All of lore.kernel.org
 help / color / mirror / Atom feed
From: Josh Poimboeuf <jpoimboe@redhat.com>
To: Jiri Kosina <jkosina@suse.cz>
Cc: Vojtech Pavlik <vojtech@suse.com>,
	Peter Zijlstra <peterz@infradead.org>,
	Andrew Morton <akpm@linux-foundation.org>,
	Ingo Molnar <mingo@redhat.com>,
	Seth Jennings <sjenning@redhat.com>,
	linux-kernel@vger.kernel.org
Subject: Re: [PATCH 1/3] sched: add sched_task_call()
Date: Thu, 19 Feb 2015 17:11:41 -0600	[thread overview]
Message-ID: <20150219231141.GE15980@treble.redhat.com> (raw)
In-Reply-To: <alpine.LNX.2.00.1502192223070.28769@pobox.suse.cz>

On Thu, Feb 19, 2015 at 10:26:09PM +0100, Jiri Kosina wrote:
> On Thu, 19 Feb 2015, Josh Poimboeuf wrote:
> 
> > How about with a TIF_IN_USERSPACE thread flag?  It could be cleared/set
> > right at the border.  Then for running tasks it's as simple as:
> > 
> > if (test_tsk_thread_flag(task, TIF_IN_USERSPACE))
> > 	klp_switch_task_universe(task);
> 
> That's in principle what CONTEXT_TRACKING is doing, i.e. the condition 
> we'd be interested in would be
> 
> 	__this_cpu_read(context_tracking.state) == IN_USER
> 
> But it has overhead.

Yeah, that does seem to pretty much do what we want.  Unfortunately it
has a much higher overhead than just setting a thread flag.

And from the Kconfig description for CONTEXT_TRACKING_FORCE, which would
enable it on all CPUs during boot, "this option brings an overhead that
you don't want in production."

Maybe that code needs a rewrite to rely on a thread flag instead.  Then
we could use it too.

-- 
Josh

  parent reply	other threads:[~2015-02-19 23:12 UTC|newest]

Thread overview: 86+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-02-16 18:52 [PATCH 0/3] prevent /proc/<pid>/stack garbage for running tasks Josh Poimboeuf
2015-02-16 18:52 ` [PATCH 1/3] sched: add sched_task_call() Josh Poimboeuf
2015-02-16 20:44   ` Peter Zijlstra
2015-02-16 22:05     ` Josh Poimboeuf
2015-02-17  9:24       ` Peter Zijlstra
2015-02-17 14:12         ` Josh Poimboeuf
2015-02-17 18:15           ` Peter Zijlstra
2015-02-17 21:25             ` Josh Poimboeuf
2015-02-18 15:21               ` Peter Zijlstra
2015-02-18 17:12                 ` Josh Poimboeuf
2015-02-19  0:20                   ` Peter Zijlstra
2015-02-19  4:17                     ` Josh Poimboeuf
2015-02-19 10:16                       ` Peter Zijlstra
2015-02-19 16:24                         ` Josh Poimboeuf
2015-02-19 16:33                           ` Vojtech Pavlik
2015-02-19 17:03                             ` Josh Poimboeuf
2015-02-19 17:08                               ` Jiri Kosina
2015-02-19 17:19                               ` Vojtech Pavlik
2015-02-19 17:32                                 ` Josh Poimboeuf
2015-02-19 17:48                                   ` Vojtech Pavlik
2015-02-19 20:40                                   ` Vojtech Pavlik
2015-02-19 21:42                                     ` Josh Poimboeuf
2015-02-20  7:46                                       ` Jiri Kosina
2015-02-20  8:49                                         ` Jiri Kosina
2015-02-20  9:50                                           ` Ingo Molnar
2015-02-20 10:02                                             ` Jiri Kosina
2015-02-20 10:44                                               ` live patching design (was: Re: [PATCH 1/3] sched: add sched_task_call()) Ingo Molnar
2015-02-20 10:58                                                 ` Jiri Kosina
2015-02-20 19:49                                                   ` Ingo Molnar
2015-02-20 21:46                                                     ` Vojtech Pavlik
2015-02-20 22:08                                                       ` Josh Poimboeuf
2015-02-21 18:30                                                         ` Ingo Molnar
2015-02-22  8:52                                                           ` Jiri Kosina
2015-02-22 10:17                                                             ` Ingo Molnar
2015-02-22 19:18                                                               ` Jiri Kosina
2015-02-23 12:43                                                           ` Jiri Kosina
2015-02-24 10:37                                                             ` Ingo Molnar
2015-02-21 18:18                                                       ` Ingo Molnar
2015-02-21 18:57                                                         ` Jiri Kosina
2015-02-21 19:16                                                           ` Ingo Molnar
2015-02-21 19:31                                                             ` Jiri Kosina
2015-02-21 19:48                                                               ` Ingo Molnar
2015-02-21 20:10                                                                 ` Jiri Kosina
2015-02-21 20:53                                                                   ` Jiri Kosina
2015-02-22  8:46                                                                   ` Ingo Molnar
2015-02-22  9:08                                                                     ` Jiri Kosina
2015-02-22  9:46                                                                     ` live kernel upgrades (was: live kernel patching design) Ingo Molnar
2015-02-22 10:34                                                                       ` Ingo Molnar
2015-02-22 10:48                                                                       ` Ingo Molnar
2015-02-22 19:13                                                                         ` Jiri Kosina
2015-02-22 23:01                                                                           ` Andrew Morton
2015-02-23  0:18                                                                             ` Dave Airlie
2015-02-23  0:44                                                                               ` Arjan van de Ven
2015-02-23  8:17                                                                                 ` Jiri Kosina
2015-02-23 10:42                                                                                   ` Richard Weinberger
2015-02-23 11:08                                                                                     ` Vojtech Pavlik
2015-02-23 11:50                                                                                 ` Pavel Machek
2015-02-24  9:16                                                                                 ` Ingo Molnar
2015-02-24 12:28                                                                                   ` Jiri Slaby
2015-03-05  0:51                                                                                     ` Ingo Molnar
2015-02-23  6:35                                                                             ` Vojtech Pavlik
2015-02-24  9:44                                                                               ` Ingo Molnar
2015-02-24 12:12                                                                                 ` Vojtech Pavlik
2015-02-24 10:53                                                                           ` Ingo Molnar
2015-02-24 12:19                                                                             ` Vojtech Pavlik
2015-02-22 14:37                                                                       ` Josh Poimboeuf
2015-02-22 16:40                                                                         ` Josh Poimboeuf
2015-02-22 19:03                                                                           ` Jiri Kosina
2015-02-24 10:23                                                                         ` Ingo Molnar
2015-02-24 11:10                                                                           ` Petr Mladek
2015-02-24 12:36                                                                           ` Vojtech Pavlik
2015-02-23 11:39                                                                       ` Pavel Machek
2015-02-24 10:25                                                                         ` Ingo Molnar
2015-02-24 12:11                                                                       ` Jiri Slaby
2015-02-24 13:18                                                                         ` live kernel upgrades Pavel Emelyanov
2015-02-20 16:12                                             ` [PATCH 1/3] sched: add sched_task_call() Josh Poimboeuf
2015-02-20 20:08                                               ` Ingo Molnar
2015-02-20 21:22                                                 ` Josh Poimboeuf
2015-02-20 17:05                                           ` Josh Poimboeuf
2015-02-19 21:26                                   ` Jiri Kosina
2015-02-19 21:38                                     ` Jiri Kosina
2015-02-19 23:11                                     ` Josh Poimboeuf [this message]
2015-02-16 18:52 ` [PATCH 2/3] stacktrace: add save_stack_trace_tsk_safe() Josh Poimboeuf
2015-02-18  0:13   ` Andrew Morton
2015-02-20  9:32     ` Jiri Kosina
2015-02-16 18:52 ` [PATCH 3/3] proc: fix /proc/<pid>/stack for running tasks Josh Poimboeuf

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=20150219231141.GE15980@treble.redhat.com \
    --to=jpoimboe@redhat.com \
    --cc=akpm@linux-foundation.org \
    --cc=jkosina@suse.cz \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@redhat.com \
    --cc=peterz@infradead.org \
    --cc=sjenning@redhat.com \
    --cc=vojtech@suse.com \
    /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
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.