Merge git://git.kernel.org/pub/scm/linux/kernel/git/dhowells/linux-2.6-fscache

* git://git.kernel.org/pub/scm/linux/kernel/git/dhowells/linux-2.6-fscache: (31 commits)
  FS-Cache: Provide nop fscache_stat_d() if CONFIG_FSCACHE_STATS=n
  SLOW_WORK: Fix GFS2 to #include <linux/module.h> before using THIS_MODULE
  SLOW_WORK: Fix CIFS to pass THIS_MODULE to slow_work_register_user()
  CacheFiles: Don't log lookup/create failing with ENOBUFS
  CacheFiles: Catch an overly long wait for an old active object
  CacheFiles: Better showing of debugging information in active object problems
  CacheFiles: Mark parent directory locks as I_MUTEX_PARENT to keep lockdep happy
  CacheFiles: Handle truncate unlocking the page we're reading
  CacheFiles: Don't write a full page if there's only a partial page to cache
  FS-Cache: Actually requeue an object when requested
  FS-Cache: Start processing an object's operations on that object's death
  FS-Cache: Make sure FSCACHE_COOKIE_LOOKING_UP cleared on lookup failure
  FS-Cache: Add a retirement stat counter
  FS-Cache: Handle pages pending storage that get evicted under OOM conditions
  FS-Cache: Handle read request vs lookup, creation or other cache failure
  FS-Cache: Don't delete pending pages from the page-store tracking tree
  FS-Cache: Fix lock misorder in fscache_write_op()
  FS-Cache: The object-available state can't rely on the cookie to be available
  FS-Cache: Permit cache retrieval ops to be interrupted in the initial wait phase
  FS-Cache: Use radix tree preload correctly in tracking of pages to be stored
  ...

1 files changed, 224 insertions, 49 deletions

diff --git a/fs/fscache/page.c b/fs/fscache/page.c
index 2568e0eb644..c598ea4c4e7 100644
--- a/fs/fscache/page.c
+++ b/fs/fscache/page.c
@@ -43,18 +43,102 @@ void __fscache_wait_on_page_write(struct fscache_cookie *cookie, struct page *pa
 EXPORT_SYMBOL(__fscache_wait_on_page_write);
 
 /*
- * note that a page has finished being written to the cache
+ * decide whether a page can be released, possibly by cancelling a store to it
+ * - we're allowed to sleep if __GFP_WAIT is flagged
  */
-static void fscache_end_page_write(struct fscache_cookie *cookie, struct page *page)
+bool __fscache_maybe_release_page(struct fscache_cookie *cookie,
+				  struct page *page,
+				  gfp_t gfp)
 {
 	struct page *xpage;
+	void *val;
+
+	_enter("%p,%p,%x", cookie, page, gfp);
+
+	rcu_read_lock();
+	val = radix_tree_lookup(&cookie->stores, page->index);
+	if (!val) {
+		rcu_read_unlock();
+		fscache_stat(&fscache_n_store_vmscan_not_storing);
+		__fscache_uncache_page(cookie, page);
+		return true;
+	}
+
+	/* see if the page is actually undergoing storage - if so we can't get
+	 * rid of it till the cache has finished with it */
+	if (radix_tree_tag_get(&cookie->stores, page->index,
+			       FSCACHE_COOKIE_STORING_TAG)) {
+		rcu_read_unlock();
+		goto page_busy;
+	}
+
+	/* the page is pending storage, so we attempt to cancel the store and
+	 * discard the store request so that the page can be reclaimed */
+	spin_lock(&cookie->stores_lock);
+	rcu_read_unlock();
+
+	if (radix_tree_tag_get(&cookie->stores, page->index,
+			       FSCACHE_COOKIE_STORING_TAG)) {
+		/* the page started to undergo storage whilst we were looking,
+		 * so now we can only wait or return */
+		spin_unlock(&cookie->stores_lock);
+		goto page_busy;
+	}
 
-	spin_lock(&cookie->lock);
 	xpage = radix_tree_delete(&cookie->stores, page->index);
-	spin_unlock(&cookie->lock);
-	ASSERT(xpage != NULL);
+	spin_unlock(&cookie->stores_lock);
+
+	if (xpage) {
+		fscache_stat(&fscache_n_store_vmscan_cancelled);
+		fscache_stat(&fscache_n_store_radix_deletes);
+		ASSERTCMP(xpage, ==, page);
+	} else {
+		fscache_stat(&fscache_n_store_vmscan_gone);
+	}
 
 	wake_up_bit(&cookie->flags, 0);
+	if (xpage)
+		page_cache_release(xpage);
+	__fscache_uncache_page(cookie, page);
+	return true;
+
+page_busy:
+	/* we might want to wait here, but that could deadlock the allocator as
+	 * the slow-work threads writing to the cache may all end up sleeping
+	 * on memory allocation */
+	fscache_stat(&fscache_n_store_vmscan_busy);
+	return false;
+}
+EXPORT_SYMBOL(__fscache_maybe_release_page);
+
+/*
+ * note that a page has finished being written to the cache
+ */
+static void fscache_end_page_write(struct fscache_object *object,
+				   struct page *page)
+{
+	struct fscache_cookie *cookie;
+	struct page *xpage = NULL;
+
+	spin_lock(&object->lock);
+	cookie = object->cookie;
+	if (cookie) {
+		/* delete the page from the tree if it is now no longer
+		 * pending */
+		spin_lock(&cookie->stores_lock);
+		radix_tree_tag_clear(&cookie->stores, page->index,
+				     FSCACHE_COOKIE_STORING_TAG);
+		if (!radix_tree_tag_get(&cookie->stores, page->index,
+					FSCACHE_COOKIE_PENDING_TAG)) {
+			fscache_stat(&fscache_n_store_radix_deletes);
+			xpage = radix_tree_delete(&cookie->stores, page->index);
+		}
+		spin_unlock(&cookie->stores_lock);
+		wake_up_bit(&cookie->flags, 0);
+	}
+	spin_unlock(&object->lock);
+	if (xpage)
+		page_cache_release(xpage);
 }
 
 /*
@@ -63,14 +147,21 @@ static void fscache_end_page_write(struct fscache_cookie *cookie, struct page *p
 static void fscache_attr_changed_op(struct fscache_operation *op)
 {
 	struct fscache_object *object = op->object;
+	int ret;
 
 	_enter("{OBJ%x OP%x}", object->debug_id, op->debug_id);
 
 	fscache_stat(&fscache_n_attr_changed_calls);
 
-	if (fscache_object_is_active(object) &&
-	    object->cache->ops->attr_changed(object) < 0)
-		fscache_abort_object(object);
+	if (fscache_object_is_active(object)) {
+		fscache_set_op_state(op, "CallFS");
+		fscache_stat(&fscache_n_cop_attr_changed);
+		ret = object->cache->ops->attr_changed(object);
+		fscache_stat_d(&fscache_n_cop_attr_changed);
+		fscache_set_op_state(op, "Done");
+		if (ret < 0)
+			fscache_abort_object(object);
+	}
 
 	_leave("");
 }
@@ -99,6 +190,7 @@ int __fscache_attr_changed(struct fscache_cookie *cookie)
 	fscache_operation_init(op, NULL);
 	fscache_operation_init_slow(op, fscache_attr_changed_op);
 	op->flags = FSCACHE_OP_SLOW | (1 << FSCACHE_OP_EXCLUSIVE);
+	fscache_set_op_name(op, "Attr");
 
 	spin_lock(&cookie->lock);
 
@@ -184,6 +276,7 @@ static struct fscache_retrieval *fscache_alloc_retrieval(
 	op->start_time	= jiffies;
 	INIT_WORK(&op->op.fast_work, fscache_retrieval_work);
 	INIT_LIST_HEAD(&op->to_do);
+	fscache_set_op_name(&op->op, "Retr");
 	return op;
 }
 
@@ -221,6 +314,43 @@ static int fscache_wait_for_deferred_lookup(struct fscache_cookie *cookie)
 }
 
 /*
+ * wait for an object to become active (or dead)
+ */
+static int fscache_wait_for_retrieval_activation(struct fscache_object *object,
+						 struct fscache_retrieval *op,
+						 atomic_t *stat_op_waits,
+						 atomic_t *stat_object_dead)
+{
+	int ret;
+
+	if (!test_bit(FSCACHE_OP_WAITING, &op->op.flags))
+		goto check_if_dead;
+
+	_debug(">>> WT");
+	fscache_stat(stat_op_waits);
+	if (wait_on_bit(&op->op.flags, FSCACHE_OP_WAITING,
+			fscache_wait_bit_interruptible,
+			TASK_INTERRUPTIBLE) < 0) {
+		ret = fscache_cancel_op(&op->op);
+		if (ret == 0)
+			return -ERESTARTSYS;
+
+		/* it's been removed from the pending queue by another party,
+		 * so we should get to run shortly */
+		wait_on_bit(&op->op.flags, FSCACHE_OP_WAITING,
+			    fscache_wait_bit, TASK_UNINTERRUPTIBLE);
+	}
+	_debug("<<< GO");
+
+check_if_dead:
+	if (unlikely(fscache_object_is_dead(object))) {
+		fscache_stat(stat_object_dead);
+		return -ENOBUFS;
+	}
+	return 0;
+}
+
+/*
  * read a page from the cache or allocate a block in which to store it
  * - we return:
  *   -ENOMEM	- out of memory, nothing done
@@ -257,6 +387,7 @@ int __fscache_read_or_alloc_page(struct fscache_cookie *cookie,
 		_leave(" = -ENOMEM");
 		return -ENOMEM;
 	}
+	fscache_set_op_name(&op->op, "RetrRA1");
 
 	spin_lock(&cookie->lock);
 
@@ -267,6 +398,9 @@ int __fscache_read_or_alloc_page(struct fscache_cookie *cookie,
 
 	ASSERTCMP(object->state, >, FSCACHE_OBJECT_LOOKING_UP);
 
+	atomic_inc(&object->n_reads);
+	set_bit(FSCACHE_OP_DEC_READ_CNT, &op->op.flags);
+
 	if (fscache_submit_op(object, &op->op) < 0)
 		goto nobufs_unlock;
 	spin_unlock(&cookie->lock);
@@ -279,23 +413,27 @@ int __fscache_read_or_alloc_page(struct fscache_cookie *cookie,
 
 	/* we wait for the operation to become active, and then process it
 	 * *here*, in this thread, and not in the thread pool */
-	if (test_bit(FSCACHE_OP_WAITING, &op->op.flags)) {
-		_debug(">>> WT");
-		fscache_stat(&fscache_n_retrieval_op_waits);
-		wait_on_bit(&op->op.flags, FSCACHE_OP_WAITING,
-			    fscache_wait_bit, TASK_UNINTERRUPTIBLE);
-		_debug("<<< GO");
-	}
+	ret = fscache_wait_for_retrieval_activation(
+		object, op,
+		__fscache_stat(&fscache_n_retrieval_op_waits),
+		__fscache_stat(&fscache_n_retrievals_object_dead));
+	if (ret < 0)
+		goto error;
 
 	/* ask the cache to honour the operation */
 	if (test_bit(FSCACHE_COOKIE_NO_DATA_YET, &object->cookie->flags)) {
+		fscache_stat(&fscache_n_cop_allocate_page);
 		ret = object->cache->ops->allocate_page(op, page, gfp);
+		fscache_stat_d(&fscache_n_cop_allocate_page);
 		if (ret == 0)
 			ret = -ENODATA;
 	} else {
+		fscache_stat(&fscache_n_cop_read_or_alloc_page);
 		ret = object->cache->ops->read_or_alloc_page(op, page, gfp);
+		fscache_stat_d(&fscache_n_cop_read_or_alloc_page);
 	}
 
+error:
 	if (ret == -ENOMEM)
 		fscache_stat(&fscache_n_retrievals_nomem);
 	else if (ret == -ERESTARTSYS)
@@ -347,7 +485,6 @@ int __fscache_read_or_alloc_pages(struct fscache_cookie *cookie,
 				  void *context,
 				  gfp_t gfp)
 {
-	fscache_pages_retrieval_func_t func;
 	struct fscache_retrieval *op;
 	struct fscache_object *object;
 	int ret;
@@ -369,6 +506,7 @@ int __fscache_read_or_alloc_pages(struct fscache_cookie *cookie,
 	op = fscache_alloc_retrieval(mapping, end_io_func, context);
 	if (!op)
 		return -ENOMEM;
+	fscache_set_op_name(&op->op, "RetrRAN");
 
 	spin_lock(&cookie->lock);
 
@@ -377,6 +515,9 @@ int __fscache_read_or_alloc_pages(struct fscache_cookie *cookie,
 	object = hlist_entry(cookie->backing_objects.first,
 			     struct fscache_object, cookie_link);
 
+	atomic_inc(&object->n_reads);
+	set_bit(FSCACHE_OP_DEC_READ_CNT, &op->op.flags);
+
 	if (fscache_submit_op(object, &op->op) < 0)
 		goto nobufs_unlock;
 	spin_unlock(&cookie->lock);
@@ -389,21 +530,27 @@ int __fscache_read_or_alloc_pages(struct fscache_cookie *cookie,
 
 	/* we wait for the operation to become active, and then process it
 	 * *here*, in this thread, and not in the thread pool */
-	if (test_bit(FSCACHE_OP_WAITING, &op->op.flags)) {
-		_debug(">>> WT");
-		fscache_stat(&fscache_n_retrieval_op_waits);
-		wait_on_bit(&op->op.flags, FSCACHE_OP_WAITING,
-			    fscache_wait_bit, TASK_UNINTERRUPTIBLE);
-		_debug("<<< GO");
-	}
+	ret = fscache_wait_for_retrieval_activation(
+		object, op,
+		__fscache_stat(&fscache_n_retrieval_op_waits),
+		__fscache_stat(&fscache_n_retrievals_object_dead));
+	if (ret < 0)
+		goto error;
 
 	/* ask the cache to honour the operation */
-	if (test_bit(FSCACHE_COOKIE_NO_DATA_YET, &object->cookie->flags))
-		func = object->cache->ops->allocate_pages;
-	else
-		func = object->cache->ops->read_or_alloc_pages;
-	ret = func(op, pages, nr_pages, gfp);
+	if (test_bit(FSCACHE_COOKIE_NO_DATA_YET, &object->cookie->flags)) {
+		fscache_stat(&fscache_n_cop_allocate_pages);
+		ret = object->cache->ops->allocate_pages(
+			op, pages, nr_pages, gfp);
+		fscache_stat_d(&fscache_n_cop_allocate_pages);
+	} else {
+		fscache_stat(&fscache_n_cop_read_or_alloc_pages);
+		ret = object->cache->ops->read_or_alloc_pages(
+			op, pages, nr_pages, gfp);
+		fscache_stat_d(&fscache_n_cop_read_or_alloc_pages);
+	}
 
+error:
 	if (ret == -ENOMEM)
 		fscache_stat(&fscache_n_retrievals_nomem);
 	else if (ret == -ERESTARTSYS)
@@ -461,6 +608,7 @@ int __fscache_alloc_page(struct fscache_cookie *cookie,
 	op = fscache_alloc_retrieval(page->mapping, NULL, NULL);
 	if (!op)
 		return -ENOMEM;
+	fscache_set_op_name(&op->op, "RetrAL1");
 
 	spin_lock(&cookie->lock);
 
@@ -475,18 +623,22 @@ int __fscache_alloc_page(struct fscache_cookie *cookie,
 
 	fscache_stat(&fscache_n_alloc_ops);
 
-	if (test_bit(FSCACHE_OP_WAITING, &op->op.flags)) {
-		_debug(">>> WT");
-		fscache_stat(&fscache_n_alloc_op_waits);
-		wait_on_bit(&op->op.flags, FSCACHE_OP_WAITING,
-			    fscache_wait_bit, TASK_UNINTERRUPTIBLE);
-		_debug("<<< GO");
-	}
+	ret = fscache_wait_for_retrieval_activation(
+		object, op,
+		__fscache_stat(&fscache_n_alloc_op_waits),
+		__fscache_stat(&fscache_n_allocs_object_dead));
+	if (ret < 0)
+		goto error;
 
 	/* ask the cache to honour the operation */
+	fscache_stat(&fscache_n_cop_allocate_page);
 	ret = object->cache->ops->allocate_page(op, page, gfp);
+	fscache_stat_d(&fscache_n_cop_allocate_page);
 
-	if (ret < 0)
+error:
+	if (ret == -ERESTARTSYS)
+		fscache_stat(&fscache_n_allocs_intr);
+	else if (ret < 0)
 		fscache_stat(&fscache_n_allocs_nobufs);
 	else
 		fscache_stat(&fscache_n_allocs_ok);
@@ -521,7 +673,7 @@ static void fscache_write_op(struct fscache_operation *_op)
 	struct fscache_storage *op =
 		container_of(_op, struct fscache_storage, op);
 	struct fscache_object *object = op->op.object;
-	struct fscache_cookie *cookie = object->cookie;
+	struct fscache_cookie *cookie;
 	struct page *page;
 	unsigned n;
 	void *results[1];
@@ -529,16 +681,19 @@ static void fscache_write_op(struct fscache_operation *_op)
 
 	_enter("{OP%x,%d}", op->op.debug_id, atomic_read(&op->op.usage));
 
-	spin_lock(&cookie->lock);
+	fscache_set_op_state(&op->op, "GetPage");
+
 	spin_lock(&object->lock);
+	cookie = object->cookie;
 
-	if (!fscache_object_is_active(object)) {
+	if (!fscache_object_is_active(object) || !cookie) {
 		spin_unlock(&object->lock);
-		spin_unlock(&cookie->lock);
 		_leave("");
 		return;
 	}
 
+	spin_lock(&cookie->stores_lock);
+
 	fscache_stat(&fscache_n_store_calls);
 
 	/* find a page to store */
@@ -549,23 +704,35 @@ static void fscache_write_op(struct fscache_operation *_op)
 		goto superseded;
 	page = results[0];
 	_debug("gang %d [%lx]", n, page->index);
-	if (page->index > op->store_limit)
+	if (page->index > op->store_limit) {
+		fscache_stat(&fscache_n_store_pages_over_limit);
 		goto superseded;
+	}
 
-	radix_tree_tag_clear(&cookie->stores, page->index,
-			     FSCACHE_COOKIE_PENDING_TAG);
+	if (page) {
+		radix_tree_tag_set(&cookie->stores, page->index,
+				   FSCACHE_COOKIE_STORING_TAG);
+		radix_tree_tag_clear(&cookie->stores, page->index,
+				     FSCACHE_COOKIE_PENDING_TAG);
+	}
 
+	spin_unlock(&cookie->stores_lock);
 	spin_unlock(&object->lock);
-	spin_unlock(&cookie->lock);
 
 	if (page) {
+		fscache_set_op_state(&op->op, "Store");
+		fscache_stat(&fscache_n_store_pages);
+		fscache_stat(&fscache_n_cop_write_page);
 		ret = object->cache->ops->write_page(op, page);
-		fscache_end_page_write(cookie, page);
-		page_cache_release(page);
-		if (ret < 0)
+		fscache_stat_d(&fscache_n_cop_write_page);
+		fscache_set_op_state(&op->op, "EndWrite");
+		fscache_end_page_write(object, page);
+		if (ret < 0) {
+			fscache_set_op_state(&op->op, "Abort");
 			fscache_abort_object(object);
-		else
+		} else {
 			fscache_enqueue_operation(&op->op);
+		}
 	}
 
 	_leave("");
@@ -575,9 +742,9 @@ superseded:
 	/* this writer is going away and there aren't any more things to
 	 * write */
 	_debug("cease");
+	spin_unlock(&cookie->stores_lock);
 	clear_bit(FSCACHE_OBJECT_PENDING_WRITE, &object->flags);
 	spin_unlock(&object->lock);
-	spin_unlock(&cookie->lock);
 	_leave("");
 }
 
@@ -634,6 +801,7 @@ int __fscache_write_page(struct fscache_cookie *cookie,
 	fscache_operation_init(&op->op, fscache_release_write_op);
 	fscache_operation_init_slow(&op->op, fscache_write_op);
 	op->op.flags = FSCACHE_OP_SLOW | (1 << FSCACHE_OP_WAITING);
+	fscache_set_op_name(&op->op, "Write1");
 
 	ret = radix_tree_preload(gfp & ~__GFP_HIGHMEM);
 	if (ret < 0)
@@ -652,6 +820,7 @@ int __fscache_write_page(struct fscache_cookie *cookie,
 	/* add the page to the pending-storage radix tree on the backing
 	 * object */
 	spin_lock(&object->lock);
+	spin_lock(&cookie->stores_lock);
 
 	_debug("store limit %llx", (unsigned long long) object->store_limit);
 
@@ -672,6 +841,7 @@ int __fscache_write_page(struct fscache_cookie *cookie,
 	if (test_and_set_bit(FSCACHE_OBJECT_PENDING_WRITE, &object->flags))
 		goto already_pending;
 
+	spin_unlock(&cookie->stores_lock);
 	spin_unlock(&object->lock);
 
 	op->op.debug_id	= atomic_inc_return(&fscache_op_debug_id);
@@ -693,6 +863,7 @@ int __fscache_write_page(struct fscache_cookie *cookie,
 already_queued:
 	fscache_stat(&fscache_n_stores_again);
 already_pending:
+	spin_unlock(&cookie->stores_lock);
 	spin_unlock(&object->lock);
 	spin_unlock(&cookie->lock);
 	radix_tree_preload_end();
@@ -702,7 +873,9 @@ already_pending:
 	return 0;
 
 submit_failed:
+	spin_lock(&cookie->stores_lock);
 	radix_tree_delete(&cookie->stores, page->index);
+	spin_unlock(&cookie->stores_lock);
 	page_cache_release(page);
 	ret = -ENOBUFS;
 	goto nobufs;
@@ -763,7 +936,9 @@ void __fscache_uncache_page(struct fscache_cookie *cookie, struct page *page)
 	if (TestClearPageFsCache(page) &&
 	    object->cache->ops->uncache_page) {
 		/* the cache backend releases the cookie lock */
+		fscache_stat(&fscache_n_cop_uncache_page);
 		object->cache->ops->uncache_page(object, page);
+		fscache_stat_d(&fscache_n_cop_uncache_page);
 		goto done;
 	}