[PATCH RFC 3/3] md/raid1: fix incorrect sector range in serialization

md/raid1: fix incorrect sector range in serialization

Fix off-by-one error in sector range calculation for serialization.
The range should be [lo, hi] inclusive, where hi = lo + sectors - 1,
not lo + sectors.

The current implementation causes false collision detection for
consecutive writes. For example, when writing:
- R1 to sectors 0-1023 (hi = 1024)
- R2 to sectors 1024-2047 (lo = 1024)

R1's hi equals R2's lo, causing raid1_rb_iter_first() to return
true and triggering unnecessary serialization, even though these
requests don't actually overlap.

Signed-off-by: Xiao Ni <[email protected]>
---
 drivers/md/raid1.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c
index 029aa458ffb2..e3d9ba71796f 100644
--- a/drivers/md/raid1.c
+++ b/drivers/md/raid1.c
@@ -62,7 +62,7 @@ static int check_and_add_serial(struct mddev *mddev, struct r1bio *r1_bio,
 	unsigned long flags;
 	int ret = 0;
 	sector_t lo = r1_bio->sector;
-	sector_t hi = lo + r1_bio->sectors;
+	sector_t hi = lo + r1_bio->sectors - 1;
 	struct serial *serial = &mddev->serial[idx];
 
 	spin_lock_irqsave(&serial->serial_lock, flags);
@@ -100,7 +100,7 @@ static void remove_serial(struct r1bio *r1_bio)
 	int found = 0;
 	struct mddev *mddev = r1_bio->mddev;
 	sector_t lo = r1_bio->sector;
-	sector_t hi = r1_bio->sector + r1_bio->sectors;
+	sector_t hi = r1_bio->sector + r1_bio->sectors - 1;
 	int idx = sector_to_idx(lo);
 	struct serial *serial = &mddev->serial[idx];
 
-- 
2.50.1 (Apple Git-155)