Messages in this thread Patch in this message | | | Date | Wed, 20 Jun 2012 18:04:35 -0700 | From | Tejun Heo <> | Subject | [GIT PULL] cgroup fixes for v3.5-rc3 |
| |
Hello, Linus.
Please pull from the following branch to receive two patches fixing a refcnt race bug during css_put(). Decing and checking the value weren't atomic and two tasks could think that they both pushed the counter to zero.
git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup.git for-3.5-fixes
Salman Qazi (2): cgroup: make sure that decisions in __css_put are atomic cgroups: Account for CSS_DEACT_BIAS in __css_put
Thanks. --- kernel/cgroup.c | 13 ++++++++++--- 1 files changed, 10 insertions(+), 3 deletions(-)
diff --git a/kernel/cgroup.c b/kernel/cgroup.c index 72fcd30..2097684 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -255,12 +255,17 @@ int cgroup_lock_is_held(void) EXPORT_SYMBOL_GPL(cgroup_lock_is_held); +static int css_unbias_refcnt(int refcnt) +{ + return refcnt >= 0 ? refcnt : refcnt - CSS_DEACT_BIAS; +} + /* the current nr of refs, always >= 0 whether @css is deactivated or not */ static int css_refcnt(struct cgroup_subsys_state *css) { int v = atomic_read(&css->refcnt); - return v >= 0 ? v : v - CSS_DEACT_BIAS; + return css_unbias_refcnt(v); } /* convenient tests for these bits */ @@ -4982,10 +4987,12 @@ EXPORT_SYMBOL_GPL(__css_tryget); void __css_put(struct cgroup_subsys_state *css) { struct cgroup *cgrp = css->cgroup; + int v; rcu_read_lock(); - atomic_dec(&css->refcnt); - switch (css_refcnt(css)) { + v = css_unbias_refcnt(atomic_dec_return(&css->refcnt)); + + switch (v) { case 1: if (notify_on_release(cgrp)) { set_bit(CGRP_RELEASABLE, &cgrp->flags);
| |