[09/12] bcache: fix for gc and write-back race

Commit Message

Coly Li Sept. 6, 2017, 6:25 a.m. UTC

From: Tang Junhui <[e-mail]>

gc and write-back get raced (see the email "bcache get stucked" I sended
before):
gc thread                               write-back thread
|                                       |bch_writeback_thread()
|bch_gc_thread()                        |
|                                       |==>read_dirty()
|==>bch_btree_gc()                      |
|==>btree_root() //get btree root       |
|                //node write locker    |
|==>bch_btree_gc_root()                 |
|                                       |==>read_dirty_submit()
|                                       |==>write_dirty()
|                                       |==>continue_at(cl,
|                                       |               write_dirty_finish,
|                                       |               system_wq);
|                                       |==>write_dirty_finish()//excute
|                                       |               //in system_wq
|                                       |==>bch_btree_insert()
|                                       |==>bch_btree_map_leaf_nodes()
|                                       |==>__bch_btree_map_nodes()
|                                       |==>btree_root //try to get btree
|                                       |              //root node read
|                                       |              //lock
|                                       |-----stuck here
|==>bch_btree_set_root()
|==>bch_journal_meta()
|==>bch_journal()
|==>journal_try_write()
|==>journal_write_unlocked() //journal_full(&c->journal)
|                            //condition satisfied
|==>continue_at(cl, journal_write, system_wq); //try to excute
|                               //journal_write in system_wq
|                               //but work queue is excuting
|                               //write_dirty_finish()
|==>closure_sync(); //wait journal_write execute
|                   //over and wake up gc,
|-------------stuck here
|==>release root node write locker

This patch alloc a separate work-queue for write-back thread to avoid such
race.

(Commit log re-organized by Coly Li to pass checkpatch.pl checking)

Signed-off-by: Tang Junhui <[e-mail]>
Acked-by: Coly Li <[e-mail]>
Cc: [e-mail]
---
 drivers/md/bcache/bcache.h    | 1 +
 drivers/md/bcache/super.c     | 2 ++
 drivers/md/bcache/writeback.c | 9 +++++++--
 3 files changed, 10 insertions(+), 2 deletions(-)

Patch

diff --git a/drivers/md/bcache/bcache.h b/drivers/md/bcache/bcache.h
index dee542fff68e..2ed9bd231d84 100644
--- a/drivers/md/bcache/bcache.h
+++ b/drivers/md/bcache/bcache.h
@@ -333,6 +333,7 @@  struct cached_dev {
 	/* Limit number of writeback bios in flight */
 	struct semaphore	in_flight;
 	struct task_struct	*writeback_thread;
+	struct workqueue_struct	*writeback_write_wq;
 
 	struct keybuf		writeback_keys;
 
diff --git a/drivers/md/bcache/super.c b/drivers/md/bcache/super.c
index 9a2c190745b6..253918972335 100644
--- a/drivers/md/bcache/super.c
+++ b/drivers/md/bcache/super.c
@@ -1059,6 +1059,8 @@  static void cached_dev_free(struct closure *cl)
 	cancel_delayed_work_sync(&dc->writeback_rate_update);
 	if (!IS_ERR_OR_NULL(dc->writeback_thread))
 		kthread_stop(dc->writeback_thread);
+	if (dc->writeback_write_wq)
+		destroy_workqueue(dc->writeback_write_wq);
 
 	mutex_lock(&bch_register_lock);
 
diff --git a/drivers/md/bcache/writeback.c b/drivers/md/bcache/writeback.c
index b533c2292ba5..323551f7cb28 100644
--- a/drivers/md/bcache/writeback.c
+++ b/drivers/md/bcache/writeback.c
@@ -187,7 +187,7 @@  static void write_dirty(struct closure *cl)
 
 	closure_bio_submit(&io->bio, cl);
 
-	continue_at(cl, write_dirty_finish, system_wq);
+	continue_at(cl, write_dirty_finish, io->dc->writeback_write_wq);
 }
 
 static void read_dirty_endio(struct bio *bio)
@@ -207,7 +207,7 @@  static void read_dirty_submit(struct closure *cl)
 
 	closure_bio_submit(&io->bio, cl);
 
-	continue_at(cl, write_dirty, system_wq);
+	continue_at(cl, write_dirty, io->dc->writeback_write_wq);
 }
 
 static void read_dirty(struct cached_dev *dc)
@@ -517,6 +517,11 @@  void bch_cached_dev_writeback_init(struct cached_dev *dc)
 
 int bch_cached_dev_writeback_start(struct cached_dev *dc)
 {
+	dc->writeback_write_wq = alloc_workqueue("bcache_writeback_wq",
+						WQ_MEM_RECLAIM, 0);
+	if (!dc->writeback_write_wq)
+		return -ENOMEM;
+
 	dc->writeback_thread = kthread_create(bch_writeback_thread, dc,
 					      "bcache_writeback");
 	if (IS_ERR(dc->writeback_thread))