On 09/07, Roland McGrath wrote:
>
> In the original code, utrace_release_task() would not be called at all when
> there are no engines attached. IMHO that is the optimization we want to
> get back to.
>
> It may still be too early in the morning for me to be thinking clearly.
and too late for me ;) will reply tomorrow.
Oleg.
> diff --git a/include/linux/tracehook.h b/include/linux/tracehook.h
> index e249770..0000000 100644
> --- a/include/linux/tracehook.h
> +++ b/include/linux/tracehook.h
> @@ -362,7 +362,12 @@ static inline void tracehook_report_vfor
> */
> static inline void tracehook_prepare_release_task(struct task_struct *task)
> {
> - utrace_release_task(task);
> + /*
> + * See kernel/utrace.c:utrace_attach_task() about this barrier.
> + */
> + smp_mb();
> + if (task_utrace_flags(task))
> + utrace_release_task(task);
> }
>
> /**
> diff --git a/kernel/utrace.c b/kernel/utrace.c
> index 76d0d84..0000000 100644
> --- a/kernel/utrace.c
> +++ b/kernel/utrace.c
> @@ -143,13 +143,13 @@ static int utrace_add_engine(struct task
>
> spin_lock(&utrace->lock);
>
> - if (utrace->reap) {
> + if (unlikely(utrace->reap)) {
> /*
> * Already entered utrace_release_task(), cannot attach now.
> */
> ret = -ESRCH;
> } else if ((flags & UTRACE_ATTACH_EXCLUSIVE) &&
> - unlikely(matching_engine(utrace, flags, ops, data))) {
> + unlikely(matching_engine(utrace, flags, ops, data))) {
> ret = -EEXIST;
> } else {
> /*
> @@ -216,16 +216,19 @@ struct utrace_engine *utrace_attach_task
>
> utrace = &target->utrace;
>
> - if (unlikely(target->exit_state == EXIT_DEAD)) {
> - /*
> - * The target has already been reaped.
> - * Check this early, though it's not synchronized.
> - * utrace_add_engine() will do the final check.
> - */
> - if (!(flags & UTRACE_ATTACH_CREATE))
> - return ERR_PTR(-ENOENT);
> - return ERR_PTR(-ESRCH);
> - }
> + /*
> + * If it already could be getting reaped, we cannot attach now.
> + * The barrier ensures this check precedes utrace_add_engine()
> + * setting ->utrace_flags. tracehook_prepare_release_task() has
> + * the corresponding barrier, after setting EXIT_DEAD and before
> + * checking ->utrace_flags to decide to call utrace_release_task().
> + */
> + ret = target->exit_state;
> + smp_mb();
> +
> + if (unlikely(ret == EXIT_DEAD))
> + return ERR_PTR((flags & UTRACE_ATTACH_CREATE) ?
> + -ESRCH : -ENOENT);
>
> if (!(flags & UTRACE_ATTACH_CREATE)) {
> spin_lock(&utrace->lock);
