| Message ID | 20230415003754.1852-3-sj@kernel.org (mailing list archive) |
|---|---|
| State | New |
| Headers | show |
| Series | mm/slab: trivial fixup for SLAB_TYPESAFE_BY_RCU example code snippet | expand |
On Sat, Apr 15, 2023 at 12:37:53AM +0000, SeongJae Park wrote: > * > + * begin: > * rcu_read_lock(); > - * again: > * obj = lockless_lookup(key); > * if (obj) { > * if (!try_get_ref(obj)) // might fail for free objects > - * goto again; > + * goto begin; > * > * if (obj->key != key) { // not the object we expected > * put_ref(obj); > + * rcu_read_unlock(); > * goto again; - * goto again; + * goto begin;
On Sat, 15 Apr 2023 03:51:54 +0100 Matthew Wilcox <willy@infradead.org> wrote: > On Sat, Apr 15, 2023 at 12:37:53AM +0000, SeongJae Park wrote: > > * > > + * begin: > > * rcu_read_lock(); > > - * again: > > * obj = lockless_lookup(key); > > * if (obj) { > > * if (!try_get_ref(obj)) // might fail for free objects > > - * goto again; > > + * goto begin; > > * > > * if (obj->key != key) { // not the object we expected > > * put_ref(obj); > > + * rcu_read_unlock(); > > * goto again; > > - * goto again; > + * goto begin; Oops, thank you for catching, Matthew! Thanks, SJ
On Sat, 15 Apr 2023 03:04:52 +0000 SeongJae Park <sj@kernel.org> wrote: > On Sat, 15 Apr 2023 03:51:54 +0100 Matthew Wilcox <willy@infradead.org> wrote: > > > On Sat, Apr 15, 2023 at 12:37:53AM +0000, SeongJae Park wrote: > > > * > > > + * begin: > > > * rcu_read_lock(); > > > - * again: > > > * obj = lockless_lookup(key); > > > * if (obj) { > > > * if (!try_get_ref(obj)) // might fail for free objects > > > - * goto again; > > > + * goto begin; I simply followed the commit da82af04352b ("doc: Update and wordsmith rculist_nulls.rst") for this change, and I think I overlooked this part. If I'm not missing something, I think we should call rcu_read_unlock() here too, just before 'goto begin'. I will post another patch for rculist_null.rst later. Thanks, SJ > > > * > > > * if (obj->key != key) { // not the object we expected > > > * put_ref(obj); > > > + * rcu_read_unlock(); > > > * goto again; > > > > - * goto again; > > + * goto begin; > > Oops, thank you for catching, Matthew! > > > Thanks, > SJ
diff --git a/include/linux/slab.h b/include/linux/slab.h index b18e56c6f06c..a2fd57322c9c 100644 --- a/include/linux/slab.h +++ b/include/linux/slab.h @@ -53,15 +53,16 @@ * stays valid, the trick to using this is relying on an independent * object validation pass. Something like: * + * begin: * rcu_read_lock(); - * again: * obj = lockless_lookup(key); * if (obj) { * if (!try_get_ref(obj)) // might fail for free objects - * goto again; + * goto begin; * * if (obj->key != key) { // not the object we expected * put_ref(obj); + * rcu_read_unlock(); * goto again; * } * }
The SLAB_TYPESAFE_BY_RCU example code snippet is having not tiny RCU read-side critical section. 'Documentation/RCU/rculist_nulls.rst' has similar example code snippet, and commit da82af04352b ("doc: Update and wordsmith rculist_nulls.rst") has broken it. Apply the change to SLAB_TYPESAFE_BY_RCU example code snippet, too. Signed-off-by: SeongJae Park <sj@kernel.org> --- include/linux/slab.h | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)