[dm-devel] [PATCH v2] md: dm-mpath: use list iterator only inside the loop

[Xiaomeng Tong]

To move the declare of list iterator into the list_for_each_entry_*()
macro in the future, it should be avoided to use the list iterator
variable after the loop body [1].

Because the cases (the list is empty or no entry is found) have been
checked prior to iterating m->priority_groups, it is fine to move the
bypass_pg(m, pg, bypassed); inside the loop when entry is found.

[1]: https://lore.kernel.org/all/CAHk-=wgRr_D8CB-D9Kg-c=EHreAsk5SqXPwr9Y7k9sA6cWXJ6w@mail.gmail.com/

Signed-off-by: Xiaomeng Tong <xiam0nd.tong at gmail.com>
---

changes since v1:
 - it is not a bug, have checked before (Mike Snitzer)
 - tweak code to avoid dead code "return -EINVAL;" (Xiaomeng Tong)

v1: https://lore.kernel.org/lkml/20220327053742.2942-1-xiam0nd.tong@gmail.com/

---
 drivers/md/dm-mpath.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/md/dm-mpath.c b/drivers/md/dm-mpath.c
index f4719b65e5e3..13c65eab32a2 100644
--- a/drivers/md/dm-mpath.c
+++ b/drivers/md/dm-mpath.c
@@ -1496,11 +1496,12 @@ static int bypass_pg_num(struct multipath *m, const char *pgstr, bool bypassed)
 	}
 
 	list_for_each_entry(pg, &m->priority_groups, list) {
-		if (!--pgnum)
+		if (!--pgnum) {
+			bypass_pg(m, pg, bypassed);
 			break;
+		}
 	}
 
-	bypass_pg(m, pg, bypassed);
 	return 0;
 }
 
-- 
2.17.1