linux-kernel - Re: [BUG]locking/rwsem: only clean RWSEM_FLAG

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]

Date:   Thu, 11 Nov 2021 16:46:31 -0500
From:   Waiman Long <longman@...hat.com>
To:     Peter Zijlstra <peterz@...radead.org>
Cc:     Hillf Danton <hdanton@...a.com>,
        马振华 <mazhenhua@...omi.com>,
        mingo <mingo@...hat.com>, will <will@...nel.org>,
        "boqun.feng" <boqun.feng@...il.com>,
        linux-kernel <linux-kernel@...r.kernel.org>
Subject: Re: [BUG]locking/rwsem: only clean RWSEM_FLAG_HANDOFF when already
 set


On 11/11/21 16:38, Peter Zijlstra wrote:
> On Thu, Nov 11, 2021 at 04:01:16PM -0500, Waiman Long wrote:
>
>>>> +			if (has_handoff || (!waiter->rt_task &&
>>>> +					    !time_after(jiffies, waiter->timeout)))
>>> Does ->rt_task really help over rt_task(current) ? I suppose there's an
>>> argument for locality, but that should be pretty much it, no?
>> Waiting for the timeout may introduce too much latency for RT task. That is
>> the only reason I am doing it. I can take it out if you think it is not
>> necessary.
> I meant simply calling rt_task(waiter->task) here, instead of mucking about
> with the extra variable.

OK, that make sense.

Cheers,
Longman