On Mon, 23 Jun 2014 02:35:09 -0700 (PDT)
David Rientjes <rient...@google.com> wrote:
> On Sat, 21 Jun 2014, Fabian Frederick wrote:
>
> > This reverts commit a2d445d44000
> >
> > Adding __init to blkcg_policy_register makes it unreliable when
> > loading cfq at runtime.
> >
> > (Similar problem to b5097e956a4d "block: add __init to elv_register"
> > reported by Damien Wyart).
> >
> > Cc: Jens Axboe <ax...@kernel.dk>
> > Cc: Damien Wyart <damien.wy...@gmail.com>
> > Cc: Andrew Morton <a...@linux-foundation.org>
> > Signed-off-by: Fabian Frederick <f...@skynet.be>
> > ---
> > block/blk-cgroup.c | 2 +-
> > block/blk-cgroup.h | 4 ++--
> > 2 files changed, 3 insertions(+), 3 deletions(-)
> >
> > diff --git a/block/blk-cgroup.c b/block/blk-cgroup.c
> > index 069bc20..9f5bce3 100644
> > --- a/block/blk-cgroup.c
> > +++ b/block/blk-cgroup.c
> > @@ -1093,7 +1093,7 @@ EXPORT_SYMBOL_GPL(blkcg_deactivate_policy);
> > * Register @pol with blkcg core. Might sleep and @pol may be modified on
> > * successful registration. Returns 0 on success and -errno on failure.
> > */
> > -int __init blkcg_policy_register(struct blkcg_policy *pol)
> > +int blkcg_policy_register(struct blkcg_policy *pol)
> > {
> > int i, ret;
> >
>
> This is the second such revert from you for __init annotations, have you
> double checked all the other patches you've sent?
I received a bug report for elv_register and after looking at my
recent __init patches I found this one with the same problem.
>
> > diff --git a/block/blk-cgroup.h b/block/blk-cgroup.h
> > index cbb7f94..d692b29 100644
> > --- a/block/blk-cgroup.h
> > +++ b/block/blk-cgroup.h
> > @@ -145,7 +145,7 @@ void blkcg_drain_queue(struct request_queue *q);
> > void blkcg_exit_queue(struct request_queue *q);
> >
> > /* Blkio controller policy registration */
> > -int __init blkcg_policy_register(struct blkcg_policy *pol);
> > +int blkcg_policy_register(struct blkcg_policy *pol);
> > void blkcg_policy_unregister(struct blkcg_policy *pol);
> > int blkcg_activate_policy(struct request_queue *q,
> > const struct blkcg_policy *pol);
> > @@ -580,7 +580,7 @@ static inline struct blkcg_gq *blkg_lookup(struct blkcg
> > *blkcg, void *key) { ret
> > static inline int blkcg_init_queue(struct request_queue *q) { return 0; }
> > static inline void blkcg_drain_queue(struct request_queue *q) { }
> > static inline void blkcg_exit_queue(struct request_queue *q) { }
> > -static inline int __init blkcg_policy_register(struct blkcg_policy *pol) {
> > return 0; }
> > +static inline int blkcg_policy_register(struct blkcg_policy *pol) { return
> > 0; }
>
> Why would you have marked an inline function as __init in the first place?
__init was initially added to mark function in case we remove inline someday as
it's only called by __init functions. Problem in this case being it would break
external modules.
Fabian
>
> > static inline void blkcg_policy_unregister(struct blkcg_policy *pol) { }
> > static inline int blkcg_activate_policy(struct request_queue *q,
> > const struct blkcg_policy *pol) {
> > return 0; }
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
