From: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
To: cpufreq <cpufreq@www.linux.org.uk>
Cc: alexey.y.starikovskiy@intel.com, Dave Jones <davej@redhat.com>,
Len Brown <lenb@unix-os.sc.intel.com>
Subject: [PATCH 1/4] minor optimizations to ondemand governor
Date: Wed, 28 Jun 2006 13:49:52 -0700 [thread overview]
Message-ID: <20060628134952.B12989@unix-os.sc.intel.com> (raw)
ondemand_remove_slowdown_06.patch
Remove slowdown from ondemand sampling path. This reduces the code path length
in dbs_check_cpu() by half. slowdown was not used by ondemand by default.
If there are any user level tools that were using this tunable, they
may report error now.
Signed-off-by: Alexey Starikovskiy <alexey.y.starikovskiy@intel.com>
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Index: linux-2.6.17/drivers/cpufreq/cpufreq_ondemand.c
===================================================================
--- linux-2.6.17.orig/drivers/cpufreq/cpufreq_ondemand.c
+++ linux-2.6.17/drivers/cpufreq/cpufreq_ondemand.c
@@ -5,6 +5,10 @@
* (C) 2003 Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>.
* Jun Nakajima <jun.nakajima@intel.com>
*
+ * Jun 2006 - Alexey Starikovskiy <alexey.y.starikovskiy@intel.com>
+ * Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
+ * - Remove slow down option and reduce the path length dbs_check_cpu
+ *
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License version 2 as
* published by the Free Software Foundation.
@@ -56,16 +60,14 @@ static unsigned int def_sampling_rate;
#define MIN_SAMPLING_RATE (def_sampling_rate / MIN_SAMPLING_RATE_RATIO)
#define MAX_SAMPLING_RATE (500 * def_sampling_rate)
#define DEF_SAMPLING_RATE_LATENCY_MULTIPLIER (1000)
-#define DEF_SAMPLING_DOWN_FACTOR (1)
-#define MAX_SAMPLING_DOWN_FACTOR (10)
#define TRANSITION_LATENCY_LIMIT (10 * 1000)
static void do_dbs_timer(void *data);
struct cpu_dbs_info_s {
+ cputime64_t prev_cpu_idle;
+ cputime64_t prev_cpu_wall;
struct cpufreq_policy *cur_policy;
- unsigned int prev_cpu_idle_up;
- unsigned int prev_cpu_idle_down;
unsigned int enable;
};
static DEFINE_PER_CPU(struct cpu_dbs_info_s, cpu_dbs_info);
@@ -87,24 +89,26 @@ static struct workqueue_struct *dbs_work
struct dbs_tuners {
unsigned int sampling_rate;
- unsigned int sampling_down_factor;
unsigned int up_threshold;
unsigned int ignore_nice;
};
static struct dbs_tuners dbs_tuners_ins = {
.up_threshold = DEF_FREQUENCY_UP_THRESHOLD,
- .sampling_down_factor = DEF_SAMPLING_DOWN_FACTOR,
.ignore_nice = 0,
};
-static inline unsigned int get_cpu_idle_time(unsigned int cpu)
+static inline cputime64_t get_cpu_idle_time(unsigned int cpu)
{
- return kstat_cpu(cpu).cpustat.idle +
- kstat_cpu(cpu).cpustat.iowait +
- ( dbs_tuners_ins.ignore_nice ?
- kstat_cpu(cpu).cpustat.nice :
- 0);
+ cputime64_t retval;
+
+ retval = cputime64_add(kstat_cpu(cpu).cpustat.idle,
+ kstat_cpu(cpu).cpustat.iowait);
+
+ if (dbs_tuners_ins.ignore_nice)
+ retval = cputime64_add(retval, kstat_cpu(cpu).cpustat.nice);
+
+ return retval;
}
/************************** sysfs interface ************************/
@@ -133,29 +137,9 @@ static ssize_t show_##file_name \
return sprintf(buf, "%u\n", dbs_tuners_ins.object); \
}
show_one(sampling_rate, sampling_rate);
-show_one(sampling_down_factor, sampling_down_factor);
show_one(up_threshold, up_threshold);
show_one(ignore_nice_load, ignore_nice);
-static ssize_t store_sampling_down_factor(struct cpufreq_policy *unused,
- const char *buf, size_t count)
-{
- unsigned int input;
- int ret;
- ret = sscanf (buf, "%u", &input);
- if (ret != 1 )
- return -EINVAL;
-
- if (input > MAX_SAMPLING_DOWN_FACTOR || input < 1)
- return -EINVAL;
-
- mutex_lock(&dbs_mutex);
- dbs_tuners_ins.sampling_down_factor = input;
- mutex_unlock(&dbs_mutex);
-
- return count;
-}
-
static ssize_t store_sampling_rate(struct cpufreq_policy *unused,
const char *buf, size_t count)
{
@@ -217,12 +201,12 @@ static ssize_t store_ignore_nice_load(st
}
dbs_tuners_ins.ignore_nice = input;
- /* we need to re-evaluate prev_cpu_idle_up and prev_cpu_idle_down */
+ /* we need to re-evaluate prev_cpu_idle */
for_each_online_cpu(j) {
- struct cpu_dbs_info_s *j_dbs_info;
- j_dbs_info = &per_cpu(cpu_dbs_info, j);
- j_dbs_info->prev_cpu_idle_up = get_cpu_idle_time(j);
- j_dbs_info->prev_cpu_idle_down = j_dbs_info->prev_cpu_idle_up;
+ struct cpu_dbs_info_s *dbs_info;
+ dbs_info = &per_cpu(cpu_dbs_info, j);
+ dbs_info->prev_cpu_idle = get_cpu_idle_time(j);
+ dbs_info->prev_cpu_wall = get_jiffies_64();
}
mutex_unlock(&dbs_mutex);
@@ -234,7 +218,6 @@ static struct freq_attr _name = \
__ATTR(_name, 0644, show_##_name, store_##_name)
define_one_rw(sampling_rate);
-define_one_rw(sampling_down_factor);
define_one_rw(up_threshold);
define_one_rw(ignore_nice_load);
@@ -242,7 +225,6 @@ static struct attribute * dbs_attributes
&sampling_rate_max.attr,
&sampling_rate_min.attr,
&sampling_rate.attr,
- &sampling_down_factor.attr,
&up_threshold.attr,
&ignore_nice_load.attr,
NULL
@@ -257,11 +239,10 @@ static struct attribute_group dbs_attr_g
static void dbs_check_cpu(int cpu)
{
- unsigned int idle_ticks, up_idle_ticks, total_ticks;
- unsigned int freq_next;
- unsigned int freq_down_sampling_rate;
- static int down_skip[NR_CPUS];
+ unsigned int idle_ticks, total_ticks;
+ unsigned int load;
struct cpu_dbs_info_s *this_dbs_info;
+ cputime64_t cur_jiffies;
struct cpufreq_policy *policy;
unsigned int j;
@@ -271,10 +252,14 @@ static void dbs_check_cpu(int cpu)
return;
policy = this_dbs_info->cur_policy;
+ cur_jiffies = jiffies64_to_cputime64(get_jiffies_64());
+ total_ticks = (unsigned int) cputime64_sub(cur_jiffies,
+ this_dbs_info->prev_cpu_wall);
+ this_dbs_info->prev_cpu_wall = cur_jiffies;
/*
* Every sampling_rate, we check, if current idle time is less
* than 20% (default), then we try to increase frequency
- * Every sampling_rate*sampling_down_factor, we look for a the lowest
+ * Every sampling_rate, we look for a the lowest
* frequency which can sustain the load while keeping idle time over
* 30%. If such a frequency exist, we try to decrease to this frequency.
*
@@ -283,36 +268,26 @@ static void dbs_check_cpu(int cpu)
* 5% (default) of current frequency
*/
- /* Check for frequency increase */
+ /* Get Idle Time */
idle_ticks = UINT_MAX;
for_each_cpu_mask(j, policy->cpus) {
- unsigned int tmp_idle_ticks, total_idle_ticks;
+ cputime64_t total_idle_ticks;
+ unsigned int tmp_idle_ticks;
struct cpu_dbs_info_s *j_dbs_info;
j_dbs_info = &per_cpu(cpu_dbs_info, j);
total_idle_ticks = get_cpu_idle_time(j);
- tmp_idle_ticks = total_idle_ticks -
- j_dbs_info->prev_cpu_idle_up;
- j_dbs_info->prev_cpu_idle_up = total_idle_ticks;
+ tmp_idle_ticks = (unsigned int) cputime64_sub(total_idle_ticks,
+ j_dbs_info->prev_cpu_idle);
+ j_dbs_info->prev_cpu_idle = total_idle_ticks;
if (tmp_idle_ticks < idle_ticks)
idle_ticks = tmp_idle_ticks;
}
+ load = (100 * (total_ticks - idle_ticks)) / total_ticks;
- /* Scale idle ticks by 100 and compare with up and down ticks */
- idle_ticks *= 100;
- up_idle_ticks = (100 - dbs_tuners_ins.up_threshold) *
- usecs_to_jiffies(dbs_tuners_ins.sampling_rate);
-
- if (idle_ticks < up_idle_ticks) {
- down_skip[cpu] = 0;
- for_each_cpu_mask(j, policy->cpus) {
- struct cpu_dbs_info_s *j_dbs_info;
-
- j_dbs_info = &per_cpu(cpu_dbs_info, j);
- j_dbs_info->prev_cpu_idle_down =
- j_dbs_info->prev_cpu_idle_up;
- }
+ /* Check for frequency increase */
+ if (load > dbs_tuners_ins.up_threshold) {
/* if we are already at full speed then break out early */
if (policy->cur == policy->max)
return;
@@ -323,50 +298,22 @@ static void dbs_check_cpu(int cpu)
}
/* Check for frequency decrease */
- down_skip[cpu]++;
- if (down_skip[cpu] < dbs_tuners_ins.sampling_down_factor)
- return;
-
- idle_ticks = UINT_MAX;
- for_each_cpu_mask(j, policy->cpus) {
- unsigned int tmp_idle_ticks, total_idle_ticks;
- struct cpu_dbs_info_s *j_dbs_info;
-
- j_dbs_info = &per_cpu(cpu_dbs_info, j);
- /* Check for frequency decrease */
- total_idle_ticks = j_dbs_info->prev_cpu_idle_up;
- tmp_idle_ticks = total_idle_ticks -
- j_dbs_info->prev_cpu_idle_down;
- j_dbs_info->prev_cpu_idle_down = total_idle_ticks;
-
- if (tmp_idle_ticks < idle_ticks)
- idle_ticks = tmp_idle_ticks;
- }
-
- down_skip[cpu] = 0;
/* if we cannot reduce the frequency anymore, break out early */
if (policy->cur == policy->min)
return;
- /* Compute how many ticks there are between two measurements */
- freq_down_sampling_rate = dbs_tuners_ins.sampling_rate *
- dbs_tuners_ins.sampling_down_factor;
- total_ticks = usecs_to_jiffies(freq_down_sampling_rate);
-
/*
* The optimal frequency is the frequency that is the lowest that
* can support the current CPU usage without triggering the up
* policy. To be safe, we focus 10 points under the threshold.
*/
- freq_next = ((total_ticks - idle_ticks) * 100) / total_ticks;
- freq_next = (freq_next * policy->cur) /
+ if (load < (dbs_tuners_ins.up_threshold - 10)) {
+ unsigned int freq_next;
+ freq_next = (policy->cur * load) /
(dbs_tuners_ins.up_threshold - 10);
- if (freq_next < policy->min)
- freq_next = policy->min;
-
- if (freq_next <= ((policy->cur * 95) / 100))
__cpufreq_driver_target(policy, freq_next, CPUFREQ_RELATION_L);
+ }
}
static void do_dbs_timer(void *data)
@@ -432,9 +379,8 @@ static int cpufreq_governor_dbs(struct c
j_dbs_info = &per_cpu(cpu_dbs_info, j);
j_dbs_info->cur_policy = policy;
- j_dbs_info->prev_cpu_idle_up = get_cpu_idle_time(j);
- j_dbs_info->prev_cpu_idle_down
- = j_dbs_info->prev_cpu_idle_up;
+ j_dbs_info->prev_cpu_idle = get_cpu_idle_time(j);
+ j_dbs_info->prev_cpu_wall = get_jiffies_64();
}
this_dbs_info->enable = 1;
sysfs_create_group(&policy->kobj, &dbs_attr_group);
Index: linux-2.6.17/include/asm-generic/cputime.h
===================================================================
--- linux-2.6.17.orig/include/asm-generic/cputime.h
+++ linux-2.6.17/include/asm-generic/cputime.h
@@ -24,7 +24,9 @@ typedef u64 cputime64_t;
#define cputime64_zero (0ULL)
#define cputime64_add(__a, __b) ((__a) + (__b))
+#define cputime64_sub(__a, __b) ((__a) - (__b))
#define cputime64_to_jiffies64(__ct) (__ct)
+#define jiffies64_to_cputime64(__jif) (__jif)
#define cputime_to_cputime64(__ct) ((u64) __ct)
next reply other threads:[~2006-06-28 20:49 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2006-06-28 20:49 Venkatesh Pallipadi [this message]
2006-06-30 5:08 ` [PATCH 1/4] minor optimizations to ondemand governor Dave Jones
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20060628134952.B12989@unix-os.sc.intel.com \
--to=venkatesh.pallipadi@intel.com \
--cc=alexey.y.starikovskiy@intel.com \
--cc=cpufreq@www.linux.org.uk \
--cc=davej@redhat.com \
--cc=lenb@unix-os.sc.intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.