diff options
| author |   Yunhong Jiang <yunhong.jiang@linux.intel.com> | 2017-03-08 23:13:28 -0800 |
|---|---|---|
| committer | Yunhong Jiang <yunhong.jiang@linux.intel.com> | 2017-03-08 23:36:15 -0800 |
| commit | 52f993b8e89487ec9ee15a7fb4979e0f09a45b27 (patch) | |
| tree | d65304486afe0bea4a311c783c0d72791c8c0aa2 /kernel/fs/ocfs2/dlmglue.c | |
| parent | c189ccac5702322ed843fe17057035b7222a59b6 (diff) | |
Upgrade to 4.4.50-rt62
The current kernel is based on rt kernel v4.4.6-rt14. We will upgrade it
to 4.4.50-rt62.
The command to achieve it is:
a) Clone a git repo from
git://git.kernel.org/pub/scm/linux/kernel/git/rt/linux-stable-rt.git
b) Get the diff between this two changesets:
git diff 640eca2901f3435e616157b11379d3223a44b391
705619beeea1b0b48219a683fd1a901a86fdaf5e
where the two commits are:
[yjiang5@jnakajim-build linux-stable-rt]$ git show --oneline --name-only
640eca2901f3435e616157b11379d3223a44b391
640eca2901f3 v4.4.6-rt14
localversion-rt
[yjiang5@jnakajim-build linux-stable-rt]$ git show --oneline --name-only
705619beeea1b0b48219a683fd1a901a86fdaf5e
705619beeea1 Linux 4.4.50-rt62
localversion-rt
c) One patch has been backported thus revert the patch before applying.
filterdiff -p1 -x scripts/package/Makefile
~/tmp/v4.4.6-rt14-4.4.50-rt62.diff |patch -p1 --dry-run
Upstream status: backport
Change-Id: I244d57a32f6066e5a5b9915f9fbf99e7bbca6e01
Signed-off-by: Yunhong Jiang <yunhong.jiang@linux.intel.com>
Diffstat (limited to 'kernel/fs/ocfs2/dlmglue.c')
| -rw-r--r-- | kernel/fs/ocfs2/dlmglue.c | 10 |
1 files changed, 10 insertions, 0 deletions
diff --git a/kernel/fs/ocfs2/dlmglue.c b/kernel/fs/ocfs2/dlmglue.c index b002acf50..60a5f1548 100644 --- a/kernel/fs/ocfs2/dlmglue.c +++ b/kernel/fs/ocfs2/dlmglue.c @@ -3321,6 +3321,16 @@ static int ocfs2_downconvert_lock(struct ocfs2_super *osb, mlog(ML_BASTS, "lockres %s, level %d => %d\n", lockres->l_name, lockres->l_level, new_level); + /* + * On DLM_LKF_VALBLK, fsdlm behaves differently with o2cb. It always + * expects DLM_LKF_VALBLK being set if the LKB has LVB, so that + * we can recover correctly from node failure. Otherwise, we may get + * invalid LVB in LKB, but without DLM_SBF_VALNOTVALID being set. + */ + if (!ocfs2_is_o2cb_active() && + lockres->l_ops->flags & LOCK_TYPE_USES_LVB) + lvb = 1; + if (lvb) dlm_flags |= DLM_LKF_VALBLK; |