* [PATCH] sched: using dst_rq instead of this_rq during load balance
@ 2012-08-06 8:41 Michael Wang
2012-08-13 17:00 ` [tip:sched/core] " tip-bot for Michael Wang
2012-08-13 17:06 ` tip-bot for Michael Wang
0 siblings, 2 replies; 3+ messages in thread
From: Michael Wang @ 2012-08-06 8:41 UTC (permalink / raw)
To: LKML; +Cc: Peter Zijlstra, Ingo Molnar
From: Michael Wang <wangyun@linux.vnet.ibm.com>
As we already have dst_rq in lb_env, using or changing "this_rq" do not
make sense.
This patch will replace "this_rq" with dst_rq in load_balance, and we
don't need to change "this_rq" while process LBF_SOME_PINNED any more.
Signed-off-by: Michael Wang <wangyun@linux.vnet.ibm.com>
---
kernel/sched/fair.c | 9 ++++-----
1 files changed, 4 insertions(+), 5 deletions(-)
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index d0cc03b..f03b99c 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -4275,7 +4275,7 @@ redo:
goto out_balanced;
}
- BUG_ON(busiest == this_rq);
+ BUG_ON(busiest == env.dst_rq);
schedstat_add(sd, lb_imbalance[idle], env.imbalance);
@@ -4295,7 +4295,7 @@ redo:
more_balance:
local_irq_save(flags);
- double_rq_lock(this_rq, busiest);
+ double_rq_lock(env.dst_rq, busiest);
if (!env.loop)
update_h_load(env.src_cpu);
@@ -4305,7 +4305,7 @@ more_balance:
*/
cur_ld_moved = move_tasks(&env);
ld_moved += cur_ld_moved;
- double_rq_unlock(this_rq, busiest);
+ double_rq_unlock(env.dst_rq, busiest);
local_irq_restore(flags);
if (env.flags & LBF_NEED_BREAK) {
@@ -4341,8 +4341,7 @@ more_balance:
if ((env.flags & LBF_SOME_PINNED) && env.imbalance > 0 &&
lb_iterations++ < max_lb_iterations) {
- this_rq = cpu_rq(env.new_dst_cpu);
- env.dst_rq = this_rq;
+ env.dst_rq = cpu_rq(env.new_dst_cpu);
env.dst_cpu = env.new_dst_cpu;
env.flags &= ~LBF_SOME_PINNED;
env.loop = 0;
--
1.7.4.1
^ permalink raw reply related [flat|nested] 3+ messages in thread* [tip:sched/core] sched: using dst_rq instead of this_rq during load balance
2012-08-06 8:41 [PATCH] sched: using dst_rq instead of this_rq during load balance Michael Wang
@ 2012-08-13 17:00 ` tip-bot for Michael Wang
2012-08-13 17:06 ` tip-bot for Michael Wang
1 sibling, 0 replies; 3+ messages in thread
From: tip-bot for Michael Wang @ 2012-08-13 17:00 UTC (permalink / raw)
To: linux-tip-commits; +Cc: linux-kernel, hpa, mingo, a.p.zijlstra, wangyun, tglx
Commit-ID: 79a7810be652f95e52325ceadc380f461545ae74
Gitweb: http://git.kernel.org/tip/79a7810be652f95e52325ceadc380f461545ae74
Author: Michael Wang <wangyun@linux.vnet.ibm.com>
AuthorDate: Mon, 6 Aug 2012 16:41:59 +0800
Committer: Thomas Gleixner <tglx@linutronix.de>
CommitDate: Mon, 13 Aug 2012 18:51:59 +0200
sched: using dst_rq instead of this_rq during load balance
As we already have dst_rq in lb_env, using or changing "this_rq" do not
make sense.
This patch will replace "this_rq" with dst_rq in load_balance, and we
don't need to change "this_rq" while process LBF_SOME_PINNED any more.
Signed-off-by: Michael Wang <wangyun@linux.vnet.ibm.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/501F8357.3070102@linux.vnet.ibm.com
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
kernel/sched/fair.c | 9 ++++-----
1 files changed, 4 insertions(+), 5 deletions(-)
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 67ac0b1..4f7fcd6 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -4275,7 +4275,7 @@ redo:
goto out_balanced;
}
- BUG_ON(busiest == this_rq);
+ BUG_ON(busiest == env.dst_rq);
schedstat_add(sd, lb_imbalance[idle], env.imbalance);
@@ -4295,7 +4295,7 @@ redo:
more_balance:
local_irq_save(flags);
- double_rq_lock(this_rq, busiest);
+ double_rq_lock(env.rq, busiest);
if (!env.loop)
update_h_load(env.src_cpu);
@@ -4305,7 +4305,7 @@ more_balance:
*/
cur_ld_moved = move_tasks(&env);
ld_moved += cur_ld_moved;
- double_rq_unlock(this_rq, busiest);
+ double_rq_unlock(env.dst_rq, busiest);
local_irq_restore(flags);
if (env.flags & LBF_NEED_BREAK) {
@@ -4341,8 +4341,7 @@ more_balance:
if ((env.flags & LBF_SOME_PINNED) && env.imbalance > 0 &&
lb_iterations++ < max_lb_iterations) {
- this_rq = cpu_rq(env.new_dst_cpu);
- env.dst_rq = this_rq;
+ env.dst_rq = cpu_rq(env.new_dst_cpu);
env.dst_cpu = env.new_dst_cpu;
env.flags &= ~LBF_SOME_PINNED;
env.loop = 0;
^ permalink raw reply related [flat|nested] 3+ messages in thread* [tip:sched/core] sched: using dst_rq instead of this_rq during load balance
2012-08-06 8:41 [PATCH] sched: using dst_rq instead of this_rq during load balance Michael Wang
2012-08-13 17:00 ` [tip:sched/core] " tip-bot for Michael Wang
@ 2012-08-13 17:06 ` tip-bot for Michael Wang
1 sibling, 0 replies; 3+ messages in thread
From: tip-bot for Michael Wang @ 2012-08-13 17:06 UTC (permalink / raw)
To: linux-tip-commits; +Cc: linux-kernel, hpa, mingo, a.p.zijlstra, wangyun, tglx
Commit-ID: 78feefc512a09165627dd534111f651b6c8e605f
Gitweb: http://git.kernel.org/tip/78feefc512a09165627dd534111f651b6c8e605f
Author: Michael Wang <wangyun@linux.vnet.ibm.com>
AuthorDate: Mon, 6 Aug 2012 16:41:59 +0800
Committer: Thomas Gleixner <tglx@linutronix.de>
CommitDate: Mon, 13 Aug 2012 18:58:15 +0200
sched: using dst_rq instead of this_rq during load balance
As we already have dst_rq in lb_env, using or changing "this_rq" do not
make sense.
This patch will replace "this_rq" with dst_rq in load_balance, and we
don't need to change "this_rq" while process LBF_SOME_PINNED any more.
Signed-off-by: Michael Wang <wangyun@linux.vnet.ibm.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/501F8357.3070102@linux.vnet.ibm.com
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
kernel/sched/fair.c | 9 ++++-----
1 files changed, 4 insertions(+), 5 deletions(-)
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 99285a8..287bfac 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -4283,7 +4283,7 @@ redo:
goto out_balanced;
}
- BUG_ON(busiest == this_rq);
+ BUG_ON(busiest == env.dst_rq);
schedstat_add(sd, lb_imbalance[idle], env.imbalance);
@@ -4304,7 +4304,7 @@ redo:
update_h_load(env.src_cpu);
more_balance:
local_irq_save(flags);
- double_rq_lock(this_rq, busiest);
+ double_rq_lock(env.dst_rq, busiest);
/*
* cur_ld_moved - load moved in current iteration
@@ -4312,7 +4312,7 @@ more_balance:
*/
cur_ld_moved = move_tasks(&env);
ld_moved += cur_ld_moved;
- double_rq_unlock(this_rq, busiest);
+ double_rq_unlock(env.dst_rq, busiest);
local_irq_restore(flags);
if (env.flags & LBF_NEED_BREAK) {
@@ -4348,8 +4348,7 @@ more_balance:
if ((env.flags & LBF_SOME_PINNED) && env.imbalance > 0 &&
lb_iterations++ < max_lb_iterations) {
- this_rq = cpu_rq(env.new_dst_cpu);
- env.dst_rq = this_rq;
+ env.dst_rq = cpu_rq(env.new_dst_cpu);
env.dst_cpu = env.new_dst_cpu;
env.flags &= ~LBF_SOME_PINNED;
env.loop = 0;
^ permalink raw reply related [flat|nested] 3+ messages in thread
end of thread, other threads:[~2012-08-13 17:06 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2012-08-06 8:41 [PATCH] sched: using dst_rq instead of this_rq during load balance Michael Wang
2012-08-13 17:00 ` [tip:sched/core] " tip-bot for Michael Wang
2012-08-13 17:06 ` tip-bot for Michael Wang
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).