Re: [PATCH v2] io_uring/net: ensure async prep handlers always initialize ->done_io

[Pavel Begunkov <asml.silence@gmail.com>]

On 3/16/24 16:32, Pavel Begunkov wrote:
> On 3/16/24 16:31, Jens Axboe wrote:
>> On 3/16/24 10:28 AM, Pavel Begunkov wrote:
>>> On 3/16/24 16:14, Jens Axboe wrote:
>>>> On 3/15/24 5:28 PM, Pavel Begunkov wrote:
>>>>> On 3/15/24 23:25, Jens Axboe wrote:
>>>>>> On 3/15/24 5:19 PM, Pavel Begunkov wrote:
>>>>>>> On 3/15/24 23:13, Pavel Begunkov wrote:
>>>>>>>> On 3/15/24 23:09, Pavel Begunkov wrote:
>>>>>>>>> On 3/15/24 22:48, Jens Axboe wrote:
>>>>>>>>>> If we get a request with IOSQE_ASYNC set, then we first run the prep
>>>>>>>>>> async handlers. But if we then fail setting it up and want to post
>>>>>>>>>> a CQE with -EINVAL, we use ->done_io. This was previously guarded with
>>>>>>>>>> REQ_F_PARTIAL_IO, and the normal setup handlers do set it up before any
>>>>>>>>>> potential errors, but we need to cover the async setup too.
>>>>>>>>>
>>>>>>>>> You can hit io_req_defer_failed() { opdef->fail(); }
>>>>>>>>> off of an early submission failure path where def->prep has
>>>>>>>>> not yet been called, I don't think the patch will fix the
>>>>>>>>> problem.
>>>>>>>>>
>>>>>>>>> ->fail() handlers are fragile, maybe we should skip them
>>>>>>>>> if def->prep() wasn't called. Not even compile tested:
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c
>>>>>>>>> index 846d67a9c72e..56eed1490571 100644
>>>>>>>>> --- a/io_uring/io_uring.c
>>>>>>>>> +++ b/io_uring/io_uring.c
>>>>>>> [...]
>>>>>>>>>              def->fail(req);
>>>>>>>>>          io_req_complete_defer(req);
>>>>>>>>>      }
>>>>>>>>> @@ -2201,8 +2201,7 @@ static int io_init_req(struct io_ring_ctx *ctx, struct io_kiocb *req,
>>>>>>>>>              }
>>>>>>>>>              req->flags |= REQ_F_CREDS;
>>>>>>>>>          }
>>>>>>>>> -
>>>>>>>>> -    return def->prep(req, sqe);
>>>>>>>>> +    return 0;
>>>>>>>>>      }
>>>>>>>>>
>>>>>>>>>      static __cold int io_submit_fail_init(const struct io_uring_sqe *sqe,
>>>>>>>>> @@ -2250,8 +2249,15 @@ static inline int io_submit_sqe(struct io_ring_ctx *ctx, struct io_kiocb *req,
>>>>>>>>>          int ret;
>>>>>>>>>
>>>>>>>>>          ret = io_init_req(ctx, req, sqe);
>>>>>>>>> -    if (unlikely(ret))
>>>>>>>>> +    if (unlikely(ret)) {
>>>>>>>>> +fail:
>>>>>>>
>>>>>>> Obvious the diff is crap, but still bugging me enough to write
>>>>>>> that the label should've been one line below, otherwise we'd
>>>>>>> flag after ->prep as well.
>>>>>>
>>>>>> It certainly needs testing :-)
>>>>>>
>>>>>> We can go either way - patch up the net thing, or do a proper EARLY_FAIL
>>>>>> and hopefully not have to worry about it again. Do you want to clean it
>>>>>> up, test it, and send it out?
>>>>>
>>>>> I'd rather leave it to you, I suspect it wouldn't fix the syzbot
>>>>> report w/o fiddling with done_io as in your patch.
>>>>
>>>> I gave this a shot, but some fail handlers do want to get called. But

Maybe I didn't get you right. I assumed you're saying "the zc's ->fail
wants to get called even if prep didn't happen. ?


>>> Which one and/or which part of it?
>>
>> send zc
> 
> I don't think so. If prep wasn't called there wouldn't be
> a notif allocated, and so no F_MORE required. If you take
> at the code path it's under REQ_F_NEED_CLEANUP, which is only
> set by opcode handlers
> 
> 
>>
>> I think the sanest is:
>>
>> 1) Opcode handlers should always initialize whatever they need before
>>     failure

Yes

>> 2) If we fail before ->prep, don't call ->fail

That's what I suggested

>> Yes that doesn't cover the case where opcode handlers do stupid things
>> like use opcode members in fail if they fail the prep, but that should
>> be the smallest part.
>>
> 

-- 
Pavel Begunkov