Skip to contentSkip to conversation

rcu: Add comment documenting how rcu_seq_snap works

PATCH Superseded Joel Fernandes (Google) 2 participants 2 messages May 12, 2018 RCU

v1 of 2 Superseded Latest is v2 →

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 rcu: Add comment documenting how rcu_seq_snap works

rcu_seq_snap may be tricky for someone looking at it for the first time. Lets document how it works with an example to make it easier.

---
 kernel/rcu/rcu.h | 23 ++++++++++++++++++++++-
 1 file changed, 22 insertions(+), 1 deletion(-)
kernel/rcu/rcu.h +22 -1
index 003671825d62..004ace3d22c2 100644
@@ -91,7 +91,28 @@ static inline void rcu_seq_end(unsigned long *sp)
 	WRITE_ONCE(*sp, rcu_seq_endval(sp));
 }
 
-/* Take a snapshot of the update side's sequence number. */
+/*
+ * Take a snapshot of the update side's sequence number.
+ *
+ * This function predicts what the grace period number will be the next
+ * time an RCU callback will be executed, given the current grace period's
+ * number. This can be gp+1 if RCU is idle, or gp+2 if a grace period is
+ * already in progress.
+ *
+ * We do this with a single addition and masking.
+ * For example, if RCU_SEQ_STATE_MASK=1 and the least significant bit (LSB) of
+ * the seq is used to track if a GP is in progress or not, its sufficient if we
+ * add (2+1) and mask with ~1. Lets see why with an example:
+ *
+ * Say the current seq is 6 which is 0x110 (gp is 3 and state bit is 0).
+ * To get the next GP number, we have to atleast add 0x10 to this (0x1 << 1) to
+ * account for the state bit. However, if the current seq is 7 (GP num is 3
+ * and state bit is 1), then it means the current grace period is already
+ * in progress so the next the callback will run is at gp+2. To account for
+ * the extra +1, we just overflow the LSB by adding another 0x1 and masking
+ * with ~0x1. Incase no GP was in progress (RCU is idle), then the adding
+ * by 0x1 and masking will have no effect. This is calculated as below.
+ */
 static inline unsigned long rcu_seq_snap(unsigned long *sp)
 {
 	unsigned long s;
Signature · 1 line
2.17.0.441.gb46fe60e1d-goog
#1 lore raw

PATCH 1/1 rcu: Add comment documenting how rcu_seq_snap works

Quoted 21 lines · Joel Fernandes (Google)

rcu_seq_snap may be tricky for someone looking at it for the first time. Lets document how it works with an example to make it easier.

---
 kernel/rcu/rcu.h | 23 ++++++++++++++++++++++-
 1 file changed, 22 insertions(+), 1 deletion(-)
kernel/rcu/rcu.h +4 -1
index 003671825d62..004ace3d22c2 100644
@@ -91,7 +91,28 @@ static inline void rcu_seq_end(unsigned long *sp)
 	WRITE_ONCE(*sp, rcu_seq_endval(sp));
 }
 
-/* Take a snapshot of the update side's sequence number. */
+/*
+ * Take a snapshot of the update side's sequence number.
+ *
+ * This function predicts what the grace period number will be the next

kernel/rcu/rcu.h static inline void rcu_seq_end(unsigned long *sp) · line 107

Quoted 10 lines
kernel/rcu/rcu.h +10 -0
+ * time an RCU callback will be executed, given the current grace period's
+ * number. This can be gp+1 if RCU is idle, or gp+2 if a grace period is
+ * already in progress.
+ *
+ * We do this with a single addition and masking.
+ * For example, if RCU_SEQ_STATE_MASK=1 and the least significant bit (LSB) of
+ * the seq is used to track if a GP is in progress or not, its sufficient if we
+ * add (2+1) and mask with ~1. Lets see why with an example:
+ *
+ * Say the current seq is 6 which is 0x110 (gp is 3 and state bit is 0).
                                        0b110
                                   or   0x6

kernel/rcu/rcu.h static inline void rcu_seq_end(unsigned long *sp) · line 108

Quoted 1 line
kernel/rcu/rcu.h +1 -0
+ * To get the next GP number, we have to atleast add 0x10 to this (0x1 << 1) to

at least add 0b10

kernel/rcu/rcu.h static inline void rcu_seq_end(unsigned long *sp) · line 111

Quoted 3 lines
kernel/rcu/rcu.h +3 -0
+ * account for the state bit. However, if the current seq is 7 (GP num is 3
+ * and state bit is 1), then it means the current grace period is already
+ * in progress so the next the callback will run is at gp+2. To account for

so the next time? the callback will run

kernel/rcu/rcu.h static inline void rcu_seq_end(unsigned long *sp) · line 113

Quoted 2 lines
kernel/rcu/rcu.h +2 -0
+ * the extra +1, we just overflow the LSB by adding another 0x1 and masking
+ * with ~0x1. Incase no GP was in progress (RCU is idle), then the adding

In case

Quoted 6 lines
kernel/rcu/rcu.h +2 -0
+ * by 0x1 and masking will have no effect. This is calculated as below.
+ */
 static inline unsigned long rcu_seq_snap(unsigned long *sp)
 {
 	unsigned long s;
Signature · 1 line
~Randy