| Message ID | 1642682627-12041-1-git-send-email-gautham.ananthakrishna@oracle.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | [PATCHC,RFC,V2,1/1] ocfs2: fix a deadlock when commit trans | expand |
Seems you still sent a patch that couldn't build. Have you checked my previous comments and tried my suggested way? Thanks, Joseph On 1/20/22 8:43 PM, Gautham Ananthakrishna wrote: > commit 6f1b228529ae49b0f85ab89bcdb6c365df401558 caused a deadlock > which was uncovered by our internal tests. The deadlock is as foollows: > > Task 1: > A1) jbd2_journal_commit_transaction() > A2) spin_lock(&jh->b_state_lock); > A3) __jbd2_journal_remove_checkpoint() > A4) jbd2_journal_put_journal_head() > A5) jbd_lock_bh_journal_head() > > Task 2: > B1) ocfs2_test_bg_bit_allocatable() > B2) jbd_lock_bh_journal_head() > B3) spin_lock(&jh->b_state_lock); > > A1->A2->A3->A4->B1->B2->B3(blocked)->A5(blocked) > > Now cause process 2 has the jbd lock, it doesn’t let process 1 to continue after A5. > Process 2 now is waiting for b_state_lock to be released by process 1. > > This patch resolves the deadlock. > > Signed-off-by: Gautham Ananthakrishna <gautham.ananthakrishna@oracle.com> > --- > fs/ocfs2/suballoc.c | 7 +++---- > 1 file changed, 3 insertions(+), 4 deletions(-) > > diff --git a/fs/ocfs2/suballoc.c b/fs/ocfs2/suballoc.c > index 481017e..a618970 100644 > --- a/fs/ocfs2/suballoc.c > +++ b/fs/ocfs2/suballoc.c > @@ -1259,9 +1259,8 @@ static int ocfs2_test_bg_bit_allocatable(struct buffer_head *bg_bh, > if (!buffer_jbd(bg_bh)) > return 1; > > - jbd_lock_bh_journal_head(bg_bh); > - if (buffer_jbd(bg_bh)) { > - jh = bh2jh(bg_bh); > + jh = jbd2_journal_grab_journal_head(bg_bh); > + if (jh) { > spin_lock(&jh->b_state_lock); > bg = (struct ocfs2_group_desc *) jh->b_committed_data; > if (bg) > @@ -1269,8 +1268,8 @@ static int ocfs2_test_bg_bit_allocatable(struct buffer_head *bg_bh, > else > ret = 1; > spin_unlock(&jh->b_state_lock); > + jbd2_journal_put_journal_head(bg_bh); > } > - jbd_unlock_bh_journal_head(bg_bh); > > return ret; > }
diff --git a/fs/ocfs2/suballoc.c b/fs/ocfs2/suballoc.c index 481017e..a618970 100644 --- a/fs/ocfs2/suballoc.c +++ b/fs/ocfs2/suballoc.c @@ -1259,9 +1259,8 @@ static int ocfs2_test_bg_bit_allocatable(struct buffer_head *bg_bh, if (!buffer_jbd(bg_bh)) return 1; - jbd_lock_bh_journal_head(bg_bh); - if (buffer_jbd(bg_bh)) { - jh = bh2jh(bg_bh); + jh = jbd2_journal_grab_journal_head(bg_bh); + if (jh) { spin_lock(&jh->b_state_lock); bg = (struct ocfs2_group_desc *) jh->b_committed_data; if (bg) @@ -1269,8 +1268,8 @@ static int ocfs2_test_bg_bit_allocatable(struct buffer_head *bg_bh, else ret = 1; spin_unlock(&jh->b_state_lock); + jbd2_journal_put_journal_head(bg_bh); } - jbd_unlock_bh_journal_head(bg_bh); return ret; }
commit 6f1b228529ae49b0f85ab89bcdb6c365df401558 caused a deadlock which was uncovered by our internal tests. The deadlock is as foollows: Task 1: A1) jbd2_journal_commit_transaction() A2) spin_lock(&jh->b_state_lock); A3) __jbd2_journal_remove_checkpoint() A4) jbd2_journal_put_journal_head() A5) jbd_lock_bh_journal_head() Task 2: B1) ocfs2_test_bg_bit_allocatable() B2) jbd_lock_bh_journal_head() B3) spin_lock(&jh->b_state_lock); A1->A2->A3->A4->B1->B2->B3(blocked)->A5(blocked) Now cause process 2 has the jbd lock, it doesn’t let process 1 to continue after A5. Process 2 now is waiting for b_state_lock to be released by process 1. This patch resolves the deadlock. Signed-off-by: Gautham Ananthakrishna <gautham.ananthakrishna@oracle.com> --- fs/ocfs2/suballoc.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-)