Re: [PATCH v2 3/4] io_uring: don't spinlock when not posting CQEs

[Hao Xu <haoxu@linux.alibaba.com>]

在 2021/11/25 上午11:48, Hao Xu 写道:
> 在 2021/11/10 下午11:49, Pavel Begunkov 写道:
>> When no of queued for the batch completion requests need to post an CQE,
>> see IOSQE_CQE_SKIP_SUCCESS, avoid grabbing ->completion_lock and other
>> commit/post.
>>
>> Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
>> ---
>>   fs/io_uring.c | 26 +++++++++++++++++---------
>>   1 file changed, 17 insertions(+), 9 deletions(-)
>>
>> diff --git a/fs/io_uring.c b/fs/io_uring.c
>> index 22572cfd6864..0c0ea3bbb50a 100644
>> --- a/fs/io_uring.c
>> +++ b/fs/io_uring.c
>> @@ -321,6 +321,7 @@ struct io_submit_state {
>>       bool            plug_started;
>>       bool            need_plug;
>> +    bool            flush_cqes;
>>       unsigned short        submit_nr;
>>       struct blk_plug        plug;
>>   };
>> @@ -1525,8 +1526,11 @@ static void io_prep_async_link(struct io_kiocb 
>> *req)
>>   static inline void io_req_add_compl_list(struct io_kiocb *req)
>>   {
>> +    struct io_ring_ctx *ctx = req->ctx;
>>       struct io_submit_state *state = &req->ctx->submit_state;
>> +    if (!(req->flags & REQ_F_CQE_SKIP))
>> +        ctx->submit_state.flush_cqes = true;
> Should we init it to false in submit_state_start()?
Never mind, I saw it in __io_submit_flush_completions() so there is no
problem..