LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
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>
Subject: Re: [patch 1/4] add basic task isolation prctl interface
Date: Wed, 28 Jul 2021 19:08:45 +0200	[thread overview]
Message-ID: <441d94dbd93443839d31fdfe5f6048de35b0d7b0.camel@redhat.com> (raw)
In-Reply-To: <20210728131610.GA11900@fuller.cnet>

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


  parent reply	other threads:[~2021-07-28 17:09 UTC|newest]

Thread overview: 26+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-07-27 10:38 [patch 0/4] prctl task isolation interface and vmstat sync Marcelo Tosatti
2021-07-27 10:38 ` [patch 1/4] add basic task isolation prctl interface Marcelo Tosatti
2021-07-27 10:48   ` nsaenzju
2021-07-27 11:00     ` Marcelo Tosatti
2021-07-27 12:38       ` nsaenzju
2021-07-27 13:06         ` Marcelo Tosatti
2021-07-27 13:08           ` Marcelo Tosatti
2021-07-27 13:09         ` Frederic Weisbecker
2021-07-27 14:52           ` Marcelo Tosatti
2021-07-27 23:45             ` Frederic Weisbecker
2021-07-28  9:37               ` Marcelo Tosatti
2021-07-28 11:45                 ` Frederic Weisbecker
2021-07-28 13:21                   ` Marcelo Tosatti
2021-07-28 21:22                     ` Frederic Weisbecker
2021-07-28 11:55                 ` nsaenzju
2021-07-28 13:16                   ` Marcelo Tosatti
     [not found]                     ` <CAFki+LkQwoqVTKmgnwLQQM8ua-ixbLp8i+jUT6xF15k6X=89mw@mail.gmail.com>
2021-07-28 16:21                       ` Marcelo Tosatti
2021-07-28 17:08                     ` nsaenzju [this message]
     [not found]                 ` <CAFki+LmHeXmSFze8YEHFNbYA5hLEtnZyk37Yjf-eyOuKa8Os4w@mail.gmail.com>
2021-07-28 16:17                   ` Marcelo Tosatti
2021-07-27 10:38 ` [patch 2/4] task isolation: sync vmstats on return to userspace Marcelo Tosatti
2021-07-27 10:38 ` [patch 3/4] mm: vmstat: move need_update Marcelo Tosatti
2021-07-27 10:38 ` [patch 4/4] mm: vmstat_refresh: avoid queueing work item if cpu stats are clean Marcelo Tosatti
2021-07-30 20:18 [patch 0/4] extensible prctl task isolation interface and vmstat sync (v2) Marcelo Tosatti
2021-07-30 20:18 ` [patch 1/4] add basic task isolation prctl interface Marcelo Tosatti
     [not found]   ` <CAFki+Lnf0cs62Se0aPubzYxP9wh7xjMXn7RXEPvrmtBdYBrsow@mail.gmail.com>
2021-07-31  0:49     ` Marcelo Tosatti
2021-07-31  7:47   ` kernel test robot
     [not found]   ` <CAFki+LkQVQOe+5aNEKWDvLdnjWjxzKWOiqOvBZzeuPWX+G=XgA@mail.gmail.com>
2021-08-02 14:16     ` Marcelo Tosatti

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=441d94dbd93443839d31fdfe5f6048de35b0d7b0.camel@redhat.com \
    --to=nsaenzju@redhat.com \
    --cc=abelits@marvell.com \
    --cc=cl@linux.com \
    --cc=frederic@kernel.org \
    --cc=juri.lelli@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mtosatti@redhat.com \
    --cc=nilal@redhat.com \
    --cc=peterx@redhat.com \
    --cc=peterz@infradead.org \
    --cc=tglx@linutronix.de \
    --subject='Re: [patch 1/4] add basic task isolation prctl interface' \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).