linux-kernel - Re: [PATCH v2 3/5] locking/qspinlock: Introduce CNA into the slow path of qspinlock

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 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, 6 Jun 2019 11:42:09 -0400
From:   Waiman Long <longman@...hat.com>
To:     Alex Kogan <alex.kogan@...cle.com>,
        Peter Zijlstra <peterz@...radead.org>
Cc:     linux@...linux.org.uk, mingo@...hat.com, will.deacon@....com,
        arnd@...db.de, linux-arch@...r.kernel.org,
        linux-arm-kernel@...ts.infradead.org, linux-kernel@...r.kernel.org,
        Thomas Gleixner <tglx@...utronix.de>, bp@...en8.de,
        hpa@...or.com, x86@...nel.org,
        Steven Sistare <steven.sistare@...cle.com>,
        Daniel Jordan <daniel.m.jordan@...cle.com>,
        dave.dice@...cle.com, Rahul Yadav <rahul.x.yadav@...cle.com>
Subject: Re: [PATCH v2 3/5] locking/qspinlock: Introduce CNA into the slow
 path of qspinlock

On 6/6/19 11:32 AM, Waiman Long wrote:
> On 6/6/19 11:21 AM, Alex Kogan wrote:
>>>> Also, the paravirt code is under arch/x86, while CNA is generic (not
>>>> x86-specific).  Do you still want to see CNA-related patching residing
>>>> under arch/x86?
>>>>
>>>> We still need a config option (something like NUMA_AWARE_SPINLOCKS) to
>>>> enable CNA patching under this config only, correct?
>>> There is the static_call() stuff that could be generic; I posted a new
>>> version of that today (x86 only for now, but IIRC there's arm64 patches
>>> for that around somewhere too).
>> The static_call technique appears as the more desirable long-term approach, but I think it would be prudent to keep the patches decoupled for now so we can move forward with less entanglements.
>> So unless anyone objects, we will work on plugging into the existing patching for pv.
>> And we will keep that code under arch/x86, but will be open for any suggestion to move it elsewhere.
>>
> If you mean making the CNV code depends on PARAVIRT_SPINLOCKS for now,
> that is fine. The code should be under kernel/locking. You shouldn't put
> it somewhere under arch/x86.

I mean the core CNV code should be under kernel/locking. The paravirt
specific code, however, should be close to the current paravirt code
which is under arch/x86.

-Longman