Hold multipath.lock around all code that iterates over the priority_groups list. This patch fixes the following crash:

general protection fault: 0000 [#1] PREEMPT SMP RIP: 0010:multipath_busy+0x77/0xd0 [dm_multipath] Call Trace: dm_mq_queue_rq+0x44/0x110 [dm_mod] blk_mq_dispatch_rq_list+0x73/0x440 blk_mq_do_dispatch_sched+0x60/0xe0 blk_mq_sched_dispatch_requests+0x11a/0x1a0 __blk_mq_run_hw_queue+0x11f/0x1c0 __blk_mq_delay_run_hw_queue+0x95/0xe0 blk_mq_run_hw_queue+0x25/0x80 blk_mq_flush_plug_list+0x197/0x420 blk_flush_plug_list+0xe4/0x270 blk_finish_plug+0x27/0x40 __do_page_cache_readahead+0x2b4/0x370 force_page_cache_readahead+0xb4/0x110 generic_file_read_iter+0x755/0x970 __vfs_read+0xd2/0x140 vfs_read+0x9b/0x140 SyS_read+0x45/0xa0 do_syscall_64+0x56/0x1a0 entry_SYSCALL64_slow_path+0x25/0x25

From the disassembly of multipath_busy (0x77 = 119):

./include/linux/blkdev.h: 992 return bdev->bd_disk->queue; /* this is never NULL */ 0x00000000000006b4 <+116>: mov (%rax),%rax 0x00000000000006b7 <+119>: mov 0xe0(%rax),%rax

Signed-off-by: Bart Van Assche bart.vanassche@wdc.com Cc: Hannes Reinecke hare@suse.com Cc: stable@vger.kernel.org --- drivers/md/dm-mpath.c | 40 ++++++++++++++++++++++++++++------------ 1 file changed, 28 insertions(+), 12 deletions(-)

diff --git a/drivers/md/dm-mpath.c b/drivers/md/dm-mpath.c index c8faa2b85842..61def92f306a 100644 --- a/drivers/md/dm-mpath.c +++ b/drivers/md/dm-mpath.c @@ -237,10 +237,19 @@ static int alloc_multipath_stage2(struct dm_target *ti, struct multipath *m)

static void free_multipath(struct multipath *m) { - struct priority_group *pg, *tmp; + struct priority_group *pg; + unsigned long flags; + + while (true) { + spin_lock_irqsave(&m->lock, flags); + pg = list_first_entry_or_null(&m->priority_groups, typeof(*pg), + list); + if (pg) + list_del(&pg->list); + spin_unlock_irqrestore(&m->lock, flags);

- list_for_each_entry_safe(pg, tmp, &m->priority_groups, list) { - list_del(&pg->list); + if (!pg) + break; free_priority_group(pg, m->ti); }

@@ -337,6 +346,7 @@ static int pg_init_all_paths(struct multipath *m) }

static void __switch_pg(struct multipath *m, struct priority_group *pg) + __must_hold(&m->lock) { m->current_pg = pg;

@@ -355,8 +365,8 @@ static void __switch_pg(struct multipath *m, struct priority_group *pg) static struct pgpath *choose_path_in_pg(struct multipath *m, struct priority_group *pg, size_t nr_bytes) + __must_hold(&m->lock) { - unsigned long flags; struct dm_path *path; struct pgpath *pgpath;

@@ -368,10 +378,8 @@ static struct pgpath *choose_path_in_pg(struct multipath *m,

if (unlikely(READ_ONCE(m->current_pg) != pg)) { /* Only update current_pgpath if pg changed */ - spin_lock_irqsave(&m->lock, flags); m->current_pgpath = pgpath; __switch_pg(m, pg); - spin_unlock_irqrestore(&m->lock, flags); }

return pgpath; @@ -381,7 +389,7 @@ static struct pgpath *choose_pgpath(struct multipath *m, size_t nr_bytes) { unsigned long flags; struct priority_group *pg; - struct pgpath *pgpath; + struct pgpath *pgpath, *res = NULL; unsigned bypassed = 1;

if (!atomic_read(&m->nr_valid_paths)) { @@ -419,6 +427,7 @@ static struct pgpath *choose_pgpath(struct multipath *m, size_t nr_bytes) * Second time we only try the ones we skipped, but set * pg_init_delay_retry so we do not hammer controllers. */ + spin_lock_irqsave(&m->lock, flags); do { list_for_each_entry(pg, &m->priority_groups, list) { if (pg->bypassed == !!bypassed) @@ -427,18 +436,22 @@ static struct pgpath *choose_pgpath(struct multipath *m, size_t nr_bytes) if (!IS_ERR_OR_NULL(pgpath)) { if (!bypassed) set_bit(MPATHF_PG_INIT_DELAY_RETRY, &m->flags); - return pgpath; + res = pgpath; + break; } } - } while (bypassed--); + } while (!res && bypassed--); + spin_unlock_irqrestore(&m->lock, flags);

failed: spin_lock_irqsave(&m->lock, flags); - m->current_pgpath = NULL; - m->current_pg = NULL; + if (!res) { + m->current_pgpath = NULL; + m->current_pg = NULL; + } spin_unlock_irqrestore(&m->lock, flags);

- return NULL; + return res; }

/* @@ -1875,6 +1888,7 @@ static int multipath_busy(struct dm_target *ti) struct multipath *m = ti->private; struct priority_group *pg, *next_pg; struct pgpath *pgpath; + unsigned long flags;

/* pg_init in progress */ if (atomic_read(&m->pg_init_in_progress)) @@ -1906,6 +1920,7 @@ static int multipath_busy(struct dm_target *ti) * will be able to select it. So we consider such a pg as not busy. */ busy = true; + spin_lock_irqsave(&m->lock, flags); list_for_each_entry(pgpath, &pg->pgpaths, list) { if (pgpath->is_active) { has_active = true; @@ -1915,6 +1930,7 @@ static int multipath_busy(struct dm_target *ti) } } } + spin_unlock_irqrestore(&m->lock, flags);

if (!has_active) { /*