Re: [PATCH 0/8] io_uring: buffer registration enhancements

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On 12/11/2020 23:00, Bijan Mottahedeh wrote:
> This patchset is the follow-on to my previous RFC which implements a
> set of enhancements to buffer registration consistent with existing file
> registration functionality:

I like the idea of generic resource handling

> 
> - buffer registration updates		IORING_REGISTER_BUFFERS_UPDATE
> 					IORING_OP_BUFFERS_UPDATE

Do you need it for something specific?

> 
> - readv/writev with fixed buffers	IOSQE_FIXED_BUFFER

Why do we need it?

> 
> - buffer registration sharing		IORING_SETUP_SHARE_BUF
> 					IORING_SETUP_ATTACH_BUF

I haven't looked it up. What's the overhead on that?
And again, do you really need it?

The set is +600 lines, so just want to know that there is
a real benefit from having it.

> 
> Patches 1,2 modularize existing buffer registration code.
> 
> Patch 3 generalizes fixed_file functionality to fixed_rsrc.
> 
> Patch 4 applies fixed_rsrc functionality for fixed buffers support.
> 
> Patch 5 generalizes files_update functionality to rsrc_update.
> 
> Patch 6 implements buffer registration update, and introduces
> IORING_REGISTER_BUFFERS_UPDATE and IORING_OP_BUFFERS_UPDATE, consistent
> with file registration update.
> 
> Patch 7 implements readv/writev support with fixed buffers, and introduces
> IOSQE_FIXED_BUFFER, consistent with fixed files.
> 
> Patch 8 implements buffer sharing among multiple rings; it works as follows:
> 
> - A new ring, A,  is setup. Since no buffers have been registered, the
>   registered buffer state is an empty set, Z. That's different from the
>   NULL state in current implementation.
> 
> - Ring B is setup, attaching to Ring A. It's also attaching to it's
>   buffer registrations, now we have two references to the same empty
>   set, Z.
> 
> - Ring A registers buffers into set Z, which is no longer empty.
> 
> - Ring B sees this immediately, since it's already sharing that set.
> 
> TBD
> 
> - I think I have to add IORING_UNREGISTER_BUFFERS to
>   io_register_op_must_quiesce() but wanted to confirm.

->fixed_file_refs trades off fast unsynchronised access (by adding
percpu_ref_get/put) to not do full quiesce. So, yes, after it's hooked
up right.

> 
> I have used liburing file-{register,update} tests as models for
> buffer-{register,update,share}, tests and they run ok.
> 
> The liburing test suite fails for "self" with/without this patchset.
> 
> Bijan Mottahedeh (8):
>   io_uring: modularize io_sqe_buffer_register
>   io_uring: modularize io_sqe_buffers_register
>   io_uring: generalize fixed file functionality
>   io_uring: implement fixed buffers registration similar to fixed files
>   io_uring: generalize files_update functionlity to rsrc_update
>   io_uring: support buffer registration updates
>   io_uring: support readv/writev with fixed buffers
>   io_uring: support buffer registration sharing
> 
>  fs/io_uring.c                 | 1021 ++++++++++++++++++++++++++++++++---------
>  include/uapi/linux/io_uring.h |   15 +-
>  2 files changed, 807 insertions(+), 229 deletions(-)
> 

-- 
Pavel Begunkov



[Index of Archives]     [Linux Samsung SoC]     [Linux Rockchip SoC]     [Linux Actions SoC]     [Linux for Synopsys ARC Processors]     [Linux NFS]     [Linux NILFS]     [Linux USB Devel]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]


  Powered by Linux