netdev - Re: [PATCH v1 0/7] reverse ifq refcount

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <c39297cf-daab-43e0-82c7-3210d570e427@kernel.dk>
Date: Thu, 6 Nov 2025 16:21:29 -0700
From: Jens Axboe <axboe@...nel.dk>
To: Pavel Begunkov <asml.silence@...il.com>, David Wei <dw@...idwei.uk>,
 io-uring@...r.kernel.org, netdev@...r.kernel.org
Subject: Re: [PATCH v1 0/7] reverse ifq refcount

On 11/6/25 9:10 AM, Pavel Begunkov wrote:
> On 11/4/25 22:44, David Wei wrote:
>> Reverse the refcount relationship between ifq and rings i.e. ring ctxs
>> and page pool memory providers hold refs on an ifq instead of the other
>> way around. This makes ifqs an independently refcounted object separate
>> to rings.
>>
>> This is split out from a larger patchset [1] that adds ifq sharing. It
>> will be needed for both ifq export and import/sharing later. Split it
>> out as to make dependency management easier.
>>
>> [1]: https://lore.kernel.org/io-uring/20251103234110.127790-1-dw@davidwei.uk/
> 
> FWIW, if 1-3 are merged I can take the rest to the mix with
> dependencies for David's work, but it should also be fine if
> all 7 go into io_uring-6.19 there shouldn't be any conflicts.

Let's just stuff them into for-6.19/io_uring and avoid the extra
roundtrip.

-- 
Jens Axboe