[PATCH 3.16 103/133] bcache: fix for gc and write-back race

From: Ben Hutchings
Date: Tue Nov 21 2017 - 21:40:44 EST


3.16.51-rc1 review patch. If anyone has any objections, please let me know.

------------------

From: Tang Junhui <tang.junhui@xxxxxxxxxx>

commit 9baf30972b5568d8b5bc8b3c46a6ec5b58100463 upstream.

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 <tang.junhui@xxxxxxxxxx>
Acked-by: Coly Li <colyli@xxxxxxx>
Signed-off-by: Jens Axboe <axboe@xxxxxxxxx>
[bwh: Backported to 3.16: adjust context]
Signed-off-by: Ben Hutchings <ben@xxxxxxxxxxxxxxx>
---
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(-)

--- a/drivers/md/bcache/bcache.h
+++ b/drivers/md/bcache/bcache.h
@@ -348,6 +348,7 @@ 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;

--- a/drivers/md/bcache/super.c
+++ b/drivers/md/bcache/super.c
@@ -1075,6 +1075,8 @@ static void cached_dev_free(struct closu
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);

--- a/drivers/md/bcache/writeback.c
+++ b/drivers/md/bcache/writeback.c
@@ -191,7 +191,7 @@ static void write_dirty(struct closure *

closure_bio_submit(&io->bio, cl, &io->dc->disk);

- 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, int error)
@@ -211,7 +211,7 @@ static void read_dirty_submit(struct clo

closure_bio_submit(&io->bio, cl, &io->dc->disk);

- 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)
@@ -518,6 +518,11 @@ int bch_cached_dev_writeback_init(struct
dc->writeback_rate_d_term = 30;
dc->writeback_rate_p_term_inverse = 6000;

+ 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))