All of lore.kernel.org
 help / color / mirror / Atom feed
From: Frederic Weisbecker <fweisbec@gmail.com>
To: "Luis R. Rodriguez" <mcgrof@suse.com>
Cc: Juergen Gross <jgross@suse.com>,
	"ksummit-discuss@lists.linuxfoundation.org"
	<ksummit-discuss@lists.linuxfoundation.org>,
	Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>,
	Peter Zijlstra <peterz@infradead.org>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	Chris Metcalf <cmetcalf@ezchip.com>,
	Jan Beulich <JBeulich@suse.com>, Christoph Lameter <cl@linux.com>,
	Ingo Molnar <mingo@kernel.org>
Subject: Re: [Ksummit-discuss] [BELATED CORE TOPIC] context tracking / nohz / RCU state
Date: Wed, 12 Aug 2015 16:27:34 +0200	[thread overview]
Message-ID: <20150812142732.GD21542@lerouge> (raw)
In-Reply-To: <20150811184258.GB30479@wotan.suse.de>

On Tue, Aug 11, 2015 at 08:42:58PM +0200, Luis R. Rodriguez wrote:
> On Tue, Aug 11, 2015 at 10:49:36AM -0700, Andy Lutomirski wrote:
> > This is a bit late, but here goes anyway.
> > 
> > Having played with the x86 context tracking hooks for awhile, I think
> > it would be nice if core code that needs to be aware of CPU context
> > (kernel, user, idle, guest, etc) could come up with single,
> > comprehensible, easily validated set of hooks that arch code is
> > supposed to call.
> > 
> > Currently we have:
> > 
> >  - RCU hooks, which come in a wide variety to notify about IRQs, NMIs, etc.
> > 
> >  - Context tracking hooks.  Only used by some arches.  Calling these
> > calls the RCU hooks for you in most cases.  They have weird
> > interactions with interrupts and they're slow.
> > 
> >  - vtime.  Beats the heck out of me.
> > 
> >  - Whatever deferred things Christoph keeps reminding us about.
> > 
> > Honestly, I don't fully understand what all these hooks are supposed
> > to do, nor do I care all that much.  From my perspective, the code
> > code should be able to do whatever it wants and rely on appropriate
> > notifications from arch code.  It would be great if we could come up
> > with something straightforward that covers everything.  For example:
> > 
> > user_mode_to_kernel_mode()
> > kernel_mode_to_user_mode()
> > kernel_mode_to_guest_mode()
> > in_a_periodic_tick()
> > starting_nmi()
> > ending_nmi()
> > may_i_turn_off_ticks_right_now()
> > or, better yet:
> > i_am_turning_off_ticks_right_now_and_register_your_own_darned_hrtimer_if_thats_a_problem()
> >
> > Some arches may need:
> > 
> > i_am_lame_and_forgot_my_previous_context()
> 
> Can all this information be generalized with some basic core hooks
> or could some of this contextual informatioin typically vary depending
> on the sequence we are in ? It sounds like its the later and that's
> the issue ?

That's what we do with context tracking. It tracks the context (user/kernel)
and stores these informations. And indeed the contextual informations can vary
depending for example if an exception triggered in userspace or kernelspace.

WARNING: multiple messages have this Message-ID (diff)
From: Frederic Weisbecker <fweisbec@gmail.com>
To: "Luis R. Rodriguez" <mcgrof@suse.com>
Cc: Andy Lutomirski <luto@amacapital.net>,
	"ksummit-discuss@lists.linuxfoundation.org" 
	<ksummit-discuss@lists.linuxfoundation.org>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	Paul McKenney <paulmck@linux.vnet.ibm.com>,
	Christoph Lameter <cl@linux.com>, Ingo Molnar <mingo@kernel.org>,
	Thomas Gleixner <tglx@linutronix.de>,
	"H. Peter Anvin" <hpa@zytor.com>,
	Peter Zijlstra <peterz@infradead.org>,
	Juergen Gross <jgross@suse.com>, Jan Beulich <JBeulich@suse.com>,
	Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>,
	Chris Metcalf <cmetcalf@ezchip.com>,
	Rik van Riel <riel@redhat.com>
Subject: Re: [Ksummit-discuss] [BELATED CORE TOPIC] context tracking / nohz / RCU state
Date: Wed, 12 Aug 2015 16:27:34 +0200	[thread overview]
Message-ID: <20150812142732.GD21542@lerouge> (raw)
In-Reply-To: <20150811184258.GB30479@wotan.suse.de>

On Tue, Aug 11, 2015 at 08:42:58PM +0200, Luis R. Rodriguez wrote:
> On Tue, Aug 11, 2015 at 10:49:36AM -0700, Andy Lutomirski wrote:
> > This is a bit late, but here goes anyway.
> > 
> > Having played with the x86 context tracking hooks for awhile, I think
> > it would be nice if core code that needs to be aware of CPU context
> > (kernel, user, idle, guest, etc) could come up with single,
> > comprehensible, easily validated set of hooks that arch code is
> > supposed to call.
> > 
> > Currently we have:
> > 
> >  - RCU hooks, which come in a wide variety to notify about IRQs, NMIs, etc.
> > 
> >  - Context tracking hooks.  Only used by some arches.  Calling these
> > calls the RCU hooks for you in most cases.  They have weird
> > interactions with interrupts and they're slow.
> > 
> >  - vtime.  Beats the heck out of me.
> > 
> >  - Whatever deferred things Christoph keeps reminding us about.
> > 
> > Honestly, I don't fully understand what all these hooks are supposed
> > to do, nor do I care all that much.  From my perspective, the code
> > code should be able to do whatever it wants and rely on appropriate
> > notifications from arch code.  It would be great if we could come up
> > with something straightforward that covers everything.  For example:
> > 
> > user_mode_to_kernel_mode()
> > kernel_mode_to_user_mode()
> > kernel_mode_to_guest_mode()
> > in_a_periodic_tick()
> > starting_nmi()
> > ending_nmi()
> > may_i_turn_off_ticks_right_now()
> > or, better yet:
> > i_am_turning_off_ticks_right_now_and_register_your_own_darned_hrtimer_if_thats_a_problem()
> >
> > Some arches may need:
> > 
> > i_am_lame_and_forgot_my_previous_context()
> 
> Can all this information be generalized with some basic core hooks
> or could some of this contextual informatioin typically vary depending
> on the sequence we are in ? It sounds like its the later and that's
> the issue ?

That's what we do with context tracking. It tracks the context (user/kernel)
and stores these informations. And indeed the contextual informations can vary
depending for example if an exception triggered in userspace or kernelspace.

  parent reply	other threads:[~2015-08-12 14:27 UTC|newest]

Thread overview: 56+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-08-11 17:49 [Ksummit-discuss] [BELATED CORE TOPIC] context tracking / nohz / RCU state Andy Lutomirski
2015-08-11 17:49 ` Andy Lutomirski
2015-08-11 18:33 ` [Ksummit-discuss] " Paul E. McKenney
2015-08-11 18:33   ` Paul E. McKenney
2015-08-11 19:07   ` [Ksummit-discuss] " Andy Lutomirski
2015-08-11 19:07     ` Andy Lutomirski
2015-08-11 21:47     ` [Ksummit-discuss] " Paul E. McKenney
2015-08-11 21:47       ` Paul E. McKenney
2015-08-11 21:52       ` [Ksummit-discuss] " Andy Lutomirski
2015-08-11 21:52         ` Andy Lutomirski
2015-08-12  0:51         ` [Ksummit-discuss] " Paul E. McKenney
2015-08-12  0:51           ` Paul E. McKenney
2015-08-12  1:16           ` [Ksummit-discuss] " Andy Lutomirski
2015-08-12  1:16             ` Andy Lutomirski
2015-08-12 13:38             ` [Ksummit-discuss] " Paul E. McKenney
2015-08-12 13:38               ` Paul E. McKenney
2015-08-12 14:52     ` [Ksummit-discuss] " Frederic Weisbecker
2015-08-12 14:52       ` Frederic Weisbecker
2015-08-12 14:38   ` [Ksummit-discuss] " Frederic Weisbecker
2015-08-12 14:38     ` Frederic Weisbecker
2015-08-12 15:59     ` [Ksummit-discuss] " Paul E. McKenney
2015-08-12 15:59       ` Paul E. McKenney
2015-08-11 18:42 ` [Ksummit-discuss] " Luis R. Rodriguez
2015-08-11 18:42   ` Luis R. Rodriguez
2015-08-11 21:50   ` Paul E. McKenney
2015-08-11 21:50     ` Paul E. McKenney
2015-08-12 20:17     ` Luis R. Rodriguez
2015-08-12 20:17       ` Luis R. Rodriguez
2015-08-12 14:27   ` Frederic Weisbecker [this message]
2015-08-12 14:27     ` Frederic Weisbecker
2015-08-12 16:03     ` Paul E. McKenney
2015-08-12 16:03       ` Paul E. McKenney
2015-08-13  1:29       ` Lai Jiangshan
2015-08-13  1:29         ` Lai Jiangshan
2015-08-13 13:07         ` Frederic Weisbecker
2015-08-13 13:07           ` Frederic Weisbecker
2015-08-13 13:03       ` Frederic Weisbecker
2015-08-13 13:03         ` Frederic Weisbecker
2015-08-11 19:31 ` josh
2015-08-11 19:31   ` josh
2015-08-11 21:32 ` Kevin Hilman
2015-08-11 21:32   ` Kevin Hilman
2015-08-12  3:56 ` [Ksummit-discuss] " Lai Jiangshan
2015-08-12  3:56   ` Lai Jiangshan
2015-08-12 14:20 ` Frederic Weisbecker
2015-08-12 14:20   ` Frederic Weisbecker
2015-10-12 18:40 ` [Ksummit-discuss] " Theodore Ts'o
2015-10-12 18:40   ` Theodore Ts'o
2015-10-12 19:55   ` Thomas Gleixner
2015-10-12 19:55     ` Thomas Gleixner
2015-10-12 20:40   ` Paul E. McKenney
2015-10-16 17:02     ` Frederic Weisbecker
2015-10-16 17:02       ` Frederic Weisbecker
2015-10-17 19:45       ` Theodore Ts'o
2015-10-17 19:45         ` Theodore Ts'o
2015-10-19 14:14         ` Frederic Weisbecker

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=20150812142732.GD21542@lerouge \
    --to=fweisbec@gmail.com \
    --cc=JBeulich@suse.com \
    --cc=cl@linux.com \
    --cc=cmetcalf@ezchip.com \
    --cc=jgross@suse.com \
    --cc=konrad.wilk@oracle.com \
    --cc=ksummit-discuss@lists.linuxfoundation.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mcgrof@suse.com \
    --cc=mingo@kernel.org \
    --cc=peterz@infradead.org \
    /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.