public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Marcelo Tosatti <mtosatti@redhat.com>
To: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org, Nitesh Lal <nilal@redhat.com>,
	Nicolas Saenz Julienne <nsaenzju@redhat.com>,
	Frederic Weisbecker <frederic@kernel.org>,
	Christoph Lameter <cl@linux.com>,
	Juri Lelli <juri.lelli@redhat.com>,
	Peter Zijlstra <peterz@infradead.org>,
	Alex Belits <abelits@belits.com>, Peter Xu <peterx@redhat.com>,
	Daniel Bristot de Oliveira <bristot@redhat.com>,
	Oscar Shiang <oscar0225@livemail.tw>
Subject: Re: [patch v12 13/13] task isolation: only TIF_TASK_ISOL if task isolation is enabled
Date: Tue, 3 May 2022 16:12:21 -0300	[thread overview]
Message-ID: <YnF+lbxJXiQMbS/a@fuller.cnet> (raw)
In-Reply-To: <87zgk7x93h.ffs@tglx>

On Wed, Apr 27, 2022 at 09:45:54AM +0200, Thomas Gleixner wrote:
> On Tue, Mar 15 2022 at 12:31, Marcelo Tosatti wrote:
> 
> $Subject does not qualify as a parseable sentence.
> 
> > This avoids processing of TIF_TASK_ISOL, when returning to userspace,
> > for tasks which do not have task isolation configured.
> 
> That's how kernel development works, right:
> 
>    1) Add half baken stuff
>    2) Apply duct tape on top
> 
> You know exactly, that this is _not_ the way it works.
> 
> This whole thing is half thought out tinkerware with [ill|un]defined
> semantics.

It seems to be inline with the remaining TIF_ bits:

                if (ti_work & _TIF_NOTIFY_RESUME)
                        tracehook_notify_resume(regs);

+               if (ti_work & _TIF_TASK_ISOL)
+                       task_isol_exit_to_user_mode();
+


And there is even:

--- a/arch/x86/include/asm/thread_info.h
+++ b/arch/x86/include/asm/thread_info.h
@@ -83,6 +83,7 @@ struct thread_info {
 #define TIF_NEED_RESCHED       3       /* rescheduling necessary */
 #define TIF_SINGLESTEP         4       /* reenable singlestep on user return*/
 #define TIF_SSBD               5       /* Speculative store bypass disable */
+#define TIF_TASK_ISOL          6       /* task isolation work pending */
 #define TIF_SPEC_IB            9       /* Indirect branch speculation mitigation */
 #define TIF_SPEC_L1D_FLUSH     10      /* Flush L1D on mm switches (processes) */
 #define TIF_USER_RETURN_NOTIFY 11      /* notify kernel of userspace return */

So the purpose of TIF_TASK_ISOL is to condense in a single bit the 
question: "is there task isolation work pending?"

By looking at the code, we see the sites where this bit is set are:

1) Task isolation configuration.

@@ -251,6 +257,11 @@ static int cfg_feat_quiesce_set(unsigned
        info->quiesce_mask = i_qctrl->quiesce_mask;
        info->oneshot_mask = i_qctrl->quiesce_oneshot_mask;
        info->conf_mask |= ISOL_F_QUIESCE;
+
+       if ((info->active_mask & ISOL_F_QUIESCE) &&
+           (info->quiesce_mask & ISOL_F_QUIESCE_VMSTATS))
+               set_thread_flag(TIF_TASK_ISOL);
+
        ret = 0;

2) copy_process (clone / fork):

@@ -303,6 +314,7 @@ int __copy_task_isol(struct task_struct
                new_info->active_mask = info->active_mask;

        tsk->task_isol_info = new_info;
+       set_ti_thread_flag(task_thread_info(tsk), TIF_TASK_ISOL);

        return 0;
 }

3) task isolation activation:

@@ -330,6 +342,10 @@ int prctl_task_isol_activate_set(unsigne
        info->active_mask = active_mask;
        ret = 0;

+       if ((info->active_mask & ISOL_F_QUIESCE) &&
+           (info->quiesce_mask & ISOL_F_QUIESCE_VMSTATS))
+               set_thread_flag(TIF_TASK_ISOL);
+
 out:
        return ret;

Would you prefer an explanation, in words, when these bits are set, when
they are cleared? 

So the meaning is:

+#define TIF_TASK_ISOL          6       /* task isolation work pending */

And the definition is "task isolation work" depends on what task
isolation features are implemented.

(honestly, seem pretty clear to me, but willing to improve...).


  reply	other threads:[~2022-05-03 21:02 UTC|newest]

Thread overview: 44+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-03-15 15:31 [patch v12 00/13] extensible prctl task isolation interface and vmstat sync Marcelo Tosatti
2022-03-15 15:31 ` [patch v12 01/13] s390: add support for TIF_TASK_ISOL Marcelo Tosatti
2022-03-15 15:31 ` [patch v12 02/13] x86: " Marcelo Tosatti
2022-03-15 15:31 ` [patch v12 03/13] add basic task isolation prctl interface Marcelo Tosatti
2022-04-25 22:23   ` Thomas Gleixner
2022-03-15 15:31 ` [patch v12 04/13] add prctl task isolation prctl docs and samples Marcelo Tosatti
2022-04-26  0:15   ` Thomas Gleixner
2022-03-15 15:31 ` [patch v12 05/13] task isolation: sync vmstats on return to userspace Marcelo Tosatti
2022-04-25 23:06   ` Thomas Gleixner
2022-04-27  6:56   ` Thomas Gleixner
2022-03-15 15:31 ` [patch v12 06/13] procfs: add per-pid task isolation state Marcelo Tosatti
2022-04-25 23:27   ` Thomas Gleixner
2022-03-15 15:31 ` [patch v12 07/13] task isolation: sync vmstats conditional on changes Marcelo Tosatti
2022-03-17 14:51   ` Frederic Weisbecker
2022-04-27  8:03   ` Thomas Gleixner
2022-03-15 15:31 ` [patch v12 08/13] task isolation: enable return to userspace processing Marcelo Tosatti
2022-03-15 15:31 ` [patch v12 09/13] task isolation: add preempt notifier to sync per-CPU vmstat dirty info to thread info Marcelo Tosatti
2022-03-16  2:41   ` Oscar Shiang
2022-04-27  7:11   ` Thomas Gleixner
2022-04-27 12:09     ` Thomas Gleixner
2022-05-04 16:32       ` Marcelo Tosatti
2022-05-04 17:39         ` Thomas Gleixner
2022-03-15 15:31 ` [patch v12 10/13] KVM: x86: process isolation work from VM-entry code path Marcelo Tosatti
2022-03-15 15:31 ` [patch v12 11/13] mm: vmstat: move need_update Marcelo Tosatti
2022-03-15 15:31 ` [patch v12 12/13] mm: vmstat_refresh: avoid queueing work item if cpu stats are clean Marcelo Tosatti
2022-04-27  7:23   ` Thomas Gleixner
2022-05-03 19:17     ` Marcelo Tosatti
2022-03-15 15:31 ` [patch v12 13/13] task isolation: only TIF_TASK_ISOL if task isolation is enabled Marcelo Tosatti
2022-04-27  7:45   ` Thomas Gleixner
2022-05-03 19:12     ` Marcelo Tosatti [this message]
2022-05-04 13:03       ` Thomas Gleixner
2022-03-17 15:08 ` [patch v12 00/13] extensible prctl task isolation interface and vmstat sync Frederic Weisbecker
2022-04-25 16:29   ` Marcelo Tosatti
2022-04-25 21:12     ` Thomas Gleixner
2022-05-03 18:57       ` Marcelo Tosatti
2022-04-27  9:19 ` Christoph Lameter
2022-05-03 18:57   ` Marcelo Tosatti
2022-05-04 13:20     ` Thomas Gleixner
2022-05-04 18:56       ` Marcelo Tosatti
2022-05-04 20:15         ` Thomas Gleixner
2022-05-05 16:52           ` Marcelo Tosatti
2022-06-01 16:14             ` Marcelo Tosatti
2022-05-04 17:01 ` Tim Chen
2022-05-04 20:08   ` 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=YnF+lbxJXiQMbS/a@fuller.cnet \
    --to=mtosatti@redhat.com \
    --cc=abelits@belits.com \
    --cc=bristot@redhat.com \
    --cc=cl@linux.com \
    --cc=frederic@kernel.org \
    --cc=juri.lelli@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=nilal@redhat.com \
    --cc=nsaenzju@redhat.com \
    --cc=oscar0225@livemail.tw \
    --cc=peterx@redhat.com \
    --cc=peterz@infradead.org \
    --cc=tglx@linutronix.de \
    /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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox