linux-kernel - Re: [External] Re: [PATCH v3 2/3] io_uring: avoid ring quiesce while registering/unregistering eventfd

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: <b06201e7-98dd-a087-cdf4-1a4ea45767f0@kernel.dk>
Date:   Thu, 3 Feb 2022 15:18:00 -0700
From:   Jens Axboe <axboe@...nel.dk>
To:     Pavel Begunkov <asml.silence@...il.com>,
        Usama Arif <usama.arif@...edance.com>,
        io-uring@...r.kernel.org, linux-kernel@...r.kernel.org
Cc:     fam.zheng@...edance.com
Subject: Re: [External] Re: [PATCH v3 2/3] io_uring: avoid ring quiesce while
 registering/unregistering eventfd

On 2/3/22 12:43 PM, Pavel Begunkov wrote:
> On 2/3/22 19:06, Jens Axboe wrote:
>> On 2/3/22 12:00 PM, Pavel Begunkov wrote:
>>> On 2/3/22 18:29, Jens Axboe wrote:
>>>> On 2/3/22 11:26 AM, Usama Arif wrote:
>>>>> Hmm, maybe i didn't understand you and Pavel correctly. Are you
>>>>> suggesting to do the below diff over patch 3? I dont think that would be
>>>>> correct, as it is possible that just after checking if ctx->io_ev_fd is
>>>>> present unregister can be called by another thread and set ctx->io_ev_fd
>>>>> to NULL that would cause a NULL pointer exception later? In the current
>>>>> patch, the check of whether ev_fd exists happens as the first thing
>>>>> after rcu_read_lock and the rcu_read_lock are extremely cheap i believe.
>>>>
>>>> They are cheap, but they are still noticeable at high requests/sec
>>>> rates. So would be best to avoid them.
>>>>
>>>> And yes it's obviously racy, there's the potential to miss an eventfd
>>>> notification if it races with registering an eventfd descriptor. But
>>>> that's not really a concern, as if you register with inflight IO
>>>> pending, then that always exists just depending on timing. The only
>>>> thing I care about here is that it's always _safe_. Hence something ala
>>>> what you did below is totally fine, as we're re-evaluating under rcu
>>>> protection.
>>>
>>> Indeed, the patch doesn't have any formal guarantees for propagation
>>> to already inflight requests, so this extra unsynchronised check
>>> doesn't change anything.
>>>
>>> I'm still more сurious why we need RCU and extra complexity when
>>> apparently there is no use case for that. If it's only about
>>> initial initialisation, then as I described there is a much
>>> simpler approach.
>>
>> Would be nice if we could get rid of the quiesce code in general, but I
>> haven't done a check to see what'd be missing after this...
> 
> Ok, I do think full quiesce is worth keeping as don't think all
> registered parts need dynamic update. E.g. zc notification dynamic
> reregistation doesn't make sense and I'd rather rely on existing
> straightforward mechanisms than adding extra bits, even if it's
> rsrc_nodes. That's not mentioning unnecessary extra overhead.
> 
> btw, I wouldn't say this eventfd specific sync is much simpler than
> the whole full quiesce.

It's easier to understand though, as it follows the usual rules of
RCU which are used throughout the kernel.

On quiesce in general, my curiosity was around whether we'd ever
get to the point where all register handlers are marked as not
needing quisce, and it seems it isn't that far off.

-- 
Jens Axboe