2020-04-19 22:01:05

by Wei Yang

[permalink] [raw]
Subject: [Patch v2] rcu: simplify the calculation of rcu_state.ncpus

There is only 1 bit set in mask, which means the difference between
oldmask and the new one would be at the position where the bit is set in
mask.

Based on this knowledge, rcu_state.ncpus could be calculated by checking
whether mask is already set in rnp->expmaskinitnext.

Signed-off-by: Wei Yang <[email protected]>

---
v2:
* trivial adjust based on Paul's suggestion.
---
kernel/rcu/tree.c | 9 +++------
1 file changed, 3 insertions(+), 6 deletions(-)

diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c
index f288477ee1c2..6d39485f7f51 100644
--- a/kernel/rcu/tree.c
+++ b/kernel/rcu/tree.c
@@ -3732,10 +3732,9 @@ void rcu_cpu_starting(unsigned int cpu)
{
unsigned long flags;
unsigned long mask;
- int nbits;
- unsigned long oldmask;
struct rcu_data *rdp;
struct rcu_node *rnp;
+ bool newcpu;

if (per_cpu(rcu_cpu_started, cpu))
return;
@@ -3747,12 +3746,10 @@ void rcu_cpu_starting(unsigned int cpu)
mask = rdp->grpmask;
raw_spin_lock_irqsave_rcu_node(rnp, flags);
WRITE_ONCE(rnp->qsmaskinitnext, rnp->qsmaskinitnext | mask);
- oldmask = rnp->expmaskinitnext;
+ newcpu = !(rnp->expmaskinitnext & mask);
rnp->expmaskinitnext |= mask;
- oldmask ^= rnp->expmaskinitnext;
- nbits = bitmap_weight(&oldmask, BITS_PER_LONG);
/* Allow lockless access for expedited grace periods. */
- smp_store_release(&rcu_state.ncpus, rcu_state.ncpus + nbits); /* ^^^ */
+ smp_store_release(&rcu_state.ncpus, rcu_state.ncpus + newcpu); /* ^^^ */
ASSERT_EXCLUSIVE_WRITER(rcu_state.ncpus);
rcu_gpnum_ovf(rnp, rdp); /* Offline-induced counter wrap? */
rdp->rcu_onl_gp_seq = READ_ONCE(rcu_state.gp_seq);
--
2.23.0


2020-04-20 01:44:56

by Paul E. McKenney

[permalink] [raw]
Subject: Re: [Patch v2] rcu: simplify the calculation of rcu_state.ncpus

On Sun, Apr 19, 2020 at 09:57:15PM +0000, Wei Yang wrote:
> There is only 1 bit set in mask, which means the difference between
> oldmask and the new one would be at the position where the bit is set in
> mask.
>
> Based on this knowledge, rcu_state.ncpus could be calculated by checking
> whether mask is already set in rnp->expmaskinitnext.
>
> Signed-off-by: Wei Yang <[email protected]>

Queued, thank you!

I updated the commit log as shown below, so please let me know if I
messed something up.

Thanx, Paul

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

commit 2ff1b8268456b1a476f8b79672c87d32d4f59024
Author: Wei Yang <[email protected]>
Date: Sun Apr 19 21:57:15 2020 +0000

rcu: Simplify the calculation of rcu_state.ncpus

There is only 1 bit set in mask, which means that the only difference
between oldmask and the new one will be at the position where the bit is
set in mask. This commit therefore updates rcu_state.ncpus by checking
whether the bit in mask is already set in rnp->expmaskinitnext.

Signed-off-by: Wei Yang <[email protected]>
Signed-off-by: Paul E. McKenney <[email protected]>

diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c
index f288477..6d39485 100644
--- a/kernel/rcu/tree.c
+++ b/kernel/rcu/tree.c
@@ -3732,10 +3732,9 @@ void rcu_cpu_starting(unsigned int cpu)
{
unsigned long flags;
unsigned long mask;
- int nbits;
- unsigned long oldmask;
struct rcu_data *rdp;
struct rcu_node *rnp;
+ bool newcpu;

if (per_cpu(rcu_cpu_started, cpu))
return;
@@ -3747,12 +3746,10 @@ void rcu_cpu_starting(unsigned int cpu)
mask = rdp->grpmask;
raw_spin_lock_irqsave_rcu_node(rnp, flags);
WRITE_ONCE(rnp->qsmaskinitnext, rnp->qsmaskinitnext | mask);
- oldmask = rnp->expmaskinitnext;
+ newcpu = !(rnp->expmaskinitnext & mask);
rnp->expmaskinitnext |= mask;
- oldmask ^= rnp->expmaskinitnext;
- nbits = bitmap_weight(&oldmask, BITS_PER_LONG);
/* Allow lockless access for expedited grace periods. */
- smp_store_release(&rcu_state.ncpus, rcu_state.ncpus + nbits); /* ^^^ */
+ smp_store_release(&rcu_state.ncpus, rcu_state.ncpus + newcpu); /* ^^^ */
ASSERT_EXCLUSIVE_WRITER(rcu_state.ncpus);
rcu_gpnum_ovf(rnp, rdp); /* Offline-induced counter wrap? */
rdp->rcu_onl_gp_seq = READ_ONCE(rcu_state.gp_seq);

2020-04-20 21:32:18

by Wei Yang

[permalink] [raw]
Subject: Re: [Patch v2] rcu: simplify the calculation of rcu_state.ncpus

On Sun, Apr 19, 2020 at 06:41:37PM -0700, Paul E. McKenney wrote:
>On Sun, Apr 19, 2020 at 09:57:15PM +0000, Wei Yang wrote:
>> There is only 1 bit set in mask, which means the difference between
>> oldmask and the new one would be at the position where the bit is set in
>> mask.
>>
>> Based on this knowledge, rcu_state.ncpus could be calculated by checking
>> whether mask is already set in rnp->expmaskinitnext.
>>
>> Signed-off-by: Wei Yang <[email protected]>
>
>Queued, thank you!
>
>I updated the commit log as shown below, so please let me know if I
>messed something up.
>

Looks pretty good.

> Thanx, Paul
>
>------------------------------------------------------------------------
>
>commit 2ff1b8268456b1a476f8b79672c87d32d4f59024
>Author: Wei Yang <[email protected]>
>Date: Sun Apr 19 21:57:15 2020 +0000
>
> rcu: Simplify the calculation of rcu_state.ncpus
>
> There is only 1 bit set in mask, which means that the only difference
> between oldmask and the new one will be at the position where the bit is
> set in mask. This commit therefore updates rcu_state.ncpus by checking
> whether the bit in mask is already set in rnp->expmaskinitnext.
>
> Signed-off-by: Wei Yang <[email protected]>
> Signed-off-by: Paul E. McKenney <[email protected]>
>
>diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c
>index f288477..6d39485 100644
>--- a/kernel/rcu/tree.c
>+++ b/kernel/rcu/tree.c
>@@ -3732,10 +3732,9 @@ void rcu_cpu_starting(unsigned int cpu)
> {
> unsigned long flags;
> unsigned long mask;
>- int nbits;
>- unsigned long oldmask;
> struct rcu_data *rdp;
> struct rcu_node *rnp;
>+ bool newcpu;
>
> if (per_cpu(rcu_cpu_started, cpu))
> return;
>@@ -3747,12 +3746,10 @@ void rcu_cpu_starting(unsigned int cpu)
> mask = rdp->grpmask;
> raw_spin_lock_irqsave_rcu_node(rnp, flags);
> WRITE_ONCE(rnp->qsmaskinitnext, rnp->qsmaskinitnext | mask);
>- oldmask = rnp->expmaskinitnext;
>+ newcpu = !(rnp->expmaskinitnext & mask);
> rnp->expmaskinitnext |= mask;
>- oldmask ^= rnp->expmaskinitnext;
>- nbits = bitmap_weight(&oldmask, BITS_PER_LONG);
> /* Allow lockless access for expedited grace periods. */
>- smp_store_release(&rcu_state.ncpus, rcu_state.ncpus + nbits); /* ^^^ */
>+ smp_store_release(&rcu_state.ncpus, rcu_state.ncpus + newcpu); /* ^^^ */
> ASSERT_EXCLUSIVE_WRITER(rcu_state.ncpus);
> rcu_gpnum_ovf(rnp, rdp); /* Offline-induced counter wrap? */
> rdp->rcu_onl_gp_seq = READ_ONCE(rcu_state.gp_seq);

--
Wei Yang
Help you, Help me