[RFC,v4,0/5] introduce io_uring_cmd_import_fixed_vec

Message

Sidong Yang March 17, 2025, 1:57 p.m. UTC

This patche series introduce io_uring_cmd_import_vec. With this function,
Multiple fixed buffer could be used in uring cmd. It's vectored version
for io_uring_cmd_import_fixed(). Also this patch series includes a usage
for new api for encoded read/write in btrfs by using uring cmd.

There was approximately 10 percent of performance improvements through benchmark.
The benchmark code is in
https://github.com/SidongYang/btrfs-encoded-io-test/blob/main/main.c

./main -l
Elapsed time: 0.598997 seconds
./main -l -f
Elapsed time: 0.540332 seconds

Additionally, There is a commit that fixed a memory bug in btrfs uring encoded
read.

v2:
 - don't export iou_vc, use bvec for btrfs
 - use io_is_compat for checking compat
 - reduce allocation/free for import fixed vec

v3:
 - add iou_vec cache in io_uring_cmd and use it
 - also encoded write fixed supported

v4:
 - add a patch that introduce io_async_cmd
 - add a patch that fixes a bug in btrfs encoded read

Sidong Yang (5):
  io_uring/cmd: introduce io_async_cmd for hide io_uring_cmd_data
  io-uring/cmd: add iou_vec field for io_uring_cmd
  io-uring/cmd: introduce io_uring_cmd_import_fixed_vec
  btrfs: ioctl: introduce btrfs_uring_import_iovec()
  btrfs: ioctl: don't free iov when -EAGAIN in uring encoded read

 fs/btrfs/ioctl.c             | 35 ++++++++++++++++-----
 include/linux/io_uring/cmd.h | 14 +++++++++
 io_uring/io_uring.c          |  4 +--
 io_uring/opdef.c             |  3 +-
 io_uring/uring_cmd.c         | 60 +++++++++++++++++++++++++++++++-----
 io_uring/uring_cmd.h         | 10 ++++++
 6 files changed, 108 insertions(+), 18 deletions(-)

Comments

Pavel Begunkov March 18, 2025, 7:30 a.m. UTC | #1

On 3/17/25 13:57, Sidong Yang wrote:
> This patche series introduce io_uring_cmd_import_vec. With this function,
> Multiple fixed buffer could be used in uring cmd. It's vectored version
> for io_uring_cmd_import_fixed(). Also this patch series includes a usage
> for new api for encoded read/write in btrfs by using uring cmd.

You're vigorously ignoring the previous comment, you can't stick
your name to my patches and send them as your own, that's not
going to work. git format-patch and other tools allow to send
other's patches in the same patch set without mutilating them.

Sidong Yang March 18, 2025, 7:41 a.m. UTC | #2

On Tue, Mar 18, 2025 at 07:30:51AM +0000, Pavel Begunkov wrote:
> On 3/17/25 13:57, Sidong Yang wrote:
> > This patche series introduce io_uring_cmd_import_vec. With this function,
> > Multiple fixed buffer could be used in uring cmd. It's vectored version
> > for io_uring_cmd_import_fixed(). Also this patch series includes a usage
> > for new api for encoded read/write in btrfs by using uring cmd.
> 
> You're vigorously ignoring the previous comment, you can't stick
> your name to my patches and send them as your own, that's not
> going to work. git format-patch and other tools allow to send
> other's patches in the same patch set without mutilating them.

I'm just not familiar with this. That wasn't my intention. Sorry, Your
patches will be included without modification.

Thanks,
Sidong
> 
> -- 
> Pavel Begunkov
>

Jens Axboe March 18, 2025, 1:19 p.m. UTC | #3

On 3/18/25 1:41 AM, Sidong Yang wrote:
> On Tue, Mar 18, 2025 at 07:30:51AM +0000, Pavel Begunkov wrote:
>> On 3/17/25 13:57, Sidong Yang wrote:
>>> This patche series introduce io_uring_cmd_import_vec. With this function,
>>> Multiple fixed buffer could be used in uring cmd. It's vectored version
>>> for io_uring_cmd_import_fixed(). Also this patch series includes a usage
>>> for new api for encoded read/write in btrfs by using uring cmd.
>>
>> You're vigorously ignoring the previous comment, you can't stick
>> your name to my patches and send them as your own, that's not
>> going to work. git format-patch and other tools allow to send
>> other's patches in the same patch set without mutilating them.
> 
> I'm just not familiar with this. That wasn't my intention. Sorry, Your
> patches will be included without modification.

Assuming these are from a git branch you have, just ensure they are
committed with Pavel as the author, and git send-email will do the right
thing when sending out the patch.