On 29/01/2020 02:51, Jens Axboe wrote:
> On 1/28/20 4:40 PM, Jens Axboe wrote:
>> On 1/28/20 4:36 PM, Pavel Begunkov wrote:
>>> On 28/01/2020 22:42, Jens Axboe wrote:
>>>> I didn't like it becoming a bit too complicated, both in terms of
>>>> implementation and use. And the fact that we'd have to jump through
>>>> hoops to make this work for a full chain.
>>>>
>>>> So I punted and just added sqe->personality and IOSQE_PERSONALITY.
>>>> This makes it way easier to use. Same branch:
>>>>
>>>> https://git.kernel.dk/cgit/linux-block/log/?h=for-5.6/io_uring-vfs-creds
>>>>
>>>> I'd feel much better with this variant for 5.6.
>>>>
>>>
>>> Checked out ("don't use static creds/mm assignments")
>>>
>>> 1. do we miscount cred refs? We grab one in get_current_cred() for each async
>>> request, but if (worker->creds != work->creds) it will never be put.
>>
>> Yeah I think you're right, that needs a bit of fixing up.
> 

Hmm, it seems it leaks it unconditionally, as it grabs in a ref in override_creds().

> I think this may have gotten fixed with the later addition posted today?
> I'll double check. But for the newer stuff, we put it for both cases
> when the request is freed.

Yeah, maybe. I got tangled trying to verify both at once and decided to start
with the old one.


>>> 2. shouldn't worker->creds be named {old,saved,etc}_creds? It's set as
>>>
>>>     worker->creds = override_creds(work->creds);
>>>
>>> Where override_creds() returns previous creds. And if so, then the following
>>> fast check looks strange:
>>>
>>>     worker->creds != work->creds
>>
>> Don't care too much about the naming, but the logic does appear off.
>> I'll take a look at both of these tonight, unless you beat me to it.

Apparently, you're faster :)

> 
> Testing this now, what a braino.
> 
> diff --git a/fs/io-wq.c b/fs/io-wq.c
> index ee49e8852d39..8fbbadf04cc3 100644
> --- a/fs/io-wq.c
> +++ b/fs/io-wq.c
> @@ -56,7 +56,8 @@ struct io_worker {
>  
>  	struct rcu_head rcu;
>  	struct mm_struct *mm;
> -	const struct cred *creds;
> +	const struct cred *cur_creds;
> +	const struct cred *saved_creds;
>  	struct files_struct *restore_files;
>  };
>  
> @@ -135,9 +136,9 @@ static bool __io_worker_unuse(struct io_wqe *wqe, struct io_worker *worker)
>  {
>  	bool dropped_lock = false;
>  
> -	if (worker->creds) {
> -		revert_creds(worker->creds);
> -		worker->creds = NULL;
> +	if (worker->saved_creds) {
> +		revert_creds(worker->saved_creds);
> +		worker->cur_creds = worker->saved_creds = NULL;
>  	}
>  
>  	if (current->files != worker->restore_files) {
> @@ -424,10 +425,11 @@ static void io_wq_switch_mm(struct io_worker *worker, struct io_wq_work *work)
>  static void io_wq_switch_creds(struct io_worker *worker,
>  			       struct io_wq_work *work)
>  {
> -	if (worker->creds)
> -		revert_creds(worker->creds);
> +	if (worker->saved_creds)
> +		revert_creds(worker->saved_creds);
>  
> -	worker->creds = override_creds(work->creds);
> +	worker->saved_creds = override_creds(work->creds);
> +	worker->cur_creds = work->creds;
>  }

How about as follows? rever_creds() is a bit heavier than put_creds().

static void io_wq_switch_creds(struct io_worker *worker,
			       struct io_wq_work *work)
{
	const struct cred *old_creds = override_creds(work->creds);

	if (worker->saved_creds)
		put_cred(old_creds);
	else
		worker->saved_creds = old;
	worker->cur_creds = work->creds;
}

>  
>  static void io_worker_handle_work(struct io_worker *worker)
> @@ -480,7 +482,7 @@ static void io_worker_handle_work(struct io_worker *worker)
>  		}
>  		if (work->mm != worker->mm)
>  			io_wq_switch_mm(worker, work);
> -		if (worker->creds != work->creds)
> +		if (worker->cur_creds != work->creds)
>  			io_wq_switch_creds(worker, work);
>  		/*
>  		 * OK to set IO_WQ_WORK_CANCEL even for uncancellable work,
> 

-- 
Pavel Begunkov