diff options
author | Ameer Hamza <[email protected]> | 2022-09-17 01:52:25 +0500 |
---|---|---|
committer | GitHub <[email protected]> | 2022-09-16 13:52:25 -0700 |
commit | 577d41d3b2e4b37f51270c399c85b2708e21238a (patch) | |
tree | d946e9b410d7009b542d4b754400fdc097e9a24d | |
parent | 8da218a7a2ee0d0c0a0741f1007ebce0bf22584a (diff) |
zfs recv hangs if max recordsize is less than received recordsize
- Some optimizations for bqueue enqueue/dequeue.
- Added a fix to prevent deadlock when both bqueue_enqueue_impl()
and bqueue_dequeue() waits for signal to be triggered.
Reviewed-by: Alexander Motin <[email protected]>
Reviewed-by: Ryan Moeller <[email protected]>
Signed-off-by: Ameer Hamza <[email protected]>
Closes #13855
-rw-r--r-- | include/sys/bqueue.h | 14 | ||||
-rw-r--r-- | include/sys/fs/zfs.h | 6 | ||||
-rw-r--r-- | module/zfs/bqueue.c | 23 |
3 files changed, 23 insertions, 20 deletions
diff --git a/include/sys/bqueue.h b/include/sys/bqueue.h index 797aecd79..b96219660 100644 --- a/include/sys/bqueue.h +++ b/include/sys/bqueue.h @@ -30,22 +30,22 @@ typedef struct bqueue { kmutex_t bq_lock; kcondvar_t bq_add_cv; kcondvar_t bq_pop_cv; - uint64_t bq_size; - uint64_t bq_maxsize; - uint64_t bq_fill_fraction; + size_t bq_size; + size_t bq_maxsize; + uint_t bq_fill_fraction; size_t bq_node_offset; } bqueue_t; typedef struct bqueue_node { list_node_t bqn_node; - uint64_t bqn_size; + size_t bqn_size; } bqueue_node_t; -int bqueue_init(bqueue_t *, uint64_t, uint64_t, size_t); +int bqueue_init(bqueue_t *, uint_t, size_t, size_t); void bqueue_destroy(bqueue_t *); -void bqueue_enqueue(bqueue_t *, void *, uint64_t); -void bqueue_enqueue_flush(bqueue_t *, void *, uint64_t); +void bqueue_enqueue(bqueue_t *, void *, size_t); +void bqueue_enqueue_flush(bqueue_t *, void *, size_t); void *bqueue_dequeue(bqueue_t *); boolean_t bqueue_empty(bqueue_t *); diff --git a/include/sys/fs/zfs.h b/include/sys/fs/zfs.h index 8cbd0e602..dedee0e7b 100644 --- a/include/sys/fs/zfs.h +++ b/include/sys/fs/zfs.h @@ -1758,9 +1758,9 @@ typedef enum { * against the cost of COWing a giant block to modify one byte, and the * large latency of reading or writing a large block. * - * Note that although blocks up to 16MB are supported, the recordsize - * property can not be set larger than zfs_max_recordsize (default 1MB). - * See the comment near zfs_max_recordsize in dsl_dataset.c for details. + * The recordsize property can not be set larger than zfs_max_recordsize + * (default 16MB on 64-bit and 1MB on 32-bit). See the comment near + * zfs_max_recordsize in dsl_dataset.c for details. * * Note that although the LSIZE field of the blkptr_t can store sizes up * to 32MB, the dnode's dn_datablkszsec can only store sizes up to diff --git a/module/zfs/bqueue.c b/module/zfs/bqueue.c index 22539efc4..ec5ce4388 100644 --- a/module/zfs/bqueue.c +++ b/module/zfs/bqueue.c @@ -42,8 +42,7 @@ obj2node(bqueue_t *q, void *data) * Return 0 on success, or -1 on failure. */ int -bqueue_init(bqueue_t *q, uint64_t fill_fraction, uint64_t size, - size_t node_offset) +bqueue_init(bqueue_t *q, uint_t fill_fraction, size_t size, size_t node_offset) { if (fill_fraction == 0) { return (-1); @@ -78,22 +77,26 @@ bqueue_destroy(bqueue_t *q) } static void -bqueue_enqueue_impl(bqueue_t *q, void *data, uint64_t item_size, - boolean_t flush) +bqueue_enqueue_impl(bqueue_t *q, void *data, size_t item_size, boolean_t flush) { ASSERT3U(item_size, >, 0); ASSERT3U(item_size, <=, q->bq_maxsize); mutex_enter(&q->bq_lock); obj2node(q, data)->bqn_size = item_size; - while (q->bq_size + item_size > q->bq_maxsize) { + while (q->bq_size && q->bq_size + item_size > q->bq_maxsize) { + /* + * Wake up bqueue_dequeue() thread if already sleeping in order + * to prevent the deadlock condition + */ + cv_signal(&q->bq_pop_cv); cv_wait_sig(&q->bq_add_cv, &q->bq_lock); } q->bq_size += item_size; list_insert_tail(&q->bq_list, data); - if (q->bq_size >= q->bq_maxsize / q->bq_fill_fraction) - cv_signal(&q->bq_pop_cv); if (flush) cv_broadcast(&q->bq_pop_cv); + else if (q->bq_size >= q->bq_maxsize / q->bq_fill_fraction) + cv_signal(&q->bq_pop_cv); mutex_exit(&q->bq_lock); } @@ -103,7 +106,7 @@ bqueue_enqueue_impl(bqueue_t *q, void *data, uint64_t item_size, * > 0. */ void -bqueue_enqueue(bqueue_t *q, void *data, uint64_t item_size) +bqueue_enqueue(bqueue_t *q, void *data, size_t item_size) { bqueue_enqueue_impl(q, data, item_size, B_FALSE); } @@ -117,7 +120,7 @@ bqueue_enqueue(bqueue_t *q, void *data, uint64_t item_size) * destroy the condvar before the enqueuing thread is done. */ void -bqueue_enqueue_flush(bqueue_t *q, void *data, uint64_t item_size) +bqueue_enqueue_flush(bqueue_t *q, void *data, size_t item_size) { bqueue_enqueue_impl(q, data, item_size, B_TRUE); } @@ -130,7 +133,7 @@ void * bqueue_dequeue(bqueue_t *q) { void *ret = NULL; - uint64_t item_size; + size_t item_size; mutex_enter(&q->bq_lock); while (q->bq_size == 0) { cv_wait_sig(&q->bq_pop_cv, &q->bq_lock); |