Re: [PATCH 6/6] mm, oom: fortify task_will_free_mem

From	Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Newsgroups	linux.kernel
Subject	Re: [PATCH 6/6] mm, oom: fortify task_will_free_mem
Date	2016-05-26 16:50 +0200
Message-ID	<rD4LL-7DF-1@gated-at.bofh.it> (permalink)
References	<rD2TD-6uy-5@gated-at.bofh.it> <rD2TE-6uy-17@gated-at.bofh.it> <rD4iK-7u2-17@gated-at.bofh.it> <rD4LL-7DF-3@gated-at.bofh.it>
Organization	linux.* mail to news gateway

Show all headers | View raw

Michal Hocko wrote:
> +/*
> + * Checks whether the given task is dying or exiting and likely to
> + * release its address space. This means that all threads and processes
> + * sharing the same mm have to be killed or exiting.
> + */
> +static inline bool task_will_free_mem(struct task_struct *task)
> +{
> +	struct mm_struct *mm = NULL;
> +	struct task_struct *p;
> +	bool ret = false;

If atomic_read(&p->mm->mm_users) <= get_nr_threads(p), this returns "false".
According to previous version, I think this is "bool ret = true;".

> +
> +	/*
> +	 * If the process has passed exit_mm we have to skip it because
> +	 * we have lost a link to other tasks sharing this mm, we do not
> +	 * have anything to reap and the task might then get stuck waiting
> +	 * for parent as zombie and we do not want it to hold TIF_MEMDIE
> +	 */
> +	p = find_lock_task_mm(task);
> +	if (!p)
> +		return false;
> +
> +	if (!__task_will_free_mem(p)) {
> +		task_unlock(p);
> +		return false;
> +	}
> +
> +	/*
> +	 * Check whether there are other processes sharing the mm - they all have
> +	 * to be killed or exiting.
> +	 */
> +	if (atomic_read(&p->mm->mm_users) > get_nr_threads(p)) {
> +		mm = p->mm;
> +		/* pin the mm to not get freed and reused */
> +		atomic_inc(&mm->mm_count);
> +	}
> +	task_unlock(p);
> +
> +	if (mm) {
> +		rcu_read_lock();
> +		for_each_process(p) {
> +			bool vfork;
> +
> +			/*
> +			 * skip over vforked tasks because they are mostly
> +			 * independent and will drop the mm soon
> +			 */
> +			task_lock(p);
> +			vfork = p->vfork_done;
> +			task_unlock(p);
> +			if (vfork)
> +				continue;
> +
> +			ret = __task_will_free_mem(p);
> +			if (!ret)
> +				break;
> +		}
> +		rcu_read_unlock();
> +		mmdrop(mm);
> +	}
> +
> +	return ret;
> +}

Thread

[PATCH 6/6] mm, oom: fortify task_will_free_mem Michal Hocko <mhocko@kernel.org> - 2016-05-26 14:50 +0200
  Re: [PATCH 6/6] mm, oom: fortify task_will_free_mem Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> - 2016-05-26 16:20 +0200
    Re: [PATCH 6/6] mm, oom: fortify task_will_free_mem Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> - 2016-05-26 16:50 +0200
      Re: [PATCH 6/6] mm, oom: fortify task_will_free_mem Michal Hocko <mhocko@kernel.org> - 2016-05-26 17:00 +0200
    Re: [PATCH 6/6] mm, oom: fortify task_will_free_mem Michal Hocko <mhocko@kernel.org> - 2016-05-26 16:50 +0200
  Re: [PATCH 6/6] mm, oom: fortify task_will_free_mem Michal Hocko <mhocko@kernel.org> - 2016-05-27 13:10 +0200

csiph-web