We could only handle the case that css exists
and css_try_get_online() fails.
Signed-off-by: Peng Wang <[email protected]>
---
kernel/cgroup/cgroup.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c
index bf9dbffd46b1..a988d77f6c6d 100644
--- a/kernel/cgroup/cgroup.c
+++ b/kernel/cgroup/cgroup.c
@@ -488,7 +488,7 @@ static struct cgroup_subsys_state *cgroup_tryget_css(struct cgroup *cgrp,
rcu_read_lock();
css = cgroup_css(cgrp, ss);
- if (!css || !css_tryget_online(css))
+ if (css && !css_tryget_online(css))
css = NULL;
rcu_read_unlock();
--
2.19.1
On Mon, Jul 08, 2019 at 09:42:43AM -0700, Tejun Heo wrote:
> On Wed, Jul 03, 2019 at 10:07:49AM +0800, Peng Wang wrote:
> > We could only handle the case that css exists
> > and css_try_get_online() fails.
>
> As css_tryget_online() can't handle NULL input, this is a bug fix.
> Can you please clarify that in the description?
- if (!css || !css_tryget_online(css))
+ if (css && !css_tryget_online(css))
If css == NULL, !css is true, and the second part of the || statement
will not be evaluated. So it's not a bug fix.
Thanks!
On Mon, Jul 08, 2019 at 05:29:49PM +0000, Roman Gushchin wrote:
> On Mon, Jul 08, 2019 at 09:42:43AM -0700, Tejun Heo wrote:
> > On Wed, Jul 03, 2019 at 10:07:49AM +0800, Peng Wang wrote:
> > > We could only handle the case that css exists
> > > and css_try_get_online() fails.
> >
> > As css_tryget_online() can't handle NULL input, this is a bug fix.
> > Can you please clarify that in the description?
>
> - if (!css || !css_tryget_online(css))
> + if (css && !css_tryget_online(css))
>
> If css == NULL, !css is true, and the second part of the || statement
> will not be evaluated. So it's not a bug fix.
Ah right, it's just confusing. Will apply after the merge window.
Thanks.
--
tejun
On Wed, Jul 03, 2019 at 10:07:49AM +0800, Peng Wang wrote:
> We could only handle the case that css exists
> and css_try_get_online() fails.
As css_tryget_online() can't handle NULL input, this is a bug fix.
Can you please clarify that in the description?
Thanks.
--
tejun
On Wed, Jul 03, 2019 at 10:07:49AM +0800, Peng Wang wrote:
> We could only handle the case that css exists
> and css_try_get_online() fails.
>
> Signed-off-by: Peng Wang <[email protected]>
Applied to cgroup/for-5.4.
Thanks.
--
tejun