRE-MISMATCH %zx != %zx-%zx %lx\n",
+			rreq->debug_id, subreq->debug_index,
+			iov_iter_count(&subreq->io_iter), subreq->len,
+			subreq->transferred, subreq->flags);
 	rreq->netfs_ops->issue_read(subreq);
 }
 
@@ -127,9 +124,10 @@ static void netfs_rreq_unmark_after_write(struct netfs_io_request *rreq,
 			/* We might have multiple writes from the same huge
 			 * folio, but we mustn't unlock a folio more than once.
 			 */
-			if (have_unlocked && folio_index(folio) <= unlocked)
+			if (have_unlocked && folio->index <= unlocked)
 				continue;
-			unlocked = folio_index(folio);
+			unlocked = folio_next_index(folio) - 1;
+			trace_netfs_folio(folio, netfs_folio_trace_end_copy);
 			folio_end_fscache(folio);
 			have_unlocked = true;
 		}
@@ -201,7 +199,7 @@ static void netfs_rreq_do_write_to_cache(struct netfs_io_request *rreq)
 		}
 
 		ret = cres->ops->prepare_write(cres, &subreq->start, &subreq->len,
-					       rreq->i_size, true);
+					       subreq->len, rreq->i_size, true);
 		if (ret < 0) {
 			trace_netfs_failure(rreq, subreq, ret, netfs_fail_prepare_write);
 			trace_netfs_sreq(subreq, netfs_sreq_trace_write_skip);
@@ -260,6 +258,30 @@ static void netfs_rreq_short_read(struct netfs_io_request *rreq,
 }
 
 /*
+ * Reset the subrequest iterator prior to resubmission.
+ */
+static void netfs_reset_subreq_iter(struct netfs_io_request *rreq,
+				    struct netfs_io_subrequest *subreq)
+{
+	size_t remaining = subreq->len - subreq->transferred;
+	size_t count = iov_iter_count(&subreq->io_iter);
+
+	if (count == remaining)
+		return;
+
+	_debug("R=%08x[%u] ITER RESUB-MISMATCH %zx != %zx-%zx-%llx %x\n",
+	       rreq->debug_id, subreq->debug_index,
+	       iov_iter_count(&subreq->io_iter), subreq->transferred,
+	       subreq->len, rreq->i_size,
+	       subreq->io_iter.iter_type);
+
+	if (count < remaining)
+		iov_iter_revert(&subreq->io_iter, remaining - count);
+	else
+		iov_iter_advance(&subreq->io_iter, count - remaining);
+}
+
+/*
  * Resubmit any short or failed operations.  Returns true if we got the rreq
  * ref back.
  */
@@ -287,6 +309,7 @@ static bool netfs_rreq_perform_resubmissions(struct netfs_io_request *rreq)
 			trace_netfs_sreq(subreq, netfs_sreq_trace_download_instead);
 			netfs_get_subrequest(subreq, netfs_sreq_trace_get_resubmit);
 			atomic_inc(&rreq->nr_outstanding);
+			netfs_reset_subreq_iter(rreq, subreq);
 			netfs_read_from_server(rreq, subreq);
 		} else if (test_bit(NETFS_SREQ_SHORT_IO, &subreq->flags)) {
 			netfs_rreq_short_read(rreq, subreq);
@@ -321,6 +344,43 @@ static void netfs_rreq_is_still_valid(struct netfs_io_request *rreq)
 }
 
 /*
+ * Determine how much we can admit to having read from a DIO read.
+ */
+static void netfs_rreq_assess_dio(struct netfs_io_request *rreq)
+{
+	struct netfs_io_subrequest *subreq;
+	unsigned int i;
+	size_t transferred = 0;
+
+	for (i = 0; i < rreq->direct_bv_count; i++)
+		flush_dcache_page(rreq->direct_bv[i].bv_page);
+
+	list_for_each_entry(subreq, &rreq->subrequests, rreq_link) {
+		if (subreq->error || subreq->transferred == 0)
+			break;
+		transferred += subreq->transferred;
+		if (subreq->transferred < subreq->len)
+			break;
+	}
+
+	for (i = 0; i < rreq->direct_bv_count; i++)
+		flush_dcache_page(rreq->direct_bv[i].bv_page);
+
+	rreq->transferred = transferred;
+	task_io_account_read(transferred);
+
+	if (rreq->iocb) {
+		rreq->iocb->ki_pos += transferred;
+		if (rreq->iocb->ki_complete)
+			rreq->iocb->ki_complete(
+				rreq->iocb, rreq->error ? rreq->error : transferred);
+	}
+	if (rreq->netfs_ops->done)
+		rreq->netfs_ops->done(rreq);
+	inode_dio_end(rreq->inode);
+}
+
+/*
  * Assess the state of a read request and decide what to do next.
  *
  * Note that we could be in an ordinary kernel thread, on a workqueue or in
@@ -340,8 +400,12 @@ again:
 		return;
 	}
 
-	netfs_rreq_unlock_folios(rreq);
+	if (rreq->origin != NETFS_DIO_READ)
+		netfs_rreq_unlock_folios(rreq);
+	else
+		netfs_rreq_assess_dio(rreq);
 
+	trace_netfs_rreq(rreq, netfs_rreq_trace_wake_ip);
 	clear_bit_unlock(NETFS_RREQ_IN_PROGRESS, &rreq->flags);
 	wake_up_bit(&rreq->flags, NETFS_RREQ_IN_PROGRESS);
 
@@ -399,9 +463,9 @@ void netfs_subreq_terminated(struct netfs_io_subrequest *subreq,
 	struct netfs_io_request *rreq = subreq->rreq;
 	int u;
 
-	_enter("[%u]{%llx,%lx},%zd",
-	       subreq->debug_index, subreq->start, subreq->flags,
-	       transferred_or_error);
+	_enter("R=%x[%x]{%llx,%lx},%zd",
+	       rreq->debug_id, subreq->debug_index,
+	       subreq->start, subreq->flags, transferred_or_error);
 
 	switch (subreq->source) {
 	case NETFS_READ_FROM_CACHE:
@@ -501,15 +565,20 @@ static enum netfs_io_source netfs_cache_prepare_read(struct netfs_io_subrequest
  */
 static enum netfs_io_source
 netfs_rreq_prepare_read(struct netfs_io_request *rreq,
-			struct netfs_io_subrequest *subreq)
+			struct netfs_io_subrequest *subreq,
+			struct iov_iter *io_iter)
 {
-	enum netfs_io_source source;
+	enum netfs_io_source source = NETFS_DOWNLOAD_FROM_SERVER;
+	struct netfs_inode *ictx = netfs_inode(rreq->inode);
+	size_t lsize;
 
 	_enter("%llx-%llx,%llx", subreq->start, subreq->start + subreq->len, rreq->i_size);
 
-	source = netfs_cache_prepare_read(subreq, rreq->i_size);
-	if (source == NETFS_INVALID_READ)
-		goto out;
+	if (rreq->origin != NETFS_DIO_READ) {
+		source = netfs_cache_prepare_read(subreq, rreq->i_size);
+		if (source == NETFS_INVALID_READ)
+			goto out;
+	}
 
 	if (source == NETFS_DOWNLOAD_FROM_SERVER) {
 		/* Call out to the netfs to let it shrink the request to fit
@@ -518,19 +587,52 @@ netfs_rreq_prepare_read(struct netfs_io_request *rreq,
 		 * to make serial calls, it can indicate a short read and then
 		 * we will call it again.
 		 */
+		if (rreq->origin != NETFS_DIO_READ) {
+			if (subreq->start >= ictx->zero_point) {
+				source = NETFS_FILL_WITH_ZEROES;
+				goto set;
+			}
+			if (subreq->len > ictx->zero_point - subreq->start)
+				subreq->len = ictx->zero_point - subreq->start;
+		}
 		if (subreq->len > rreq->i_size - subreq->start)
 			subreq->len = rreq->i_size - subreq->start;
+		if (rreq->rsize && subreq->len > rreq->rsize)
+			subreq->len = rreq->rsize;
 
 		if (rreq->netfs_ops->clamp_length &&
 		    !rreq->netfs_ops->clamp_length(subreq)) {
 			source = NETFS_INVALID_READ;
 			goto out;
 		}
+
+		if (subreq->max_nr_segs) {
+			lsize = netfs_limit_iter(io_iter, 0, subreq->len,
+						 subreq->max_nr_segs);
+			if (subreq->len > lsize) {
+				subreq->len = lsize;
+				trace_netfs_sreq(subreq, netfs_sreq_trace_limited);
+			}
+		}
 	}
 
-	if (WARN_ON(subreq->len == 0))
+set:
+	if (subreq->len > rreq->len)
+		pr_warn("R=%08x[%u] SREQ>RREQ %zx > %zx\n",
+			rreq->debug_id, subreq->debug_index,
+			subreq->len, rreq->len);
+
+	if (WARN_ON(subreq->len == 0)) {
 		source = NETFS_INVALID_READ;
+		goto out;
+	}
 
+	subreq->source = source;
+	trace_netfs_sreq(subreq, netfs_sreq_trace_prepare);
+
+	subreq->io_iter = *io_iter;
+	iov_iter_truncate(&subreq->io_iter, subreq->len);
+	iov_iter_advance(io_iter, subreq->len);
 out:
 	subreq->source = source;
 	trace_netfs_sreq(subreq, netfs_sreq_trace_prepare);
@@ -541,6 +643,7 @@ out:
  * Slice off a piece of a read request and submit an I/O request for it.
  */
 static bool netfs_rreq_submit_slice(struct netfs_io_request *rreq,
+				    struct iov_iter *io_iter,
 				    unsigned int *_debug_index)
 {
 	struct netfs_io_subrequest *subreq;
@@ -552,7 +655,7 @@ static bool netfs_rreq_submit_slice(struct netfs_io_request *rreq,
 
 	subreq->debug_index	= (*_debug_index)++;
 	subreq->start		= rreq->start + rreq->submitted;
-	subreq->len		= rreq->len   - rreq->submitted;
+	subreq->len		= io_iter->count;
 
 	_debug("slice %llx,%zx,%zx", subreq->start, subreq->len, rreq->submitted);
 	list_add_tail(&subreq->rreq_link, &rreq->subrequests);
@@ -565,7 +668,7 @@ static bool netfs_rreq_submit_slice(struct netfs_io_request *rreq,
 	 * (the starts must coincide), in which case, we go around the loop
 	 * again and ask it to download the next piece.
 	 */
-	source = netfs_rreq_prepare_read(rreq, subreq);
+	source = netfs_rreq_prepare_read(rreq, subreq, io_iter);
 	if (source == NETFS_INVALID_READ)
 		goto subreq_failed;
 
@@ -603,6 +706,7 @@ subreq_failed:
  */
 int netfs_begin_read(struct netfs_io_request *rreq, bool sync)
 {
+	struct iov_iter io_iter;
 	unsigned int debug_index = 0;
 	int ret;
 
@@ -611,50 +715,71 @@ int netfs_begin_read(struct netfs_io_request *rreq, bool sync)
 
 	if (rreq->len == 0) {
 		pr_err("Zero-sized read [R=%x]\n", rreq->debug_id);
-		netfs_put_request(rreq, false, netfs_rreq_trace_put_zero_len);
 		return -EIO;
 	}
 
-	INIT_WORK(&rreq->work, netfs_rreq_work);
+	if (rreq->origin == NETFS_DIO_READ)
+		inode_dio_begin(rreq->inode);
 
-	if (sync)
-		netfs_get_request(rreq, netfs_rreq_trace_get_hold);
+	// TODO: Use bounce buffer if requested
+	rreq->io_iter = rreq->iter;
+
+	INIT_WORK(&rreq->work, netfs_rreq_work);
 
 	/* Chop the read into slices according to what the cache and the netfs
 	 * want and submit each one.
 	 */
+	netfs_get_request(rreq, netfs_rreq_trace_get_for_outstanding);
 	atomic_set(&rreq->nr_outstanding, 1);
+	io_iter = rreq->io_iter;
 	do {
-		if (!netfs_rreq_submit_slice(rreq, &debug_index))
+		_debug("submit %llx + %zx >= %llx",
+		       rreq->start, rreq->submitted, rreq->i_size);
+		if (rreq->origin == NETFS_DIO_READ &&
+		    rreq->start + rreq->submitted >= rreq->i_size)
+			break;
+		if (!netfs_rreq_submit_slice(rreq, &io_iter, &debug_index))
+			break;
+		if (test_bit(NETFS_RREQ_BLOCKED, &rreq->flags) &&
+		    test_bit(NETFS_RREQ_NONBLOCK, &rreq->flags))
 			break;
 
 	} while (rreq->submitted < rreq->len);
 
+	if (!rreq->submitted) {
+		netfs_put_request(rreq, false, netfs_rreq_trace_put_no_submit);
+		ret = 0;
+		goto out;
+	}
+
 	if (sync) {
-		/* Keep nr_outstanding incremented so that the ref always belongs to
-		 * us, and the service code isn't punted off to a random thread pool to
-		 * process.
+		/* Keep nr_outstanding incremented so that the ref always
+		 * belongs to us, and the service code isn't punted off to a
+		 * random thread pool to process.  Note that this might start
+		 * further work, such as writing to the cache.
 		 */
-		for (;;) {
-			wait_var_event(&rreq->nr_outstanding,
-				       atomic_read(&rreq->nr_outstanding) == 1);
+		wait_var_event(&rreq->nr_outstanding,
+			       atomic_read(&rreq->nr_outstanding) == 1);
+		if (atomic_dec_and_test(&rreq->nr_outstanding))
 			netfs_rreq_assess(rreq, false);
-			if (!test_bit(NETFS_RREQ_IN_PROGRESS, &rreq->flags))
-				break;
-			cond_resched();
-		}
+
+		trace_netfs_rreq(rreq, netfs_rreq_trace_wait_ip);
+		wait_on_bit(&rreq->flags, NETFS_RREQ_IN_PROGRESS,
+			    TASK_UNINTERRUPTIBLE);
 
 		ret = rreq->error;
-		if (ret == 0 && rreq->submitted < rreq->len) {
+		if (ret == 0 && rreq->submitted < rreq->len &&
+		    rreq->origin != NETFS_DIO_READ) {
 			trace_netfs_failure(rreq, NULL, ret, netfs_fail_short_read);
 			ret = -EIO;
 		}
-		netfs_put_request(rreq, false, netfs_rreq_trace_put_hold);
 	} else {
 		/* If we decrement nr_outstanding to 0, the ref belongs to us. */
 		if (atomic_dec_and_test(&rreq->nr_outstanding))
 			netfs_rreq_assess(rreq, false);
-		ret = 0;
+		ret = -EIOCBQUEUED;
 	}
+
+out:
 	return ret;
 }
diff --git a/fs/netfs/iterator.c b/fs/netfs/iterator.c
index 2ff07ba655a0..b781bbbf1d8d 100644
--- a/fs/netfs/iterator.c
+++ b/fs/netfs/iterator.c
@@ -101,3 +101,100 @@ ssize_t netfs_extract_user_iter(struct iov_iter *orig, size_t orig_len,
 	return npages;
 }
 EXPORT_SYMBOL_GPL(netfs_extract_user_iter);
+
+/*
+ * Select the span of a bvec iterator we're going to use.  Limit it by both maximum
+ * size and maximum number of segments.  Returns the size of the span in bytes.
+ */
+static size_t netfs_limit_bvec(const struct iov_iter *iter, size_t start_offset,
+			       size_t max_size, size_t max_segs)
+{
+	const struct bio_vec *bvecs = iter->bvec;
+	unsigned int nbv = iter->nr_segs, ix = 0, nsegs = 0;
+	size_t len, span = 0, n = iter->count;
+	size_t skip = iter->iov_offset + start_offset;
+
+	if (WARN_ON(!iov_iter_is_bvec(iter)) ||
+	    WARN_ON(start_offset > n) ||
+	    n == 0)
+		return 0;
+
+	while (n && ix < nbv && skip) {
+		len = bvecs[ix].bv_len;
+		if (skip < len)
+			break;
+		skip -= len;
+		n -= len;
+		ix++;
+	}
+
+	while (n && ix < nbv) {
+		len = min3(n, bvecs[ix].bv_len - skip, max_size);
+		span += len;
+		nsegs++;
+		ix++;
+		if (span >= max_size || nsegs >= max_segs)
+			break;
+		skip = 0;
+		n -= len;
+	}
+
+	return min(span, max_size);
+}
+
+/*
+ * Select the span of an xarray iterator we're going to use.  Limit it by both
+ * maximum size and maximum number of segments.  It is assumed that segments
+ * can be larger than a page in size, provided they're physically contiguous.
+ * Returns the size of the span in bytes.
+ */
+static size_t netfs_limit_xarray(const struct iov_iter *iter, size_t start_offset,
+				 size_t max_size, size_t max_segs)
+{
+	struct folio *folio;
+	unsigned int nsegs = 0;
+	loff_t pos = iter->xarray_start + iter->iov_offset;
+	pgoff_t index = pos / PAGE_SIZE;
+	size_t span = 0, n = iter->count;
+
+	XA_STATE(xas, iter->xarray, index);
+
+	if (WARN_ON(!iov_iter_is_xarray(iter)) ||
+	    WARN_ON(start_offset > n) ||
+	    n == 0)
+		return 0;
+	max_size = min(max_size, n - start_offset);
+
+	rcu_read_lock();
+	xas_for_each(&xas, folio, ULONG_MAX) {
+		size_t offset, flen, len;
+		if (xas_retry(&xas, folio))
+			continue;
+		if (WARN_ON(xa_is_value(folio)))
+			break;
+		if (WARN_ON(folio_test_hugetlb(folio)))
+			break;
+
+		flen = folio_size(folio);
+		offset = offset_in_folio(folio, pos);
+		len = min(max_size, flen - offset);
+		span += len;
+		nsegs++;
+		if (span >= max_size || nsegs >= max_segs)
+			break;
+	}
+
+	rcu_read_unlock();
+	return min(span, max_size);
+}
+
+size_t netfs_limit_iter(const struct iov_iter *iter, size_t start_offset,
+			size_t max_size, size_t max_segs)
+{
+	if (iov_iter_is_bvec(iter))
+		return netfs_limit_bvec(iter, start_offset, max_size, max_segs);
+	if (iov_iter_is_xarray(iter))
+		return netfs_limit_xarray(iter, start_offset, max_size, max_segs);
+	BUG();
+}
+EXPORT_SYMBOL(netfs_limit_iter);
diff --git a/fs/netfs/locking.c b/fs/netfs/locking.c
new file mode 100644
index 000000000000..75dc52a49b3a
--- /dev/null
+++ b/fs/netfs/locking.c
@@ -0,0 +1,216 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * I/O and data path helper functionality.
+ *
+ * Borrowed from NFS Copyright (c) 2016 Trond Myklebust
+ */
+
+#include <linux/kernel.h>
+#include <linux/netfs.h>
+#include "internal.h"
+
+/*
+ * inode_dio_wait_interruptible - wait for outstanding DIO requests to finish
+ * @inode: inode to wait for
+ *
+ * Waits for all pending direct I/O requests to finish so that we can
+ * proceed with a truncate or equivalent operation.
+ *
+ * Must be called under a lock that serializes taking new references
+ * to i_dio_count, usually by inode->i_mutex.
+ */
+static int inode_dio_wait_interruptible(struct inode *inode)
+{
+	if (!atomic_read(&inode->i_dio_count))
+		return 0;
+
+	wait_queue_head_t *wq = bit_waitqueue(&inode->i_state, __I_DIO_WAKEUP);
+	DEFINE_WAIT_BIT(q, &inode->i_state, __I_DIO_WAKEUP);
+
+	for (;;) {
+		prepare_to_wait(wq, &q.wq_entry, TASK_INTERRUPTIBLE);
+		if (!atomic_read(&inode->i_dio_count))
+			break;
+		if (signal_pending(current))
+			break;
+		schedule();
+	}
+	finish_wait(wq, &q.wq_entry);
+
+	return atomic_read(&inode->i_dio_count) ? -ERESTARTSYS : 0;
+}
+
+/* Call with exclusively locked inode->i_rwsem */
+static int netfs_block_o_direct(struct netfs_inode *ictx)
+{
+	if (!test_bit(NETFS_ICTX_ODIRECT, &ictx->flags))
+		return 0;
+	clear_bit(NETFS_ICTX_ODIRECT, &ictx->flags);
+	return inode_dio_wait_interruptible(&ictx->inode);
+}
+
+/**
+ * netfs_start_io_read - declare the file is being used for buffered reads
+ * @inode: file inode
+ *
+ * Declare that a buffered read operation is about to start, and ensure
+ * that we block all direct I/O.
+ * On exit, the function ensures that the NETFS_ICTX_ODIRECT flag is unset,
+ * and holds a shared lock on inode->i_rwsem to ensure that the flag
+ * cannot be changed.
+ * In practice, this means that buffered read operations are allowed to
+ * execute in parallel, thanks to the shared lock, whereas direct I/O
+ * operations need to wait to grab an exclusive lock in order to set
+ * NETFS_ICTX_ODIRECT.
+ * Note that buffered writes and truncates both take a write lock on
+ * inode->i_rwsem, meaning that those are serialised w.r.t. the reads.
+ */
+int netfs_start_io_read(struct inode *inode)
+	__acquires(inode->i_rwsem)
+{
+	struct netfs_inode *ictx = netfs_inode(inode);
+
+	/* Be an optimist! */
+	if (down_read_interruptible(&inode->i_rwsem) < 0)
+		return -ERESTARTSYS;
+	if (test_bit(NETFS_ICTX_ODIRECT, &ictx->flags) == 0)
+		return 0;
+	up_read(&inode->i_rwsem);
+
+	/* Slow path.... */
+	if (down_write_killable(&inode->i_rwsem) < 0)
+		return -ERESTARTSYS;
+	if (netfs_block_o_direct(ictx) < 0) {
+		up_write(&inode->i_rwsem);
+		return -ERESTARTSYS;
+	}
+	downgrade_write(&inode->i_rwsem);
+	return 0;
+}
+EXPORT_SYMBOL(netfs_start_io_read);
+
+/**
+ * netfs_end_io_read - declare that the buffered read operation is done
+ * @inode: file inode
+ *
+ * Declare that a buffered read operation is done, and release the shared
+ * lock on inode->i_rwsem.
+ */
+void netfs_end_io_read(struct inode *inode)
+	__releases(inode->i_rwsem)
+{
+	up_read(&inode->i_rwsem);
+}
+EXPORT_SYMBOL(netfs_end_io_read);
+
+/**
+ * netfs_start_io_write - declare the file is being used for buffered writes
+ * @inode: file inode
+ *
+ * Declare that a buffered read operation is about to start, and ensure
+ * that we block all direct I/O.
+ */
+int netfs_start_io_write(struct inode *inode)
+	__acquires(inode->i_rwsem)
+{
+	struct netfs_inode *ictx = netfs_inode(inode);
+
+	if (down_write_killable(&inode->i_rwsem) < 0)
+		return -ERESTARTSYS;
+	if (netfs_block_o_direct(ictx) < 0) {
+		up_write(&inode->i_rwsem);
+		return -ERESTARTSYS;
+	}
+	return 0;
+}
+EXPORT_SYMBOL(netfs_start_io_write);
+
+/**
+ * netfs_end_io_write - declare that the buffered write operation is done
+ * @inode: file inode
+ *
+ * Declare that a buffered write operation is done, and release the
+ * lock on inode->i_rwsem.
+ */
+void netfs_end_io_write(struct inode *inode)
+	__releases(inode->i_rwsem)
+{
+	up_write(&inode->i_rwsem);
+}
+EXPORT_SYMBOL(netfs_end_io_write);
+
+/* Call with exclusively locked inode->i_rwsem */
+static int netfs_block_buffered(struct inode *inode)
+{
+	struct netfs_inode *ictx = netfs_inode(inode);
+	int ret;
+
+	if (!test_bit(NETFS_ICTX_ODIRECT, &ictx->flags)) {
+		set_bit(NETFS_ICTX_ODIRECT, &ictx->flags);
+		if (inode->i_mapping->nrpages != 0) {
+			unmap_mapping_range(inode->i_mapping, 0, 0, 0);
+			ret = filemap_fdatawait(inode->i_mapping);
+			if (ret < 0) {
+				clear_bit(NETFS_ICTX_ODIRECT, &ictx->flags);
+				return ret;
+			}
+		}
+	}
+	return 0;
+}
+
+/**
+ * netfs_start_io_direct - declare the file is being used for direct i/o
+ * @inode: file inode
+ *
+ * Declare that a direct I/O operation is about to start, and ensure
+ * that we block all buffered I/O.
+ * On exit, the function ensures that the NETFS_ICTX_ODIRECT flag is set,
+ * and holds a shared lock on inode->i_rwsem to ensure that the flag
+ * cannot be changed.
+ * In practice, this means that direct I/O operations are allowed to
+ * execute in parallel, thanks to the shared lock, whereas buffered I/O
+ * operations need to wait to grab an exclusive lock in order to clear
+ * NETFS_ICTX_ODIRECT.
+ * Note that buffered writes and truncates both take a write lock on
+ * inode->i_rwsem, meaning that those are serialised w.r.t. O_DIRECT.
+ */
+int netfs_start_io_direct(struct inode *inode)
+	__acquires(inode->i_rwsem)
+{
+	struct netfs_inode *ictx = netfs_inode(inode);
+	int ret;
+
+	/* Be an optimist! */
+	if (down_read_interruptible(&inode->i_rwsem) < 0)
+		return -ERESTARTSYS;
+	if (test_bit(NETFS_ICTX_ODIRECT, &ictx->flags) != 0)
+		return 0;
+	up_read(&inode->i_rwsem);
+
+	/* Slow path.... */
+	if (down_write_killable(&inode->i_rwsem) < 0)
+		return -ERESTARTSYS;
+	ret = netfs_block_buffered(inode);
+	if (ret < 0) {
+		up_write(&inode->i_rwsem);
+		return ret;
+	}
+	downgrade_write(&inode->i_rwsem);
+	return 0;
+}
+EXPORT_SYMBOL(netfs_start_io_direct);
+
+/**
+ * netfs_end_io_direct - declare that the direct i/o operation is done
+ * @inode: file inode
+ *
+ * Declare that a direct I/O operation is done, and release the shared
+ * lock on inode->i_rwsem.
+ */
+void netfs_end_io_direct(struct inode *inode)
+	__releases(inode->i_rwsem)
+{
+	up_read(&inode->i_rwsem);
+}
+EXPORT_SYMBOL(netfs_end_io_direct);
diff --git a/fs/netfs/main.c b/fs/netfs/main.c
index 068568702957..5e77618a7940 100644
--- a/fs/netfs/main.c
+++ b/fs/netfs/main.c
@@ -7,6 +7,8 @@
 
 #include <linux/module.h>
 #include <linux/export.h>
+#include <linux/proc_fs.h>
+#include <linux/seq_file.h>
 #include "internal.h"
 #define CREATE_TRACE_POINTS
 #include <trace/events/netfs.h>
@@ -15,6 +17,113 @@ MODULE_DESCRIPTION("Network fs support");
 MODULE_AUTHOR("Red Hat, Inc.");
 MODULE_LICENSE("GPL");
 
+EXPORT_TRACEPOINT_SYMBOL(netfs_sreq);
+
 unsigned netfs_debug;
 module_param_named(debug, netfs_debug, uint, S_IWUSR | S_IRUGO);
 MODULE_PARM_DESC(netfs_debug, "Netfs support debugging mask");
+
+#ifdef CONFIG_PROC_FS
+LIST_HEAD(netfs_io_requests);
+DEFINE_SPINLOCK(netfs_proc_lock);
+
+static const char *netfs_origins[nr__netfs_io_origin] = {
+	[NETFS_READAHEAD]		= "RA",
+	[NETFS_READPAGE]		= "RP",
+	[NETFS_READ_FOR_WRITE]		= "RW",
+	[NETFS_WRITEBACK]		= "WB",
+	[NETFS_WRITETHROUGH]		= "WT",
+	[NETFS_LAUNDER_WRITE]		= "LW",
+	[NETFS_UNBUFFERED_WRITE]	= "UW",
+	[NETFS_DIO_READ]		= "DR",
+	[NETFS_DIO_WRITE]		= "DW",
+};
+
+/*
+ * Generate a list of I/O requests in /proc/fs/netfs/requests
+ */
+static int netfs_requests_seq_show(struct seq_file *m, void *v)
+{
+	struct netfs_io_request *rreq;
+
+	if (v == &netfs_io_requests) {
+		seq_puts(m,
+			 "REQUEST  OR REF FL ERR  OPS COVERAGE\n"
+			 "======== == === == ==== === =========\n"
+			 );
+		return 0;
+	}
+
+	rreq = list_entry(v, struct netfs_io_request, proc_link);
+	seq_printf(m,
+		   "%08x %s %3d %2lx %4d %3d @%04llx %zx/%zx",
+		   rreq->debug_id,
+		   netfs_origins[rreq->origin],
+		   refcount_read(&rreq->ref),
+		   rreq->flags,
+		   rreq->error,
+		   atomic_read(&rreq->nr_outstanding),
+		   rreq->start, rreq->submitted, rreq->len);
+	seq_putc(m, '\n');
+	return 0;
+}
+
+static void *netfs_requests_seq_start(struct seq_file *m, loff_t *_pos)
+	__acquires(rcu)
+{
+	rcu_read_lock();
+	return seq_list_start_head(&netfs_io_requests, *_pos);
+}
+
+static void *netfs_requests_seq_next(struct seq_file *m, void *v, loff_t *_pos)
+{
+	return seq_list_next(v, &netfs_io_requests, _pos);
+}
+
+static void netfs_requests_seq_stop(struct seq_file *m, void *v)
+	__releases(rcu)
+{
+	rcu_read_unlock();
+}
+
+static const struct seq_operations netfs_requests_seq_ops = {
+	.start  = netfs_requests_seq_start,
+	.next   = netfs_requests_seq_next,
+	.stop   = netfs_requests_seq_stop,
+	.show   = netfs_requests_seq_show,
+};
+#endif /* CONFIG_PROC_FS */
+
+static int __init netfs_init(void)
+{
+	int ret = -ENOMEM;
+
+	if (!proc_mkdir("fs/netfs", NULL))
+		goto error;
+	if (!proc_create_seq("fs/netfs/requests", S_IFREG | 0444, NULL,
+			     &netfs_requests_seq_ops))
+		goto error_proc;
+#ifdef CONFIG_FSCACHE_STATS
+	if (!proc_create_single("fs/netfs/stats", S_IFREG | 0444, NULL,
+				netfs_stats_show))
+		goto error_proc;
+#endif
+
+	ret = fscache_init();
+	if (ret < 0)
+		goto error_proc;
+	return 0;
+
+error_proc:
+	remove_proc_entry("fs/netfs", NULL);
+error:
+	return ret;
+}
+fs_initcall(netfs_init);
+
+static void __exit netfs_exit(void)
+{
+	fscache_exit();
+	remove_proc_entry("fs/netfs", NULL);
+}
+module_exit(netfs_exit);
diff --git a/fs/netfs/misc.c b/fs/netfs/misc.c
new file mode 100644
index 000000000000..90051ced8e2a
--- /dev/null
+++ b/fs/netfs/misc.c
@@ -0,0 +1,260 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Miscellaneous routines.
+ *
+ * Copyright (C) 2023 Red Hat, Inc. All Rights Reserved.
+ * Written by David Howells (dhowells@redhat.com)
+ */
+
+#include <linux/swap.h>
+#include "internal.h"
+
+/*
+ * Attach a folio to the buffer and maybe set marks on it to say that we need
+ * to put the folio later and twiddle the pagecache flags.
+ */
+int netfs_xa_store_and_mark(struct xarray *xa, unsigned long index,
+			    struct folio *folio, unsigned int flags,
+			    gfp_t gfp_mask)
+{
+	XA_STATE_ORDER(xas, xa, index, folio_order(folio));
+
+retry:
+	xas_lock(&xas);
+	for (;;) {
+		xas_store(&xas, folio);
+		if (!xas_error(&xas))
+			break;
+		xas_unlock(&xas);
+		if (!xas_nomem(&xas, gfp_mask))
+			return xas_error(&xas);
+		goto retry;
+	}
+
+	if (flags & NETFS_FLAG_PUT_MARK)
+		xas_set_mark(&xas, NETFS_BUF_PUT_MARK);
+	if (flags & NETFS_FLAG_PAGECACHE_MARK)
+		xas_set_mark(&xas, NETFS_BUF_PAGECACHE_MARK);
+	xas_unlock(&xas);
+	return xas_error(&xas);
+}
+
+/*
+ * Create the specified range of folios in the buffer attached to the read
+ * request.  The folios are marked with NETFS_BUF_PUT_MARK so that we know that
+ * these need freeing later.
+ */
+int netfs_add_folios_to_buffer(struct xarray *buffer,
+			       struct address_space *mapping,
+			       pgoff_t index, pgoff_t to, gfp_t gfp_mask)
+{
+	struct folio *folio;
+	int ret;
+
+	if (to + 1 == index) /* Page range is inclusive */
+		return 0;
+
+	do {
+		/* TODO: Figure out what order folio can be allocated here */
+		folio = filemap_alloc_folio(readahead_gfp_mask(mapping), 0);
+		if (!folio)
+			return -ENOMEM;
+		folio->index = index;
+		ret = netfs_xa_store_and_mark(buffer, index, folio,
+					      NETFS_FLAG_PUT_MARK, gfp_mask);
+		if (ret < 0) {
+			folio_put(folio);
+			return ret;
+		}
+
+		index += folio_nr_pages(folio);
+	} while (index <= to && index != 0);
+
+	return 0;
+}
+
+/*
+ * Clear an xarray buffer, putting a ref on the folios that have
+ * NETFS_BUF_PUT_MARK set.
+ */
+void netfs_clear_buffer(struct xarray *buffer)
+{
+	struct folio *folio;
+	XA_STATE(xas, buffer, 0);
+
+	rcu_read_lock();
+	xas_for_each_marked(&xas, folio, ULONG_MAX, NETFS_BUF_PUT_MARK) {
+		folio_put(folio);
+	}
+	rcu_read_unlock();
+	xa_destroy(buffer);
+}
+
+/**
+ * netfs_dirty_folio - Mark folio dirty and pin a cache object for writeback
+ * @mapping: The mapping the folio belongs to.
+ * @folio: The folio being dirtied.
+ *
+ * Set the dirty flag on a folio and pin an in-use cache object in memory so
+ * that writeback can later write to it.  This is intended to be called from
+ * the filesystem's ->dirty_folio() method.
+ *
+ * Return: true if the dirty flag was set on the folio, false otherwise.
+ */
+bool netfs_dirty_folio(struct address_space *mapping, struct folio *folio)
+{
+	struct inode *inode = mapping->host;
+	struct netfs_inode *ictx = netfs_inode(inode);
+	struct fscache_cookie *cookie = netfs_i_cookie(ictx);
+	bool need_use = false;
+
+	_enter("");
+
+	if (!filemap_dirty_folio(mapping, folio))
+		return false;
+	if (!fscache_cookie_valid(cookie))
+		return true;
+
+	if (!(inode->i_state & I_PINNING_NETFS_WB)) {
+		spin_lock(&inode->i_lock);
+		if (!(inode->i_state & I_PINNING_NETFS_WB)) {
+			inode->i_state |= I_PINNING_NETFS_WB;
+			need_use = true;
+		}
+		spin_unlock(&inode->i_lock);
+
+		if (need_use)
+			fscache_use_cookie(cookie, true);
+	}
+	return true;
+}
+EXPORT_SYMBOL(netfs_dirty_folio);
+
+/**
+ * netfs_unpin_writeback - Unpin writeback resources
+ * @inode: The inode on which the cookie resides
+ * @wbc: The writeback control
+ *
+ * Unpin the writeback resources pinned by netfs_dirty_folio().  This is
+ * intended to be called as/by the netfs's ->write_inode() method.
+ */
+int netfs_unpin_writeback(struct inode *inode, struct writeback_control *wbc)
+{
+	struct fscache_cookie *cookie = netfs_i_cookie(netfs_inode(inode));
+
+	if (wbc->unpinned_netfs_wb)
+		fscache_unuse_cookie(cookie, NULL, NULL);
+	return 0;
+}
+EXPORT_SYMBOL(netfs_unpin_writeback);
+
+/**
+ * netfs_clear_inode_writeback - Clear writeback resources pinned by an inode
+ * @inode: The inode to clean up
+ * @aux: Auxiliary data to apply to the inode
+ *
+ * Clear any writeback resources held by an inode when the inode is evicted.
+ * This must be called before clear_inode() is called.
+ */
+void netfs_clear_inode_writeback(struct inode *inode, const void *aux)
+{
+	struct fscache_cookie *cookie = netfs_i_cookie(netfs_inode(inode));
+
+	if (inode->i_state & I_PINNING_NETFS_WB) {
+		loff_t i_size = i_size_read(inode);
+		fscache_unuse_cookie(cookie, aux, &i_size);
+	}
+}
+EXPORT_SYMBOL(netfs_clear_inode_writeback);
+
+/**
+ * netfs_invalidate_folio - Invalidate or partially invalidate a folio
+ * @folio: Folio proposed for release
+ * @offset: Offset of the invalidated region
+ * @length: Length of the invalidated region
+ *
+ * Invalidate part or all of a folio for a network filesystem.  The folio will
+ * be removed afterwards if the invalidated region covers the entire folio.
+ */
+void netfs_invalidate_folio(struct folio *folio, size_t offset, size_t length)
+{
+	struct netfs_folio *finfo = NULL;
+	size_t flen = folio_size(folio);
+
+	_enter("{%lx},%zx,%zx", folio->index, offset, length);
+
+	folio_wait_fscache(folio);
+
+	if (!folio_test_private(folio))
+		return;
+
+	finfo = netfs_folio_info(folio);
+
+	if (offset == 0 && length >= flen)
+		goto erase_completely;
+
+	if (finfo) {
+		/* We have a partially uptodate page from a streaming write. */
+		unsigned int fstart = finfo->dirty_offset;
+		unsigned int fend = fstart + finfo->dirty_len;
+		unsigned int end = offset + length;
+
+		if (offset >= fend)
+			return;
+		if (end <= fstart)
+			return;
+		if (offset <= fstart && end >= fend)
+			goto erase_completely;
+		if (offset <= fstart && end > fstart)
+			goto reduce_len;
+		if (offset > fstart && end >= fend)
+			goto move_start;
+		/* A partial write was split.  The caller has already zeroed
+		 * it, so just absorb the hole.
+		 */
+	}
+	return;
+
+erase_completely:
+	netfs_put_group(netfs_folio_group(folio));
+	folio_detach_private(folio);
+	folio_clear_uptodate(folio);
+	kfree(finfo);
+	return;
+reduce_len:
+	finfo->dirty_len = offset + length - finfo->dirty_offset;
+	return;
+move_start:
+	finfo->dirty_len -= offset - finfo->dirty_offset;
+	finfo->dirty_offset = offset;
+}
+EXPORT_SYMBOL(netfs_invalidate_folio);
+
+/**
+ * netfs_release_folio - Try to release a folio
+ * @folio: Folio proposed for release
+ * @gfp: Flags qualifying the release
+ *
+ * Request release of a folio and clean up its private state if it's not busy.
+ * Returns true if the folio can now be released, false if not
+ */
+bool netfs_release_folio(struct folio *folio, gfp_t gfp)
+{
+	struct netfs_inode *ctx = netfs_inode(folio_inode(folio));
+	unsigned long long end;
+
+	end = folio_pos(folio) + folio_size(folio);
+	if (end > ctx->zero_point)
+		ctx->zero_point = end;
+
+	if (folio_test_private(folio))
+		return false;
+	if (folio_test_fscache(folio)) {
+		if (current_is_kswapd() || !(gfp & __GFP_FS))
+			return false;
+		folio_wait_fscache(folio);
+	}
+
+	fscache_note_page_release(netfs_i_cookie(ctx));
+	return true;
+}
+EXPORT_SYMBOL(netfs_release_folio);
diff --git a/fs/netfs/objects.c b/fs/netfs/objects.c
index e17cdf53f6a7..610ceb5bd86c 100644
--- a/fs/netfs/objects.c
+++ b/fs/netfs/objects.c
@@ -20,14 +20,20 @@ struct netfs_io_request *netfs_alloc_request(struct address_space *mapping,
 	struct inode *inode = file ? file_inode(file) : mapping->host;
 	struct netfs_inode *ctx = netfs_inode(inode);
 	struct netfs_io_request *rreq;
+	bool is_unbuffered = (origin == NETFS_UNBUFFERED_WRITE ||
+			      origin == NETFS_DIO_READ ||
+			      origin == NETFS_DIO_WRITE);
+	bool cached = !is_unbuffered && netfs_is_cache_enabled(ctx);
 	int ret;
 
-	rreq = kzalloc(sizeof(struct netfs_io_request), GFP_KERNEL);
+	rreq = kzalloc(ctx->ops->io_request_size ?: sizeof(struct netfs_io_request),
+		       GFP_KERNEL);
 	if (!rreq)
 		return ERR_PTR(-ENOMEM);
 
 	rreq->start	= start;
 	rreq->len	= len;
+	rreq->upper_len	= len;
 	rreq->origin	= origin;
 	rreq->netfs_ops	= ctx->ops;
 	rreq->mapping	= mapping;
@@ -35,8 +41,14 @@ struct netfs_io_request *netfs_alloc_request(struct address_space *mapping,
 	rreq->i_size	= i_size_read(inode);
 	rreq->debug_id	= atomic_inc_return(&debug_ids);
 	INIT_LIST_HEAD(&rreq->subrequests);
+	INIT_WORK(&rreq->work, NULL);
 	refcount_set(&rreq->ref, 1);
+
 	__set_bit(NETFS_RREQ_IN_PROGRESS, &rreq->flags);
+	if (cached)
+		__set_bit(NETFS_RREQ_WRITE_TO_CACHE, &rreq->flags);
+	if (file && file->f_flags & O_NONBLOCK)
+		__set_bit(NETFS_RREQ_NONBLOCK, &rreq->flags);
 	if (rreq->netfs_ops->init_request) {
 		ret = rreq->netfs_ops->init_request(rreq, file);
 		if (ret < 0) {
@@ -45,6 +57,8 @@ struct netfs_io_request *netfs_alloc_request(struct address_space *mapping,
 		}
 	}
 
+	trace_netfs_rreq_ref(rreq->debug_id, 1, netfs_rreq_trace_new);
+	netfs_proc_add_rreq(rreq);
 	netfs_stat(&netfs_n_rh_rreq);
 	return rreq;
 }
@@ -74,33 +88,47 @@ static void netfs_free_request(struct work_struct *work)
 {
 	struct netfs_io_request *rreq =
 		container_of(work, struct netfs_io_request, work);
+	unsigned int i;
 
 	trace_netfs_rreq(rreq, netfs_rreq_trace_free);
+	netfs_proc_del_rreq(rreq);
 	netfs_clear_subrequests(rreq, false);
 	if (rreq->netfs_ops->free_request)
 		rreq->netfs_ops->free_request(rreq);
 	if (rreq->cache_resources.ops)
 		rreq->cache_resources.ops->end_operation(&rreq->cache_resources);
-	kfree(rreq);
+	if (rreq->direct_bv) {
+		for (i = 0; i < rreq->direct_bv_count; i++) {
+			if (rreq->direct_bv[i].bv_page) {
+				if (rreq->direct_bv_unpin)
+					unpin_user_page(rreq->direct_bv[i].bv_page);
+			}
+		}
+		kvfree(rreq->direct_bv);
+	}
+	kfree_rcu(rreq, rcu);
 	netfs_stat_d(&netfs_n_rh_rreq);
 }
 
 void netfs_put_request(struct netfs_io_request *rreq, bool was_async,
 		       enum netfs_rreq_ref_trace what)
 {
-	unsigned int debug_id = rreq->debug_id;
+	unsigned int debug_id;
 	bool dead;
 	int r;
 
-	dead = __refcount_dec_and_test(&rreq->ref, &r);
-	trace_netfs_rreq_ref(debug_id, r - 1, what);
-	if (dead) {
-		if (was_async) {
-			rreq->work.func = netfs_free_request;
-			if (!queue_work(system_unbound_wq, &rreq->work))
-				BUG();
-		} else {
-			netfs_free_request(&rreq->work);
+	if (rreq) {
+		debug_id = rreq->debug_id;
+		dead = __refcount_dec_and_test(&rreq->ref, &r);
+		trace_netfs_rreq_ref(debug_id, r - 1, what);
+		if (dead) {
+			if (was_async) {
+				rreq->work.func = netfs_free_request;
+				if (!queue_work(system_unbound_wq, &rreq->work))
+					BUG();
+			} else {
+				netfs_free_request(&rreq->work);
+			}
 		}
 	}
 }
@@ -112,8 +140,11 @@ struct netfs_io_subrequest *netfs_alloc_subrequest(struct netfs_io_request *rreq
 {
 	struct netfs_io_subrequest *subreq;
 
-	subreq = kzalloc(sizeof(struct netfs_io_subrequest), GFP_KERNEL);
+	subreq = kzalloc(rreq->netfs_ops->io_subrequest_size ?:
+			 sizeof(struct netfs_io_subrequest),
+			 GFP_KERNEL);
 	if (subreq) {
+		INIT_WORK(&subreq->work, NULL);
 		INIT_LIST_HEAD(&subreq->rreq_link);
 		refcount_set(&subreq->ref, 2);
 		subreq->rreq = rreq;
@@ -140,6 +171,8 @@ static void netfs_free_subrequest(struct netfs_io_subrequest *subreq,
 	struct netfs_io_request *rreq = subreq->rreq;
 
 	trace_netfs_sreq(subreq, netfs_sreq_trace_free);
+	if (rreq->netfs_ops->free_subrequest)
+		rreq->netfs_ops->free_subrequest(subreq);
 	kfree(subreq);
 	netfs_stat_d(&netfs_n_rh_sreq);
 	netfs_put_request(rreq, was_async, netfs_rreq_trace_put_subreq);
diff --git a/fs/netfs/output.c b/fs/netfs/output.c
new file mode 100644
index 000000000000..625eb68f3e5a
--- /dev/null
+++ b/fs/netfs/output.c
@@ -0,0 +1,478 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Network filesystem high-level write support.
+ *
+ * Copyright (C) 2023 Red Hat, Inc. All Rights Reserved.
+ * Written by David Howells (dhowells@redhat.com)
+ */
+
+#include <linux/fs.h>
+#include <linux/mm.h>
+#include <linux/pagemap.h>
+#include <linux/slab.h>
+#include <linux/writeback.h>
+#include <linux/pagevec.h>
+#include "internal.h"
+
+/**
+ * netfs_create_write_request - Create a write operation.
+ * @wreq: The write request this is storing from.
+ * @dest: The destination type
+ * @start: Start of the region this write will modify
+ * @len: Length of the modification
+ * @worker: The worker function to handle the write(s)
+ *
+ * Allocate a write operation, set it up and add it to the list on a write
+ * request.
+ */
+struct netfs_io_subrequest *netfs_create_write_request(struct netfs_io_request *wreq,
+						       enum netfs_io_source dest,
+						       loff_t start, size_t len,
+						       work_func_t worker)
+{
+	struct netfs_io_subrequest *subreq;
+
+	subreq = netfs_alloc_subrequest(wreq);
+	if (subreq) {
+		INIT_WORK(&subreq->work, worker);
+		subreq->source	= dest;
+		subreq->start	= start;
+		subreq->len	= len;
+		subreq->debug_index = wreq->subreq_counter++;
+
+		switch (subreq->source) {
+		case NETFS_UPLOAD_TO_SERVER:
+			netfs_stat(&netfs_n_wh_upload);
+			break;
+		case NETFS_WRITE_TO_CACHE:
+			netfs_stat(&netfs_n_wh_write);
+			break;
+		default:
+			BUG();
+		}
+
+		subreq->io_iter = wreq->io_iter;
+		iov_iter_advance(&subreq->io_iter, subreq->start - wreq->start);
+		iov_iter_truncate(&subreq->io_iter, subreq->len);
+
+		trace_netfs_sreq_ref(wreq->debug_id, subreq->debug_index,
+				     refcount_read(&subreq->ref),
+				     netfs_sreq_trace_new);
+		atomic_inc(&wreq->nr_outstanding);
+		list_add_tail(&subreq->rreq_link, &wreq->subrequests);
+		trace_netfs_sreq(subreq, netfs_sreq_trace_prepare);
+	}
+
+	return subreq;
+}
+EXPORT_SYMBOL(netfs_create_write_request);
+
+/*
+ * Process a completed write request once all the component operations have
+ * been completed.
+ */
+static void netfs_write_terminated(struct netfs_io_request *wreq, bool was_async)
+{
+	struct netfs_io_subrequest *subreq;
+	struct netfs_inode *ctx = netfs_inode(wreq->inode);
+	size_t transferred = 0;
+
+	_enter("R=%x[]", wreq->debug_id);
+
+	trace_netfs_rreq(wreq, netfs_rreq_trace_write_done);
+
+	list_for_each_entry(subreq, &wreq->subrequests, rreq_link) {
+		if (subreq->error || subreq->transferred == 0)
+			break;
+		transferred += subreq->transferred;
+		if (subreq->transferred < subreq->len)
+			break;
+	}
+	wreq->transferred = transferred;
+
+	list_for_each_entry(subreq, &wreq->subrequests, rreq_link) {
+		if (!subreq->error)
+			continue;
+		switch (subreq->source) {
+		case NETFS_UPLOAD_TO_SERVER:
+			/* Depending on the type of failure, this may prevent
+			 * writeback completion unless we're in disconnected
+			 * mode.
+			 */
+			if (!wreq->error)
+				wreq->error = subreq->error;
+			break;
+
+		case NETFS_WRITE_TO_CACHE:
+			/* Failure doesn't prevent writeback completion unless
+			 * we're in disconnected mode.
+			 */
+			if (subreq->error != -ENOBUFS)
+				ctx->ops->invalidate_cache(wreq);
+			break;
+
+		default:
+			WARN_ON_ONCE(1);
+			if (!wreq->error)
+				wreq->error = -EIO;
+			return;
+		}
+	}
+
+	wreq->cleanup(wreq);
+
+	if (wreq->origin == NETFS_DIO_WRITE &&
+	    wreq->mapping->nrpages) {
+		pgoff_t first = wreq->start >> PAGE_SHIFT;
+		pgoff_t last = (wreq->start + wreq->transferred - 1) >> PAGE_SHIFT;
+		invalidate_inode_pages2_range(wreq->mapping, first, last);
+	}
+
+	if (wreq->origin == NETFS_DIO_WRITE)
+		inode_dio_end(wreq->inode);
+
+	_debug("finished");
+	trace_netfs_rreq(wreq, netfs_rreq_trace_wake_ip);
+	clear_bit_unlock(NETFS_RREQ_IN_PROGRESS, &wreq->flags);
+	wake_up_bit(&wreq->flags, NETFS_RREQ_IN_PROGRESS);
+
+	if (wreq->iocb) {
+		wreq->iocb->ki_pos += transferred;
+		if (wreq->iocb->ki_complete)
+			wreq->iocb->ki_complete(
+				wreq->iocb, wreq->error ? wreq->error : transferred);
+	}
+
+	netfs_clear_subrequests(wreq, was_async);
+	netfs_put_request(wreq, was_async, netfs_rreq_trace_put_complete);
+}
+
+/*
+ * Deal with the completion of writing the data to the cache.
+ */
+void netfs_write_subrequest_terminated(void *_op, ssize_t transferred_or_error,
+				       bool was_async)
+{
+	struct netfs_io_subrequest *subreq = _op;
+	struct netfs_io_request *wreq = subreq->rreq;
+	unsigned int u;
+
+	_enter("%x[%x] %zd", wreq->debug_id, subreq->debug_index, transferred_or_error);
+
+	switch (subreq->source) {
+	case NETFS_UPLOAD_TO_SERVER:
+		netfs_stat(&netfs_n_wh_upload_done);
+		break;
+	case NETFS_WRITE_TO_CACHE:
+		netfs_stat(&netfs_n_wh_write_done);
+		break;
+	case NETFS_INVALID_WRITE:
+		break;
+	default:
+		BUG();
+	}
+
+	if (IS_ERR_VALUE(transferred_or_error)) {
+		subreq->error = transferred_or_error;
+		trace_netfs_failure(wreq, subreq, transferred_or_error,
+				    netfs_fail_write);
+		goto failed;
+	}
+
+	if (WARN(transferred_or_error > subreq->len - subreq->transferred,
+		 "Subreq excess write: R%x[%x] %zd > %zu - %zu",
+		 wreq->debug_id, subreq->debug_index,
+		 transferred_or_error, subreq->len, subreq->transferred))
+		transferred_or_error = subreq->len - subreq->transferred;
+
+	subreq->error = 0;
+	subreq->transferred += transferred_or_error;
+
+	if (iov_iter_count(&subreq->io_iter) != subreq->len - subreq->transferred)
+		pr_warn("R=%08x[%u] ITER POST-MISMATCH %zx != %zx-%zx %x\n",
+			wreq->debug_id, subreq->debug_index,
+			iov_iter_count(&subreq->io_iter), subreq->len,
+			subreq->transferred, subreq->io_iter.iter_type);
+
+	if (subreq->transferred < subreq->len)
+		goto incomplete;
+
+	__clear_bit(NETFS_SREQ_NO_PROGRESS, &subreq->flags);
+out:
+	trace_netfs_sreq(subreq, netfs_sreq_trace_terminated);
+
+	/* If we decrement nr_outstanding to 0, the ref belongs to us. */
+	u = atomic_dec_return(&wreq->nr_outstanding);
+	if (u == 0)
+		netfs_write_terminated(wreq, was_async);
+	else if (u == 1)
+		wake_up_var(&wreq->nr_outstanding);
+
+	netfs_put_subrequest(subreq, was_async, netfs_sreq_trace_put_terminated);
+	return;
+
+incomplete:
+	if (transferred_or_error == 0) {
+		if (__test_and_set_bit(NETFS_SREQ_NO_PROGRESS, &subreq->flags)) {
+			subreq->error = -ENODATA;
+			goto failed;
+		}
+	} else {
+		__clear_bit(NETFS_SREQ_NO_PROGRESS, &subreq->flags);
+	}
+
+	__set_bit(NETFS_SREQ_SHORT_IO, &subreq->flags);
+	set_bit(NETFS_RREQ_INCOMPLETE_IO, &wreq->flags);
+	goto out;
+
+failed:
+	switch (subreq->source) {
+	case NETFS_WRITE_TO_CACHE:
+		netfs_stat(&netfs_n_wh_write_failed);
+		set_bit(NETFS_RREQ_INCOMPLETE_IO, &wreq->flags);
+		break;
+	case NETFS_UPLOAD_TO_SERVER:
+		netfs_stat(&netfs_n_wh_upload_failed);
+		set_bit(NETFS_RREQ_FAILED, &wreq->flags);
+		wreq->error = subreq->error;
+		break;
+	default:
+		break;
+	}
+	goto out;
+}
+EXPORT_SYMBOL(netfs_write_subrequest_terminated);
+
+static void netfs_write_to_cache_op(struct netfs_io_subrequest *subreq)
+{
+	struct netfs_io_request *wreq = subreq->rreq;
+	struct netfs_cache_resources *cres = &wreq->cache_resources;
+
+	trace_netfs_sreq(subreq, netfs_sreq_trace_submit);
+
+	cres->ops->write(cres, subreq->start, &subreq->io_iter,
+			 netfs_write_subrequest_terminated, subreq);
+}
+
+static void netfs_write_to_cache_op_worker(struct work_struct *work)
+{
+	struct netfs_io_subrequest *subreq =
+		container_of(work, struct netfs_io_subrequest, work);
+
+	netfs_write_to_cache_op(subreq);
+}
+
+/**
+ * netfs_queue_write_request - Queue a write request for attention
+ * @subreq: The write request to be queued
+ *
+ * Queue the specified write request for processing by a worker thread.  We
+ * pass the caller's ref on the request to the worker thread.
+ */
+void netfs_queue_write_request(struct netfs_io_subrequest *subreq)
+{
+	if (!queue_work(system_unbound_wq, &subreq->work))
+		netfs_put_subrequest(subreq, false, netfs_sreq_trace_put_wip);
+}
+EXPORT_SYMBOL(netfs_queue_write_request);
+
+/*
+ * Set up a op for writing to the cache.
+ */
+static void netfs_set_up_write_to_cache(struct netfs_io_request *wreq)
+{
+	struct netfs_cache_resources *cres = &wreq->cache_resources;
+	struct netfs_io_subrequest *subreq;
+	struct netfs_inode *ctx = netfs_inode(wreq->inode);
+	struct fscache_cookie *cookie = netfs_i_cookie(ctx);
+	loff_t start = wreq->start;
+	size_t len = wreq->len;
+	int ret;
+
+	if (!fscache_cookie_enabled(cookie)) {
+		clear_bit(NETFS_RREQ_WRITE_TO_CACHE, &wreq->flags);
+		return;
+	}
+
+	_debug("write to cache");
+	ret = fscache_begin_write_operation(cres, cookie);
+	if (ret < 0)
+		return;
+
+	ret = cres->ops->prepare_write(cres, &start, &len, wreq->upper_len,
+				       i_size_read(wreq->inode), true);
+	if (ret < 0)
+		return;
+
+	subreq = netfs_create_write_request(wreq, NETFS_WRITE_TO_CACHE, start, len,
+					    netfs_write_to_cache_op_worker);
+	if (!subreq)
+		return;
+
+	netfs_write_to_cache_op(subreq);
+}
+
+/*
+ * Begin the process of writing out a chunk of data.
+ *
+ * We are given a write request that holds a series of dirty regions and
+ * (partially) covers a sequence of folios, all of which are present.  The
+ * pages must have been marked as writeback as appropriate.
+ *
+ * We need to perform the following steps:
+ *
+ * (1) If encrypting, create an output buffer and encrypt each block of the
+ *     data into it, otherwise the output buffer will point to the original
+ *     folios.
+ *
+ * (2) If the data is to be cached, set up a write op for the entire output
+ *     buffer to the cache, if the cache wants to accept it.
+ *
+ * (3) If the data is to be uploaded (ie. not merely cached):
+ *
+ *     (a) If the data is to be compressed, create a compression buffer and
+ *         compress the data into it.
+ *
+ *     (b) For each destination we want to upload to, set up write ops to write
+ *         to that destination.  We may need multiple writes if the data is not
+ *         contiguous or the span exceeds wsize for a server.
+ */
+int netfs_begin_write(struct netfs_io_request *wreq, bool may_wait,
+		      enum netfs_write_trace what)
+{
+	struct netfs_inode *ctx = netfs_inode(wreq->inode);
+
+	_enter("R=%x %llx-%llx f=%lx",
+	       wreq->debug_id, wreq->start, wreq->start + wreq->len - 1,
+	       wreq->flags);
+
+	trace_netfs_write(wreq, what);
+	if (wreq->len == 0 || wreq->iter.count == 0) {
+		pr_err("Zero-sized write [R=%x]\n", wreq->debug_id);
+		return -EIO;
+	}
+
+	if (wreq->origin == NETFS_DIO_WRITE)
+		inode_dio_begin(wreq->inode);
+
+	wreq->io_iter = wreq->iter;
+
+	/* ->outstanding > 0 carries a ref */
+	netfs_get_request(wreq, netfs_rreq_trace_get_for_outstanding);
+	atomic_set(&wreq->nr_outstanding, 1);
+
+	/* Start the encryption/compression going.  We can do that in the
+	 * background whilst we generate a list of write ops that we want to
+	 * perform.
+	 */
+	// TODO: Encrypt or compress the region as appropriate
+
+	/* We need to write all of the region to the cache */
+	if (test_bit(NETFS_RREQ_WRITE_TO_CACHE, &wreq->flags))
+		netfs_set_up_write_to_cache(wreq);
+
+	/* However, we don't necessarily write all of the region to the server.
+	 * Caching of reads is being managed this way also.
+	 */
+	if (test_bit(NETFS_RREQ_UPLOAD_TO_SERVER, &wreq->flags))
+		ctx->ops->create_write_requests(wreq, wreq->start, wreq->len);
+
+	if (atomic_dec_and_test(&wreq->nr_outstanding))
+		netfs_write_terminated(wreq, false);
+
+	if (!may_wait)
+		return -EIOCBQUEUED;
+
+	wait_on_bit(&wreq->flags, NETFS_RREQ_IN_PROGRESS,
+		    TASK_UNINTERRUPTIBLE);
+	return wreq->error;
+}
+
+/*
+ * Begin a write operation for writing through the pagecache.
+ */
+struct netfs_io_request *netfs_begin_writethrough(struct kiocb *iocb, size_t len)
+{
+	struct netfs_io_request *wreq;
+	struct file *file = iocb->ki_filp;
+
+	wreq = netfs_alloc_request(file->f_mapping, file, iocb->ki_pos, len,
+				   NETFS_WRITETHROUGH);
+	if (IS_ERR(wreq))
+		return wreq;
+
+	trace_netfs_write(wreq, netfs_write_trace_writethrough);
+
+	__set_bit(NETFS_RREQ_UPLOAD_TO_SERVER, &wreq->flags);
+	iov_iter_xarray(&wreq->iter, ITER_SOURCE, &wreq->mapping->i_pages, wreq->start, 0);
+	wreq->io_iter = wreq->iter;
+
+	/* ->outstanding > 0 carries a ref */
+	netfs_get_request(wreq, netfs_rreq_trace_get_for_outstanding);
+	atomic_set(&wreq->nr_outstanding, 1);
+	return wreq;
+}
+
+static void netfs_submit_writethrough(struct netfs_io_request *wreq, bool final)
+{
+	struct netfs_inode *ictx = netfs_inode(wreq->inode);
+	unsigned long long start;
+	size_t len;
+
+	if (!test_bit(NETFS_RREQ_UPLOAD_TO_SERVER, &wreq->flags))
+		return;
+
+	start = wreq->start + wreq->submitted;
+	len = wreq->iter.count - wreq->submitted;
+	if (!final) {
+		len /= wreq->wsize; /* Round to number of maximum packets */
+		len *= wreq->wsize;
+	}
+
+	ictx->ops->create_write_requests(wreq, start, len);
+	wreq->submitted += len;
+}
+
+/*
+ * Advance the state of the write operation used when writing through the
+ * pagecache.  Data has been copied into the pagecache that we need to append
+ * to the request.  If we've added more than wsize then we need to create a new
+ * subrequest.
+ */
+int netfs_advance_writethrough(struct netfs_io_request *wreq, size_t copied, bool to_page_end)
+{
+	_enter("ic=%zu sb=%zu ws=%u cp=%zu tp=%u",
+	       wreq->iter.count, wreq->submitted, wreq->wsize, copied, to_page_end);
+
+	wreq->iter.count += copied;
+	wreq->io_iter.count += copied;
+	if (to_page_end && wreq->io_iter.count - wreq->submitted >= wreq->wsize)
+		netfs_submit_writethrough(wreq, false);
+
+	return wreq->error;
+}
+
+/*
+ * End a write operation used when writing through the pagecache.
+ */
+int netfs_end_writethrough(struct netfs_io_request *wreq, struct kiocb *iocb)
+{
+	int ret = -EIOCBQUEUED;
+
+	_enter("ic=%zu sb=%zu ws=%u",
+	       wreq->iter.count, wreq->submitted, wreq->wsize);
+
+	if (wreq->submitted < wreq->io_iter.count)
+		netfs_submit_writethrough(wreq, true);
+
+	if (atomic_dec_and_test(&wreq->nr_outstanding))
+		netfs_write_terminated(wreq, false);
+
+	if (is_sync_kiocb(iocb)) {
+		wait_on_bit(&wreq->flags, NETFS_RREQ_IN_PROGRESS,
+			    TASK_UNINTERRUPTIBLE);
+		ret = wreq->error;
+	}
+
+	netfs_put_request(wreq, false, netfs_rreq_trace_put_return);
+	return ret;
+}
diff --git a/fs/netfs/stats.c b/fs/netfs/stats.c
index 5510a7a14a40..deeba9f9dcf5 100644
--- a/fs/netfs/stats.c
+++ b/fs/netfs/stats.c
@@ -9,6 +9,8 @@
 #include <linux/seq_file.h>
 #include "internal.h"
 
+atomic_t netfs_n_rh_dio_read;
+atomic_t netfs_n_rh_dio_write;
 atomic_t netfs_n_rh_readahead;
 atomic_t netfs_n_rh_readpage;
 atomic_t netfs_n_rh_rreq;
@@ -27,32 +29,48 @@ atomic_t netfs_n_rh_write_begin;
 atomic_t netfs_n_rh_write_done;
 atomic_t netfs_n_rh_write_failed;
 atomic_t netfs_n_rh_write_zskip;
+atomic_t netfs_n_wh_wstream_conflict;
+atomic_t netfs_n_wh_upload;
+atomic_t netfs_n_wh_upload_done;
+atomic_t netfs_n_wh_upload_failed;
+atomic_t netfs_n_wh_write;
+atomic_t netfs_n_wh_write_done;
+atomic_t netfs_n_wh_write_failed;
 
-void netfs_stats_show(struct seq_file *m)
+int netfs_stats_show(struct seq_file *m, void *v)
 {
-	seq_printf(m, "RdHelp : RA=%u RP=%u WB=%u WBZ=%u rr=%u sr=%u\n",
+	seq_printf(m, "Netfs  : DR=%u DW=%u RA=%u RP=%u WB=%u WBZ=%u\n",
+		   atomic_read(&netfs_n_rh_dio_read),
+		   atomic_read(&netfs_n_rh_dio_write),
 		   atomic_read(&netfs_n_rh_readahead),
 		   atomic_read(&netfs_n_rh_readpage),
 		   atomic_read(&netfs_n_rh_write_begin),
-		   atomic_read(&netfs_n_rh_write_zskip),
-		   atomic_read(&netfs_n_rh_rreq),
-		   atomic_read(&netfs_n_rh_sreq));
-	seq_printf(m, "RdHelp : ZR=%u sh=%u sk=%u\n",
+		   atomic_read(&netfs_n_rh_write_zskip));
+	seq_printf(m, "Netfs  : ZR=%u sh=%u sk=%u\n",
 		   atomic_read(&netfs_n_rh_zero),
 		   atomic_read(&netfs_n_rh_short_read),
 		   atomic_read(&netfs_n_rh_write_zskip));
-	seq_printf(m, "RdHelp : DL=%u ds=%u df=%u di=%u\n",
+	seq_printf(m, "Netfs  : DL=%u ds=%u df=%u di=%u\n",
 		   atomic_read(&netfs_n_rh_download),
 		   atomic_read(&netfs_n_rh_download_done),
 		   atomic_read(&netfs_n_rh_download_failed),
 		   atomic_read(&netfs_n_rh_download_instead));
-	seq_printf(m, "RdHelp : RD=%u rs=%u rf=%u\n",
+	seq_printf(m, "Netfs  : RD=%u rs=%u rf=%u\n",
 		   atomic_read(&netfs_n_rh_read),
 		   atomic_read(&netfs_n_rh_read_done),
 		   atomic_read(&netfs_n_rh_read_failed));
-	seq_printf(m, "RdHelp : WR=%u ws=%u wf=%u\n",
-		   atomic_read(&netfs_n_rh_write),
-		   atomic_read(&netfs_n_rh_write_done),
-		   atomic_read(&netfs_n_rh_write_failed));
+	seq_printf(m, "Netfs  : UL=%u us=%u uf=%u\n",
+		   atomic_read(&netfs_n_wh_upload),
+		   atomic_read(&netfs_n_wh_upload_done),
+		   atomic_read(&netfs_n_wh_upload_failed));
+	seq_printf(m, "Netfs  : WR=%u ws=%u wf=%u\n",
+		   atomic_read(&netfs_n_wh_write),
+		   atomic_read(&netfs_n_wh_write_done),
+		   atomic_read(&netfs_n_wh_write_failed));
+	seq_printf(m, "Netfs  : rr=%u sr=%u wsc=%u\n",
+		   atomic_read(&netfs_n_rh_rreq),
+		   atomic_read(&netfs_n_rh_sreq),
+		   atomic_read(&netfs_n_wh_wstream_conflict));
+	return fscache_stats_show(m);
 }
 EXPORT_SYMBOL(netfs_stats_show);
diff --git a/fs/nfs/Kconfig b/fs/nfs/Kconfig
index 01ac733a6320..f7e32d76e34d 100644
--- a/fs/nfs/Kconfig
+++ b/fs/nfs/Kconfig
@@ -169,8 +169,8 @@ config ROOT_NFS
 
 config NFS_FSCACHE
 	bool "Provide NFS client caching support"
-	depends on NFS_FS=m && FSCACHE || NFS_FS=y && FSCACHE=y
-	select NETFS_SUPPORT
+	depends on NFS_FS=m && NETFS_SUPPORT || NFS_FS=y && NETFS_SUPPORT=y
+	select FSCACHE
 	help
 	  Say Y here if you want NFS data to be cached locally on disc through
 	  the general filesystem cache manager
diff --git a/fs/nfs/fscache.c b/fs/nfs/fscache.c
index b05717fe0d4e..2d1bfee225c3 100644
--- a/fs/nfs/fscache.c
+++ b/fs/nfs/fscache.c
@@ -274,12 +274,6 @@ static void nfs_netfs_free_request(struct netfs_io_request *rreq)
 	put_nfs_open_context(rreq->netfs_priv);
 }
 
-static inline int nfs_netfs_begin_cache_operation(struct netfs_io_request *rreq)
-{
-	return fscache_begin_read_operation(&rreq->cache_resources,
-					    netfs_i_cookie(netfs_inode(rreq->inode)));
-}
-
 static struct nfs_netfs_io_data *nfs_netfs_alloc(struct netfs_io_subrequest *sreq)
 {
 	struct nfs_netfs_io_data *netfs;
@@ -387,7 +381,6 @@ void nfs_netfs_read_completion(struct nfs_pgio_header *hdr)
 const struct netfs_request_ops nfs_netfs_ops = {
 	.init_request		= nfs_netfs_init_request,
 	.free_request		= nfs_netfs_free_request,
-	.begin_cache_operation	= nfs_netfs_begin_cache_operation,
 	.issue_read		= nfs_netfs_issue_read,
 	.clamp_length		= nfs_netfs_clamp_length
 };
diff --git a/fs/nfs/fscache.h b/fs/nfs/fscache.h
index 5407ab8c8783..e3cb4923316b 100644
--- a/fs/nfs/fscache.h
+++ b/fs/nfs/fscache.h
@@ -80,7 +80,7 @@ static inline void nfs_netfs_put(struct nfs_netfs_io_data *netfs)
 }
 static inline void nfs_netfs_inode_init(struct nfs_inode *nfsi)
 {
-	netfs_inode_init(&nfsi->netfs, &nfs_netfs_ops);
+	netfs_inode_init(&nfsi->netfs, &nfs_netfs_ops, false);
 }
 extern void nfs_netfs_initiate_read(struct nfs_pgio_header *hdr);
 extern void nfs_netfs_read_completion(struct nfs_pgio_header *hdr);
diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index 2fa54cfd4882..6dc6340e2852 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -7911,14 +7911,16 @@ check_for_locks(struct nfs4_file *fp, struct nfs4_lockowner *lowner)
 {
 	struct file_lock *fl;
 	int status = false;
-	struct nfsd_file *nf = find_any_file(fp);
+	struct nfsd_file *nf;
 	struct inode *inode;
 	struct file_lock_context *flctx;
 
+	spin_lock(&fp->fi_lock);
+	nf = find_any_file_locked(fp);
 	if (!nf) {
 		/* Any valid lock stateid should have some sort of access */
 		WARN_ON_ONCE(1);
-		return status;
+		goto out;
 	}
 
 	inode = file_inode(nf->nf_file);
@@ -7934,7 +7936,8 @@ check_for_locks(struct nfs4_file *fp, struct nfs4_lockowner *lowner)
 		}
 		spin_unlock(&flctx->flc_lock);
 	}
-	nfsd_file_put(nf);
+out:
+	spin_unlock(&fp->fi_lock);
 	return status;
 }
 
@@ -7944,10 +7947,8 @@ check_for_locks(struct nfs4_file *fp, struct nfs4_lockowner *lowner)
  * @cstate: NFSv4 COMPOUND state
  * @u: RELEASE_LOCKOWNER arguments
  *
- * The lockowner's so_count is bumped when a lock record is added
- * or when copying a conflicting lock. The latter case is brief,
- * but can lead to fleeting false positives when looking for
- * locks-in-use.
+ * Check if theree are any locks still held and if not - free the lockowner
+ * and any lock state that is owned.
  *
  * Return values:
  *   %nfs_ok: lockowner released or not found
@@ -7983,10 +7984,13 @@ nfsd4_release_lockowner(struct svc_rqst *rqstp,
 		spin_unlock(&clp->cl_lock);
 		return nfs_ok;
 	}
-	if (atomic_read(&lo->lo_owner.so_count) != 2) {
-		spin_unlock(&clp->cl_lock);
-		nfs4_put_stateowner(&lo->lo_owner);
-		return nfserr_locks_held;
+
+	list_for_each_entry(stp, &lo->lo_owner.so_stateids, st_perstateowner) {
+		if (check_for_locks(stp->st_stid.sc_file, lo)) {
+			spin_unlock(&clp->cl_lock);
+			nfs4_put_stateowner(&lo->lo_owner);
+			return nfserr_locks_held;
+		}
 	}
 	unhash_lockowner_locked(lo);
 	while (!list_empty(&lo->lo_owner.so_stateids)) {
diff --git a/fs/overlayfs/namei.c b/fs/overlayfs/namei.c
index 984ffdaeed6c..5764f91d283e 100644
--- a/fs/overlayfs/namei.c
+++ b/fs/overlayfs/namei.c
@@ -18,10 +18,11 @@
 
 struct ovl_lookup_data {
 	struct super_block *sb;
-	struct vfsmount *mnt;
+	const struct ovl_layer *layer;
 	struct qstr name;
 	bool is_dir;
 	bool opaque;
+	bool xwhiteouts;
 	bool stop;
 	bool last;
 	char *redirect;
@@ -201,17 +202,13 @@ struct dentry *ovl_decode_real_fh(struct ovl_fs *ofs, struct ovl_fh *fh,
 	return real;
 }
 
-static bool ovl_is_opaquedir(struct ovl_fs *ofs, const struct path *path)
-{
-	return ovl_path_check_dir_xattr(ofs, path, OVL_XATTR_OPAQUE);
-}
-
 static struct dentry *ovl_lookup_positive_unlocked(struct ovl_lookup_data *d,
 						   const char *name,
 						   struct dentry *base, int len,
 						   bool drop_negative)
 {
-	struct dentry *ret = lookup_one_unlocked(mnt_idmap(d->mnt), name, base, len);
+	struct dentry *ret = lookup_one_unlocked(mnt_idmap(d->layer->mnt), name,
+						 base, len);
 
 	if (!IS_ERR(ret) && d_flags_negative(smp_load_acquire(&ret->d_flags))) {
 		if (drop_negative && ret->d_lockref.count == 1) {
@@ -232,10 +229,13 @@ static int ovl_lookup_single(struct dentry *base, struct ovl_lookup_data *d,
 			     size_t prelen, const char *post,
 			     struct dentry **ret, bool drop_negative)
 {
+	struct ovl_fs *ofs = OVL_FS(d->sb);
 	struct dentry *this;
 	struct path path;
 	int err;
 	bool last_element = !post[0];
+	bool is_upper = d->layer->idx == 0;
+	char val;
 
 	this = ovl_lookup_positive_unlocked(d, name, base, namelen, drop_negative);
 	if (IS_ERR(this)) {
@@ -253,8 +253,8 @@ static int ovl_lookup_single(struct dentry *base, struct ovl_lookup_data *d,
 	}
 
 	path.dentry = this;
-	path.mnt = d->mnt;
-	if (ovl_path_is_whiteout(OVL_FS(d->sb), &path)) {
+	path.mnt = d->layer->mnt;
+	if (ovl_path_is_whiteout(ofs, &path)) {
 		d->stop = d->opaque = true;
 		goto put_and_out;
 	}
@@ -272,7 +272,7 @@ static int ovl_lookup_single(struct dentry *base, struct ovl_lookup_data *d,
 			d->stop = true;
 			goto put_and_out;
 		}
-		err = ovl_check_metacopy_xattr(OVL_FS(d->sb), &path, NULL);
+		err = ovl_check_metacopy_xattr(ofs, &path, NULL);
 		if (err < 0)
 			goto out_err;
 
@@ -292,7 +292,12 @@ static int ovl_lookup_single(struct dentry *base, struct ovl_lookup_data *d,
 		if (d->last)
 			goto out;
 
-		if (ovl_is_opaquedir(OVL_FS(d->sb), &path)) {
+		/* overlay.opaque=x means xwhiteouts directory */
+		val = ovl_get_opaquedir_val(ofs, &path);
+		if (last_element && !is_upper && val == 'x') {
+			d->xwhiteouts = true;
+			ovl_layer_set_xwhiteouts(ofs, d->layer);
+		} else if (val == 'y') {
 			d->stop = true;
 			if (last_element)
 				d->opaque = true;
@@ -863,7 +868,8 @@ fail:
  * Returns next layer in stack starting from top.
  * Returns -1 if this is the last layer.
  */
-int ovl_path_next(int idx, struct dentry *dentry, struct path *path)
+int ovl_path_next(int idx, struct dentry *dentry, struct path *path,
+		  const struct ovl_layer **layer)
 {
 	struct ovl_entry *oe = OVL_E(dentry);
 	struct ovl_path *lowerstack = ovl_lowerstack(oe);
@@ -871,13 +877,16 @@ int ovl_path_next(int idx, struct dentry *dentry, struct path *path)
 	BUG_ON(idx < 0);
 	if (idx == 0) {
 		ovl_path_upper(dentry, path);
-		if (path->dentry)
+		if (path->dentry) {
+			*layer = &OVL_FS(dentry->d_sb)->layers[0];
 			return ovl_numlower(oe) ? 1 : -1;
+		}
 		idx++;
 	}
 	BUG_ON(idx > ovl_numlower(oe));
 	path->dentry = lowerstack[idx - 1].dentry;
-	path->mnt = lowerstack[idx - 1].layer->mnt;
+	*layer = lowerstack[idx - 1].layer;
+	path->mnt = (*layer)->mnt;
 
 	return (idx < ovl_numlower(oe)) ? idx + 1 : -1;
 }
@@ -1055,7 +1064,7 @@ struct dentry *ovl_lookup(struct inode *dir, struct dentry *dentry,
 	old_cred = ovl_override_creds(dentry->d_sb);
 	upperdir = ovl_dentry_upper(dentry->d_parent);
 	if (upperdir) {
-		d.mnt = ovl_upper_mnt(ofs);
+		d.layer = &ofs->layers[0];
 		err = ovl_lookup_layer(upperdir, &d, &upperdentry, true);
 		if (err)
 			goto out;
@@ -1111,7 +1120,7 @@ struct dentry *ovl_lookup(struct inode *dir, struct dentry *dentry,
 		else if (d.is_dir || !ofs->numdatalayer)
 			d.last = lower.layer->idx == ovl_numlower(roe);
 
-		d.mnt = lower.layer->mnt;
+		d.layer = lower.layer;
 		err = ovl_lookup_layer(lower.dentry, &d, &this, false);
 		if (err)
 			goto out_put;
@@ -1278,6 +1287,8 @@ struct dentry *ovl_lookup(struct inode *dir, struct dentry *dentry,
 
 	if (upperopaque)
 		ovl_dentry_set_opaque(dentry);
+	if (d.xwhiteouts)
+		ovl_dentry_set_xwhiteouts(dentry);
 
 	if (upperdentry)
 		ovl_dentry_set_upper_alias(dentry);
diff --git a/fs/overlayfs/overlayfs.h b/fs/overlayfs/overlayfs.h
index 5ba11eb43767..ee949f3e7c77 100644
--- a/fs/overlayfs/overlayfs.h
+++ b/fs/overlayfs/overlayfs.h
@@ -50,7 +50,6 @@ enum ovl_xattr {
 	OVL_XATTR_METACOPY,
 	OVL_XATTR_PROTATTR,
 	OVL_XATTR_XWHITEOUT,
-	OVL_XATTR_XWHITEOUTS,
 };
 
 enum ovl_inode_flag {
@@ -70,6 +69,8 @@ enum ovl_entry_flag {
 	OVL_E_UPPER_ALIAS,
 	OVL_E_OPAQUE,
 	OVL_E_CONNECTED,
+	/* Lower stack may contain xwhiteout entries */
+	OVL_E_XWHITEOUTS,
 };
 
 enum {
@@ -477,6 +478,10 @@ bool ovl_dentry_test_flag(unsigned long flag, struct dentry *dentry);
 bool ovl_dentry_is_opaque(struct dentry *dentry);
 bool ovl_dentry_is_whiteout(struct dentry *dentry);
 void ovl_dentry_set_opaque(struct dentry *dentry);
+bool ovl_dentry_has_xwhiteouts(struct dentry *dentry);
+void ovl_dentry_set_xwhiteouts(struct dentry *dentry);
+void ovl_layer_set_xwhiteouts(struct ovl_fs *ofs,
+			      const struct ovl_layer *layer);
 bool ovl_dentry_has_upper_alias(struct dentry *dentry);
 void ovl_dentry_set_upper_alias(struct dentry *dentry);
 bool ovl_dentry_needs_data_copy_up(struct dentry *dentry, int flags);
@@ -494,11 +499,10 @@ struct file *ovl_path_open(const struct path *path, int flags);
 int ovl_copy_up_start(struct dentry *dentry, int flags);
 void ovl_copy_up_end(struct dentry *dentry);
 bool ovl_already_copied_up(struct dentry *dentry, int flags);
-bool ovl_path_check_dir_xattr(struct ovl_fs *ofs, const struct path *path,
-			      enum ovl_xattr ox);
+char ovl_get_dir_xattr_val(struct ovl_fs *ofs, const struct path *path,
+			   enum ovl_xattr ox);
 bool ovl_path_check_origin_xattr(struct ovl_fs *ofs, const struct path *path);
 bool ovl_path_check_xwhiteout_xattr(struct ovl_fs *ofs, const struct path *path);
-bool ovl_path_check_xwhiteouts_xattr(struct ovl_fs *ofs, const struct path *path);
 bool ovl_init_uuid_xattr(struct super_block *sb, struct ovl_fs *ofs,
 			 const struct path *upperpath);
 
@@ -573,7 +577,13 @@ static inline bool ovl_is_impuredir(struct super_block *sb,
 		.mnt = ovl_upper_mnt(ofs),
 	};
 
-	return ovl_path_check_dir_xattr(ofs, &upperpath, OVL_XATTR_IMPURE);
+	return ovl_get_dir_xattr_val(ofs, &upperpath, OVL_XATTR_IMPURE) == 'y';
+}
+
+static inline char ovl_get_opaquedir_val(struct ovl_fs *ofs,
+					 const struct path *path)
+{
+	return ovl_get_dir_xattr_val(ofs, path, OVL_XATTR_OPAQUE);
 }
 
 static inline bool ovl_redirect_follow(struct ovl_fs *ofs)
@@ -680,7 +690,8 @@ int ovl_get_index_name(struct ovl_fs *ofs, struct dentry *origin,
 struct dentry *ovl_get_index_fh(struct ovl_fs *ofs, struct ovl_fh *fh);
 struct dentry *ovl_lookup_index(struct ovl_fs *ofs, struct dentry *upper,
 				struct dentry *origin, bool verify);
-int ovl_path_next(int idx, struct dentry *dentry, struct path *path);
+int ovl_path_next(int idx, struct dentry *dentry, struct path *path,
+		  const struct ovl_layer **layer);
 int ovl_verify_lowerdata(struct dentry *dentry);
 struct dentry *ovl_lookup(struct inode *dir, struct dentry *dentry,
 			  unsigned int flags);
diff --git a/fs/overlayfs/ovl_entry.h b/fs/overlayfs/ovl_entry.h
index 5fa9c58af65f..cb449ab310a7 100644
--- a/fs/overlayfs/ovl_entry.h
+++ b/fs/overlayfs/ovl_entry.h
@@ -40,6 +40,8 @@ struct ovl_layer {
 	int idx;
 	/* One fsid per unique underlying sb (upper fsid == 0) */
 	int fsid;
+	/* xwhiteouts were found on this layer */
+	bool has_xwhiteouts;
 };
 
 struct ovl_path {
@@ -59,7 +61,7 @@ struct ovl_fs {
 	unsigned int numfs;
 	/* Number of data-only lower layers */
 	unsigned int numdatalayer;
-	const struct ovl_layer *layers;
+	struct ovl_layer *layers;
 	struct ovl_sb *fs;
 	/* workbasedir is the path at workdir= mount option */
 	struct dentry *workbasedir;
diff --git a/fs/overlayfs/readdir.c b/fs/overlayfs/readdir.c
index e71156baa7bc..0ca8af060b0c 100644
--- a/fs/overlayfs/readdir.c
+++ b/fs/overlayfs/readdir.c
@@ -305,8 +305,6 @@ static inline int ovl_dir_read(const struct path *realpath,
 	if (IS_ERR(realfile))
 		return PTR_ERR(realfile);
 
-	rdd->in_xwhiteouts_dir = rdd->dentry &&
-		ovl_path_check_xwhiteouts_xattr(OVL_FS(rdd->dentry->d_sb), realpath);
 	rdd->first_maybe_whiteout = NULL;
 	rdd->ctx.pos = 0;
 	do {
@@ -359,10 +357,13 @@ static int ovl_dir_read_merged(struct dentry *dentry, struct list_head *list,
 		.is_lowest = false,
 	};
 	int idx, next;
+	const struct ovl_layer *layer;
 
 	for (idx = 0; idx != -1; idx = next) {
-		next = ovl_path_next(idx, dentry, &realpath);
+		next = ovl_path_next(idx, dentry, &realpath, &layer);
 		rdd.is_upper = ovl_dentry_upper(dentry) == realpath.dentry;
+		rdd.in_xwhiteouts_dir = layer->has_xwhiteouts &&
+					ovl_dentry_has_xwhiteouts(dentry);
 
 		if (next != -1) {
 			err = ovl_dir_read(&realpath, &rdd);
diff --git a/fs/overlayfs/super.c b/fs/overlayfs/super.c
index 4ab66e3d4cff..2eef6c70b2ae 100644
--- a/fs/overlayfs/super.c
+++ b/fs/overlayfs/super.c
@@ -1249,6 +1249,7 @@ static struct dentry *ovl_get_root(struct super_block *sb,
 				   struct ovl_entry *oe)
 {
 	struct dentry *root;
+	struct ovl_fs *ofs = OVL_FS(sb);
 	struct ovl_path *lowerpath = ovl_lowerstack(oe);
 	unsigned long ino = d_inode(lowerpath->dentry)->i_ino;
 	int fsid = lowerpath->layer->fsid;
@@ -1270,6 +1271,20 @@ static struct dentry *ovl_get_root(struct super_block *sb,
 			ovl_set_flag(OVL_IMPURE, d_inode(root));
 	}
 
+	/* Look for xwhiteouts marker except in the lowermost layer */
+	for (int i = 0; i < ovl_numlower(oe) - 1; i++, lowerpath++) {
+		struct path path = {
+			.mnt = lowerpath->layer->mnt,
+			.dentry = lowerpath->dentry,
+		};
+
+		/* overlay.opaque=x means xwhiteouts directory */
+		if (ovl_get_opaquedir_val(ofs, &path) == 'x') {
+			ovl_layer_set_xwhiteouts(ofs, lowerpath->layer);
+			ovl_dentry_set_xwhiteouts(root);
+		}
+	}
+
 	/* Root is always merge -> can have whiteouts */
 	ovl_set_flag(OVL_WHITEOUTS, d_inode(root));
 	ovl_dentry_set_flag(OVL_E_CONNECTED, root);
diff --git a/fs/overlayfs/util.c b/fs/overlayfs/util.c
index 0217094c23ea..a8e17f14d7a2 100644
--- a/fs/overlayfs/util.c
+++ b/fs/overlayfs/util.c
@@ -461,6 +461,33 @@ void ovl_dentry_set_opaque(struct dentry *dentry)
 	ovl_dentry_set_flag(OVL_E_OPAQUE, dentry);
 }
 
+bool ovl_dentry_has_xwhiteouts(struct dentry *dentry)
+{
+	return ovl_dentry_test_flag(OVL_E_XWHITEOUTS, dentry);
+}
+
+void ovl_dentry_set_xwhiteouts(struct dentry *dentry)
+{
+	ovl_dentry_set_flag(OVL_E_XWHITEOUTS, dentry);
+}
+
+/*
+ * ovl_layer_set_xwhiteouts() is called before adding the overlay dir
+ * dentry to dcache, while readdir of that same directory happens after
+ * the overlay dir dentry is in dcache, so if some cpu observes that
+ * ovl_dentry_is_xwhiteouts(), it will also observe layer->has_xwhiteouts
+ * for the layers where xwhiteouts marker was found in that merge dir.
+ */
+void ovl_layer_set_xwhiteouts(struct ovl_fs *ofs,
+			      const struct ovl_layer *layer)
+{
+	if (layer->has_xwhiteouts)
+		return;
+
+	/* Write once to read-mostly layer properties */
+	ofs->layers[layer->idx].has_xwhiteouts = true;
+}
+
 /*
  * For hard links and decoded file handles, it's possible for ovl_dentry_upper()
  * to return positive, while there's no actual upper alias for the inode.
@@ -739,19 +766,6 @@ bool ovl_path_check_xwhiteout_xattr(struct ovl_fs *ofs, const struct path *path)
 	return res >= 0;
 }
 
-bool ovl_path_check_xwhiteouts_xattr(struct ovl_fs *ofs, const struct path *path)
-{
-	struct dentry *dentry = path->dentry;
-	int res;
-
-	/* xattr.whiteouts must be a directory */
-	if (!d_is_dir(dentry))
-		return false;
-
-	res = ovl_path_getxattr(ofs, path, OVL_XATTR_XWHITEOUTS, NULL, 0);
-	return res >= 0;
-}
-
 /*
  * Load persistent uuid from xattr into s_uuid if found, or store a new
  * random generated value in s_uuid and in xattr.
@@ -811,20 +825,17 @@ fail:
 	return false;
 }
 
-bool ovl_path_check_dir_xattr(struct ovl_fs *ofs, const struct path *path,
-			       enum ovl_xattr ox)
+char ovl_get_dir_xattr_val(struct ovl_fs *ofs, const struct path *path,
+			   enum ovl_xattr ox)
 {
 	int res;
 	char val;
 
 	if (!d_is_dir(path->dentry))
-		return false;
+		return 0;
 
 	res = ovl_path_getxattr(ofs, path, ox, &val, 1);
-	if (res == 1 && val == 'y')
-		return true;
-
-	return false;
+	return res == 1 ? val : 0;
 }
 
 #define OVL_XATTR_OPAQUE_POSTFIX	"opaque"
@@ -837,7 +848,6 @@ bool ovl_path_check_dir_xattr(struct ovl_fs *ofs, const struct path *path,
 #define OVL_XATTR_METACOPY_POSTFIX	"metacopy"
 #define OVL_XATTR_PROTATTR_POSTFIX	"protattr"
 #define OVL_XATTR_XWHITEOUT_POSTFIX	"whiteout"
-#define OVL_XATTR_XWHITEOUTS_POSTFIX	"whiteouts"
 
 #define OVL_XATTR_TAB_ENTRY(x) \
 	[x] = { [false] = OVL_XATTR_TRUSTED_PREFIX x ## _POSTFIX, \
@@ -854,7 +864,6 @@ const char *const ovl_xattr_table[][2] = {
 	OVL_XATTR_TAB_ENTRY(OVL_XATTR_METACOPY),
 	OVL_XATTR_TAB_ENTRY(OVL_XATTR_PROTATTR),
 	OVL_XATTR_TAB_ENTRY(OVL_XATTR_XWHITEOUT),
-	OVL_XATTR_TAB_ENTRY(OVL_XATTR_XWHITEOUTS),
 };
 
 int ovl_check_setxattr(struct ovl_fs *ofs, struct dentry *upperdentry,
diff --git a/fs/smb/client/cached_dir.c b/fs/smb/client/cached_dir.c
index d64a306a414b..971892620504 100644
--- a/fs/smb/client/cached_dir.c
+++ b/fs/smb/client/cached_dir.c
@@ -151,7 +151,7 @@ int open_cached_dir(unsigned int xid, struct cifs_tcon *tcon,
 		return -EOPNOTSUPP;
 
 	ses = tcon->ses;
-	server = ses->server;
+	server = cifs_pick_channel(ses);
 	cfids = tcon->cfids;
 
 	if (!server->ops->new_lease_key)
diff --git a/fs/smb/client/cifs_debug.c b/fs/smb/client/cifs_debug.c
index 60027f5aebe8..3e4209f41c18 100644
--- a/fs/smb/client/cifs_debug.c
+++ b/fs/smb/client/cifs_debug.c
@@ -659,6 +659,7 @@ static ssize_t cifs_stats_proc_write(struct file *file,
 					spin_lock(&tcon->stat_lock);
 					tcon->bytes_read = 0;
 					tcon->bytes_written = 0;
+					tcon->stats_from_time = ktime_get_real_seconds();
 					spin_unlock(&tcon->stat_lock);
 					if (server->ops->clear_stats)
 						server->ops->clear_stats(tcon);
@@ -737,8 +738,9 @@ static int cifs_stats_proc_show(struct seq_file *m, void *v)
 				seq_printf(m, "\n%d) %s", i, tcon->tree_name);
 				if (tcon->need_reconnect)
 					seq_puts(m, "\tDISCONNECTED ");
-				seq_printf(m, "\nSMBs: %d",
-					   atomic_read(&tcon->num_smbs_sent));
+				seq_printf(m, "\nSMBs: %d since %ptTs UTC",
+					   atomic_read(&tcon->num_smbs_sent),
+					   &tcon->stats_from_time);
 				if (server->ops->print_stats)
 					server->ops->print_stats(m, tcon);
 			}
diff --git a/fs/smb/client/cifsfs.c b/fs/smb/client/cifsfs.c
index 99b0ade833aa..e902de4e475a 100644
--- a/fs/smb/client/cifsfs.c
+++ b/fs/smb/client/cifsfs.c
@@ -430,7 +430,7 @@ static void
 cifs_evict_inode(struct inode *inode)
 {
 	truncate_inode_pages_final(&inode->i_data);
-	if (inode->i_state & I_PINNING_FSCACHE_WB)
+	if (inode->i_state & I_PINNING_NETFS_WB)
 		cifs_fscache_unuse_inode_cookie(inode, true);
 	cifs_fscache_release_inode_cookie(inode);
 	clear_inode(inode);
@@ -681,6 +681,8 @@ cifs_show_options(struct seq_file *s, struct dentry *root)
 		seq_printf(s, ",rasize=%u", cifs_sb->ctx->rasize);
 	if (tcon->ses->server->min_offload)
 		seq_printf(s, ",esize=%u", tcon->ses->server->min_offload);
+	if (tcon->ses->server->retrans)
+		seq_printf(s, ",retrans=%u", tcon->ses->server->retrans);
 	seq_printf(s, ",echo_interval=%lu",
 			tcon->ses->server->echo_interval / HZ);
 
@@ -793,8 +795,7 @@ static int cifs_show_stats(struct seq_file *s, struct dentry *root)
 
 static int cifs_write_inode(struct inode *inode, struct writeback_control *wbc)
 {
-	fscache_unpin_writeback(wbc, cifs_inode_cookie(inode));
-	return 0;
+	return netfs_unpin_writeback(inode, wbc);
 }
 
 static int cifs_drop_inode(struct inode *inode)
@@ -1222,7 +1223,7 @@ static int cifs_precopy_set_eof(struct inode *src_inode, struct cifsInodeInfo *s
 	if (rc < 0)
 		goto set_failed;
 
-	netfs_resize_file(&src_cifsi->netfs, src_end);
+	netfs_resize_file(&src_cifsi->netfs, src_end, true);
 	fscache_resize_cookie(cifs_inode_cookie(src_inode), src_end);
 	return 0;
 
@@ -1353,7 +1354,7 @@ static loff_t cifs_remap_file_range(struct file *src_file, loff_t off,
 			smb_file_src, smb_file_target, off, len, destoff);
 		if (rc == 0 && new_size > i_size_read(target_inode)) {
 			truncate_setsize(target_inode, new_size);
-			netfs_resize_file(&target_cifsi->netfs, new_size);
+			netfs_resize_file(&target_cifsi->netfs, new_size, true);
 			fscache_resize_cookie(cifs_inode_cookie(target_inode),
 					      new_size);
 		}
diff --git a/fs/smb/client/cifsglob.h b/fs/smb/client/cifsglob.h
index 879d5ef8a66e..20036fb16cec 100644
--- a/fs/smb/client/cifsglob.h
+++ b/fs/smb/client/cifsglob.h
@@ -204,6 +204,8 @@ struct cifs_open_info_data {
 		};
 	} reparse;
 	char *symlink_target;
+	struct cifs_sid posix_owner;
+	struct cifs_sid posix_group;
 	union {
 		struct smb2_file_all_info fi;
 		struct smb311_posix_qinfo posix_fi;
@@ -751,6 +753,7 @@ struct TCP_Server_Info {
 	unsigned int	max_read;
 	unsigned int	max_write;
 	unsigned int	min_offload;
+	unsigned int	retrans;
 	__le16	compress_algorithm;
 	__u16	signing_algorithm;
 	__le16	cipher_type;
@@ -1207,6 +1210,7 @@ struct cifs_tcon {
 	__u64    bytes_read;
 	__u64    bytes_written;
 	spinlock_t stat_lock;  /* protects the two fields above */
+	time64_t stats_from_time;
 	FILE_SYSTEM_DEVICE_INFO fsDevInfo;
 	FILE_SYSTEM_ATTRIBUTE_INFO fsAttrInfo; /* ok if fs name truncated */
 	FILE_SYSTEM_UNIX_INFO fsUnixInfo;
diff --git a/fs/smb/client/connect.c b/fs/smb/client/connect.c
index 3052a208c6ca..bfd568f89710 100644
--- a/fs/smb/client/connect.c
+++ b/fs/smb/client/connect.c
@@ -1574,6 +1574,9 @@ static int match_server(struct TCP_Server_Info *server,
 	if (server->min_offload != ctx->min_offload)
 		return 0;
 
+	if (server->retrans != ctx->retrans)
+		return 0;
+
 	return 1;
 }
 
@@ -1798,6 +1801,7 @@ smbd_connected:
 		goto out_err_crypto_release;
 	}
 	tcp_ses->min_offload = ctx->min_offload;
+	tcp_ses->retrans = ctx->retrans;
 	/*
 	 * at this point we are the only ones with the pointer
 	 * to the struct since the kernel thread not created yet
diff --git a/fs/smb/client/file.c b/fs/smb/client/file.c
index 1b4262aff8fa..90da81d0372a 100644
--- a/fs/smb/client/file.c
+++ b/fs/smb/client/file.c
@@ -87,7 +87,7 @@ void cifs_pages_written_back(struct inode *inode, loff_t start, unsigned int len
 			continue;
 		if (!folio_test_writeback(folio)) {
 			WARN_ONCE(1, "bad %x @%llx page %lx %lx\n",
-				  len, start, folio_index(folio), end);
+				  len, start, folio->index, end);
 			continue;
 		}
 
@@ -120,7 +120,7 @@ void cifs_pages_write_failed(struct inode *inode, loff_t start, unsigned int len
 			continue;
 		if (!folio_test_writeback(folio)) {
 			WARN_ONCE(1, "bad %x @%llx page %lx %lx\n",
-				  len, start, folio_index(folio), end);
+				  len, start, folio->index, end);
 			continue;
 		}
 
@@ -151,7 +151,7 @@ void cifs_pages_write_redirty(struct inode *inode, loff_t start, unsigned int le
 	xas_for_each(&xas, folio, end) {
 		if (!folio_test_writeback(folio)) {
 			WARN_ONCE(1, "bad %x @%llx page %lx %lx\n",
-				  len, start, folio_index(folio), end);
+				  len, start, folio->index, end);
 			continue;
 		}
 
@@ -2651,7 +2651,7 @@ static void cifs_extend_writeback(struct address_space *mapping,
 				continue;
 			if (xa_is_value(folio))
 				break;
-			if (folio_index(folio) != index)
+			if (folio->index != index)
 				break;
 			if (!folio_try_get_rcu(folio)) {
 				xas_reset(&xas);
@@ -2899,7 +2899,7 @@ redo_folio:
 					goto skip_write;
 			}
 
-			if (folio_mapping(folio) != mapping ||
+			if (folio->mapping != mapping ||
 			    !folio_test_dirty(folio)) {
 				start += folio_size(folio);
 				folio_unlock(folio);
@@ -5043,27 +5043,13 @@ static void cifs_swap_deactivate(struct file *file)
 	/* do we need to unpin (or unlock) the file */
 }
 
-/*
- * Mark a page as having been made dirty and thus needing writeback.  We also
- * need to pin the cache object to write back to.
- */
-#ifdef CONFIG_CIFS_FSCACHE
-static bool cifs_dirty_folio(struct address_space *mapping, struct folio *folio)
-{
-	return fscache_dirty_folio(mapping, folio,
-					cifs_inode_cookie(mapping->host));
-}
-#else
-#define cifs_dirty_folio filemap_dirty_folio
-#endif
-
 const struct address_space_operations cifs_addr_ops = {
 	.read_folio = cifs_read_folio,
 	.readahead = cifs_readahead,
 	.writepages = cifs_writepages,
 	.write_begin = cifs_write_begin,
 	.write_end = cifs_write_end,
-	.dirty_folio = cifs_dirty_folio,
+	.dirty_folio = netfs_dirty_folio,
 	.release_folio = cifs_release_folio,
 	.direct_IO = cifs_direct_io,
 	.invalidate_folio = cifs_invalidate_folio,
@@ -5087,7 +5073,7 @@ const struct address_space_operations cifs_addr_ops_smallbuf = {
 	.writepages = cifs_writepages,
 	.write_begin = cifs_write_begin,
 	.write_end = cifs_write_end,
-	.dirty_folio = cifs_dirty_folio,
+	.dirty_folio = netfs_dirty_folio,
 	.release_folio = cifs_release_folio,
 	.invalidate_folio = cifs_invalidate_folio,
 	.launder_folio = cifs_launder_folio,
diff --git a/fs/smb/client/fs_context.c b/fs/smb/client/fs_context.c
index a3493da12ad1..52cbef2eeb28 100644
--- a/fs/smb/client/fs_context.c
+++ b/fs/smb/client/fs_context.c
@@ -139,6 +139,7 @@ const struct fs_parameter_spec smb3_fs_parameters[] = {
 	fsparam_u32("dir_mode", Opt_dirmode),
 	fsparam_u32("port", Opt_port),
 	fsparam_u32("min_enc_offload", Opt_min_enc_offload),
+	fsparam_u32("retrans", Opt_retrans),
 	fsparam_u32("esize", Opt_min_enc_offload),
 	fsparam_u32("bsize", Opt_blocksize),
 	fsparam_u32("rasize", Opt_rasize),
@@ -1064,6 +1065,9 @@ static int smb3_fs_context_parse_param(struct fs_context *fc,
 	case Opt_min_enc_offload:
 		ctx->min_offload = result.uint_32;
 		break;
+	case Opt_retrans:
+		ctx->retrans = result.uint_32;
+		break;
 	case Opt_blocksize:
 		/*
 		 * inode blocksize realistically should never need to be
@@ -1619,6 +1623,8 @@ int smb3_init_fs_context(struct fs_context *fc)
 	ctx->backupuid_specified = false; /* no backup intent for a user */
 	ctx->backupgid_specified = false; /* no backup intent for a group */
 
+	ctx->retrans = 1;
+
 /*
  *	short int override_uid = -1;
  *	short int override_gid = -1;
diff --git a/fs/smb/client/fs_context.h b/fs/smb/client/fs_context.h
index cf46916286d0..182ce11cbe93 100644
--- a/fs/smb/client/fs_context.h
+++ b/fs/smb/client/fs_context.h
@@ -118,6 +118,7 @@ enum cifs_param {
 	Opt_file_mode,
 	Opt_dirmode,
 	Opt_min_enc_offload,
+	Opt_retrans,
 	Opt_blocksize,
 	Opt_rasize,
 	Opt_rsize,
@@ -245,6 +246,7 @@ struct smb3_fs_context {
 	unsigned int rsize;
 	unsigned int wsize;
 	unsigned int min_offload;
+	unsigned int retrans;
 	bool sockopt_tcp_nodelay:1;
 	/* attribute cache timemout for files and directories in jiffies */
 	unsigned long acregmax;
diff --git a/fs/smb/client/fscache.c b/fs/smb/client/fscache.c
index e5cad149f5a2..c4a3cb736881 100644
--- a/fs/smb/client/fscache.c
+++ b/fs/smb/client/fscache.c
@@ -180,7 +180,7 @@ static int fscache_fallback_write_pages(struct inode *inode, loff_t start, size_
 	if (ret < 0)
 		return ret;
 
-	ret = cres.ops->prepare_write(&cres, &start, &len, i_size_read(inode),
+	ret = cres.ops->prepare_write(&cres, &start, &len, len, i_size_read(inode),
 				      no_space_allocated_yet);
 	if (ret == 0)
 		ret = fscache_write(&cres, start, &iter, NULL, NULL);
diff --git a/fs/smb/client/inode.c b/fs/smb/client/inode.c
index 9f37c1758f73..f0989484f2c6 100644
--- a/fs/smb/client/inode.c
+++ b/fs/smb/client/inode.c
@@ -665,8 +665,6 @@ static int cifs_sfu_mode(struct cifs_fattr *fattr, const unsigned char *path,
 /* Fill a cifs_fattr struct with info from POSIX info struct */
 static void smb311_posix_info_to_fattr(struct cifs_fattr *fattr,
 				       struct cifs_open_info_data *data,
-				       struct cifs_sid *owner,
-				       struct cifs_sid *group,
 				       struct super_block *sb)
 {
 	struct smb311_posix_qinfo *info = &data->posix_fi;
@@ -722,8 +720,8 @@ out_reparse:
 		fattr->cf_symlink_target = data->symlink_target;
 		data->symlink_target = NULL;
 	}
-	sid_to_id(cifs_sb, owner, fattr, SIDOWNER);
-	sid_to_id(cifs_sb, group, fattr, SIDGROUP);
+	sid_to_id(cifs_sb, &data->posix_owner, fattr, SIDOWNER);
+	sid_to_id(cifs_sb, &data->posix_group, fattr, SIDGROUP);
 
 	cifs_dbg(FYI, "POSIX query info: mode 0x%x uniqueid 0x%llx nlink %d\n",
 		fattr->cf_mode, fattr->cf_uniqueid, fattr->cf_nlink);
@@ -1070,9 +1068,7 @@ static int reparse_info_to_fattr(struct cifs_open_info_data *data,
 				 const unsigned int xid,
 				 struct cifs_tcon *tcon,
 				 const char *full_path,
-				 struct cifs_fattr *fattr,
-				 struct cifs_sid *owner,
-				 struct cifs_sid *group)
+				 struct cifs_fattr *fattr)
 {
 	struct TCP_Server_Info *server = tcon->ses->server;
 	struct cifs_sb_info *cifs_sb = CIFS_SB(sb);
@@ -1117,7 +1113,7 @@ static int reparse_info_to_fattr(struct cifs_open_info_data *data,
 	}
 
 	if (tcon->posix_extensions)
-		smb311_posix_info_to_fattr(fattr, data, owner, group, sb);
+		smb311_posix_info_to_fattr(fattr, data, sb);
 	else
 		cifs_open_info_to_fattr(fattr, data, sb);
 out:
@@ -1171,8 +1167,7 @@ static int cifs_get_fattr(struct cifs_open_info_data *data,
 		 */
 		if (cifs_open_data_reparse(data)) {
 			rc = reparse_info_to_fattr(data, sb, xid, tcon,
-						   full_path, fattr,
-						   NULL, NULL);
+						   full_path, fattr);
 		} else {
 			cifs_open_info_to_fattr(fattr, data, sb);
 		}
@@ -1317,10 +1312,10 @@ static int smb311_posix_get_fattr(struct cifs_open_info_data *data,
 				  const unsigned int xid)
 {
 	struct cifs_open_info_data tmp_data = {};
+	struct TCP_Server_Info *server;
 	struct cifs_sb_info *cifs_sb = CIFS_SB(sb);
 	struct cifs_tcon *tcon;
 	struct tcon_link *tlink;
-	struct cifs_sid owner, group;
 	int tmprc;
 	int rc = 0;
 
@@ -1328,14 +1323,14 @@ static int smb311_posix_get_fattr(struct cifs_open_info_data *data,
 	if (IS_ERR(tlink))
 		return PTR_ERR(tlink);
 	tcon = tlink_tcon(tlink);
+	server = tcon->ses->server;
 
 	/*
 	 * 1. Fetch file metadata if not provided (data)
 	 */
 	if (!data) {
-		rc = smb311_posix_query_path_info(xid, tcon, cifs_sb,
-						  full_path, &tmp_data,
-						  &owner, &group);
+		rc = server->ops->query_path_info(xid, tcon, cifs_sb,
+						  full_path, &tmp_data);
 		data = &tmp_data;
 	}
 
@@ -1347,11 +1342,9 @@ static int smb311_posix_get_fattr(struct cifs_open_info_data *data,
 	case 0:
 		if (cifs_open_data_reparse(data)) {
 			rc = reparse_info_to_fattr(data, sb, xid, tcon,
-						   full_path, fattr,
-						   &owner, &group);
+						   full_path, fattr);
 		} else {
-			smb311_posix_info_to_fattr(fattr, data,
-						   &owner, &group, sb);
+			smb311_posix_info_to_fattr(fattr, data, sb);
 		}
 		break;
 	case -EREMOTE:
diff --git a/fs/smb/client/misc.c b/fs/smb/client/misc.c
index c2137ea3c253..0748d7b757b9 100644
--- a/fs/smb/client/misc.c
+++ b/fs/smb/client/misc.c
@@ -140,6 +140,7 @@ tcon_info_alloc(bool dir_leases_enabled)
 	spin_lock_init(&ret_buf->stat_lock);
 	atomic_set(&ret_buf->num_local_opens, 0);
 	atomic_set(&ret_buf->num_remote_opens, 0);
+	ret_buf->stats_from_time = ktime_get_real_seconds();
 #ifdef CONFIG_CIFS_DFS_UPCALL
 	INIT_LIST_HEAD(&ret_buf->dfs_ses_list);
 #endif
diff --git a/fs/smb/client/readdir.c b/fs/smb/client/readdir.c
index 056cae1ddcce..94255401b38d 100644
--- a/fs/smb/client/readdir.c
+++ b/fs/smb/client/readdir.c
@@ -133,14 +133,14 @@ retry:
 				 * Query dir responses don't provide enough
 				 * information about reparse points other than
 				 * their reparse tags.  Save an invalidation by
-				 * not clobbering the existing mode, size and
-				 * symlink target (if any) when reparse tag and
-				 * ctime haven't changed.
+				 * not clobbering some existing attributes when
+				 * reparse tag and ctime haven't changed.
 				 */
 				rc = 0;
 				if (fattr->cf_cifsattrs & ATTR_REPARSE) {
 					if (likely(reparse_inode_match(inode, fattr))) {
 						fattr->cf_mode = inode->i_mode;
+						fattr->cf_rdev = inode->i_rdev;
 						fattr->cf_eof = CIFS_I(inode)->server_eof;
 						fattr->cf_symlink_target = NULL;
 					} else {
@@ -645,10 +645,10 @@ static int cifs_entry_is_dot(struct cifs_dirent *de, bool is_unicode)
 static int is_dir_changed(struct file *file)
 {
 	struct inode *inode = file_inode(file);
-	struct cifsInodeInfo *cifsInfo = CIFS_I(inode);
+	struct cifsInodeInfo *cifs_inode_info = CIFS_I(inode);
 
-	if (cifsInfo->time == 0)
-		return 1; /* directory was changed, perhaps due to unlink */
+	if (cifs_inode_info->time == 0)
+		return 1; /* directory was changed, e.g. unlink or new file */
 	else
 		return 0;
 
diff --git a/fs/smb/client/smb2inode.c b/fs/smb/client/smb2inode.c
index 5053a5550abe..a652200540c8 100644
--- a/fs/smb/client/smb2inode.c
+++ b/fs/smb/client/smb2inode.c
@@ -56,6 +56,35 @@ static inline __u32 file_create_options(struct dentry *dentry)
 	return 0;
 }
 
+/* Parse owner and group from SMB3.1.1 POSIX query info */
+static int parse_posix_sids(struct cifs_open_info_data *data,
+			    struct kvec *rsp_iov)
+{
+	struct smb2_query_info_rsp *qi = rsp_iov->iov_base;
+	unsigned int out_len = le32_to_cpu(qi->OutputBufferLength);
+	unsigned int qi_len = sizeof(data->posix_fi);
+	int owner_len, group_len;
+	u8 *sidsbuf, *sidsbuf_end;
+
+	if (out_len <= qi_len)
+		return -EINVAL;
+
+	sidsbuf = (u8 *)qi + le16_to_cpu(qi->OutputBufferOffset) + qi_len;
+	sidsbuf_end = sidsbuf + out_len - qi_len;
+
+	owner_len = posix_info_sid_size(sidsbuf, sidsbuf_end);
+	if (owner_len == -1)
+		return -EINVAL;
+
+	memcpy(&data->posix_owner, sidsbuf, owner_len);
+	group_len = posix_info_sid_size(sidsbuf + owner_len, sidsbuf_end);
+	if (group_len == -1)
+		return -EINVAL;
+
+	memcpy(&data->posix_group, sidsbuf + owner_len, group_len);
+	return 0;
+}
+
 /*
  * note: If cfile is passed, the reference to it is dropped here.
  * So make sure that you do not reuse cfile after return from this func.
@@ -69,7 +98,6 @@ static int smb2_compound_op(const unsigned int xid, struct cifs_tcon *tcon,
 			    __u32 desired_access, __u32 create_disposition,
 			    __u32 create_options, umode_t mode, struct kvec *in_iov,
 			    int *cmds, int num_cmds, struct cifsFileInfo *cfile,
-			    __u8 **extbuf, size_t *extbuflen,
 			    struct kvec *out_iov, int *out_buftype)
 {
 
@@ -494,21 +522,9 @@ finished:
 					&rsp_iov[i + 1], sizeof(idata->posix_fi) /* add SIDs */,
 					(char *)&idata->posix_fi);
 			}
-			if (rc == 0) {
-				unsigned int length = le32_to_cpu(qi_rsp->OutputBufferLength);
-
-				if (length > sizeof(idata->posix_fi)) {
-					char *base = (char *)rsp_iov[i + 1].iov_base +
-						le16_to_cpu(qi_rsp->OutputBufferOffset) +
-						sizeof(idata->posix_fi);
-					*extbuflen = length - sizeof(idata->posix_fi);
-					*extbuf = kmemdup(base, *extbuflen, GFP_KERNEL);
-					if (!*extbuf)
-						rc = -ENOMEM;
-				} else {
-					rc = -EINVAL;
-				}
-			}
+			if (rc == 0)
+				rc = parse_posix_sids(idata, &rsp_iov[i + 1]);
+
 			SMB2_query_info_free(&rqst[num_rqst++]);
 			if (rc)
 				trace_smb3_posix_query_info_compound_err(xid,  ses->Suid,
@@ -662,7 +678,7 @@ int smb2_query_path_info(const unsigned int xid,
 	struct smb2_hdr *hdr;
 	struct kvec in_iov[2], out_iov[3] = {};
 	int out_buftype[3] = {};
-	int cmds[2] = { SMB2_OP_QUERY_INFO,  };
+	int cmds[2];
 	bool islink;
 	int i, num_cmds;
 	int rc, rc2;
@@ -670,20 +686,36 @@ int smb2_query_path_info(const unsigned int xid,
 	data->adjust_tz = false;
 	data->reparse_point = false;
 
-	if (strcmp(full_path, ""))
-		rc = -ENOENT;
-	else
-		rc = open_cached_dir(xid, tcon, full_path, cifs_sb, false, &cfid);
-	/* If it is a root and its handle is cached then use it */
-	if (!rc) {
-		if (cfid->file_all_info_is_valid) {
-			memcpy(&data->fi, &cfid->file_all_info, sizeof(data->fi));
+	/*
+	 * BB TODO: Add support for using cached root handle in SMB3.1.1 POSIX.
+	 * Create SMB2_query_posix_info worker function to do non-compounded
+	 * query when we already have an open file handle for this. For now this
+	 * is fast enough (always using the compounded version).
+	 */
+	if (!tcon->posix_extensions) {
+		if (*full_path) {
+			rc = -ENOENT;
 		} else {
-			rc = SMB2_query_info(xid, tcon, cfid->fid.persistent_fid,
-					     cfid->fid.volatile_fid, &data->fi);
+			rc = open_cached_dir(xid, tcon, full_path,
+					     cifs_sb, false, &cfid);
+		}
+		/* If it is a root and its handle is cached then use it */
+		if (!rc) {
+			if (cfid->file_all_info_is_valid) {
+				memcpy(&data->fi, &cfid->file_all_info,
+				       sizeof(data->fi));
+			} else {
+				rc = SMB2_query_info(xid, tcon,
+						     cfid->fid.persistent_fid,
+						     cfid->fid.volatile_fid,
+						     &data->fi);
+			}
+			close_cached_dir(cfid);
+			return rc;
 		}
-		close_cached_dir(cfid);
-		return rc;
+		cmds[0] = SMB2_OP_QUERY_INFO;
+	} else {
+		cmds[0] = SMB2_OP_POSIX_QUERY_INFO;
 	}
 
 	in_iov[0].iov_base = data;
@@ -693,9 +725,8 @@ int smb2_query_path_info(const unsigned int xid,
 	cifs_get_readable_path(tcon, full_path, &cfile);
 	rc = smb2_compound_op(xid, tcon, cifs_sb, full_path,
 			      FILE_READ_ATTRIBUTES, FILE_OPEN,
-			      create_options, ACL_NO_MODE,
-			      in_iov, cmds, 1, cfile,
-			      NULL, NULL, out_iov, out_buftype);
+			      create_options, ACL_NO_MODE, in_iov,
+			      cmds, 1, cfile, out_iov, out_buftype);
 	hdr = out_iov[0].iov_base;
 	/*
 	 * If first iov is unset, then SMB session was dropped or we've got a
@@ -707,6 +738,10 @@ int smb2_query_path_info(const unsigned int xid,
 	switch (rc) {
 	case 0:
 	case -EOPNOTSUPP:
+		/*
+		 * BB TODO: When support for special files added to Samba
+		 * re-verify this path.
+		 */
 		rc = parse_create_response(data, cifs_sb, &out_iov[0]);
 		if (rc || !data->reparse_point)
 			goto out;
@@ -722,8 +757,8 @@ int smb2_query_path_info(const unsigned int xid,
 		cifs_get_readable_path(tcon, full_path, &cfile);
 		rc = smb2_compound_op(xid, tcon, cifs_sb, full_path,
 				      FILE_READ_ATTRIBUTES, FILE_OPEN,
-				      create_options, ACL_NO_MODE, in_iov, cmds,
-				      num_cmds, cfile, NULL, NULL, NULL, NULL);
+				      create_options, ACL_NO_MODE, in_iov,
+				      cmds, num_cmds, cfile, NULL, NULL);
 		break;
 	case -EREMOTE:
 		break;
@@ -746,101 +781,6 @@ out:
 	return rc;
 }
 
-int smb311_posix_query_path_info(const unsigned int xid,
-				 struct cifs_tcon *tcon,
-				 struct cifs_sb_info *cifs_sb,
-				 const char *full_path,
-				 struct cifs_open_info_data *data,
-				 struct cifs_sid *owner,
-				 struct cifs_sid *group)
-{
-	int rc;
-	__u32 create_options = 0;
-	struct cifsFileInfo *cfile;
-	struct kvec in_iov[2], out_iov[3] = {};
-	int out_buftype[3] = {};
-	__u8 *sidsbuf = NULL;
-	__u8 *sidsbuf_end = NULL;
-	size_t sidsbuflen = 0;
-	size_t owner_len, group_len;
-	int cmds[2] = { SMB2_OP_POSIX_QUERY_INFO,  };
-	int i, num_cmds;
-
-	data->adjust_tz = false;
-	data->reparse_point = false;
-
-	/*
-	 * BB TODO: Add support for using the cached root handle.
-	 * Create SMB2_query_posix_info worker function to do non-compounded query
-	 * when we already have an open file handle for this. For now this is fast enough
-	 * (always using the compounded version).
-	 */
-	in_iov[0].iov_base = data;
-	in_iov[0].iov_len = sizeof(*data);
-	in_iov[1] = in_iov[0];
-
-	cifs_get_readable_path(tcon, full_path, &cfile);
-	rc = smb2_compound_op(xid, tcon, cifs_sb, full_path,
-			      FILE_READ_ATTRIBUTES, FILE_OPEN,
-			      create_options, ACL_NO_MODE, in_iov, cmds, 1,
-			      cfile, &sidsbuf, &sidsbuflen, out_iov, out_buftype);
-	/*
-	 * If first iov is unset, then SMB session was dropped or we've got a
-	 * cached open file (@cfile).
-	 */
-	if (!out_iov[0].iov_base || out_buftype[0] == CIFS_NO_BUFFER)
-		goto out;
-
-	switch (rc) {
-	case 0:
-	case -EOPNOTSUPP:
-		/* BB TODO: When support for special files added to Samba re-verify this path */
-		rc = parse_create_response(data, cifs_sb, &out_iov[0]);
-		if (rc || !data->reparse_point)
-			goto out;
-
-		if (data->reparse.tag == IO_REPARSE_TAG_SYMLINK) {
-			/* symlink already parsed in create response */
-			num_cmds = 1;
-		} else {
-			cmds[1] = SMB2_OP_GET_REPARSE;
-			num_cmds = 2;
-		}
-		create_options |= OPEN_REPARSE_POINT;
-		cifs_get_readable_path(tcon, full_path, &cfile);
-		rc = smb2_compound_op(xid, tcon, cifs_sb, full_path,
-				      FILE_READ_ATTRIBUTES, FILE_OPEN,
-				      create_options, ACL_NO_MODE, in_iov, cmds,
-				      num_cmds, cfile, &sidsbuf, &sidsbuflen, NULL, NULL);
-		break;
-	}
-
-out:
-	if (rc == 0) {
-		sidsbuf_end = sidsbuf + sidsbuflen;
-
-		owner_len = posix_info_sid_size(sidsbuf, sidsbuf_end);
-		if (owner_len == -1) {
-			rc = -EINVAL;
-			goto out;
-		}
-		memcpy(owner, sidsbuf, owner_len);
-
-		group_len = posix_info_sid_size(
-			sidsbuf + owner_len, sidsbuf_end);
-		if (group_len == -1) {
-			rc = -EINVAL;
-			goto out;
-		}
-		memcpy(group, sidsbuf + owner_len, group_len);
-	}
-
-	kfree(sidsbuf);
-	for (i = 0; i < ARRAY_SIZE(out_buftype); i++)
-		free_rsp_buf(out_buftype[i], out_iov[i].iov_base);
-	return rc;
-}
-
 int
 smb2_mkdir(const unsigned int xid, struct inode *parent_inode, umode_t mode,
 	   struct cifs_tcon *tcon, const char *name,
@@ -848,9 +788,9 @@ smb2_mkdir(const unsigned int xid, struct inode *parent_inode, umode_t mode,
 {
 	return smb2_compound_op(xid, tcon, cifs_sb, name,
 				FILE_WRITE_ATTRIBUTES, FILE_CREATE,
-				CREATE_NOT_FILE, mode, NULL,
-				&(int){SMB2_OP_MKDIR}, 1,
-				NULL, NULL, NULL, NULL, NULL);
+				CREATE_NOT_FILE, mode,
+				NULL, &(int){SMB2_OP_MKDIR}, 1,
+				NULL, NULL, NULL);
 }
 
 void
@@ -875,7 +815,7 @@ smb2_mkdir_setinfo(struct inode *inode, const char *name,
 				 FILE_WRITE_ATTRIBUTES, FILE_CREATE,
 				 CREATE_NOT_FILE, ACL_NO_MODE, &in_iov,
 				 &(int){SMB2_OP_SET_INFO}, 1,
-				 cfile, NULL, NULL, NULL, NULL);
+				 cfile, NULL, NULL);
 	if (tmprc == 0)
 		cifs_i->cifsAttrs = dosattrs;
 }
@@ -887,8 +827,9 @@ smb2_rmdir(const unsigned int xid, struct cifs_tcon *tcon, const char *name,
 	drop_cached_dir_by_name(xid, tcon, name, cifs_sb);
 	return smb2_compound_op(xid, tcon, cifs_sb, name,
 				DELETE, FILE_OPEN, CREATE_NOT_FILE,
-				ACL_NO_MODE, NULL, &(int){SMB2_OP_RMDIR}, 1,
-				NULL, NULL, NULL, NULL, NULL);
+				ACL_NO_MODE, NULL,
+				&(int){SMB2_OP_RMDIR}, 1,
+				NULL, NULL, NULL);
 }
 
 int
@@ -897,8 +838,9 @@ smb2_unlink(const unsigned int xid, struct cifs_tcon *tcon, const char *name,
 {
 	return smb2_compound_op(xid, tcon, cifs_sb, name, DELETE, FILE_OPEN,
 				CREATE_DELETE_ON_CLOSE | OPEN_REPARSE_POINT,
-				ACL_NO_MODE, NULL, &(int){SMB2_OP_DELETE}, 1,
-				NULL, NULL, NULL, NULL, NULL);
+				ACL_NO_MODE, NULL,
+				&(int){SMB2_OP_DELETE}, 1,
+				NULL, NULL, NULL);
 }
 
 static int smb2_set_path_attr(const unsigned int xid, struct cifs_tcon *tcon,
@@ -919,8 +861,8 @@ static int smb2_set_path_attr(const unsigned int xid, struct cifs_tcon *tcon,
 	in_iov.iov_base = smb2_to_name;
 	in_iov.iov_len = 2 * UniStrnlen((wchar_t *)smb2_to_name, PATH_MAX);
 	rc = smb2_compound_op(xid, tcon, cifs_sb, from_name, access,
-			      FILE_OPEN, create_options, ACL_NO_MODE, &in_iov,
-			      &command, 1, cfile, NULL, NULL, NULL, NULL);
+			      FILE_OPEN, create_options, ACL_NO_MODE,
+			      &in_iov, &command, 1, cfile, NULL, NULL);
 smb2_rename_path:
 	kfree(smb2_to_name);
 	return rc;
@@ -971,7 +913,7 @@ smb2_set_path_size(const unsigned int xid, struct cifs_tcon *tcon,
 				FILE_WRITE_DATA, FILE_OPEN,
 				0, ACL_NO_MODE, &in_iov,
 				&(int){SMB2_OP_SET_EOF}, 1,
-				cfile, NULL, NULL, NULL, NULL);
+				cfile, NULL, NULL);
 }
 
 int
@@ -999,8 +941,8 @@ smb2_set_file_info(struct inode *inode, const char *full_path,
 	rc = smb2_compound_op(xid, tcon, cifs_sb, full_path,
 			      FILE_WRITE_ATTRIBUTES, FILE_OPEN,
 			      0, ACL_NO_MODE, &in_iov,
-			      &(int){SMB2_OP_SET_INFO}, 1, cfile,
-			      NULL, NULL, NULL, NULL);
+			      &(int){SMB2_OP_SET_INFO}, 1,
+			      cfile, NULL, NULL);
 	cifs_put_tlink(tlink);
 	return rc;
 }
@@ -1035,7 +977,7 @@ struct inode *smb2_get_reparse_inode(struct cifs_open_info_data *data,
 		cifs_get_writable_path(tcon, full_path, FIND_WR_ANY, &cfile);
 		rc = smb2_compound_op(xid, tcon, cifs_sb, full_path,
 				      da, cd, co, ACL_NO_MODE, in_iov,
-				      cmds, 2, cfile, NULL, NULL, NULL, NULL);
+				      cmds, 2, cfile, NULL, NULL);
 		if (!rc) {
 			rc = smb311_posix_get_inode_info(&new, full_path,
 							 data, sb, xid);
@@ -1045,7 +987,7 @@ struct inode *smb2_get_reparse_inode(struct cifs_open_info_data *data,
 		cifs_get_writable_path(tcon, full_path, FIND_WR_ANY, &cfile);
 		rc = smb2_compound_op(xid, tcon, cifs_sb, full_path,
 				      da, cd, co, ACL_NO_MODE, in_iov,
-				      cmds, 2, cfile, NULL, NULL, NULL, NULL);
+				      cmds, 2, cfile, NULL, NULL);
 		if (!rc) {
 			rc = cifs_get_inode_info(&new, full_path,
 						 data, sb, xid, NULL);
@@ -1072,8 +1014,8 @@ int smb2_query_reparse_point(const unsigned int xid,
 	rc = smb2_compound_op(xid, tcon, cifs_sb, full_path,
 			      FILE_READ_ATTRIBUTES, FILE_OPEN,
 			      OPEN_REPARSE_POINT, ACL_NO_MODE, &in_iov,
-			      &(int){SMB2_OP_GET_REPARSE}, 1, cfile,
-			      NULL, NULL, NULL, NULL);
+			      &(int){SMB2_OP_GET_REPARSE}, 1,
+			      cfile, NULL, NULL);
 	if (rc)
 		goto out;
 
diff --git a/fs/smb/client/smb2maperror.c b/fs/smb/client/smb2maperror.c
index 1a90dd78b238..ac1895358908 100644
--- a/fs/smb/client/smb2maperror.c
+++ b/fs/smb/client/smb2maperror.c
@@ -1210,6 +1210,8 @@ static const struct status_to_posix_error smb2_error_map_table[] = {
 	{STATUS_INVALID_TASK_INDEX, -EIO, "STATUS_INVALID_TASK_INDEX"},
 	{STATUS_THREAD_ALREADY_IN_TASK, -EIO, "STATUS_THREAD_ALREADY_IN_TASK"},
 	{STATUS_CALLBACK_BYPASS, -EIO, "STATUS_CALLBACK_BYPASS"},
+	{STATUS_SERVER_UNAVAILABLE, -EAGAIN, "STATUS_SERVER_UNAVAILABLE"},
+	{STATUS_FILE_NOT_AVAILABLE, -EAGAIN, "STATUS_FILE_NOT_AVAILABLE"},
 	{STATUS_PORT_CLOSED, -EIO, "STATUS_PORT_CLOSED"},
 	{STATUS_MESSAGE_LOST, -EIO, "STATUS_MESSAGE_LOST"},
 	{STATUS_INVALID_MESSAGE, -EIO, "STATUS_INVALID_MESSAGE"},
diff --git a/fs/smb/client/smb2ops.c b/fs/smb/client/smb2ops.c
index 01a5bd7e6a30..d9553c2556a2 100644
--- a/fs/smb/client/smb2ops.c
+++ b/fs/smb/client/smb2ops.c
@@ -614,7 +614,8 @@ parse_server_interfaces(struct network_interface_info_ioctl_rsp *buf,
 				 "multichannel not available\n"
 				 "Empty network interface list returned by server %s\n",
 				 ses->server->hostname);
-		rc = -EINVAL;
+		rc = -EOPNOTSUPP;
+		ses->iface_last_update = jiffies;
 		goto out;
 	}
 
@@ -712,7 +713,6 @@ parse_server_interfaces(struct network_interface_info_ioctl_rsp *buf,
 
 		ses->iface_count++;
 		spin_unlock(&ses->iface_lock);
-		ses->iface_last_update = jiffies;
 next_iface:
 		nb_iface++;
 		next = le32_to_cpu(p->Next);
@@ -734,11 +734,7 @@ next_iface:
 	if ((bytes_left > 8) || p->Next)
 		cifs_dbg(VFS, "%s: incomplete interface info\n", __func__);
 
-
-	if (!ses->iface_count) {
-		rc = -EINVAL;
-		goto out;
-	}
+	ses->iface_last_update = jiffies;
 
 out:
 	/*
diff --git a/fs/smb/client/smb2pdu.c b/fs/smb/client/smb2pdu.c
index bd25c34dc398..288199f0b987 100644
--- a/fs/smb/client/smb2pdu.c
+++ b/fs/smb/client/smb2pdu.c
@@ -156,6 +156,57 @@ out:
 	return;
 }
 
+/* helper function for code reuse */
+static int
+cifs_chan_skip_or_disable(struct cifs_ses *ses,
+			  struct TCP_Server_Info *server,
+			  bool from_reconnect)
+{
+	struct TCP_Server_Info *pserver;
+	unsigned int chan_index;
+
+	if (SERVER_IS_CHAN(server)) {
+		cifs_dbg(VFS,
+			"server %s does not support multichannel anymore. Skip secondary channel\n",
+			 ses->server->hostname);
+
+		spin_lock(&ses->chan_lock);
+		chan_index = cifs_ses_get_chan_index(ses, server);
+		if (chan_index == CIFS_INVAL_CHAN_INDEX) {
+			spin_unlock(&ses->chan_lock);
+			goto skip_terminate;
+		}
+
+		ses->chans[chan_index].server = NULL;
+		spin_unlock(&ses->chan_lock);
+
+		/*
+		 * the above reference of server by channel
+		 * needs to be dropped without holding chan_lock
+		 * as cifs_put_tcp_session takes a higher lock
+		 * i.e. cifs_tcp_ses_lock
+		 */
+		cifs_put_tcp_session(server, from_reconnect);
+
+		server->terminate = true;
+		cifs_signal_cifsd_for_reconnect(server, false);
+
+		/* mark primary server as needing reconnect */
+		pserver = server->primary_server;
+		cifs_signal_cifsd_for_reconnect(pserver, false);
+skip_terminate:
+		mutex_unlock(&ses->session_mutex);
+		return -EHOSTDOWN;
+	}
+
+	cifs_server_dbg(VFS,
+		"server does not support multichannel anymore. Disable all other channels\n");
+	cifs_disable_secondary_channels(ses);
+
+
+	return 0;
+}
+
 static int
 smb2_reconnect(__le16 smb2_command, struct cifs_tcon *tcon,
 	       struct TCP_Server_Info *server, bool from_reconnect)
@@ -164,8 +215,6 @@ smb2_reconnect(__le16 smb2_command, struct cifs_tcon *tcon,
 	struct nls_table *nls_codepage = NULL;
 	struct cifs_ses *ses;
 	int xid;
-	struct TCP_Server_Info *pserver;
-	unsigned int chan_index;
 
 	/*
 	 * SMB2s NegProt, SessSetup, Logoff do not have tcon yet so
@@ -310,44 +359,11 @@ again:
 		 */
 		if (ses->chan_count > 1 &&
 		    !(server->capabilities & SMB2_GLOBAL_CAP_MULTI_CHANNEL)) {
-			if (SERVER_IS_CHAN(server)) {
-				cifs_dbg(VFS, "server %s does not support " \
-					 "multichannel anymore. skipping secondary channel\n",
-					 ses->server->hostname);
-
-				spin_lock(&ses->chan_lock);
-				chan_index = cifs_ses_get_chan_index(ses, server);
-				if (chan_index == CIFS_INVAL_CHAN_INDEX) {
-					spin_unlock(&ses->chan_lock);
-					goto skip_terminate;
-				}
-
-				ses->chans[chan_index].server = NULL;
-				spin_unlock(&ses->chan_lock);
-
-				/*
-				 * the above reference of server by channel
-				 * needs to be dropped without holding chan_lock
-				 * as cifs_put_tcp_session takes a higher lock
-				 * i.e. cifs_tcp_ses_lock
-				 */
-				cifs_put_tcp_session(server, from_reconnect);
-
-				server->terminate = true;
-				cifs_signal_cifsd_for_reconnect(server, false);
-
-				/* mark primary server as needing reconnect */
-				pserver = server->primary_server;
-				cifs_signal_cifsd_for_reconnect(pserver, false);
-
-skip_terminate:
+			rc = cifs_chan_skip_or_disable(ses, server,
+						       from_reconnect);
+			if (rc) {
 				mutex_unlock(&ses->session_mutex);
-				rc = -EHOSTDOWN;
 				goto out;
-			} else {
-				cifs_server_dbg(VFS, "does not support " \
-					 "multichannel anymore. disabling all other channels\n");
-				cifs_disable_secondary_channels(ses);
 			}
 		}
 
@@ -395,20 +411,35 @@ skip_sess_setup:
 		rc = SMB3_request_interfaces(xid, tcon, false);
 		free_xid(xid);
 
-		if (rc)
+		if (rc == -EOPNOTSUPP) {
+			/*
+			 * some servers like Azure SMB server do not advertise
+			 * that multichannel has been disabled with server
+			 * capabilities, rather return STATUS_NOT_IMPLEMENTED.
+			 * treat this as server not supporting multichannel
+			 */
+
+			rc = cifs_chan_skip_or_disable(ses, server,
+						       from_reconnect);
+			goto skip_add_channels;
+		} else if (rc)
 			cifs_dbg(FYI, "%s: failed to query server interfaces: %d\n",
 				 __func__, rc);
 
 		if (ses->chan_max > ses->chan_count &&
+		    ses->iface_count &&
 		    !SERVER_IS_CHAN(server)) {
 			if (ses->chan_count == 1)
 				cifs_server_dbg(VFS, "supports multichannel now\n");
 
 			cifs_try_adding_channels(ses);
+			queue_delayed_work(cifsiod_wq, &tcon->query_interfaces,
+					   (SMB_INTERFACE_POLL_INTERVAL * HZ));
 		}
 	} else {
 		mutex_unlock(&ses->session_mutex);
 	}
+skip_add_channels:
 
 	if (smb2_command != SMB2_INTERNAL_CMD)
 		mod_delayed_work(cifsiod_wq, &server->reconnect, 0);
@@ -1958,10 +1989,7 @@ SMB2_tcon(const unsigned int xid, struct cifs_ses *ses, const char *tree,
 	__le16 *unc_path = NULL;
 	int flags = 0;
 	unsigned int total_len;
-	struct TCP_Server_Info *server;
-
-	/* always use master channel */
-	server = ses->server;
+	struct TCP_Server_Info *server = cifs_pick_channel(ses);
 
 	cifs_dbg(FYI, "TCON\n");
 
@@ -2094,6 +2122,7 @@ SMB2_tdis(const unsigned int xid, struct cifs_tcon *tcon)
 	struct smb2_tree_disconnect_req *req; /* response is trivial */
 	int rc = 0;
 	struct cifs_ses *ses = tcon->ses;
+	struct TCP_Server_Info *server = cifs_pick_channel(ses);
 	int flags = 0;
 	unsigned int total_len;
 	struct kvec iov[1];
@@ -2116,7 +2145,7 @@ SMB2_tdis(const unsigned int xid, struct cifs_tcon *tcon)
 
 	invalidate_all_cached_dirs(tcon);
 
-	rc = smb2_plain_req_init(SMB2_TREE_DISCONNECT, tcon, ses->server,
+	rc = smb2_plain_req_init(SMB2_TREE_DISCONNECT, tcon, server,
 				 (void **) &req,
 				 &total_len);
 	if (rc)
@@ -2134,7 +2163,7 @@ SMB2_tdis(const unsigned int xid, struct cifs_tcon *tcon)
 	rqst.rq_iov = iov;
 	rqst.rq_nvec = 1;
 
-	rc = cifs_send_recv(xid, ses, ses->server,
+	rc = cifs_send_recv(xid, ses, server,
 			    &rqst, &resp_buf_type, flags, &rsp_iov);
 	cifs_small_buf_release(req);
 	if (rc) {
@@ -2279,7 +2308,7 @@ int smb2_parse_contexts(struct TCP_Server_Info *server,
 
 		noff = le16_to_cpu(cc->NameOffset);
 		nlen = le16_to_cpu(cc->NameLength);
-		if (noff + nlen >= doff)
+		if (noff + nlen > doff)
 			return -EINVAL;
 
 		name = (char *)cc + noff;
@@ -3918,7 +3947,7 @@ void smb2_reconnect_server(struct work_struct *work)
 	struct cifs_ses *ses, *ses2;
 	struct cifs_tcon *tcon, *tcon2;
 	struct list_head tmp_list, tmp_ses_list;
-	bool tcon_exist = false, ses_exist = false;
+	bool ses_exist = false;
 	bool tcon_selected = false;
 	int rc;
 	bool resched = false;
@@ -3964,7 +3993,7 @@ void smb2_reconnect_server(struct work_struct *work)
 			if (tcon->need_reconnect || tcon->need_reopen_files) {
 				tcon->tc_count++;
 				list_add_tail(&tcon->rlist, &tmp_list);
-				tcon_selected = tcon_exist = true;
+				tcon_selected = true;
 			}
 		}
 		/*
@@ -3973,7 +4002,7 @@ void smb2_reconnect_server(struct work_struct *work)
 		 */
 		if (ses->tcon_ipc && ses->tcon_ipc->need_reconnect) {
 			list_add_tail(&ses->tcon_ipc->rlist, &tmp_list);
-			tcon_selected = tcon_exist = true;
+			tcon_selected = true;
 			cifs_smb_ses_inc_refcount(ses);
 		}
 		/*
diff --git a/fs/smb/client/smb2proto.h b/fs/smb/client/smb2proto.h
index 343ada691e76..0034b537b0b3 100644
--- a/fs/smb/client/smb2proto.h
+++ b/fs/smb/client/smb2proto.h
@@ -299,9 +299,7 @@ int smb311_posix_query_path_info(const unsigned int xid,
 				 struct cifs_tcon *tcon,
 				 struct cifs_sb_info *cifs_sb,
 				 const char *full_path,
-				 struct cifs_open_info_data *data,
-				 struct cifs_sid *owner,
-				 struct cifs_sid *group);
+				 struct cifs_open_info_data *data);
 int posix_info_parse(const void *beg, const void *end,
 		     struct smb2_posix_info_parsed *out);
 int posix_info_sid_size(const void *beg, const void *end);
diff --git a/fs/smb/client/smb2status.h b/fs/smb/client/smb2status.h
index a9e958166fc5..9c6d79b0bd49 100644
--- a/fs/smb/client/smb2status.h
+++ b/fs/smb/client/smb2status.h
@@ -982,6 +982,8 @@ struct ntstatus {
 #define STATUS_INVALID_TASK_INDEX cpu_to_le32(0xC0000501)
 #define STATUS_THREAD_ALREADY_IN_TASK cpu_to_le32(0xC0000502)
 #define STATUS_CALLBACK_BYPASS cpu_to_le32(0xC0000503)
+#define STATUS_SERVER_UNAVAILABLE cpu_to_le32(0xC0000466)
+#define STATUS_FILE_NOT_AVAILABLE cpu_to_le32(0xC0000467)
 #define STATUS_PORT_CLOSED cpu_to_le32(0xC0000700)
 #define STATUS_MESSAGE_LOST cpu_to_le32(0xC0000701)
 #define STATUS_INVALID_MESSAGE cpu_to_le32(0xC0000702)
diff --git a/fs/smb/server/asn1.c b/fs/smb/server/asn1.c
index 4a4b2b03ff33..b931a99ab9c8 100644
--- a/fs/smb/server/asn1.c
+++ b/fs/smb/server/asn1.c
@@ -214,10 +214,15 @@ static int ksmbd_neg_token_alloc(void *context, size_t hdrlen,
 {
 	struct ksmbd_conn *conn = context;
 
+	if (!vlen)
+		return -EINVAL;
+
 	conn->mechToken = kmemdup_nul(value, vlen, GFP_KERNEL);
 	if (!conn->mechToken)
 		return -ENOMEM;
 
+	conn->mechTokenLen = (unsigned int)vlen;
+
 	return 0;
 }
 
diff --git a/fs/smb/server/connection.c b/fs/smb/server/connection.c
index d311c2ee10bd..09e1e7771592 100644
--- a/fs/smb/server/connection.c
+++ b/fs/smb/server/connection.c
@@ -416,13 +416,7 @@ static void stop_sessions(void)
 again:
 	down_read(&conn_list_lock);
 	list_for_each_entry(conn, &conn_list, conns_list) {
-		struct task_struct *task;
-
 		t = conn->transport;
-		task = t->handler;
-		if (task)
-			ksmbd_debug(CONN, "Stop session handler %s/%d\n",
-				    task->comm, task_pid_nr(task));
 		ksmbd_conn_set_exiting(conn);
 		if (t->ops->shutdown) {
 			up_read(&conn_list_lock);
diff --git a/fs/smb/server/connection.h b/fs/smb/server/connection.h
index 3c005246a32e..0e04cf8b1d89 100644
--- a/fs/smb/server/connection.h
+++ b/fs/smb/server/connection.h
@@ -88,6 +88,7 @@ struct ksmbd_conn {
 	__u16				dialect;
 
 	char				*mechToken;
+	unsigned int			mechTokenLen;
 
 	struct ksmbd_conn_ops	*conn_ops;
 
@@ -134,7 +135,6 @@ struct ksmbd_transport_ops {
 struct ksmbd_transport {
 	struct ksmbd_conn		*conn;
 	struct ksmbd_transport_ops	*ops;
-	struct task_struct		*handler;
 };
 
 #define KSMBD_TCP_RECV_TIMEOUT	(7 * HZ)
diff --git a/fs/smb/server/oplock.c b/fs/smb/server/oplock.c
index 001926d3b348..53dfaac425c6 100644
--- a/fs/smb/server/oplock.c
+++ b/fs/smb/server/oplock.c
@@ -1197,6 +1197,12 @@ int smb_grant_oplock(struct ksmbd_work *work, int req_op_level, u64 pid,
 	bool prev_op_has_lease;
 	__le32 prev_op_state = 0;
 
+	/* Only v2 leases handle the directory */
+	if (S_ISDIR(file_inode(fp->filp)->i_mode)) {
+		if (!lctx || lctx->version != 2)
+			return 0;
+	}
+
 	opinfo = alloc_opinfo(work, pid, tid);
 	if (!opinfo)
 		return -ENOMEM;
diff --git a/fs/smb/server/smb2pdu.c b/fs/smb/server/smb2pdu.c
index 3143819935dc..ba7a72a6a4f4 100644
--- a/fs/smb/server/smb2pdu.c
+++ b/fs/smb/server/smb2pdu.c
@@ -1414,7 +1414,10 @@ static struct ksmbd_user *session_user(struct ksmbd_conn *conn,
 	char *name;
 	unsigned int name_off, name_len, secbuf_len;
 
-	secbuf_len = le16_to_cpu(req->SecurityBufferLength);
+	if (conn->use_spnego && conn->mechToken)
+		secbuf_len = conn->mechTokenLen;
+	else
+		secbuf_len = le16_to_cpu(req->SecurityBufferLength);
 	if (secbuf_len < sizeof(struct authenticate_message)) {
 		ksmbd_debug(SMB, "blob len %d too small\n", secbuf_len);
 		return NULL;
@@ -1505,7 +1508,10 @@ static int ntlm_authenticate(struct ksmbd_work *work,
 		struct authenticate_message *authblob;
 
 		authblob = user_authblob(conn, req);
-		sz = le16_to_cpu(req->SecurityBufferLength);
+		if (conn->use_spnego && conn->mechToken)
+			sz = conn->mechTokenLen;
+		else
+			sz = le16_to_cpu(req->SecurityBufferLength);
 		rc = ksmbd_decode_ntlmssp_auth_blob(authblob, sz, conn, sess);
 		if (rc) {
 			set_user_flag(sess->user, KSMBD_USER_FLAG_BAD_PASSWORD);
@@ -1778,8 +1784,7 @@ int smb2_sess_setup(struct ksmbd_work *work)
 
 	negblob_off = le16_to_cpu(req->SecurityBufferOffset);
 	negblob_len = le16_to_cpu(req->SecurityBufferLength);
-	if (negblob_off < offsetof(struct smb2_sess_setup_req, Buffer) ||
-	    negblob_len < offsetof(struct negotiate_message, NegotiateFlags)) {
+	if (negblob_off < offsetof(struct smb2_sess_setup_req, Buffer)) {
 		rc = -EINVAL;
 		goto out_err;
 	}
@@ -1788,8 +1793,15 @@ int smb2_sess_setup(struct ksmbd_work *work)
 			negblob_off);
 
 	if (decode_negotiation_token(conn, negblob, negblob_len) == 0) {
-		if (conn->mechToken)
+		if (conn->mechToken) {
 			negblob = (struct negotiate_message *)conn->mechToken;
+			negblob_len = conn->mechTokenLen;
+		}
+	}
+
+	if (negblob_len < offsetof(struct negotiate_message, NegotiateFlags)) {
+		rc = -EINVAL;
+		goto out_err;
 	}
 
 	if (server_conf.auth_mechs & conn->auth_mechs) {
diff --git a/fs/smb/server/transport_rdma.c b/fs/smb/server/transport_rdma.c
index c5629a68c8b7..8faa25c6e129 100644
--- a/fs/smb/server/transport_rdma.c
+++ b/fs/smb/server/transport_rdma.c
@@ -2039,6 +2039,7 @@ static bool rdma_frwr_is_supported(struct ib_device_attr *attrs)
 static int smb_direct_handle_connect_request(struct rdma_cm_id *new_cm_id)
 {
 	struct smb_direct_transport *t;
+	struct task_struct *handler;
 	int ret;
 
 	if (!rdma_frwr_is_supported(&new_cm_id->device->attrs)) {
@@ -2056,11 +2057,11 @@ static int smb_direct_handle_connect_request(struct rdma_cm_id *new_cm_id)
 	if (ret)
 		goto out_err;
 
-	KSMBD_TRANS(t)->handler = kthread_run(ksmbd_conn_handler_loop,
-					      KSMBD_TRANS(t)->conn, "ksmbd:r%u",
-					      smb_direct_port);
-	if (IS_ERR(KSMBD_TRANS(t)->handler)) {
-		ret = PTR_ERR(KSMBD_TRANS(t)->handler);
+	handler = kthread_run(ksmbd_conn_handler_loop,
+			      KSMBD_TRANS(t)->conn, "ksmbd:r%u",
+			      smb_direct_port);
+	if (IS_ERR(handler)) {
+		ret = PTR_ERR(handler);
 		pr_err("Can't start thread\n");
 		goto out_err;
 	}
diff --git a/fs/smb/server/transport_tcp.c b/fs/smb/server/transport_tcp.c
index eff7a1d793f0..9d4222154dcc 100644
--- a/fs/smb/server/transport_tcp.c
+++ b/fs/smb/server/transport_tcp.c
@@ -185,6 +185,7 @@ static int ksmbd_tcp_new_connection(struct socket *client_sk)
 	struct sockaddr *csin;
 	int rc = 0;
 	struct tcp_transport *t;
+	struct task_struct *handler;
 
 	t = alloc_transport(client_sk);
 	if (!t) {
@@ -199,13 +200,13 @@ static int ksmbd_tcp_new_connection(struct socket *client_sk)
 		goto out_error;
 	}
 
-	KSMBD_TRANS(t)->handler = kthread_run(ksmbd_conn_handler_loop,
-					      KSMBD_TRANS(t)->conn,
-					      "ksmbd:%u",
-					      ksmbd_tcp_get_port(csin));
-	if (IS_ERR(KSMBD_TRANS(t)->handler)) {
+	handler = kthread_run(ksmbd_conn_handler_loop,
+			      KSMBD_TRANS(t)->conn,
+			      "ksmbd:%u",
+			      ksmbd_tcp_get_port(csin));
+	if (IS_ERR(handler)) {
 		pr_err("cannot start conn thread\n");
-		rc = PTR_ERR(KSMBD_TRANS(t)->handler);
+		rc = PTR_ERR(handler);
 		free_transport(t);
 	}
 	return rc;
diff --git a/fs/tracefs/event_inode.c b/fs/tracefs/event_inode.c
index 6795fda2af19..6b211522a13e 100644
--- a/fs/tracefs/event_inode.c
+++ b/fs/tracefs/event_inode.c
@@ -34,7 +34,15 @@ static DEFINE_MUTEX(eventfs_mutex);
 
 /* Choose something "unique" ;-) */
 #define EVENTFS_FILE_INODE_INO		0x12c4e37
-#define EVENTFS_DIR_INODE_INO		0x134b2f5
+
+/* Just try to make something consistent and unique */
+static int eventfs_dir_ino(struct eventfs_inode *ei)
+{
+	if (!ei->ino)
+		ei->ino = get_next_ino();
+
+	return ei->ino;
+}
 
 /*
  * The eventfs_inode (ei) itself is protected by SRCU. It is released from
@@ -396,7 +404,7 @@ static struct dentry *create_dir(struct eventfs_inode *ei, struct dentry *parent
 	inode->i_fop = &eventfs_file_operations;
 
 	/* All directories will have the same inode number */
-	inode->i_ino = EVENTFS_DIR_INODE_INO;
+	inode->i_ino = eventfs_dir_ino(ei);
 
 	ti = get_tracefs(inode);
 	ti->flags |= TRACEFS_EVENT_INODE;
@@ -802,7 +810,7 @@ static int eventfs_iterate(struct file *file, struct dir_context *ctx)
 
 		name = ei_child->name;
 
-		ino = EVENTFS_DIR_INODE_INO;
+		ino = eventfs_dir_ino(ei_child);
 
 		if (!dir_emit(ctx, name, strlen(name), ino, DT_DIR))
 			goto out_dec;
diff --git a/fs/tracefs/internal.h b/fs/tracefs/internal.h
index 12b7d0150ae9..45397df9bb65 100644
--- a/fs/tracefs/internal.h
+++ b/fs/tracefs/internal.h
@@ -55,6 +55,10 @@ struct eventfs_inode {
 	struct eventfs_attr		*entry_attrs;
 	struct eventfs_attr		attr;
 	void				*data;
+	unsigned int			is_freed:1;
+	unsigned int			is_events:1;
+	unsigned int			nr_entries:30;
+	unsigned int			ino;
 	/*
 	 * Union - used for deletion
 	 * @llist:	for calling dput() if needed after RCU
@@ -64,9 +68,6 @@ struct eventfs_inode {
 		struct llist_node	llist;
 		struct rcu_head		rcu;
 	};
-	unsigned int			is_freed:1;
-	unsigned int			is_events:1;
-	unsigned int			nr_entries:30;
 };
 
 static inline struct tracefs_inode *get_tracefs(const struct inode *inode)
diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c
index 98aaca933bdd..f362345467fa 100644
--- a/fs/xfs/libxfs/xfs_bmap.c
+++ b/fs/xfs/libxfs/xfs_bmap.c
@@ -3277,7 +3277,7 @@ xfs_bmap_alloc_account(
 	struct xfs_bmalloca	*ap)
 {
 	bool			isrt = XFS_IS_REALTIME_INODE(ap->ip) &&
-					(ap->flags & XFS_BMAPI_ATTRFORK);
+					!(ap->flags & XFS_BMAPI_ATTRFORK);
 	uint			fld;
 
 	if (ap->flags & XFS_BMAPI_COWFORK) {