Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
MIME-Version: 1.0
In-Reply-To: <CF849A07-B7CE-4DE9-8246-53AC5A53A705@surriel.com>
References: <20180716190337.26133-1-riel@surriel.com> <20180716190337.26133-5-riel@surriel.com>
 <CALCETrUMntRe_LYX5wj9YD4xZ+84QExK+ZNb3yxEBDEKa7nePQ@mail.gmail.com>
 <B6566AA4-2E8A-43BD-A224-0F6D4747B8FB@surriel.com> <F93A3335-4830-4E36-9179-D0FA5D640C8A@amacapital.net>
 <B976CC13-D014-433A-83DE-F8DF9AB4F421@surriel.com> <CALCETrWDiMhgiR3f8n0jdWcW31EDJ+Waq0wh5sMDutfigANGnA@mail.gmail.com>
 <CF849A07-B7CE-4DE9-8246-53AC5A53A705@surriel.com>
From:   Andy Lutomirski <luto@amacapital.net>
Date:   Thu, 19 Jul 2018 10:18:58 -0700
Message-ID: <CALCETrXhkcMyMo_-dmrQ3M+VJqLS_me5TKA688bdsQo5m1ddzg@mail.gmail.com>
Subject: Re: [PATCH 4/7] x86,tlb: make lazy TLB mode lazier
To:     Rik van Riel <riel@surriel.com>
Cc:     Peter Zijlstra <peterz@infradead.org>,
        Vitaly Kuznetsov <vkuznets@redhat.com>,
        Andy Lutomirski <luto@kernel.org>,
        LKML <linux-kernel@vger.kernel.org>, X86 ML <x86@kernel.org>,
        Mike Galbraith <efault@gmx.de>,
        kernel-team <kernel-team@fb.com>, Ingo Molnar <mingo@kernel.org>,
        Dave Hansen <dave.hansen@intel.com>
Content-Type: text/plain; charset="UTF-8"
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Thu, Jul 19, 2018 at 10:15 AM, Rik van Riel <riel@surriel.com> wrote:
>
>
> Given that CPUs in lazy TLB mode stay part of the mm_cpumask,
> that WARN_ON seems misplaced. You are right though, that the
> mm_cpumask alone should provide enough information for us to
> avoid a need for both tsk->active_mm and the refcounting.
>

If you do this extra shootdown after freeing pagetables, it would be
odd if mm_cpumask() wasn't empty.  But you're right, the warn is
probably silly.  And if you move it into arch_exit_mmap(), the warn is
definitely wrong.

>
> Does all that make sense?  Basically, as I understand it, the
> expensive atomic ops you're seeing are all pointless because they're
> enabling an optimization that hasn't actually worked for a long time,
> if ever.
>
>
> Our benchmark results suggest that lazy TLB mode works, and makes
> a measurable performance difference. Getting rid of the atomic ops
> should make it a little better, though :)
>

I'm not saying lazy mode is useless.  I'm saying that active_mm isn't
needed for x86's lazy mode :)