Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Message-ID: <441d94dbd93443839d31fdfe5f6048de35b0d7b0.camel@redhat.com>
Subject: Re: [patch 1/4] add basic task isolation prctl interface
From:   nsaenzju@redhat.com
To:     Marcelo Tosatti <mtosatti@redhat.com>
Cc:     Frederic Weisbecker <frederic@kernel.org>,
        linux-kernel@vger.kernel.org, Nitesh Lal <nilal@redhat.com>,
        Christoph Lameter <cl@linux.com>,
        Juri Lelli <juri.lelli@redhat.com>,
        Peter Zijlstra <peterz@infradead.org>,
        Alex Belits <abelits@marvell.com>,
        Peter Xu <peterx@redhat.com>,
        Thomas Gleixner <tglx@linutronix.de>
Date:   Wed, 28 Jul 2021 19:08:45 +0200
In-Reply-To: <20210728131610.GA11900@fuller.cnet>
References: <20210727103803.464432924@fuller.cnet>
         <20210727104119.551607458@fuller.cnet>
         <7b2d6bf91d30c007e19a7d2cbddcb2460e72d163.camel@redhat.com>
         <20210727110050.GA502360@fuller.cnet>
         <a020a45ddea10956938f59bd235b88fe873d0e98.camel@redhat.com>
         <20210727130930.GB283787@lothringen> <20210727145209.GA518735@fuller.cnet>
         <20210727234539.GH283787@lothringen> <20210728093707.GA3242@fuller.cnet>
         <e0135b88dad323d0abd1ce05081e0b554421af7c.camel@redhat.com>
         <20210728131610.GA11900@fuller.cnet>
Content-Type: text/plain; charset="UTF-8"
User-Agent: Evolution 3.40.3 (3.40.3-1.fc34) 
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
Precedence: bulk

On Wed, 2021-07-28 at 10:16 -0300, Marcelo Tosatti wrote:
> > For example, let's say we introduce ISOL_F_QUIESCE_DEFER_TLB_FLUSH, this will
> > defer relatively short IPIs on isolated CPUs in exchange for a longer flush
> > whenever we enter the kernel (syscall, IRQs, NMI, etc...). 
> 
> Why the flush has to be longer when you enter the kernel?

What I had in mind was cost of rapid partial flushes (IPIs) vs full flushes on
entry, although I haven't really measured anything so the extra latency cost
might as well be zero.

> ISOL_F_QUIESCE_DEFER_TLB_FLUSH might collapse multiple IPIs 
> into a single IPI, so the behaviour might be beneficial 
> for "standard" types of application as well.
> 
> > A latency sensitive
> > application might be OK with the former but not with the latter.
> 
> Two alternatives:
> 
> 1) The pattern above, where particular subsystems that might interrupt 
> the kernel are enabled automatically if the kernel supports it.
> 
> Pros: 
> Applications which implement this only need to be changed once,
> and can benefit from new kernel features.
>
> Applications can disable particular features if they turn
> out to be problematic.
> 
> Cons: 
> New features might break applications.
>
> 2) Force applications to enable each new feature individually.
> 
> Pros: Won't cause regressions, kernel behaviour is explicitly 
> controlled by userspace.
> 
> Cons: Apps won't benefit from new features automatically.
> 
> ---
> 
> It seems to me 1) is preferred. Can also add a sysfs control to
> have a "default_isolation_feature" flag, which can be changed
> by a sysadmin in case a new feature is undesired.
> 
> Thoughts?

I'd still take option 2. Nitesh has a very good point, latency requirements are
hit or miss. What's the benefit of enabling new features on an already valid
application vs the potential regression?

That said I see value in providing means for users that want all
features/modes, but it should be an through an explicit action on their part.

--
Nicolás Sáenz