Merge patch series "netfs: Read performance improvements and "single-blob" support"

David Howells <[email protected]> says:

This set of patches is primarily about two things: improving read
performance and supporting monolithic single-blob objects that have to be
read/written as such (e.g. AFS directory contents).  The implementation of
the two parts is interwoven as each makes the other possible.

READ PERFORMANCE
================

The read performance improvements are intended to speed up some loss of
performance detected in cifs and to a lesser extend in afs.  The problem is
that we queue too many work items during the collection of read results:
each individual subrequest is collected by its own work item, and then they
have to interact with each other when a series of subrequests don't exactly
align with the pattern of folios that are being read by the overall
request.

Whilst the processing of the pages covered by individual subrequests as
they complete potentially allows folios to be woken in parallel and with
minimum delay, it can shuffle wakeups for sequential reads out of order -
and that is the most common I/O pattern.

The final assessment and cleanup of an operation is then held up until the
last I/O completes - and for a synchronous sequential operation, this means
the bouncing around of work items just adds latency.

Two changes have been made to make this work:

 (1) All collection is now done in a single "work item" that works
     progressively through the subrequests as they complete (and also
     dispatches retries as necessary).

 (2) For readahead and AIO, this work item be done on a workqueue and can
     run in parallel with the ultimate consumer of the data; for
     synchronous direct or unbuffered reads, the collection is run in the
     application thread and not offloaded.

Functions such as smb2_readv_callback() then just tell netfslib that the
subrequest has terminated; netfslib does a minimal bit of processing on the
spot - stat counting and tracing mostly - and then queues/wakes up the
worker.  This simplifies the logic as the collector just walks sequentially
through the subrequests as they complete and walks through the folios, if
buffered, unlocking them as it goes.  It also keeps to a minimum the amount
of latency injected into the filesystem's low-level I/O handling

The way netfs supports filesystems using the deprecated PG_private_2 flag
is changed: folios are flagged and added to a write request as they
complete and that takes care of scheduling the writes to the cache.  The
originating read request can then just unlock the pages whatever happens.

SINGLE-BLOB OBJECT SUPPORT
==========================

Single-blob objects are files for which the content of the file must be
read from or written to the server in a single operation because reading
them in parts may yield inconsistent results.  AFS directories are an
example of this as there exists the possibility that the contents are
generated on the fly and would differ between reads or might change due to
third party interference.

Such objects will be written to and retrieved from the cache if one is
present, though we allow/may need to propose multiple subrequests to do so.
The important part is that read from/write to the *server* is monolithic.

Single blob reading is, for the moment, fully synchronous and does result
collection in the application thread and, also for the moment, the API is
supplied the buffer in the form of a folio_queue chain rather than using
the pagecache.

AFS CHANGES
===========

This series makes a number of changes to the kafs filesystem, primarily in
the area of directory handling:

 (1) AFS's FetchData RPC reply processing is made partially asynchronous
     which allows the netfs_io_request's outstanding operation counter to
     be removed as part of reducing the collection to a single work item.

 (2) Directory and symlink reading are plumbed through netfslib using the
     single-blob object API and are now cacheable with fscache.  This also
     allows the afs_read struct to be eliminated and netfs_io_subrequest to
     be used directly instead.

 (3) Directory and symlink content are now stored in a folio_queue buffer
     rather than in the pagecache.  This means we don't require the RCU
     read lock and xarray iteration to access it, and folios won't randomly
     disappear under us because the VM wants them back.

     There are some downsides to this, though: the storage folios are no
     longer known to the VM, drop_caches can't flush them, the folios are
     not migrateable.  The inode must also be marked dirty manually to get
     the data written to the cache in the background.

 (4) The vnode operation lock is changed from a mutex struct to a private
     lock implementation.  The problem is that the lock now needs to be
     dropped in a separate thread and mutexes don't permit that.

 (5) When a new directory or symlink is created, we now initialise it
     locally and mark it valid rather than downloading it (we know what
     it's likely to look like).

 (6) We now use the in-directory hashtable to reduce the number of entries
     we need to scan when doing a lookup.  The edit routines have to
     maintain the hash chains.

 (7) Cancellation (e.g. by signal) of an async call after the rxrpc_call
     has been set up is now offloaded to the worker thread as there will be
     a notification from rxrpc upon completion.  This avoids a double
     cleanup.

SUPPORTING CHANGES
==================

To support the above some other changes are also made:

 (1) A "rolling buffer" implementation is created to abstract out the two
     separate folio_queue chaining implementations I had (one for read and
     one for write).

 (2) Functions are provided to create/extend a buffer in a folio_queue
     chain and tear it down again.  This is used to handle AFS directories,
     but could also be used to create bounce buffers for content crypto and
     transport crypto.

 (3) The was_async argument is dropped from netfs_read_subreq_terminated().
     Instead we wake the read collection work item by either queuing it or
     waking up the app thread.

 (4) We don't need to use BH-excluding locks when communicating between the
     issuing thread and the collection thread as neither of them now run in
     BH context.

MISCELLANY
==========

Also included are a number of new tracepoints; a split of the netfslib
write collection code to put retrying into its own file (it gets more
complicated with content encryption).

There are also some minor fixes AFS included, including fixing the AFS
directory format struct layout, reducing some directory over-invalidation
and making afs_mkdir() translate EEXIST to ENOTEMPY (which is not available
on all systems the servers support).

Finally, there's a patch to try and detect entry into the folio unlock
function with no folio_queue structs in the buffer (which isn't allowed in
the cases that can get there).  This is a debugging patch, but should be
minimal overhead.

* patches from https://lore.kernel.org/r/[email protected]: (31 commits)
  netfs: Report on NULL folioq in netfs_writeback_unlock_folios()
  afs: Add a tracepoint for afs_read_receive()
  afs: Locally initialise the contents of a new symlink on creation
  afs: Use the contained hashtable to search a directory
  afs: Make afs_mkdir() locally initialise a new directory's content
  netfs: Change the read result collector to only use one work item
  afs: Make {Y,}FS.FetchData an asynchronous operation
  afs: Fix cleanup of immediately failed async calls
  afs: Eliminate afs_read
  afs: Use netfslib for symlinks, allowing them to be cached
  afs: Use netfslib for directories
  afs: Make afs_init_request() get a key if not given a file
  netfs: Add support for caching single monolithic objects such as AFS dirs
  netfs: Add functions to build/clean a buffer in a folio_queue
  afs: Add more tracepoints to do with tracking validity
  cachefiles: Add auxiliary data trace
  cachefiles: Add some subrequest tracepoints
  netfs: Remove some extraneous directory invalidations
  afs: Fix directory format encoding struct
  afs: Fix EEXIST error returned from afs_rmdir() to be ENOTEMPTY
  ...

Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Christian Brauner <[email protected]>

1 files changed, 147 insertions, 111 deletions

diff --git a/fs/afs/file.c b/fs/afs/file.c
index 6762eff97517..fc15497608c6 100644
--- a/fs/afs/file.c
+++ b/fs/afs/file.c
@@ -20,7 +20,6 @@
 #include "internal.h"
 
 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 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,
@@ -61,13 +60,6 @@ const struct address_space_operations afs_file_aops = {
 	.writepages	= afs_writepages,
 };
 
-const struct address_space_operations afs_symlink_aops = {
-	.read_folio	= afs_symlink_read_folio,
-	.release_folio	= netfs_release_folio,
-	.invalidate_folio = netfs_invalidate_folio,
-	.migrate_folio	= filemap_migrate_folio,
-};
-
 static const struct vm_operations_struct afs_vm_ops = {
 	.open		= afs_vm_open,
 	.close		= afs_vm_close,
@@ -208,49 +200,12 @@ int afs_release(struct inode *inode, struct file *file)
 	return ret;
 }
 
-/*
- * Allocate a new read record.
- */
-struct afs_read *afs_alloc_read(gfp_t gfp)
-{
-	struct afs_read *req;
-
-	req = kzalloc(sizeof(struct afs_read), gfp);
-	if (req)
-		refcount_set(&req->usage, 1);
-
-	return req;
-}
-
-/*
- * Dispose of a ref to a read record.
- */
-void afs_put_read(struct afs_read *req)
-{
-	if (refcount_dec_and_test(&req->usage)) {
-		if (req->cleanup)
-			req->cleanup(req);
-		key_put(req->key);
-		kfree(req);
-	}
-}
-
 static void afs_fetch_data_notify(struct afs_operation *op)
 {
-	struct afs_read *req = op->fetch.req;
-	struct netfs_io_subrequest *subreq = req->subreq;
-	int error = afs_op_error(op);
+	struct netfs_io_subrequest *subreq = op->fetch.subreq;
 
-	req->error = error;
-	if (subreq) {
-		subreq->rreq->i_size = req->file_size;
-		if (req->pos + req->actual_len >= req->file_size)
-			__set_bit(NETFS_SREQ_HIT_EOF, &subreq->flags);
-		netfs_read_subreq_terminated(subreq, error, false);
-		req->subreq = NULL;
-	} else if (req->done) {
-		req->done(req);
-	}
+	subreq->error = afs_op_error(op);
+	netfs_read_subreq_terminated(subreq);
 }
 
 static void afs_fetch_data_success(struct afs_operation *op)
@@ -260,7 +215,7 @@ static void afs_fetch_data_success(struct afs_operation *op)
 	_enter("op=%08x", op->debug_id);
 	afs_vnode_commit_status(op, &op->file[0]);
 	afs_stat_v(vnode, n_fetches);
-	atomic_long_add(op->fetch.req->actual_len, &op->net->n_fetch_bytes);
+	atomic_long_add(op->fetch.subreq->transferred, &op->net->n_fetch_bytes);
 	afs_fetch_data_notify(op);
 }
 
@@ -270,107 +225,188 @@ static void afs_fetch_data_aborted(struct afs_operation *op)
 	afs_fetch_data_notify(op);
 }
 
-static void afs_fetch_data_put(struct afs_operation *op)
-{
-	op->fetch.req->error = afs_op_error(op);
-	afs_put_read(op->fetch.req);
-}
-
-static const struct afs_operation_ops afs_fetch_data_operation = {
+const struct afs_operation_ops afs_fetch_data_operation = {
 	.issue_afs_rpc	= afs_fs_fetch_data,
 	.issue_yfs_rpc	= yfs_fs_fetch_data,
 	.success	= afs_fetch_data_success,
 	.aborted	= afs_fetch_data_aborted,
 	.failed		= afs_fetch_data_notify,
-	.put		= afs_fetch_data_put,
 };
 
+static void afs_issue_read_call(struct afs_operation *op)
+{
+	op->call_responded = false;
+	op->call_error = 0;
+	op->call_abort_code = 0;
+	if (test_bit(AFS_SERVER_FL_IS_YFS, &op->server->flags))
+		yfs_fs_fetch_data(op);
+	else
+		afs_fs_fetch_data(op);
+}
+
+static void afs_end_read(struct afs_operation *op)
+{
+	if (op->call_responded && op->server)
+		set_bit(AFS_SERVER_FL_RESPONDING, &op->server->flags);
+
+	if (!afs_op_error(op))
+		afs_fetch_data_success(op);
+	else if (op->cumul_error.aborted)
+		afs_fetch_data_aborted(op);
+	else
+		afs_fetch_data_notify(op);
+
+	afs_end_vnode_operation(op);
+	afs_put_operation(op);
+}
+
+/*
+ * Perform I/O processing on an asynchronous call.  The work item carries a ref
+ * to the call struct that we either need to release or to pass on.
+ */
+static void afs_read_receive(struct afs_call *call)
+{
+	struct afs_operation *op = call->op;
+	enum afs_call_state state;
+
+	_enter("");
+
+	state = READ_ONCE(call->state);
+	if (state == AFS_CALL_COMPLETE)
+		return;
+	trace_afs_read_recv(op, call);
+
+	while (state < AFS_CALL_COMPLETE && READ_ONCE(call->need_attention)) {
+		WRITE_ONCE(call->need_attention, false);
+		afs_deliver_to_call(call);
+		state = READ_ONCE(call->state);
+	}
+
+	if (state < AFS_CALL_COMPLETE) {
+		netfs_read_subreq_progress(op->fetch.subreq);
+		if (rxrpc_kernel_check_life(call->net->socket, call->rxcall))
+			return;
+		/* rxrpc terminated the call. */
+		afs_set_call_complete(call, call->error, call->abort_code);
+	}
+
+	op->call_abort_code	= call->abort_code;
+	op->call_error		= call->error;
+	op->call_responded	= call->responded;
+	op->call		= NULL;
+	call->op		= NULL;
+	afs_put_call(call);
+
+	/* If the call failed, then we need to crank the server rotation
+	 * handle and try the next.
+	 */
+	if (afs_select_fileserver(op)) {
+		afs_issue_read_call(op);
+		return;
+	}
+
+	afs_end_read(op);
+}
+
+void afs_fetch_data_async_rx(struct work_struct *work)
+{
+	struct afs_call *call = container_of(work, struct afs_call, async_work);
+
+	afs_read_receive(call);
+	afs_put_call(call);
+}
+
+void afs_fetch_data_immediate_cancel(struct afs_call *call)
+{
+	if (call->async) {
+		afs_get_call(call, afs_call_trace_wake);
+		if (!queue_work(afs_async_calls, &call->async_work))
+			afs_deferred_put_call(call);
+		flush_work(&call->async_work);
+	}
+}
+
 /*
  * Fetch file data from the volume.
  */
-int afs_fetch_data(struct afs_vnode *vnode, struct afs_read *req)
+static void afs_issue_read(struct netfs_io_subrequest *subreq)
 {
 	struct afs_operation *op;
+	struct afs_vnode *vnode = AFS_FS_I(subreq->rreq->inode);
+	struct key *key = subreq->rreq->netfs_priv;
 
 	_enter("%s{%llx:%llu.%u},%x,,,",
 	       vnode->volume->name,
 	       vnode->fid.vid,
 	       vnode->fid.vnode,
 	       vnode->fid.unique,
-	       key_serial(req->key));
+	       key_serial(key));
 
-	op = afs_alloc_operation(req->key, vnode->volume);
+	op = afs_alloc_operation(key, vnode->volume);
 	if (IS_ERR(op)) {
-		if (req->subreq)
-			netfs_read_subreq_terminated(req->subreq, PTR_ERR(op), false);
-		return PTR_ERR(op);
+		subreq->error = PTR_ERR(op);
+		netfs_read_subreq_terminated(subreq);
+		return;
 	}
 
 	afs_op_set_vnode(op, 0, vnode);
 
-	op->fetch.req	= afs_get_read(req);
+	op->fetch.subreq = subreq;
 	op->ops		= &afs_fetch_data_operation;
-	return afs_do_sync_operation(op);
-}
-
-static void afs_read_worker(struct work_struct *work)
-{
-	struct netfs_io_subrequest *subreq = container_of(work, struct netfs_io_subrequest, work);
-	struct afs_vnode *vnode = AFS_FS_I(subreq->rreq->inode);
-	struct afs_read *fsreq;
-
-	fsreq = afs_alloc_read(GFP_NOFS);
-	if (!fsreq)
-		return netfs_read_subreq_terminated(subreq, -ENOMEM, false);
-
-	fsreq->subreq	= subreq;
-	fsreq->pos	= subreq->start + subreq->transferred;
-	fsreq->len	= subreq->len   - subreq->transferred;
-	fsreq->key	= key_get(subreq->rreq->netfs_priv);
-	fsreq->vnode	= vnode;
-	fsreq->iter	= &subreq->io_iter;
 
 	trace_netfs_sreq(subreq, netfs_sreq_trace_submit);
-	afs_fetch_data(fsreq->vnode, fsreq);
-	afs_put_read(fsreq);
-}
 
-static void afs_issue_read(struct netfs_io_subrequest *subreq)
-{
-	INIT_WORK(&subreq->work, afs_read_worker);
-	queue_work(system_long_wq, &subreq->work);
-}
-
-static int afs_symlink_read_folio(struct file *file, struct folio *folio)
-{
-	struct afs_vnode *vnode = AFS_FS_I(folio->mapping->host);
-	struct afs_read *fsreq;
-	int ret;
+	if (subreq->rreq->origin == NETFS_READAHEAD ||
+	    subreq->rreq->iocb) {
+		op->flags |= AFS_OPERATION_ASYNC;
 
-	fsreq = afs_alloc_read(GFP_NOFS);
-	if (!fsreq)
-		return -ENOMEM;
+		if (!afs_begin_vnode_operation(op)) {
+			subreq->error = afs_put_operation(op);
+			netfs_read_subreq_terminated(subreq);
+			return;
+		}
 
-	fsreq->pos	= folio_pos(folio);
-	fsreq->len	= folio_size(folio);
-	fsreq->vnode	= vnode;
-	fsreq->iter	= &fsreq->def_iter;
-	iov_iter_xarray(&fsreq->def_iter, ITER_DEST, &folio->mapping->i_pages,
-			fsreq->pos, fsreq->len);
+		if (!afs_select_fileserver(op)) {
+			afs_end_read(op);
+			return;
+		}
 
-	ret = afs_fetch_data(fsreq->vnode, fsreq);
-	if (ret == 0)
-		folio_mark_uptodate(folio);
-	folio_unlock(folio);
-	return ret;
+		afs_issue_read_call(op);
+	} else {
+		afs_do_sync_operation(op);
+	}
 }
 
 static int afs_init_request(struct netfs_io_request *rreq, struct file *file)
 {
+	struct afs_vnode *vnode = AFS_FS_I(rreq->inode);
+
 	if (file)
 		rreq->netfs_priv = key_get(afs_file_key(file));
 	rreq->rsize = 256 * 1024;
 	rreq->wsize = 256 * 1024 * 1024;
+
+	switch (rreq->origin) {
+	case NETFS_READ_SINGLE:
+		if (!file) {
+			struct key *key = afs_request_key(vnode->volume->cell);
+
+			if (IS_ERR(key))
+				return PTR_ERR(key);
+			rreq->netfs_priv = key;
+		}
+		break;
+	case NETFS_WRITEBACK:
+	case NETFS_WRITETHROUGH:
+	case NETFS_UNBUFFERED_WRITE:
+	case NETFS_DIO_WRITE:
+		if (S_ISREG(rreq->inode->i_mode))
+			rreq->io_streams[0].avail = true;
+		break;
+	case NETFS_WRITEBACK_SINGLE:
+	default:
+		break;
+	}
 	return 0;
 }