Skip to contentSkip to conversation

rcu/tree: Remove CONFIG_PREMPT_RCU check in force_qs_rnp

PATCH In review Neeraj Upadhyay 2 participants 3 messages Jun 22, 2020 – Jun 23, 2020 RCU

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 rcu/tree: Remove CONFIG_PREMPT_RCU check in force_qs_rnp

Remove CONFIG_PREMPT_RCU check in force_qs_rnp(). Originally, this check was required to skip executing fqs failsafe for rcu-sched, which was added in commit a77da14ce9af ("rcu:
Yet another fix for preemption and CPU hotplug"). However, this failsafe has been removed, since then. So, cleanup the code to avoid any confusion around the need for boosting, for !CONFIG_PREMPT_RCU.

---
 kernel/rcu/tree.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)
kernel/rcu/tree.c +1 -2
index 6226bfb..57c904b 100644
@@ -2514,8 +2514,7 @@ static void force_qs_rnp(int (*f)(struct rcu_data *rdp))
 		raw_spin_lock_irqsave_rcu_node(rnp, flags);
 		rcu_state.cbovldnext |= !!rnp->cbovldmask;
 		if (rnp->qsmask == 0) {
-			if (!IS_ENABLED(CONFIG_PREEMPT_RCU) ||
-			    rcu_preempt_blocked_readers_cgp(rnp)) {
+			if (rcu_preempt_blocked_readers_cgp(rnp)) {
 				/*
 				 * No point in scanning bits because they
 				 * are all zero.  But we might need to
Signature · 2 lines
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum,
a Linux Foundation Collaborative Project
maintainer
#1 lore raw
Quoted 9 lines · Neeraj Upadhyay

Remove CONFIG_PREMPT_RCU check in force_qs_rnp(). Originally, this check was required to skip executing fqs failsafe for rcu-sched, which was added in commit a77da14ce9af ("rcu:
Yet another fix for preemption and CPU hotplug"). However, this failsafe has been removed, since then. So, cleanup the code to avoid any confusion around the need for boosting, for !CONFIG_PREMPT_RCU.

Good point, there is a !PREEMPT definition of the function rcu_preempt_blocked_readers_cgp() that unconditionally returns zero. And if !PREEMPT kernels, the same things happens in the "if" body as after it, so behavior is not changed.

I have queued and pushed this with an upgraded commit log as shown below.

Thanx, Paul

Quoted 21 lines
---
 kernel/rcu/tree.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)
kernel/rcu/tree.c +1 -2
index 6226bfb..57c904b 100644
@@ -2514,8 +2514,7 @@ static void force_qs_rnp(int (*f)(struct rcu_data *rdp))
 		raw_spin_lock_irqsave_rcu_node(rnp, flags);
 		rcu_state.cbovldnext |= !!rnp->cbovldmask;
 		if (rnp->qsmask == 0) {
-			if (!IS_ENABLED(CONFIG_PREEMPT_RCU) ||
-			    rcu_preempt_blocked_readers_cgp(rnp)) {
+			if (rcu_preempt_blocked_readers_cgp(rnp)) {
 				/*
 				 * No point in scanning bits because they
 				 * are all zero.  But we might need to
Signature · 2 lines
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum,
a Linux Foundation Collaborative Project

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

commit a4600389c35010aef414b89e2817d4a527e751b5

Author: Neeraj Upadhyay <[email protected]>
Date:   Mon Jun 22 23:37:03 2020 +0530

rcu/tree: Remove CONFIG_PREMPT_RCU check in force_qs_rnp()

Originally, the call to rcu_preempt_blocked_readers_cgp() from force_qs_rnp() had to be conditioned on CONFIG_PREEMPT_RCU=y, as in commit a77da14ce9af ("rcu: Yet another fix for preemption and CPU hotplug"). However, there is now a CONFIG_PREEMPT_RCU=n definition of rcu_preempt_blocked_readers_cgp() that unconditionally returns zero, so invoking it is now safe. In addition, the CONFIG_PREEMPT_RCU=n definition of rcu_initiate_boost() simply releases the rcu_node structure's ->lock, which is what happens when the "if" condition evaluates to false.

This commit therefore drops the IS_ENABLED(CONFIG_PREEMPT_RCU) check, so that rcu_initiate_boost() is called only in CONFIG_PREEMPT_RCU=y kernels when there are readers blocking the current grace period. This does not change the behavior, but reduces code-reader confusion by eliminating non-CONFIG_PREEMPT_RCU=y calls to rcu_initiate_boost().

kernel/rcu/tree.c +1 -2
index 6226bfb..57c904b 100644
@@ -2514,8 +2514,7 @@ static void force_qs_rnp(int (*f)(struct rcu_data *rdp))
 		raw_spin_lock_irqsave_rcu_node(rnp, flags);
 		rcu_state.cbovldnext |= !!rnp->cbovldmask;
 		if (rnp->qsmask == 0) {
-			if (!IS_ENABLED(CONFIG_PREEMPT_RCU) ||
-			    rcu_preempt_blocked_readers_cgp(rnp)) {
+			if (rcu_preempt_blocked_readers_cgp(rnp)) {
 				/*
 				 * No point in scanning bits because they
 				 * are all zero.  But we might need to
#2 lore raw

Hi Paul,

Quoted 16 lines · Paul E. McKenney
Quoted 9 lines · Neeraj Upadhyay

Remove CONFIG_PREMPT_RCU check in force_qs_rnp(). Originally, this check was required to skip executing fqs failsafe for rcu-sched, which was added in commit a77da14ce9af ("rcu:
Yet another fix for preemption and CPU hotplug"). However, this failsafe has been removed, since then. So, cleanup the code to avoid any confusion around the need for boosting, for !CONFIG_PREMPT_RCU.

Good point, there is a !PREEMPT definition of the function rcu_preempt_blocked_readers_cgp() that unconditionally returns zero. And if !PREEMPT kernels, the same things happens in the "if" body as after it, so behavior is not changed.

Quoted 5 lines

I have queued and pushed this with an upgraded commit log as shown below.

Thanx, Paul

Thanks! patch looks good to me!

Thanks
Neeraj

Quoted 63 lines
Quoted 21 lines
---
  kernel/rcu/tree.c | 3 +--
  1 file changed, 1 insertion(+), 2 deletions(-)
kernel/rcu/tree.c +1 -2
index 6226bfb..57c904b 100644
@@ -2514,8 +2514,7 @@ static void force_qs_rnp(int (*f)(struct rcu_data *rdp))
  		raw_spin_lock_irqsave_rcu_node(rnp, flags);
  		rcu_state.cbovldnext |= !!rnp->cbovldmask;
  		if (rnp->qsmask == 0) {
-			if (!IS_ENABLED(CONFIG_PREEMPT_RCU) ||
-			    rcu_preempt_blocked_readers_cgp(rnp)) {
+			if (rcu_preempt_blocked_readers_cgp(rnp)) {
  				/*
  				 * No point in scanning bits because they
  				 * are all zero.  But we might need to
Signature · 2 lines
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum,
a Linux Foundation Collaborative Project

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

commit a4600389c35010aef414b89e2817d4a527e751b5

Author: Neeraj Upadhyay <[email protected]>
Date:   Mon Jun 22 23:37:03 2020 +0530

rcu/tree: Remove CONFIG_PREMPT_RCU check in force_qs_rnp()

Originally, the call to rcu_preempt_blocked_readers_cgp() from force_qs_rnp() had to be conditioned on CONFIG_PREEMPT_RCU=y, as in commit a77da14ce9af ("rcu: Yet another fix for preemption and CPU hotplug"). However, there is now a CONFIG_PREEMPT_RCU=n definition of rcu_preempt_blocked_readers_cgp() that unconditionally returns zero, so invoking it is now safe. In addition, the CONFIG_PREEMPT_RCU=n definition of rcu_initiate_boost() simply releases the rcu_node structure's ->lock, which is what happens when the "if" condition evaluates to false.

This commit therefore drops the IS_ENABLED(CONFIG_PREEMPT_RCU) check, so that rcu_initiate_boost() is called only in CONFIG_PREEMPT_RCU=y kernels when there are readers blocking the current grace period. This does not change the behavior, but reduces code-reader confusion by eliminating non-CONFIG_PREEMPT_RCU=y calls to rcu_initiate_boost().

kernel/rcu/tree.c +1 -2
index 6226bfb..57c904b 100644
@@ -2514,8 +2514,7 @@ static void force_qs_rnp(int (*f)(struct rcu_data *rdp))
  		raw_spin_lock_irqsave_rcu_node(rnp, flags);
  		rcu_state.cbovldnext |= !!rnp->cbovldmask;
  		if (rnp->qsmask == 0) {
-			if (!IS_ENABLED(CONFIG_PREEMPT_RCU) ||
-			    rcu_preempt_blocked_readers_cgp(rnp)) {
+			if (rcu_preempt_blocked_readers_cgp(rnp)) {
  				/*
  				 * No point in scanning bits because they
  				 * are all zero.  But we might need to
Signature · 1 line
QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum, hosted by The Linux Foundation