IO-Uring Archive on lore.kernel.org
 help / color / Atom feed
From: Pavel Begunkov <asml.silence@gmail.com>
To: Jens Axboe <axboe@kernel.dk>,
	io-uring@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH 1/3] io_uring: pass sqe for link head
Date: Thu, 6 Feb 2020 00:43:56 +0300
Message-ID: <332a2afd-cdfc-3b7f-c228-abc5bdc7646a@gmail.com> (raw)
In-Reply-To: <19f6aa0e-158f-5125-9df9-39ae95c72962@gmail.com>

[-- Attachment #1.1: Type: text/plain, Size: 3822 bytes --]

On 06/02/2020 00:39, Pavel Begunkov wrote:
> On 05/02/2020 22:07, Pavel Begunkov wrote:
>> Save an sqe for a head of a link, so it doesn't go through switch in
>> io_req_defer_prep() nor allocating an async context in advance.
>>
>> Also, it's fixes potenial memleak for double-preparing head requests.
>> E.g. prep in io_submit_sqe() and then prep in io_req_defer(),
>> which leaks iovec for vectored read/writes.
> 
> Looking through -rc1, remembered that Jens already fixed this. So, this may be
> striked out.

Just to clarify, I was talking about removing the last argument in the patch
message.

> 
> 
>>
>> Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
>> ---
>>  fs/io_uring.c | 19 ++++++++++---------
>>  1 file changed, 10 insertions(+), 9 deletions(-)
>>
>> diff --git a/fs/io_uring.c b/fs/io_uring.c
>> index f00c2c9c67c0..e18056af5672 100644
>> --- a/fs/io_uring.c
>> +++ b/fs/io_uring.c
>> @@ -4721,20 +4721,22 @@ static void io_queue_sqe(struct io_kiocb *req, const struct io_uring_sqe *sqe)
>>  	}
>>  }
>>  
>> -static inline void io_queue_link_head(struct io_kiocb *req)
>> +static inline void io_queue_link_head(struct io_kiocb *req,
>> +				      const struct io_uring_sqe *sqe)
>>  {
>>  	if (unlikely(req->flags & REQ_F_FAIL_LINK)) {
>>  		io_cqring_add_event(req, -ECANCELED);
>>  		io_double_put_req(req);
>>  	} else
>> -		io_queue_sqe(req, NULL);
>> +		io_queue_sqe(req, sqe);
>>  }
>>  
>>  #define SQE_VALID_FLAGS	(IOSQE_FIXED_FILE|IOSQE_IO_DRAIN|IOSQE_IO_LINK|	\
>>  				IOSQE_IO_HARDLINK | IOSQE_ASYNC)
>>  
>>  static bool io_submit_sqe(struct io_kiocb *req, const struct io_uring_sqe *sqe,
>> -			  struct io_submit_state *state, struct io_kiocb **link)
>> +			  struct io_submit_state *state, struct io_kiocb **link,
>> +			  const struct io_uring_sqe **link_sqe)
>>  {
>>  	const struct cred *old_creds = NULL;
>>  	struct io_ring_ctx *ctx = req->ctx;
>> @@ -4812,7 +4814,7 @@ static bool io_submit_sqe(struct io_kiocb *req, const struct io_uring_sqe *sqe,
>>  
>>  		/* last request of a link, enqueue the link */
>>  		if (!(sqe_flags & (IOSQE_IO_LINK|IOSQE_IO_HARDLINK))) {
>> -			io_queue_link_head(head);
>> +			io_queue_link_head(head, *link_sqe);
>>  			*link = NULL;
>>  		}
>>  	} else {
>> @@ -4823,10 +4825,8 @@ static bool io_submit_sqe(struct io_kiocb *req, const struct io_uring_sqe *sqe,
>>  		if (sqe_flags & (IOSQE_IO_LINK|IOSQE_IO_HARDLINK)) {
>>  			req->flags |= REQ_F_LINK;
>>  			INIT_LIST_HEAD(&req->link_list);
>> -			ret = io_req_defer_prep(req, sqe);
>> -			if (ret)
>> -				req->flags |= REQ_F_FAIL_LINK;
>>  			*link = req;
>> +			*link_sqe = sqe;
>>  		} else {
>>  			io_queue_sqe(req, sqe);
>>  		}
>> @@ -4924,6 +4924,7 @@ static int io_submit_sqes(struct io_ring_ctx *ctx, unsigned int nr,
>>  	struct io_kiocb *link = NULL;
>>  	int i, submitted = 0;
>>  	bool mm_fault = false;
>> +	const struct io_uring_sqe *link_sqe = NULL;
>>  
>>  	/* if we have a backlog and couldn't flush it all, return BUSY */
>>  	if (test_bit(0, &ctx->sq_check_overflow)) {
>> @@ -4983,7 +4984,7 @@ static int io_submit_sqes(struct io_ring_ctx *ctx, unsigned int nr,
>>  		req->needs_fixed_file = async;
>>  		trace_io_uring_submit_sqe(ctx, req->opcode, req->user_data,
>>  						true, async);
>> -		if (!io_submit_sqe(req, sqe, statep, &link))
>> +		if (!io_submit_sqe(req, sqe, statep, &link, &link_sqe))
>>  			break;
>>  	}
>>  
>> @@ -4993,7 +4994,7 @@ static int io_submit_sqes(struct io_ring_ctx *ctx, unsigned int nr,
>>  		percpu_ref_put_many(&ctx->refs, nr - ref_used);
>>  	}
>>  	if (link)
>> -		io_queue_link_head(link);
>> +		io_queue_link_head(link, link_sqe);
>>  	if (statep)
>>  		io_submit_state_end(&state);
>>  
>>
> 

-- 
Pavel Begunkov


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

  reply index

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-02-05 19:07 [PATCH 0/3] io_uring: clean wq path Pavel Begunkov
2020-02-05 19:07 ` [PATCH 1/3] io_uring: pass sqe for link head Pavel Begunkov
2020-02-05 21:39   ` Pavel Begunkov
2020-02-05 21:43     ` Pavel Begunkov [this message]
2020-02-05 19:07 ` [PATCH 2/3] io_uring: deduce force_nonblock in io_issue_sqe() Pavel Begunkov
2020-02-05 19:07 ` [PATCH 3/3] io_uring: pass submission ref to async Pavel Begunkov
2020-02-05 22:29 ` [PATCH 0/3] io_uring: clean wq path Pavel Begunkov
2020-02-06  2:50   ` Jens Axboe
2020-02-06  9:51     ` Pavel Begunkov
2020-02-06 14:26       ` Jens Axboe

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=332a2afd-cdfc-3b7f-c228-abc5bdc7646a@gmail.com \
    --to=asml.silence@gmail.com \
    --cc=axboe@kernel.dk \
    --cc=io-uring@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

IO-Uring Archive on lore.kernel.org

Archives are clonable:
	git clone --mirror https://lore.kernel.org/io-uring/0 io-uring/git/0.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 io-uring io-uring/ https://lore.kernel.org/io-uring \
		io-uring@vger.kernel.org
	public-inbox-index io-uring

Example config snippet for mirrors

Newsgroup available over NNTP:
	nntp://nntp.lore.kernel.org/org.kernel.vger.io-uring


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git