Patches contributed by Eötvös Lorand University
commit f9153ee6c71cb9ab38de3b8ed66b1c3fa27c3f7d
Author: Ingo Molnar <mingo@elte.hu>
Date: Mon Jul 16 09:46:30 2007 +0200
[PATCH] sched: improve weight-array comments
improve the comments around the wmult array (which controls the weight
of niced tasks). Clarify that to achieve a 10% difference in CPU
utilization, a weight multiplier of 1.25 has to be used.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
diff --git a/kernel/sched.c b/kernel/sched.c
index 3332bbb5d5cf..a7284bc79cdf 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -736,7 +736,9 @@ static void update_curr_load(struct rq *rq, u64 now)
*
* The "10% effect" is relative and cumulative: from _any_ nice level,
* if you go up 1 level, it's -10% CPU usage, if you go down 1 level
- * it's +10% CPU usage.
+ * it's +10% CPU usage. (to achieve that we use a multiplier of 1.25.
+ * If a task goes up by ~10% and another task goes down by ~10% then
+ * the relative distance between them is ~25%.)
*/
static const int prio_to_weight[40] = {
/* -20 */ 88818, 71054, 56843, 45475, 36380, 29104, 23283, 18626, 14901, 11921,
commit 5926c50b83b626991c8c38efbca2020ee96b215f
Author: Ingo Molnar <mingo@elte.hu>
Date: Mon Jul 16 09:46:30 2007 +0200
[PATCH] sched: remove dead code from task_stime()
Alexey Dobriyan noticed that task_stime() contains a piece of dead code.
(which is a remnant of earlier versions of this code) Remove that code.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
diff --git a/fs/proc/array.c b/fs/proc/array.c
index 98e78e2f18d6..c6977796fafd 100644
--- a/fs/proc/array.c
+++ b/fs/proc/array.c
@@ -332,7 +332,7 @@ static clock_t task_utime(struct task_struct *p)
static clock_t task_stime(struct task_struct *p)
{
- clock_t stime = cputime_to_clock_t(p->stime);
+ clock_t stime;
/*
* Use CFS's precise accounting. (we subtract utime from
commit e6c9116d1dc984cb7ecf1b0fe26ca4a8ab36bb57
Author: Ingo Molnar <mingo@elte.hu>
Date: Sat Jul 14 18:50:15 2007 -0700
[RFKILL]: fix net/rfkill/rfkill-input.c bug on 64-bit systems
Subject: [patch] net/input: fix net/rfkill/rfkill-input.c bug on 64-bit systems
this recent commit:
commit cf4328cd949c2086091c62c5685f1580fe9b55e4
Author: Ivo van Doorn <IvDoorn@gmail.com>
Date: Mon May 7 00:34:20 2007 -0700
[NET]: rfkill: add support for input key to control wireless radio
added this 64-bit bug:
....
unsigned int flags;
spin_lock_irqsave(&task->lock, flags);
....
irq 'flags' must be unsigned long, not unsigned int. The -rt tree has
strict checks about this on 64-bit so this triggered a build failure.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: David S. Miller <davem@davemloft.net>
diff --git a/net/rfkill/rfkill-input.c b/net/rfkill/rfkill-input.c
index e5c840c30284..230e35c59786 100644
--- a/net/rfkill/rfkill-input.c
+++ b/net/rfkill/rfkill-input.c
@@ -55,7 +55,7 @@ static void rfkill_task_handler(struct work_struct *work)
static void rfkill_schedule_toggle(struct rfkill_task *task)
{
- unsigned int flags;
+ unsigned long flags;
spin_lock_irqsave(&task->lock, flags);
commit f787a50306680c187cf2896a8017937c1bf6dc7e
Author: Ingo Molnar <mingo@elte.hu>
Date: Wed Jul 11 21:21:47 2007 +0200
[PATCH] sched: small topology.h cleanup
trivial cleanup: LOCAL_DISTANCE and REMOTE_DISTANCE are only used in
topology.h and inside an #ifndef section - limit their existence to
that #ifndef.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
diff --git a/include/linux/topology.h b/include/linux/topology.h
index da6c39b2d051..d0890a7e5bab 100644
--- a/include/linux/topology.h
+++ b/include/linux/topology.h
@@ -50,10 +50,10 @@
for_each_online_node(node) \
if (nr_cpus_node(node))
-#ifndef node_distance
/* Conform to ACPI 2.0 SLIT distance definitions */
#define LOCAL_DISTANCE 10
#define REMOTE_DISTANCE 20
+#ifndef node_distance
#define node_distance(from,to) ((from) == (to) ? LOCAL_DISTANCE : REMOTE_DISTANCE)
#endif
#ifndef RECLAIM_DISTANCE
commit 4bd77321a833077c5c9ac7b9d284e261e4a8906e
Author: Ingo Molnar <mingo@elte.hu>
Date: Wed Jul 11 21:21:47 2007 +0200
[PATCH] sched: fix show_task()/show_tasks() output
fix show_task()/show_tasks() output:
- there's no sibling info anymore
- the fields were not aligned properly with the description
- get rid of the lazy-TLB output: it's been quite some time since
we last had a bug there, and when we had a bug it wasnt helped a
bit by this debug output.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
diff --git a/kernel/sched.c b/kernel/sched.c
index 9088c2d97148..0559665a3a0b 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -4647,14 +4647,14 @@ static void show_task(struct task_struct *p)
state = p->state ? __ffs(p->state) + 1 : 0;
printk("%-13.13s %c", p->comm,
state < sizeof(stat_nam) - 1 ? stat_nam[state] : '?');
-#if (BITS_PER_LONG == 32)
+#if BITS_PER_LONG == 32
if (state == TASK_RUNNING)
- printk(" running ");
+ printk(" running ");
else
- printk(" %08lX ", thread_saved_pc(p));
+ printk(" %08lx ", thread_saved_pc(p));
#else
if (state == TASK_RUNNING)
- printk(" running task ");
+ printk(" running task ");
else
printk(" %016lx ", thread_saved_pc(p));
#endif
@@ -4666,11 +4666,7 @@ static void show_task(struct task_struct *p)
free = (unsigned long)n - (unsigned long)end_of_stack(p);
}
#endif
- printk("%5lu %5d %6d", free, p->pid, p->parent->pid);
- if (!p->mm)
- printk(" (L-TLB)\n");
- else
- printk(" (NOTLB)\n");
+ printk("%5lu %5d %6d\n", free, p->pid, p->parent->pid);
if (state != TASK_RUNNING)
show_stack(p, NULL);
@@ -4680,14 +4676,12 @@ void show_state_filter(unsigned long state_filter)
{
struct task_struct *g, *p;
-#if (BITS_PER_LONG == 32)
- printk("\n"
- " free sibling\n");
- printk(" task PC stack pid father child younger older\n");
+#if BITS_PER_LONG == 32
+ printk(KERN_INFO
+ " task PC stack pid father\n");
#else
- printk("\n"
- " free sibling\n");
- printk(" task PC stack pid father child younger older\n");
+ printk(KERN_INFO
+ " task PC stack pid father\n");
#endif
read_lock(&tasklist_lock);
do_each_thread(g, p) {
commit 45f384a64f0769bb9a3caf0516de88a629f48e61
Author: Ingo Molnar <mingo@elte.hu>
Date: Wed Jul 11 21:21:47 2007 +0200
[PATCH] sched: remove stale version info from kernel/sched_debug.c
kernel/sched_debug.c referred to CFS -v20, but there's no CFS versioning
needed within the upstream kernel.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
diff --git a/kernel/sched_debug.c b/kernel/sched_debug.c
index 1baf87cceb7c..29f2c21e7da2 100644
--- a/kernel/sched_debug.c
+++ b/kernel/sched_debug.c
@@ -171,7 +171,7 @@ static int sched_debug_show(struct seq_file *m, void *v)
u64 now = ktime_to_ns(ktime_get());
int cpu;
- SEQ_printf(m, "Sched Debug Version: v0.04, cfs-v20, %s %.*s\n",
+ SEQ_printf(m, "Sched Debug Version: v0.05, %s %.*s\n",
init_utsname()->release,
(int)strcspn(init_utsname()->version, " "),
init_utsname()->version);
commit a5968df8737eda477d9d1038f5428ebd4d0884e1
Author: Ingo Molnar <mingo@elte.hu>
Date: Wed Jul 11 21:21:47 2007 +0200
[PATCH] sched: allow larger granularity
Allow granularity up to 100 msecs, instead of 10 msecs.
(needed on larger boxes)
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
diff --git a/kernel/sched.c b/kernel/sched.c
index 2ab7fa8039ae..9088c2d97148 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -4778,7 +4778,7 @@ cpumask_t nohz_cpu_mask = CPU_MASK_NONE;
static inline void sched_init_granularity(void)
{
unsigned int factor = 1 + ilog2(num_online_cpus());
- const unsigned long gran_limit = 10000000;
+ const unsigned long gran_limit = 100000000;
sysctl_sched_granularity *= factor;
if (sysctl_sched_granularity > gran_limit)
commit c31f2e8a42c41efa46397732656ddf48cc77593e
Author: Ingo Molnar <mingo@elte.hu>
Date: Mon Jul 9 18:52:01 2007 +0200
sched: add CFS credits
add credits for recent major scheduler contributions:
Con Kolivas, for pioneering the fair-scheduling approach
Peter Williams, for smpnice
Mike Galbraith, for interactivity tuning of CFS
Srivatsa Vaddagiri, for group scheduling enhancements
Signed-off-by: Ingo Molnar <mingo@elte.hu>
diff --git a/kernel/sched.c b/kernel/sched.c
index 0e3caf742ae3..9fbced64bfee 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -16,6 +16,12 @@
* by Davide Libenzi, preemptible kernel bits by Robert Love.
* 2003-09-03 Interactivity tuning by Con Kolivas.
* 2004-04-02 Scheduler domains code by Nick Piggin
+ * 2007-04-15 Work begun on replacing all interactivity tuning with a
+ * fair scheduling design by Con Kolivas.
+ * 2007-05-05 Load balancing (smp-nice) and other improvements
+ * by Peter Williams
+ * 2007-05-06 Interactivity improvements to CFS by Mike Galbraith
+ * 2007-07-01 Group scheduling enhancements by Srivatsa Vaddagiri
*/
#include <linux/mm.h>
commit 6fb43d7b50e49a36f8be3199141bec473e5ecb00
Author: Ingo Molnar <mingo@elte.hu>
Date: Mon Jul 9 18:52:01 2007 +0200
sched: micro-optimize mmdrop()
micro-optimize mmdrop(). Improves schedule()'s assembly a bit.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 785ec8465bd3..cfb680585ab8 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1514,7 +1514,7 @@ extern struct mm_struct * mm_alloc(void);
extern void FASTCALL(__mmdrop(struct mm_struct *));
static inline void mmdrop(struct mm_struct * mm)
{
- if (atomic_dec_and_test(&mm->mm_count))
+ if (unlikely(atomic_dec_and_test(&mm->mm_count)))
__mmdrop(mm);
}
commit 9c4801cebc2add1fe514bc8eb201b16372eee11a
Author: Ingo Molnar <mingo@elte.hu>
Date: Mon Jul 9 18:52:01 2007 +0200
sched: more agressive idle balancing
the Linux scheduler is starving a number of workloads. So default
to more agressive idle-balancing. This hurts lmbench context-switching
numbers (which was the main reason we sucked at idle-balancing for
such a long time) but the lmbench numbers are fine once the system is
minimally utilized.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
diff --git a/include/linux/topology.h b/include/linux/topology.h
index a9d1f049cc15..da6c39b2d051 100644
--- a/include/linux/topology.h
+++ b/include/linux/topology.h
@@ -98,7 +98,7 @@
.cache_nice_tries = 0, \
.busy_idx = 0, \
.idle_idx = 0, \
- .newidle_idx = 1, \
+ .newidle_idx = 0, \
.wake_idx = 0, \
.forkexec_idx = 0, \
.flags = SD_LOAD_BALANCE \
@@ -128,14 +128,15 @@
.imbalance_pct = 125, \
.cache_nice_tries = 1, \
.busy_idx = 2, \
- .idle_idx = 1, \
- .newidle_idx = 2, \
+ .idle_idx = 0, \
+ .newidle_idx = 0, \
.wake_idx = 1, \
.forkexec_idx = 1, \
.flags = SD_LOAD_BALANCE \
| SD_BALANCE_NEWIDLE \
| SD_BALANCE_EXEC \
| SD_WAKE_AFFINE \
+ | SD_WAKE_IDLE \
| SD_SHARE_PKG_RESOURCES\
| BALANCE_FOR_MC_POWER, \
.last_balance = jiffies, \
@@ -158,14 +159,15 @@
.imbalance_pct = 125, \
.cache_nice_tries = 1, \
.busy_idx = 2, \
- .idle_idx = 1, \
- .newidle_idx = 2, \
+ .idle_idx = 0, \
+ .newidle_idx = 0, \
.wake_idx = 1, \
.forkexec_idx = 1, \
.flags = SD_LOAD_BALANCE \
| SD_BALANCE_NEWIDLE \
| SD_BALANCE_EXEC \
| SD_WAKE_AFFINE \
+ | SD_WAKE_IDLE \
| BALANCE_FOR_PKG_POWER,\
.last_balance = jiffies, \
.balance_interval = 1, \