rcu/tree: Default jiffies_to_sched_qs to jiffies_till_sched_qs
PATCH 1/1 rcu/tree: Default jiffies_to_sched_qs to jiffies_till_sched_qs
Current code does not call adjust_jiffies_till_sched_qs(), if jiffies_till_sched_qs is specified. For the case, where jiffies_till_first_fqs and jiffies_till_next_fqs are default, jiffies_to_sched_qs won't be a correct adjustment of jiffies_till_sched_qs.
--- kernel/rcu/tree.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-)
kernel/rcu/tree.c +1 -2
@@ -3259,8 +3259,7 @@ static void __init rcu_init_geometry(void) jiffies_till_first_fqs = d;if (jiffies_till_next_fqs ==ULONG_MAX ) jiffies_till_next_fqs = d; -if (jiffies_till_sched_qs ==ULONG_MAX ) -adjust_jiffies_till_sched_qs (); +adjust_jiffies_till_sched_qs ();/* If the compile-time values are accurate, just leave. */ if (rcu_fanout_leaf == RCU_FANOUT_LEAF &&
Signature · 2 lines
QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum, hosted by The Linux Foundation
A message that is not in the archive (sent off-list or not yet received): [email protected]
This time keeping the CC list...
Thanx, Paul
Quoted 61 lines · Paul E. McKenney
Quoted 7 lines · Neeraj Upadhyay
Current code does not call
adjust_jiffies_till_sched_qs(), ifjiffies_till_sched_qsis specified. For the case, wherejiffies_till_first_fqsandjiffies_till_next_fqsare default,jiffies_to_sched_qswon't be a correct adjustment ofjiffies_till_sched_qs.Signed-off-by Neeraj Upadhyay <[email protected]>Good catch! Queued and pushed. Please see below for updated commit log. On future patches, could you please first describe the problem and consequences, then what the fix is? This approach makes it much easier for people later on who will be trying to figure out what is going on, and who might or might not have much understanding of RCU. (For example, they might be doing a bisection or some such.)
Not a big deal, as I can touch this up, but a good habit to get into.
And no, rcutorture currently does not specify non-default values for
jiffies_till_sched_qs. Which should probably be fixed. I could makercu_implicit_dynticks_qs()complain ifjiffies_to_sched_qsis zero, but that feels a bit hacky and specific. :-/Thanx, Paul
------------------------------------------------------------------------
commit
ee474b85fa0815be940ed89a91e0d84a110a0a92Author: Neeraj Upadhyay <[email protected]> Date: Mon Mar 11 15:16:11 2019 +0530rcu: Default
jiffies_to_sched_qstojiffies_till_sched_qsThe current code only calls
adjust_jiffies_till_sched_qs()ifjiffies_till_sched_qsis left at its default value, so when thejiffies_till_sched_qskernel-boot parameter actually is specified,jiffies_to_sched_qswill be left with the value zero, which will result in useless slowdowns ofcond_resched(). This commit therefore changesrcu_init_geometry()to unconditionally invokeadjust_jiffies_till_sched_qs(), which ensures thatjiffies_to_sched_qswill be initialized in all cases, thus maintaining goodcond_resched()performance.Signed-off-by Neeraj Upadhyay <[email protected]>Signed-off-by Paul E. McKenney <[email protected]>kernel/rcu/tree.c +1
-2@@ -3259,8 +3259,7 @@ static void __init rcu_init_geometry(void) jiffies_till_first_fqs = d;if (jiffies_till_next_fqs ==ULONG_MAX ) jiffies_till_next_fqs = d; -if (jiffies_till_sched_qs ==ULONG_MAX ) -adjust_jiffies_till_sched_qs (); +adjust_jiffies_till_sched_qs ();/* If the compile-time values are accurate, just leave. */ if (rcu_fanout_leaf == RCU_FANOUT_LEAF &&
Quoted 64 lines · Paul E. McKenney
This time keeping the CC list...
Thanx, Paul
Quoted 59 lines · Paul E. McKenney
Quoted 7 lines · Neeraj Upadhyay
Current code does not call
adjust_jiffies_till_sched_qs(), ifjiffies_till_sched_qsis specified. For the case, wherejiffies_till_first_fqsandjiffies_till_next_fqsare default,jiffies_to_sched_qswon't be a correct adjustment ofjiffies_till_sched_qs.Signed-off-by Neeraj Upadhyay <[email protected]>Good catch! Queued and pushed. Please see below for updated commit log. On future patches, could you please first describe the problem and consequences, then what the fix is? This approach makes it much easier for people later on who will be trying to figure out what is going on, and who might or might not have much understanding of RCU. (For example, they might be doing a bisection or some such.)
Not a big deal, as I can touch this up, but a good habit to get into.
And no, rcutorture currently does not specify non-default values for
jiffies_till_sched_qs. Which should probably be fixed. I could makercu_implicit_dynticks_qs()complain ifjiffies_to_sched_qsis zero, but that feels a bit hacky and specific. :-/Thanx, Paul
------------------------------------------------------------------------
commit
ee474b85fa0815be940ed89a91e0d84a110a0a92Author: Neeraj Upadhyay <[email protected]> Date: Mon Mar 11 15:16:11 2019 +0530rcu: Default
jiffies_to_sched_qstojiffies_till_sched_qsThe current code only calls
adjust_jiffies_till_sched_qs()ifjiffies_till_sched_qsis left at its default value, so when thejiffies_till_sched_qskernel-boot parameter actually is specified,jiffies_to_sched_qswill be left with the value zero, which will result in useless slowdowns ofcond_resched(). This commit therefore changesrcu_init_geometry()to unconditionally invokeadjust_jiffies_till_sched_qs(), which ensures thatjiffies_to_sched_qswill be initialized in all cases, thus maintaining goodcond_resched()performance.Signed-off-by Neeraj Upadhyay <[email protected]>Signed-off-by Paul E. McKenney <[email protected]>kernel/rcu/tree.c +1
-2@@ -3259,8 +3259,7 @@ static void __init rcu_init_geometry(void) jiffies_till_first_fqs = d;if (jiffies_till_next_fqs ==ULONG_MAX ) jiffies_till_next_fqs = d; -if (jiffies_till_sched_qs ==ULONG_MAX ) -adjust_jiffies_till_sched_qs (); +adjust_jiffies_till_sched_qs ();
Quoted 3 lines
kernel/rcu/tree.c static void __init rcu_init_geometry(void) · line 3265
Quoted 2 lines
kernel/rcu/tree.c +0
-0/* If the compile-time values are accurate, just leave. */ if (rcu_fanout_leaf == RCU_FANOUT_LEAF &&
Makes sense to me.
Also the comment here needs an update too I think:
static ulong jiffies_to_sched_qs; /* Adjusted version of above if not default */
Seems to me, after your patch jiffies_to_sched_qs will always be an adjusted value of some sort, unless jiffies_till_sched_qs is specified.
Comment should be some thing like this then?
/* Either the above, or an adjusted default version based on
* jiffies_till_{first,next}_fqs if it is not specified */
thanks,
- Joel
Quoted 75 lines · Joel Fernandes
Quoted 64 lines · Paul E. McKenney
This time keeping the CC list...
Thanx, Paul
Quoted 59 lines · Paul E. McKenney
Quoted 7 lines · Neeraj Upadhyay
Current code does not call
adjust_jiffies_till_sched_qs(), ifjiffies_till_sched_qsis specified. For the case, wherejiffies_till_first_fqsandjiffies_till_next_fqsare default,jiffies_to_sched_qswon't be a correct adjustment ofjiffies_till_sched_qs.Signed-off-by Neeraj Upadhyay <[email protected]>Good catch! Queued and pushed. Please see below for updated commit log. On future patches, could you please first describe the problem and consequences, then what the fix is? This approach makes it much easier for people later on who will be trying to figure out what is going on, and who might or might not have much understanding of RCU. (For example, they might be doing a bisection or some such.)
Not a big deal, as I can touch this up, but a good habit to get into.
And no, rcutorture currently does not specify non-default values for
jiffies_till_sched_qs. Which should probably be fixed. I could makercu_implicit_dynticks_qs()complain ifjiffies_to_sched_qsis zero, but that feels a bit hacky and specific. :-/Thanx, Paul
------------------------------------------------------------------------
commit
ee474b85fa0815be940ed89a91e0d84a110a0a92Author: Neeraj Upadhyay <[email protected]> Date: Mon Mar 11 15:16:11 2019 +0530rcu: Default
jiffies_to_sched_qstojiffies_till_sched_qsThe current code only calls
adjust_jiffies_till_sched_qs()ifjiffies_till_sched_qsis left at its default value, so when thejiffies_till_sched_qskernel-boot parameter actually is specified,jiffies_to_sched_qswill be left with the value zero, which will result in useless slowdowns ofcond_resched(). This commit therefore changesrcu_init_geometry()to unconditionally invokeadjust_jiffies_till_sched_qs(), which ensures thatjiffies_to_sched_qswill be initialized in all cases, thus maintaining goodcond_resched()performance.Signed-off-by Neeraj Upadhyay <[email protected]>Signed-off-by Paul E. McKenney <[email protected]>kernel/rcu/tree.c +1
-2@@ -3259,8 +3259,7 @@ static void __init rcu_init_geometry(void) jiffies_till_first_fqs = d;if (jiffies_till_next_fqs ==ULONG_MAX ) jiffies_till_next_fqs = d; -if (jiffies_till_sched_qs ==ULONG_MAX ) -adjust_jiffies_till_sched_qs (); +adjust_jiffies_till_sched_qs ();Quoted 3 lines
kernel/rcu/tree.c static void __init rcu_init_geometry(void) · line 3265
Quoted 2 lines
kernel/rcu/tree.c +0
-0/* If the compile-time values are accurate, just leave. */ if (rcu_fanout_leaf == RCU_FANOUT_LEAF &&Makes sense to me.
Also the comment here needs an update too I think:
static ulong
jiffies_to_sched_qs; /* Adjusted version of above if not default */
Quoted 7 lines
Seems to me, after your patch
jiffies_to_sched_qswill always be an adjusted value of some sort, unlessjiffies_till_sched_qsis specified.Comment should be some thing like this then?
/* Either the above, or an adjusted default version based on
* jiffies_till_{first,next}_fqs if it is not specified */
Good point, but how about the patch below?
Thanx, Paul
------------------------------------------------------------------------
commit b1d5aaf8eff7872a63531f35aa0490f2fc8118d6
Author: Paul E. McKenney <[email protected]> Date: Mon Mar 11 15:45:13 2019 -0700
rcu: Update jiffies_to_sched_qs and adjust_jiffies_till_sched_qs() comments
This commit better documents the jiffies_to_sched_qs default-value strategy used by adjust_jiffies_till_sched_qs()
kernel/rcu/tree.c +2 -1
@@ -400,7 +400,7 @@ static bool rcu_kick_kthreads;*/ static ulong jiffies_till_sched_qs =ULONG_MAX ;module_param (jiffies_till_sched_qs, ulong,0444 ); -static ulong jiffies_to_sched_qs;/* Adjusted version of above if not default */ +static ulong jiffies_to_sched_qs;/* See adjust_jiffies_till_sched_qs(). */ module_param (jiffies_to_sched_qs, ulong,0444 );/* Display only! */ /*
@@ -418,6 +418,7 @@ static void adjust_jiffies_till_sched_qs(void)WRITE_ONCE (jiffies_to_sched_qs, jiffies_till_sched_qs);return ; } +/* Otherwise, set to third fqs scan, but bound below on large system. */ j =READ_ONCE (jiffies_till_first_fqs) +2 *READ_ONCE (jiffies_till_next_fqs);if (j <HZ /10 + nr_cpu_ids /RCU_JIFFIES_FQS_DIV )
Quoted 79 lines · Paul E. McKenney
Quoted 78 lines · Joel Fernandes
Quoted 64 lines · Paul E. McKenney
This time keeping the CC list...
Thanx, Paul
Quoted 59 lines · Paul E. McKenney
Quoted 7 lines · Neeraj Upadhyay
Current code does not call
adjust_jiffies_till_sched_qs(), ifjiffies_till_sched_qsis specified. For the case, wherejiffies_till_first_fqsandjiffies_till_next_fqsare default,jiffies_to_sched_qswon't be a correct adjustment ofjiffies_till_sched_qs.Signed-off-by Neeraj Upadhyay <[email protected]>Good catch! Queued and pushed. Please see below for updated commit log. On future patches, could you please first describe the problem and consequences, then what the fix is? This approach makes it much easier for people later on who will be trying to figure out what is going on, and who might or might not have much understanding of RCU. (For example, they might be doing a bisection or some such.)
Not a big deal, as I can touch this up, but a good habit to get into.
And no, rcutorture currently does not specify non-default values for
jiffies_till_sched_qs. Which should probably be fixed. I could makercu_implicit_dynticks_qs()complain ifjiffies_to_sched_qsis zero, but that feels a bit hacky and specific. :-/Thanx, Paul
------------------------------------------------------------------------
commit
ee474b85fa0815be940ed89a91e0d84a110a0a92Author: Neeraj Upadhyay <[email protected]> Date: Mon Mar 11 15:16:11 2019 +0530rcu: Default
jiffies_to_sched_qstojiffies_till_sched_qsThe current code only calls
adjust_jiffies_till_sched_qs()ifjiffies_till_sched_qsis left at its default value, so when thejiffies_till_sched_qskernel-boot parameter actually is specified,jiffies_to_sched_qswill be left with the value zero, which will result in useless slowdowns ofcond_resched(). This commit therefore changesrcu_init_geometry()to unconditionally invokeadjust_jiffies_till_sched_qs(), which ensures thatjiffies_to_sched_qswill be initialized in all cases, thus maintaining goodcond_resched()performance.Signed-off-by Neeraj Upadhyay <[email protected]>Signed-off-by Paul E. McKenney <[email protected]>kernel/rcu/tree.c +1
-2@@ -3259,8 +3259,7 @@ static void __init rcu_init_geometry(void) jiffies_till_first_fqs = d;if (jiffies_till_next_fqs ==ULONG_MAX ) jiffies_till_next_fqs = d; -if (jiffies_till_sched_qs ==ULONG_MAX ) -adjust_jiffies_till_sched_qs (); +adjust_jiffies_till_sched_qs ();Quoted 3 lines
kernel/rcu/tree.c static void __init rcu_init_geometry(void) · line 3265
Quoted 2 lines
kernel/rcu/tree.c +0
-0/* If the compile-time values are accurate, just leave. */ if (rcu_fanout_leaf == RCU_FANOUT_LEAF &&Makes sense to me.
Also the comment here needs an update too I think:
static ulong
jiffies_to_sched_qs; /* Adjusted version of above if not default */Seems to me, after your patch
jiffies_to_sched_qswill always be an adjusted value of some sort, unlessjiffies_till_sched_qsis specified.
Quoted 6 lines
Quoted 4 lines
Comment should be some thing like this then?
/* Either the above, or an adjusted default version based on
* jiffies_till_{first,next}_fqs if it is not specified */Good point, but how about the patch below?
Yes, your comment is better :-) thanks,
- Joel
Quoted 38 lines
Thanx, Paul
------------------------------------------------------------------------
commit
b1d5aaf8eff7872a63531f35aa0490f2fc8118d6Author: Paul E. McKenney <[email protected]> Date: Mon Mar 11 15:45:13 2019 -0700rcu: Update
jiffies_to_sched_qsandadjust_jiffies_till_sched_qs()commentsThis commit better documents the
jiffies_to_sched_qsdefault-value strategy used byadjust_jiffies_till_sched_qs()Reported-by Joel Fernandes <[email protected]>Signed-off-by Paul E. McKenney <[email protected]>kernel/rcu/tree.c +2
-1@@ -400,7 +400,7 @@ static bool rcu_kick_kthreads;*/ static ulong jiffies_till_sched_qs =ULONG_MAX ;module_param (jiffies_till_sched_qs, ulong,0444 ); -static ulong jiffies_to_sched_qs;/* Adjusted version of above if not default */ +static ulong jiffies_to_sched_qs;/* See adjust_jiffies_till_sched_qs(). */ module_param (jiffies_to_sched_qs, ulong,0444 );/* Display only! */ /* @@ -418,6 +418,7 @@ static void adjust_jiffies_till_sched_qs(void)WRITE_ONCE (jiffies_to_sched_qs, jiffies_till_sched_qs);return ; } +/* Otherwise, set to third fqs scan, but bound below on large system. */ j =READ_ONCE (jiffies_till_first_fqs) +2 *READ_ONCE (jiffies_till_next_fqs);if (j <HZ /10 + nr_cpu_ids /RCU_JIFFIES_FQS_DIV )