Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Tue, 2 Oct 2018 14:07:41 -0700
From:   Tejun Heo <tj@kernel.org>
To:     "Michael Kerrisk (man-pages)" <mtk.manpages@gmail.com>
Cc:     lkml <linux-kernel@vger.kernel.org>,
        "open list:CONTROL GROUP (CGROUP)" <cgroups@vger.kernel.org>,
        "Serge E. Hallyn" <serge@hallyn.com>,
        Amin Jamali <ajamali@pivotal.io>,
        Joao De Almeida Pereira <jpereira@pivotal.io>
Subject: Re: Cgroup v2 bug: "domain invalid" node can't be converted to
 "threaded"
Message-ID: <20181002210741.GJ270328@devbig004.ftw2.facebook.com>
References: <CAKgNAkhHYCMn74TCNiMJ=ccLd7DcmXSbvw3CbZ1YREeG7iJM5g@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <CAKgNAkhHYCMn74TCNiMJ=ccLd7DcmXSbvw3CbZ1YREeG7iJM5g@mail.gmail.com>
User-Agent: Mutt/1.5.21 (2010-09-15)
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

Hello, Michael.

Great catch.  Can you please verify whether the following patch fixes
the issue?

Thanks.

------ 8< ------
Subject: cgroup: Fix dom_cgrp propagation when enabling threaded mode

A cgroup which is already a threaded domain may be converted into a
threaded cgroup if the prerequisite conditions are met.  When this
happens, all threaded descendant should also have their ->dom_cgrp
updated to the new threaded domain cgroup.  Unfortunately, this
propagation was missing leading to the following failure.

  # cd /sys/fs/cgroup/unified
  # cat cgroup.subtree_control    # show that no controllers are enabled

  # mkdir -p mycgrp/a/b/c
  # echo threaded > mycgrp/a/b/cgroup.type

  At this point, the hierarchy looks as follows:

      mycgrp [d]
	  a [dt]
	      b [t]
		  c [inv]

  Now let's make node "a" threaded (and thus "mycgrp" s made "domain threaded"):

  # echo threaded > mycgrp/a/cgroup.type

  By this point, we now have a hierarchy that looks as follows:

      mycgrp [dt]
	  a [t]
	      b [t]
		  c [inv]

  But, when we try to convert the node "c" from "domain invalid" to
  "threaded", we get ENOTSUP on the write():

  # echo threaded > mycgrp/a/b/c/cgroup.type
  sh: echo: write error: Operation not supported

This patch fixes the problem by

* Moving the opencoded ->dom_cgrp save and restoration in
  cgroup_enable_threaded() into cgroup_{save|restore}_control() so
  that mulitple cgroups can be handled.
  
* Updating all threaded descendants' ->dom_cgrp to point to the new
  dom_cgrp when enabling threaded mode.

Signed-off-by: Tejun Heo <tj@kernel.org>
Reported-by: "Michael Kerrisk (man-pages)" <mtk.manpages@gmail.com>
Link: https://lore.kernel.org/r/CAKgNAkhHYCMn74TCNiMJ=ccLd7DcmXSbvw3CbZ1YREeG7iJM5g@mail.gmail.com
Fixes: 454000adaa2a ("cgroup: introduce cgroup->dom_cgrp and threaded css_set handling")
Cc: stable@vger.kernel.org # v4.14+
---
 kernel/cgroup/cgroup.c |   25 ++++++++++++++++---------
 1 file changed, 16 insertions(+), 9 deletions(-)

--- a/kernel/cgroup/cgroup.c
+++ b/kernel/cgroup/cgroup.c
@@ -2836,11 +2836,12 @@ restart:
 }
 
 /**
- * cgroup_save_control - save control masks of a subtree
+ * cgroup_save_control - save control masks and dom_cgrp of a subtree
  * @cgrp: root of the target subtree
  *
- * Save ->subtree_control and ->subtree_ss_mask to the respective old_
- * prefixed fields for @cgrp's subtree including @cgrp itself.
+ * Save ->subtree_control, ->subtree_ss_mask and ->dom_cgrp to the
+ * respective old_ prefixed fields for @cgrp's subtree including @cgrp
+ * itself.
  */
 static void cgroup_save_control(struct cgroup *cgrp)
 {
@@ -2850,6 +2851,7 @@ static void cgroup_save_control(struct c
 	cgroup_for_each_live_descendant_pre(dsct, d_css, cgrp) {
 		dsct->old_subtree_control = dsct->subtree_control;
 		dsct->old_subtree_ss_mask = dsct->subtree_ss_mask;
+		dsct->old_dom_cgrp = dsct->dom_cgrp;
 	}
 }
 
@@ -2875,11 +2877,12 @@ static void cgroup_propagate_control(str
 }
 
 /**
- * cgroup_restore_control - restore control masks of a subtree
+ * cgroup_restore_control - restore control masks and dom_cgrp of a subtree
  * @cgrp: root of the target subtree
  *
- * Restore ->subtree_control and ->subtree_ss_mask from the respective old_
- * prefixed fields for @cgrp's subtree including @cgrp itself.
+ * Restore ->subtree_control, ->subtree_ss_mask and ->dom_cgrp from the
+ * respective old_ prefixed fields for @cgrp's subtree including @cgrp
+ * itself.
  */
 static void cgroup_restore_control(struct cgroup *cgrp)
 {
@@ -2889,6 +2892,7 @@ static void cgroup_restore_control(struc
 	cgroup_for_each_live_descendant_post(dsct, d_css, cgrp) {
 		dsct->subtree_control = dsct->old_subtree_control;
 		dsct->subtree_ss_mask = dsct->old_subtree_ss_mask;
+		dsct->dom_cgrp = dsct->old_dom_cgrp;
 	}
 }
 
@@ -3196,6 +3200,8 @@ static int cgroup_enable_threaded(struct
 {
 	struct cgroup *parent = cgroup_parent(cgrp);
 	struct cgroup *dom_cgrp = parent->dom_cgrp;
+	struct cgroup *dsct;
+	struct cgroup_subsys_state *d_css;
 	int ret;
 
 	lockdep_assert_held(&cgroup_mutex);
@@ -3225,12 +3231,13 @@ static int cgroup_enable_threaded(struct
 	 */
 	cgroup_save_control(cgrp);
 
-	cgrp->dom_cgrp = dom_cgrp;
+	cgroup_for_each_live_descendant_pre(dsct, d_css, cgrp)
+		if (dsct == cgrp || cgroup_is_threaded(dsct))
+			dsct->dom_cgrp = dom_cgrp;
+
 	ret = cgroup_apply_control(cgrp);
 	if (!ret)
 		parent->nr_threaded_children++;
-	else
-		cgrp->dom_cgrp = cgrp;
 
 	cgroup_finalize_control(cgrp, ret);
 	return ret;