Re: [PATCH 1/1] io_uring/net: send retry for zerocopy

[Jens Axboe <axboe@kernel.dk>]

On 8/14/22 3:31 AM, Stefan Metzmacher wrote:
> Hi Pavel,
> 
>> io_uring handles short sends/recvs for stream sockets when MSG_WAITALL
>> is set, however new zerocopy send is inconsistent in this regard, which
>> might be confusing. Handle short sends.
>>
>> Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
>> ---
>>   io_uring/net.c | 20 +++++++++++++++++---
>>   1 file changed, 17 insertions(+), 3 deletions(-)
>>
>> diff --git a/io_uring/net.c b/io_uring/net.c
>> index 32fc3da04e41..f9f080b3cc1e 100644
>> --- a/io_uring/net.c
>> +++ b/io_uring/net.c
>> @@ -70,6 +70,7 @@ struct io_sendzc {
>>       unsigned            flags;
>>       unsigned            addr_len;
>>       void __user            *addr;
>> +    size_t                done_io;
>>   };
>>     #define IO_APOLL_MULTI_POLLED (REQ_F_APOLL_MULTISHOT | REQ_F_POLLED)
>> @@ -878,6 +879,7 @@ int io_sendzc_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
>>         zc->addr = u64_to_user_ptr(READ_ONCE(sqe->addr2));
>>       zc->addr_len = READ_ONCE(sqe->addr_len);
>> +    zc->done_io = 0;
>>     #ifdef CONFIG_COMPAT
>>       if (req->ctx->compat)
>> @@ -1012,11 +1014,23 @@ int io_sendzc(struct io_kiocb *req, unsigned int issue_flags)
>>       if (unlikely(ret < min_ret)) {
>>           if (ret == -EAGAIN && (issue_flags & IO_URING_F_NONBLOCK))
>>               return -EAGAIN;
>> -        return ret == -ERESTARTSYS ? -EINTR : ret;
>> +        if (ret > 0 && io_net_retry(sock, msg.msg_flags)) {
>> +            zc->len -= ret;
>> +            zc->buf += ret;
>> +            zc->done_io += ret;
>> +            req->flags |= REQ_F_PARTIAL_IO;
> 
> Don't we need a prep_async function and/or something like
> io_setup_async_msg() here to handle address?

I don't think so, it's a non-vectored interface, so all the state is
already in io_sendzc.

-- 
Jens Axboe