在 2023/5/27 5:38, Song Liu 写道:

On Fri, May 26, 2023 at 12:47 AM [off-list ref] wrote:

quoted

From: Li Nan <redacted>

need_replace will be set to 1 if no-Faulty mreplace exists, and mreplace
will be deref later. However, the latter check of mreplace might set
mreplace to NULL, null-ptr-deref occurs if need_replace is 1 at this time.

Fix it by merging two checks into one. And replace 'need_replace' with
'mreplace' because their values are always the same.

Fixes: ee37d7314a32 ("md/raid10: Fix raid10 replace hang when new added disk faulty")
Signed-off-by: Li Nan <redacted>
---
  drivers/md/raid10.c | 13 +++++--------
  1 file changed, 5 insertions(+), 8 deletions(-)

diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c
index 4fcfcb350d2b..e21502c03b45 100644
--- a/drivers/md/raid10.c
+++ b/drivers/md/raid10.c

@@ -3438,7 +3438,6 @@ static sector_t raid10_sync_request(struct mddev *mddev, sector_t sector_nr,
                         int must_sync;
                         int any_working;
                         int need_recover = 0;
-                       int need_replace = 0;
                         struct raid10_info *mirror = &conf->mirrors[i];
                         struct md_rdev *mrdev, *mreplace;

@@ -3451,10 +3450,10 @@ static sector_t raid10_sync_request(struct mddev *mddev, sector_t sector_nr,
                             !test_bit(In_sync, &mrdev->flags))
                                 need_recover = 1;
                         if (mreplace != NULL &&
-                           !test_bit(Faulty, &mreplace->flags))
-                               need_replace = 1;
+                           test_bit(Faulty, &mreplace->flags))
+                               mreplace = NULL;

-                       if (!need_recover && !need_replace) {
+                       if (!need_recover && !mreplace) {
                                 rcu_read_unlock();
                                 continue;
                         }

@@ -3470,8 +3469,6 @@ static sector_t raid10_sync_request(struct mddev *mddev, sector_t sector_nr,
                                 rcu_read_unlock();
                                 continue;
                         }

To make sure I understand the issue correctly:

The null-ptr-deref only happens when the Faulty bit was set after the
last check and before this check below, right?

Yes. I will improve log in next version.

-- 
Thanks,
Nan