public inbox for io-uring@vger.kernel.org
 help / color / mirror / Atom feed
From: Christoph Hellwig <hch@lst.de>
To: Caleb Sander Mateos <csander@purestorage.com>
Cc: Jens Axboe <axboe@kernel.dk>, Miklos Szeredi <miklos@szeredi.hu>,
	Ming Lei <ming.lei@redhat.com>, Keith Busch <kbusch@kernel.org>,
	Christoph Hellwig <hch@lst.de>, Sagi Grimberg <sagi@grimberg.me>,
	Chris Mason <clm@fb.com>, David Sterba <dsterba@suse.com>,
	io-uring@vger.kernel.org, linux-fsdevel@vger.kernel.org,
	linux-block@vger.kernel.org, linux-nvme@lists.infradead.org,
	linux-btrfs@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH 3/3] io_uring/uring_cmd: avoid double indirect call in task work dispatch
Date: Thu, 23 Oct 2025 15:40:47 +0200	[thread overview]
Message-ID: <20251023134047.GA24570@lst.de> (raw)
In-Reply-To: <20251022231326.2527838-4-csander@purestorage.com>

On Wed, Oct 22, 2025 at 05:13:26PM -0600, Caleb Sander Mateos wrote:
> io_uring task work dispatch makes an indirect call to struct io_kiocb's
> io_task_work.func field to allow running arbitrary task work functions.
> In the uring_cmd case, this calls io_uring_cmd_work(), which immediately
> makes another indirect call to struct io_uring_cmd's task_work_cb field.
> Introduce a macro DEFINE_IO_URING_CMD_TASK_WORK() to define a
> io_req_tw_func_t function wrapping an io_uring_cmd_tw_t. Convert the
> io_uring_cmd_tw_t function to the io_req_tw_func_t function in
> io_uring_cmd_complete_in_task() and io_uring_cmd_do_in_task_lazy().
> Use DEFINE_IO_URING_CMD_TASK_WORK() to define a io_req_tw_func_t
> function for each existing io_uring_cmd_tw_t function. Now uring_cmd
> task work dispatch makes a single indirect call to the io_req_tw_func_t
> wrapper function, which can inline the io_uring_cmd_tw_t function. This
> also allows removing the task_work_cb field from struct io_uring_cmd,
> freeing up some additional storage space.

Please just open code the logic instead of the symbol-hiding multi-level
macro indirection.  Everyone who will have to touch the code in the
future will thank you.


  parent reply	other threads:[~2025-10-23 13:40 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-10-22 23:13 [PATCH 0/3] io_uring/uring_cmd: avoid double indirect call in task work dispatch Caleb Sander Mateos
2025-10-22 23:13 ` [PATCH 1/3] io_uring: expose io_should_terminate_tw() Caleb Sander Mateos
2025-10-22 23:13 ` [PATCH 2/3] io_uring/uring_cmd: call io_should_terminate_tw() when needed Caleb Sander Mateos
2025-10-22 23:13 ` [PATCH 3/3] io_uring/uring_cmd: avoid double indirect call in task work dispatch Caleb Sander Mateos
2025-10-23 12:05   ` kernel test robot
2025-10-23 13:40   ` Christoph Hellwig [this message]
2025-10-23 19:12     ` Caleb Sander Mateos
2025-10-23 19:49   ` kernel test robot

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=20251023134047.GA24570@lst.de \
    --to=hch@lst.de \
    --cc=axboe@kernel.dk \
    --cc=clm@fb.com \
    --cc=csander@purestorage.com \
    --cc=dsterba@suse.com \
    --cc=io-uring@vger.kernel.org \
    --cc=kbusch@kernel.org \
    --cc=linux-block@vger.kernel.org \
    --cc=linux-btrfs@vger.kernel.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-nvme@lists.infradead.org \
    --cc=miklos@szeredi.hu \
    --cc=ming.lei@redhat.com \
    --cc=sagi@grimberg.me \
    /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
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox