[v3] rcu: Further comment and explain the state space of GP sequences

Message ID Y85868Pei9VutSiW@lothringen
State New
Headers
Series [v3] rcu: Further comment and explain the state space of GP sequences |

Commit Message

Frederic Weisbecker Jan. 23, 2023, 12:26 p.m. UTC
  On Thu, Jan 19, 2023 at 05:06:27PM +0000, Joel Fernandes wrote:
> On Thu, Jan 19, 2023 at 2:15 PM Frederic Weisbecker <frederic@kernel.org> wrote:
> >
> > On Thu, Jan 19, 2023 at 03:11:35PM +0100, Frederic Weisbecker wrote:
> > > The state space of the GP sequence number isn't documented and the
> > > definitions of its special values are scattered. Try to gather some
> > > common knowledge near the GP seq headers.
> > >
> > > Signed-off-by: Frederic Weisbecker <frederic@kernel.org>
> > > ---
> > >  kernel/rcu/rcu.h | 33 +++++++++++++++++++++++++++++++++
> > >  1 file changed, 33 insertions(+)
> > >
> > > diff --git a/kernel/rcu/rcu.h b/kernel/rcu/rcu.h
> > > index 115616ac3bfa..fb95de039596 100644
> > > --- a/kernel/rcu/rcu.h
> > > +++ b/kernel/rcu/rcu.h
> > > @@ -14,6 +14,39 @@
> > >
> > >  /*
> > >   * Grace-period counter management.
> > > + *
> > > + * The two lowest significant bits gather the control flags.
> > > + * The higher bits form the RCU sequence counter.
> > > + *
> > > + * About the control flags, a common value of 0 means that no GP is in progress.
> > > + * A value of 1 means that a grace period has started and is in progress. When
> > > + * the grace period completes, the control flags are reset to 0 and the sequence
> > > + * counter is incremented.
> > > + *
> > > + * However some specific RCU usages make use of custom values.
> > > + *
> > > + * SRCU special control values:
> > > + *
> > > + *   SRCU_SNP_INIT_SEQ       :       Invalid/init value set when SRCU node
> > > + *                                                   is initialized.
> > > + *
> > > + *   SRCU_STATE_IDLE         :       No SRCU gp is in progress
> > > + *
> > > + *   SRCU_STATE_SCAN1        :       State set by rcu_seq_start(). Indicates
> > > + *                                                           we are scanning the inactive readers
> > > + *                                                           index.
> 
> The term "inactive reader" is confusing. The readers can very much be
> active during scans. During a scan stage, there might be a reader on
> any of the 2 indexes that can be right in the middle of their critical
> section (and we don't know which index because they could have got
> preempted, right after sampling idx). Maybe "inactive slot" is a
> better term? And define "inactive slot" as the slot which is no longer
> going to be sampled by new readers.

That's why I used "inactive readers index".
I guess I should have written "inactive readers' index" to disambiguate
the fact that inactive refers to "index" and not "readers" but I almost
never observe plural genitive written that way these days.

As for the gory details of "inactive" being actually "bound to become
inactive", I'm not sure that belongs here but here is what I can do:

---
From: Frederic Weisbecker <frederic@kernel.org>
Date: Thu, 19 Jan 2023 14:29:34 +0100
Subject: [PATCH] rcu: Further comment and explain the state space of GP
 sequences

The state space of the GP sequence number isn't documented and the
definitions of its special values are scattered. Try to gather some
common knowledge near the GP seq headers.

Signed-off-by: Frederic Weisbecker <frederic@kernel.org>
---
 kernel/rcu/rcu.h | 37 +++++++++++++++++++++++++++++++++++++
 1 file changed, 37 insertions(+)
  

Comments

Joel Fernandes Jan. 23, 2023, 2:07 p.m. UTC | #1
> On Jan 23, 2023, at 7:26 AM, Frederic Weisbecker <frederic@kernel.org> wrote:
> 
> On Thu, Jan 19, 2023 at 05:06:27PM +0000, Joel Fernandes wrote:
>>> On Thu, Jan 19, 2023 at 2:15 PM Frederic Weisbecker <frederic@kernel.org> wrote:
>>> 
>>> On Thu, Jan 19, 2023 at 03:11:35PM +0100, Frederic Weisbecker wrote:
>>>> The state space of the GP sequence number isn't documented and the
>>>> definitions of its special values are scattered. Try to gather some
>>>> common knowledge near the GP seq headers.
>>>> 
>>>> Signed-off-by: Frederic Weisbecker <frederic@kernel.org>
>>>> ---
>>>> kernel/rcu/rcu.h | 33 +++++++++++++++++++++++++++++++++
>>>> 1 file changed, 33 insertions(+)
>>>> 
>>>> diff --git a/kernel/rcu/rcu.h b/kernel/rcu/rcu.h
>>>> index 115616ac3bfa..fb95de039596 100644
>>>> --- a/kernel/rcu/rcu.h
>>>> +++ b/kernel/rcu/rcu.h
>>>> @@ -14,6 +14,39 @@
>>>> 
>>>> /*
>>>>  * Grace-period counter management.
>>>> + *
>>>> + * The two lowest significant bits gather the control flags.
>>>> + * The higher bits form the RCU sequence counter.
>>>> + *
>>>> + * About the control flags, a common value of 0 means that no GP is in progress.
>>>> + * A value of 1 means that a grace period has started and is in progress. When
>>>> + * the grace period completes, the control flags are reset to 0 and the sequence
>>>> + * counter is incremented.
>>>> + *
>>>> + * However some specific RCU usages make use of custom values.
>>>> + *
>>>> + * SRCU special control values:
>>>> + *
>>>> + *   SRCU_SNP_INIT_SEQ       :       Invalid/init value set when SRCU node
>>>> + *                                                   is initialized.
>>>> + *
>>>> + *   SRCU_STATE_IDLE         :       No SRCU gp is in progress
>>>> + *
>>>> + *   SRCU_STATE_SCAN1        :       State set by rcu_seq_start(). Indicates
>>>> + *                                                           we are scanning the inactive readers
>>>> + *                                                           index.
>> 
>> The term "inactive reader" is confusing. The readers can very much be
>> active during scans. During a scan stage, there might be a reader on
>> any of the 2 indexes that can be right in the middle of their critical
>> section (and we don't know which index because they could have got
>> preempted, right after sampling idx). Maybe "inactive slot" is a
>> better term? And define "inactive slot" as the slot which is no longer
>> going to be sampled by new readers.
> 
> That's why I used "inactive readers index".
> I guess I should have written "inactive readers' index" to disambiguate
> the fact that inactive refers to "index" and not "readers" but I almost
> never observe plural genitive written that way these days.
> 
> As for the gory details of "inactive" being actually "bound to become
> inactive", I'm not sure that belongs here but here is what I can do:

Sounds good and the below change looks good to me.

Reviewed-by: Joel Fernandes (Google) <joel@joelfernandes.org>

Thanks,

 - Joel



> 
> ---
> From: Frederic Weisbecker <frederic@kernel.org>
> Date: Thu, 19 Jan 2023 14:29:34 +0100
> Subject: [PATCH] rcu: Further comment and explain the state space of GP
> sequences
> 
> The state space of the GP sequence number isn't documented and the
> definitions of its special values are scattered. Try to gather some
> common knowledge near the GP seq headers.
> 
> Signed-off-by: Frederic Weisbecker <frederic@kernel.org>
> ---
> kernel/rcu/rcu.h | 37 +++++++++++++++++++++++++++++++++++++
> 1 file changed, 37 insertions(+)
> 
> diff --git a/kernel/rcu/rcu.h b/kernel/rcu/rcu.h
> index 115616ac3bfa..5be983598b5a 100644
> --- a/kernel/rcu/rcu.h
> +++ b/kernel/rcu/rcu.h
> @@ -14,6 +14,43 @@
> 
> /*
>  * Grace-period counter management.
> + *
> + * The two lowest significant bits gather the control flags.
> + * The higher bits form the RCU sequence counter.
> + *
> + * About the control flags, a common value of 0 means that no GP is in progress.
> + * A value of 1 means that a grace period has started and is in progress. When
> + * the grace period completes, the control flags are reset to 0 and the sequence
> + * counter is incremented.
> + *
> + * However some specific RCU usages make use of custom values.
> + *
> + * SRCU special control values:
> + *
> + *    SRCU_SNP_INIT_SEQ    :    Invalid/init value set when SRCU node
> + *                    is initialized.
> + *
> + *    SRCU_STATE_IDLE        :    No SRCU gp is in progress
> + *
> + *    SRCU_STATE_SCAN1    :    State set by rcu_seq_start(). Indicates
> + *                    we are scanning the readers on the slot
> + *                    defined as inactive (though there might
> + *                    be pending readers there but their number
> + *                    is bound).
> + *
> + *    SRCU_STATE_SCAN2    :    State set manually via rcu_seq_set_state()
> + *                    Indicates we are flipping the readers
> + *                    index and then scanning the readers on the
> + *                    slot newly set as inactive (again there
> + *                    might remain a bound amount of pending
> + *                    readers there).
> + *
> + * RCU polled GP special control value:
> + *
> + *    RCU_GET_STATE_COMPLETED :    State value indicating that a polled GP
> + *                    has completed. It's an absolute value
> + *                    covering both the state and the counter of
> + *                    the GP sequence.
>  */
> 
> #define RCU_SEQ_CTR_SHIFT    2
> -- 
> 2.34.1
>
  
Paul E. McKenney Jan. 23, 2023, 5:30 p.m. UTC | #2
On Mon, Jan 23, 2023 at 09:07:43AM -0500, Joel Fernandes wrote:
> 
> 
> > On Jan 23, 2023, at 7:26 AM, Frederic Weisbecker <frederic@kernel.org> wrote:
> > 
> > On Thu, Jan 19, 2023 at 05:06:27PM +0000, Joel Fernandes wrote:
> >>> On Thu, Jan 19, 2023 at 2:15 PM Frederic Weisbecker <frederic@kernel.org> wrote:
> >>> 
> >>> On Thu, Jan 19, 2023 at 03:11:35PM +0100, Frederic Weisbecker wrote:
> >>>> The state space of the GP sequence number isn't documented and the
> >>>> definitions of its special values are scattered. Try to gather some
> >>>> common knowledge near the GP seq headers.
> >>>> 
> >>>> Signed-off-by: Frederic Weisbecker <frederic@kernel.org>
> >>>> ---
> >>>> kernel/rcu/rcu.h | 33 +++++++++++++++++++++++++++++++++
> >>>> 1 file changed, 33 insertions(+)
> >>>> 
> >>>> diff --git a/kernel/rcu/rcu.h b/kernel/rcu/rcu.h
> >>>> index 115616ac3bfa..fb95de039596 100644
> >>>> --- a/kernel/rcu/rcu.h
> >>>> +++ b/kernel/rcu/rcu.h
> >>>> @@ -14,6 +14,39 @@
> >>>> 
> >>>> /*
> >>>>  * Grace-period counter management.
> >>>> + *
> >>>> + * The two lowest significant bits gather the control flags.
> >>>> + * The higher bits form the RCU sequence counter.
> >>>> + *
> >>>> + * About the control flags, a common value of 0 means that no GP is in progress.
> >>>> + * A value of 1 means that a grace period has started and is in progress. When
> >>>> + * the grace period completes, the control flags are reset to 0 and the sequence
> >>>> + * counter is incremented.
> >>>> + *
> >>>> + * However some specific RCU usages make use of custom values.
> >>>> + *
> >>>> + * SRCU special control values:
> >>>> + *
> >>>> + *   SRCU_SNP_INIT_SEQ       :       Invalid/init value set when SRCU node
> >>>> + *                                                   is initialized.
> >>>> + *
> >>>> + *   SRCU_STATE_IDLE         :       No SRCU gp is in progress
> >>>> + *
> >>>> + *   SRCU_STATE_SCAN1        :       State set by rcu_seq_start(). Indicates
> >>>> + *                                                           we are scanning the inactive readers
> >>>> + *                                                           index.
> >> 
> >> The term "inactive reader" is confusing. The readers can very much be
> >> active during scans. During a scan stage, there might be a reader on
> >> any of the 2 indexes that can be right in the middle of their critical
> >> section (and we don't know which index because they could have got
> >> preempted, right after sampling idx). Maybe "inactive slot" is a
> >> better term? And define "inactive slot" as the slot which is no longer
> >> going to be sampled by new readers.
> > 
> > That's why I used "inactive readers index".
> > I guess I should have written "inactive readers' index" to disambiguate
> > the fact that inactive refers to "index" and not "readers" but I almost
> > never observe plural genitive written that way these days.
> > 
> > As for the gory details of "inactive" being actually "bound to become
> > inactive", I'm not sure that belongs here but here is what I can do:
> 
> Sounds good and the below change looks good to me.
> 
> Reviewed-by: Joel Fernandes (Google) <joel@joelfernandes.org>

Thank you both!

I queued the following wordsmithed version, so please let me know if I
messed anything up.

							Thanx, Paul

------------------------------------------------------------------------

commit f31701817185b58a6095b74180d1a5b05369f3c3
Author: Frederic Weisbecker <frederic@kernel.org>
Date:   Thu Jan 19 14:29:34 2023 +0100

    rcu: Further comment and explain the state space of GP sequences
    
    The state space of the GP sequence number isn't documented and the
    definitions of its special values are scattered.  This commit therefore
    gathers some common knowledge near the grace-period sequence-number
    definitions.
    
    Signed-off-by: Frederic Weisbecker <frederic@kernel.org>
    Reviewed-by: Joel Fernandes (Google) <joel@joelfernandes.org>
    Signed-off-by: Paul E. McKenney <paulmck@kernel.org>

diff --git a/kernel/rcu/rcu.h b/kernel/rcu/rcu.h
index 115616ac3bfa6..a3adcf9a9919b 100644
--- a/kernel/rcu/rcu.h
+++ b/kernel/rcu/rcu.h
@@ -14,6 +14,43 @@
 
 /*
  * Grace-period counter management.
+ *
+ * The two least significant bits contain the control flags.
+ * The most significant bits contain the grace-period sequence counter.
+ *
+ * When both control flags are zero, no grace period is in progress.
+ * When either bit is non-zero, a grace period has started and is in
+ * progress. When the grace period completes, the control flags are reset
+ * to 0 and the grace-period sequence counter is incremented.
+ *
+ * However some specific RCU usages make use of custom values.
+ *
+ * SRCU special control values:
+ *
+ *	SRCU_SNP_INIT_SEQ	:	Invalid/init value set when SRCU node
+ *					is initialized.
+ *
+ *	SRCU_STATE_IDLE		:	No SRCU gp is in progress
+ *
+ *	SRCU_STATE_SCAN1	:	State set by rcu_seq_start(). Indicates
+ *					we are scanning the readers on the slot
+ *					defined as inactive (there might well
+ *					be pending readers that will use that
+ *					index, but their number is bounded).
+ *
+ *	SRCU_STATE_SCAN2	:	State set manually via rcu_seq_set_state()
+ *					Indicates we are flipping the readers
+ *					index and then scanning the readers on the
+ *					slot newly designated as inactive (again,
+ *					the number of pending readers that will use
+ *					this inactive index is bounded).
+ *
+ * RCU polled GP special control value:
+ *
+ *	RCU_GET_STATE_COMPLETED :	State value indicating an already-completed
+ *					polled GP has completed.  This value covers
+ *					both the state and the counter of the
+ *					grace-period sequence number.
  */
 
 #define RCU_SEQ_CTR_SHIFT	2
  

Patch

diff --git a/kernel/rcu/rcu.h b/kernel/rcu/rcu.h
index 115616ac3bfa..5be983598b5a 100644
--- a/kernel/rcu/rcu.h
+++ b/kernel/rcu/rcu.h
@@ -14,6 +14,43 @@ 
 
 /*
  * Grace-period counter management.
+ *
+ * The two lowest significant bits gather the control flags.
+ * The higher bits form the RCU sequence counter.
+ *
+ * About the control flags, a common value of 0 means that no GP is in progress.
+ * A value of 1 means that a grace period has started and is in progress. When
+ * the grace period completes, the control flags are reset to 0 and the sequence
+ * counter is incremented.
+ *
+ * However some specific RCU usages make use of custom values.
+ *
+ * SRCU special control values:
+ *
+ *	SRCU_SNP_INIT_SEQ	:	Invalid/init value set when SRCU node
+ *					is initialized.
+ *
+ *	SRCU_STATE_IDLE		:	No SRCU gp is in progress
+ *
+ *	SRCU_STATE_SCAN1	:	State set by rcu_seq_start(). Indicates
+ *					we are scanning the readers on the slot
+ *					defined as inactive (though there might
+ *					be pending readers there but their number
+ *					is bound).
+ *
+ *	SRCU_STATE_SCAN2	:	State set manually via rcu_seq_set_state()
+ *					Indicates we are flipping the readers
+ *					index and then scanning the readers on the
+ *					slot newly set as inactive (again there
+ *					might remain a bound amount of pending
+ *					readers there).
+ *
+ * RCU polled GP special control value:
+ *
+ *	RCU_GET_STATE_COMPLETED :	State value indicating that a polled GP
+ *					has completed. It's an absolute value
+ *					covering both the state and the counter of
+ *					the GP sequence.
  */
 
 #define RCU_SEQ_CTR_SHIFT	2