linux/drivers/cpufreq/cpufreq_stats.c

427 lines
10 KiB
C
Raw Normal View History

/*
* drivers/cpufreq/cpufreq_stats.c
*
* Copyright (C) 2003-2004 Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>.
* (C) 2004 Zou Nan hai <nanhai.zou@intel.com>.
*
* 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.
*/
#include <linux/kernel.h>
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h percpu.h is included by sched.h and module.h and thus ends up being included when building most .c files. percpu.h includes slab.h which in turn includes gfp.h making everything defined by the two files universally available and complicating inclusion dependencies. percpu.h -> slab.h dependency is about to be removed. Prepare for this change by updating users of gfp and slab facilities include those headers directly instead of assuming availability. As this conversion needs to touch large number of source files, the following script is used as the basis of conversion. http://userweb.kernel.org/~tj/misc/slabh-sweep.py The script does the followings. * Scan files for gfp and slab usages and update includes such that only the necessary includes are there. ie. if only gfp is used, gfp.h, if slab is used, slab.h. * When the script inserts a new include, it looks at the include blocks and try to put the new include such that its order conforms to its surrounding. It's put in the include block which contains core kernel includes, in the same order that the rest are ordered - alphabetical, Christmas tree, rev-Xmas-tree or at the end if there doesn't seem to be any matching order. * If the script can't find a place to put a new include (mostly because the file doesn't have fitting include block), it prints out an error message indicating which .h file needs to be added to the file. The conversion was done in the following steps. 1. The initial automatic conversion of all .c files updated slightly over 4000 files, deleting around 700 includes and adding ~480 gfp.h and ~3000 slab.h inclusions. The script emitted errors for ~400 files. 2. Each error was manually checked. Some didn't need the inclusion, some needed manual addition while adding it to implementation .h or embedding .c file was more appropriate for others. This step added inclusions to around 150 files. 3. The script was run again and the output was compared to the edits from #2 to make sure no file was left behind. 4. Several build tests were done and a couple of problems were fixed. e.g. lib/decompress_*.c used malloc/free() wrappers around slab APIs requiring slab.h to be added manually. 5. The script was run on all .h files but without automatically editing them as sprinkling gfp.h and slab.h inclusions around .h files could easily lead to inclusion dependency hell. Most gfp.h inclusion directives were ignored as stuff from gfp.h was usually wildly available and often used in preprocessor macros. Each slab.h inclusion directive was examined and added manually as necessary. 6. percpu.h was updated not to include slab.h. 7. Build test were done on the following configurations and failures were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my distributed build env didn't work with gcov compiles) and a few more options had to be turned off depending on archs to make things build (like ipr on powerpc/64 which failed due to missing writeq). * x86 and x86_64 UP and SMP allmodconfig and a custom test config. * powerpc and powerpc64 SMP allmodconfig * sparc and sparc64 SMP allmodconfig * ia64 SMP allmodconfig * s390 SMP allmodconfig * alpha SMP allmodconfig * um on x86_64 SMP allmodconfig 8. percpu.h modifications were reverted so that it could be applied as a separate patch and serve as bisection point. Given the fact that I had only a couple of failures from tests on step 6, I'm fairly confident about the coverage of this conversion patch. If there is a breakage, it's likely to be something in one of the arch headers which should be easily discoverable easily on most builds of the specific arch. Signed-off-by: Tejun Heo <tj@kernel.org> Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 16:04:11 +08:00
#include <linux/slab.h>
#include <linux/cpu.h>
#include <linux/sysfs.h>
#include <linux/cpufreq.h>
#include <linux/module.h>
#include <linux/jiffies.h>
#include <linux/percpu.h>
#include <linux/kobject.h>
#include <linux/spinlock.h>
[PATCH] create and destroy cpufreq sysfs entries based on cpu notifiers cpufreq entries in sysfs should only be populated when CPU is online state. When we either boot with maxcpus=x and then boot the other cpus by echoing to sysfs online file, these entries should be created and destroyed when CPU_DEAD is notified. Same treatement as cache entries under sysfs. We place the processor in the lowest frequency, so hw managed P-State transitions can still work on the other threads to save power. Primary goal was to just make these directories appear/disapper dynamically. There is one in this patch i had to do, which i really dont like myself but probably best if someone handling the cpufreq infrastructure could give this code right treatment if this is not acceptable. I guess its probably good for the first cut. - Converting lock_cpu_hotplug()/unlock_cpu_hotplug() to disable/enable preempt. The locking was smack in the middle of the notification path, when the hotplug is already holding the lock. I tried another solution to avoid this so avoid taking locks if we know we are from notification path. The solution was getting very ugly and i decided this was probably good for this iteration until someone who understands cpufreq could do a better job than me. (akpm: export cpucontrol to GPL modules: drivers/cpufreq/cpufreq_stats.c now does lock_cpu_hotplug()) Signed-off-by: Ashok Raj <ashok.raj@intel.com> Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com> Cc: Dave Jones <davej@codemonkey.org.uk> Cc: Zwane Mwaikambo <zwane@holomorphy.com> Cc: Greg KH <greg@kroah.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-31 06:59:54 +08:00
#include <linux/notifier.h>
#include <asm/cputime.h>
static spinlock_t cpufreq_stats_lock;
struct cpufreq_stats {
unsigned int cpu;
unsigned int total_trans;
unsigned long long last_time;
unsigned int max_state;
unsigned int state_num;
unsigned int last_index;
u64 *time_in_state;
unsigned int *freq_table;
#ifdef CONFIG_CPU_FREQ_STAT_DETAILS
unsigned int *trans_table;
#endif
};
static DEFINE_PER_CPU(struct cpufreq_stats *, cpufreq_stats_table);
struct cpufreq_stats_attribute {
struct attribute attr;
ssize_t(*show) (struct cpufreq_stats *, char *);
};
static int cpufreq_stats_update(unsigned int cpu)
{
struct cpufreq_stats *stat;
unsigned long long cur_time;
cur_time = get_jiffies_64();
spin_lock(&cpufreq_stats_lock);
stat = per_cpu(cpufreq_stats_table, cpu);
if (stat->time_in_state)
stat->time_in_state[stat->last_index] +=
cur_time - stat->last_time;
stat->last_time = cur_time;
spin_unlock(&cpufreq_stats_lock);
return 0;
}
static ssize_t show_total_trans(struct cpufreq_policy *policy, char *buf)
{
struct cpufreq_stats *stat = per_cpu(cpufreq_stats_table, policy->cpu);
if (!stat)
return 0;
return sprintf(buf, "%d\n",
per_cpu(cpufreq_stats_table, stat->cpu)->total_trans);
}
static ssize_t show_time_in_state(struct cpufreq_policy *policy, char *buf)
{
ssize_t len = 0;
int i;
struct cpufreq_stats *stat = per_cpu(cpufreq_stats_table, policy->cpu);
if (!stat)
return 0;
cpufreq_stats_update(stat->cpu);
for (i = 0; i < stat->state_num; i++) {
len += sprintf(buf + len, "%u %llu\n", stat->freq_table[i],
(unsigned long long)
cputime64_to_clock_t(stat->time_in_state[i]));
}
return len;
}
#ifdef CONFIG_CPU_FREQ_STAT_DETAILS
static ssize_t show_trans_table(struct cpufreq_policy *policy, char *buf)
{
ssize_t len = 0;
int i, j;
struct cpufreq_stats *stat = per_cpu(cpufreq_stats_table, policy->cpu);
if (!stat)
return 0;
cpufreq_stats_update(stat->cpu);
len += snprintf(buf + len, PAGE_SIZE - len, " From : To\n");
len += snprintf(buf + len, PAGE_SIZE - len, " : ");
for (i = 0; i < stat->state_num; i++) {
if (len >= PAGE_SIZE)
break;
len += snprintf(buf + len, PAGE_SIZE - len, "%9u ",
stat->freq_table[i]);
}
if (len >= PAGE_SIZE)
return PAGE_SIZE;
len += snprintf(buf + len, PAGE_SIZE - len, "\n");
for (i = 0; i < stat->state_num; i++) {
if (len >= PAGE_SIZE)
break;
len += snprintf(buf + len, PAGE_SIZE - len, "%9u: ",
stat->freq_table[i]);
for (j = 0; j < stat->state_num; j++) {
if (len >= PAGE_SIZE)
break;
len += snprintf(buf + len, PAGE_SIZE - len, "%9u ",
stat->trans_table[i*stat->max_state+j]);
}
if (len >= PAGE_SIZE)
break;
len += snprintf(buf + len, PAGE_SIZE - len, "\n");
}
if (len >= PAGE_SIZE)
return PAGE_SIZE;
return len;
}
cpufreq_freq_attr_ro(trans_table);
#endif
cpufreq_freq_attr_ro(total_trans);
cpufreq_freq_attr_ro(time_in_state);
static struct attribute *default_attrs[] = {
&total_trans.attr,
&time_in_state.attr,
#ifdef CONFIG_CPU_FREQ_STAT_DETAILS
&trans_table.attr,
#endif
NULL
};
static struct attribute_group stats_attr_group = {
.attrs = default_attrs,
.name = "stats"
};
static int freq_table_get_index(struct cpufreq_stats *stat, unsigned int freq)
{
int index;
for (index = 0; index < stat->max_state; index++)
if (stat->freq_table[index] == freq)
return index;
return -1;
}
/* should be called late in the CPU removal sequence so that the stats
* memory is still available in case someone tries to use it.
*/
static void cpufreq_stats_free_table(unsigned int cpu)
{
struct cpufreq_stats *stat = per_cpu(cpufreq_stats_table, cpu);
if (stat) {
pr_debug("%s: Free stat table\n", __func__);
kfree(stat->time_in_state);
kfree(stat);
per_cpu(cpufreq_stats_table, cpu) = NULL;
}
}
/* must be called early in the CPU removal sequence (before
* cpufreq_remove_dev) so that policy is still valid.
*/
static void cpufreq_stats_free_sysfs(unsigned int cpu)
{
struct cpufreq_policy *policy = cpufreq_cpu_get(cpu);
if (!cpufreq_frequency_get_table(cpu))
return;
if (policy && !policy_is_shared(policy)) {
pr_debug("%s: Free sysfs stat\n", __func__);
sysfs_remove_group(&policy->kobj, &stats_attr_group);
}
if (policy)
cpufreq_cpu_put(policy);
}
static int cpufreq_stats_create_table(struct cpufreq_policy *policy,
struct cpufreq_frequency_table *table)
{
unsigned int i, j, count = 0, ret = 0;
struct cpufreq_stats *stat;
struct cpufreq_policy *data;
unsigned int alloc_size;
unsigned int cpu = policy->cpu;
if (per_cpu(cpufreq_stats_table, cpu))
return -EBUSY;
stat = kzalloc(sizeof(struct cpufreq_stats), GFP_KERNEL);
if ((stat) == NULL)
return -ENOMEM;
data = cpufreq_cpu_get(cpu);
if (data == NULL) {
ret = -EINVAL;
goto error_get_fail;
}
ret = sysfs_create_group(&data->kobj, &stats_attr_group);
if (ret)
goto error_out;
stat->cpu = cpu;
per_cpu(cpufreq_stats_table, cpu) = stat;
for (i = 0; table[i].frequency != CPUFREQ_TABLE_END; i++) {
unsigned int freq = table[i].frequency;
if (freq == CPUFREQ_ENTRY_INVALID)
continue;
count++;
}
alloc_size = count * sizeof(int) + count * sizeof(u64);
#ifdef CONFIG_CPU_FREQ_STAT_DETAILS
alloc_size += count * count * sizeof(int);
#endif
stat->max_state = count;
stat->time_in_state = kzalloc(alloc_size, GFP_KERNEL);
if (!stat->time_in_state) {
ret = -ENOMEM;
goto error_out;
}
stat->freq_table = (unsigned int *)(stat->time_in_state + count);
#ifdef CONFIG_CPU_FREQ_STAT_DETAILS
stat->trans_table = stat->freq_table + count;
#endif
j = 0;
for (i = 0; table[i].frequency != CPUFREQ_TABLE_END; i++) {
unsigned int freq = table[i].frequency;
if (freq == CPUFREQ_ENTRY_INVALID)
continue;
if (freq_table_get_index(stat, freq) == -1)
stat->freq_table[j++] = freq;
}
stat->state_num = j;
spin_lock(&cpufreq_stats_lock);
stat->last_time = get_jiffies_64();
stat->last_index = freq_table_get_index(stat, policy->cur);
spin_unlock(&cpufreq_stats_lock);
cpufreq_cpu_put(data);
return 0;
error_out:
cpufreq_cpu_put(data);
error_get_fail:
kfree(stat);
per_cpu(cpufreq_stats_table, cpu) = NULL;
return ret;
}
static void cpufreq_stats_update_policy_cpu(struct cpufreq_policy *policy)
{
struct cpufreq_stats *stat = per_cpu(cpufreq_stats_table,
policy->last_cpu);
pr_debug("Updating stats_table for new_cpu %u from last_cpu %u\n",
policy->cpu, policy->last_cpu);
per_cpu(cpufreq_stats_table, policy->cpu) = per_cpu(cpufreq_stats_table,
policy->last_cpu);
per_cpu(cpufreq_stats_table, policy->last_cpu) = NULL;
stat->cpu = policy->cpu;
}
static int cpufreq_stat_notifier_policy(struct notifier_block *nb,
unsigned long val, void *data)
{
int ret;
struct cpufreq_policy *policy = data;
struct cpufreq_frequency_table *table;
unsigned int cpu = policy->cpu;
if (val == CPUFREQ_UPDATE_POLICY_CPU) {
cpufreq_stats_update_policy_cpu(policy);
return 0;
}
if (val != CPUFREQ_NOTIFY)
return 0;
table = cpufreq_frequency_get_table(cpu);
if (!table)
return 0;
ret = cpufreq_stats_create_table(policy, table);
if (ret)
return ret;
return 0;
}
static int cpufreq_stat_notifier_trans(struct notifier_block *nb,
unsigned long val, void *data)
{
struct cpufreq_freqs *freq = data;
struct cpufreq_stats *stat;
int old_index, new_index;
if (val != CPUFREQ_POSTCHANGE)
return 0;
stat = per_cpu(cpufreq_stats_table, freq->cpu);
if (!stat)
return 0;
old_index = stat->last_index;
new_index = freq_table_get_index(stat, freq->new);
[CPUFREQ] Don't set stat->last_index to -1 if the pol->cur has incorrect value. If the driver submitted an non-existing pol>cur value (say it used the default initialized value of zero), when the cpufreq stats tries to setup its initial values it incorrectly sets stat->last_index to -1 (or 0xfffff...). And cpufreq_stats_update tries to update at that index location and fails. This can be caused by: stat->last_index = freq_table_get_index(stat, policy->cur); not finding the appropiate frequency in the table (b/c the policy->cur is wrong) and we end up crashing. The fix however is concentrated in the 'cpufreq_stats_update' as the last_index (and old_index) are updated there. Which means it can reset the last_index to -1 again and on the next iteration cause a crash. Without this patch, the following crash is observed: powernow-k8: Found 1 AMD Athlon(tm) 64 Processor 3700+ (1 cpu cores) (version 2.20.00) powernow-k8: fid 0x2 (1000 MHz), vid 0x12 powernow-k8: fid 0xa (1800 MHz), vid 0xa powernow-k8: fid 0xc (2000 MHz), vid 0x8 powernow-k8: fid 0xe (2200 MHz), vid 0x8 Marking TSC unstable due to cpufreq changes powernow-k8: fid trans failed, fid 0x2, curr 0x0 BUG: unable to handle kernel paging request at ffff880807e07b78 IP: [<ffffffff81479163>] cpufreq_stats_update+0x46/0x5b .. snip.. Pid: 1, comm: swapper Not tainted 3.0.0-rc2 #45 MICRO-STAR INTERNATIONAL CO., LTD MS-7094/MS-7094 ..snip.. Call Trace: [<ffffffff81479248>] cpufreq_stat_notifier_trans+0x48/0x7c [<ffffffff81095d68>] notifier_call_chain+0x32/0x5e [<ffffffff81095e6b>] __srcu_notifier_call_chain+0x47/0x63 [<ffffffff81095e96>] srcu_notifier_call_chain+0xf/0x11 [<ffffffff81477e7a>] cpufreq_notify_transition+0x111/0x134 [<ffffffff8147b0d4>] powernowk8_target+0x53b/0x617 [<ffffffff8147723a>] __cpufreq_driver_target+0x2e/0x30 [<ffffffff8147a127>] cpufreq_governor_dbs+0x339/0x356 [<ffffffff81477394>] __cpufreq_governor+0xa8/0xe9 [<ffffffff81477525>] __cpufreq_set_policy+0x132/0x13e [<ffffffff8147848d>] cpufreq_add_dev_interface+0x272/0x28c Reported-by: Tobias Diedrich <ranma+xen@tdiedrich.de> Tested-by: Tobias Diedrich <ranma+xen@tdiedrich.de> Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Signed-off-by: Dave Jones <davej@redhat.com>
2011-06-17 03:36:38 +08:00
/* We can't do stat->time_in_state[-1]= .. */
if (old_index == -1 || new_index == -1)
return 0;
[CPUFREQ] Don't set stat->last_index to -1 if the pol->cur has incorrect value. If the driver submitted an non-existing pol>cur value (say it used the default initialized value of zero), when the cpufreq stats tries to setup its initial values it incorrectly sets stat->last_index to -1 (or 0xfffff...). And cpufreq_stats_update tries to update at that index location and fails. This can be caused by: stat->last_index = freq_table_get_index(stat, policy->cur); not finding the appropiate frequency in the table (b/c the policy->cur is wrong) and we end up crashing. The fix however is concentrated in the 'cpufreq_stats_update' as the last_index (and old_index) are updated there. Which means it can reset the last_index to -1 again and on the next iteration cause a crash. Without this patch, the following crash is observed: powernow-k8: Found 1 AMD Athlon(tm) 64 Processor 3700+ (1 cpu cores) (version 2.20.00) powernow-k8: fid 0x2 (1000 MHz), vid 0x12 powernow-k8: fid 0xa (1800 MHz), vid 0xa powernow-k8: fid 0xc (2000 MHz), vid 0x8 powernow-k8: fid 0xe (2200 MHz), vid 0x8 Marking TSC unstable due to cpufreq changes powernow-k8: fid trans failed, fid 0x2, curr 0x0 BUG: unable to handle kernel paging request at ffff880807e07b78 IP: [<ffffffff81479163>] cpufreq_stats_update+0x46/0x5b .. snip.. Pid: 1, comm: swapper Not tainted 3.0.0-rc2 #45 MICRO-STAR INTERNATIONAL CO., LTD MS-7094/MS-7094 ..snip.. Call Trace: [<ffffffff81479248>] cpufreq_stat_notifier_trans+0x48/0x7c [<ffffffff81095d68>] notifier_call_chain+0x32/0x5e [<ffffffff81095e6b>] __srcu_notifier_call_chain+0x47/0x63 [<ffffffff81095e96>] srcu_notifier_call_chain+0xf/0x11 [<ffffffff81477e7a>] cpufreq_notify_transition+0x111/0x134 [<ffffffff8147b0d4>] powernowk8_target+0x53b/0x617 [<ffffffff8147723a>] __cpufreq_driver_target+0x2e/0x30 [<ffffffff8147a127>] cpufreq_governor_dbs+0x339/0x356 [<ffffffff81477394>] __cpufreq_governor+0xa8/0xe9 [<ffffffff81477525>] __cpufreq_set_policy+0x132/0x13e [<ffffffff8147848d>] cpufreq_add_dev_interface+0x272/0x28c Reported-by: Tobias Diedrich <ranma+xen@tdiedrich.de> Tested-by: Tobias Diedrich <ranma+xen@tdiedrich.de> Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Signed-off-by: Dave Jones <davej@redhat.com>
2011-06-17 03:36:38 +08:00
cpufreq_stats_update(freq->cpu);
if (old_index == new_index)
return 0;
spin_lock(&cpufreq_stats_lock);
stat->last_index = new_index;
#ifdef CONFIG_CPU_FREQ_STAT_DETAILS
stat->trans_table[old_index * stat->max_state + new_index]++;
#endif
stat->total_trans++;
spin_unlock(&cpufreq_stats_lock);
return 0;
}
static int __cpuinit cpufreq_stat_cpu_callback(struct notifier_block *nfb,
unsigned long action,
void *hcpu)
[PATCH] create and destroy cpufreq sysfs entries based on cpu notifiers cpufreq entries in sysfs should only be populated when CPU is online state. When we either boot with maxcpus=x and then boot the other cpus by echoing to sysfs online file, these entries should be created and destroyed when CPU_DEAD is notified. Same treatement as cache entries under sysfs. We place the processor in the lowest frequency, so hw managed P-State transitions can still work on the other threads to save power. Primary goal was to just make these directories appear/disapper dynamically. There is one in this patch i had to do, which i really dont like myself but probably best if someone handling the cpufreq infrastructure could give this code right treatment if this is not acceptable. I guess its probably good for the first cut. - Converting lock_cpu_hotplug()/unlock_cpu_hotplug() to disable/enable preempt. The locking was smack in the middle of the notification path, when the hotplug is already holding the lock. I tried another solution to avoid this so avoid taking locks if we know we are from notification path. The solution was getting very ugly and i decided this was probably good for this iteration until someone who understands cpufreq could do a better job than me. (akpm: export cpucontrol to GPL modules: drivers/cpufreq/cpufreq_stats.c now does lock_cpu_hotplug()) Signed-off-by: Ashok Raj <ashok.raj@intel.com> Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com> Cc: Dave Jones <davej@codemonkey.org.uk> Cc: Zwane Mwaikambo <zwane@holomorphy.com> Cc: Greg KH <greg@kroah.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-31 06:59:54 +08:00
{
unsigned int cpu = (unsigned long)hcpu;
switch (action) {
case CPU_ONLINE:
case CPU_ONLINE_FROZEN:
[PATCH] create and destroy cpufreq sysfs entries based on cpu notifiers cpufreq entries in sysfs should only be populated when CPU is online state. When we either boot with maxcpus=x and then boot the other cpus by echoing to sysfs online file, these entries should be created and destroyed when CPU_DEAD is notified. Same treatement as cache entries under sysfs. We place the processor in the lowest frequency, so hw managed P-State transitions can still work on the other threads to save power. Primary goal was to just make these directories appear/disapper dynamically. There is one in this patch i had to do, which i really dont like myself but probably best if someone handling the cpufreq infrastructure could give this code right treatment if this is not acceptable. I guess its probably good for the first cut. - Converting lock_cpu_hotplug()/unlock_cpu_hotplug() to disable/enable preempt. The locking was smack in the middle of the notification path, when the hotplug is already holding the lock. I tried another solution to avoid this so avoid taking locks if we know we are from notification path. The solution was getting very ugly and i decided this was probably good for this iteration until someone who understands cpufreq could do a better job than me. (akpm: export cpucontrol to GPL modules: drivers/cpufreq/cpufreq_stats.c now does lock_cpu_hotplug()) Signed-off-by: Ashok Raj <ashok.raj@intel.com> Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com> Cc: Dave Jones <davej@codemonkey.org.uk> Cc: Zwane Mwaikambo <zwane@holomorphy.com> Cc: Greg KH <greg@kroah.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-31 06:59:54 +08:00
cpufreq_update_policy(cpu);
break;
case CPU_DOWN_PREPARE:
case CPU_DOWN_PREPARE_FROZEN:
cpufreq_stats_free_sysfs(cpu);
break;
[PATCH] create and destroy cpufreq sysfs entries based on cpu notifiers cpufreq entries in sysfs should only be populated when CPU is online state. When we either boot with maxcpus=x and then boot the other cpus by echoing to sysfs online file, these entries should be created and destroyed when CPU_DEAD is notified. Same treatement as cache entries under sysfs. We place the processor in the lowest frequency, so hw managed P-State transitions can still work on the other threads to save power. Primary goal was to just make these directories appear/disapper dynamically. There is one in this patch i had to do, which i really dont like myself but probably best if someone handling the cpufreq infrastructure could give this code right treatment if this is not acceptable. I guess its probably good for the first cut. - Converting lock_cpu_hotplug()/unlock_cpu_hotplug() to disable/enable preempt. The locking was smack in the middle of the notification path, when the hotplug is already holding the lock. I tried another solution to avoid this so avoid taking locks if we know we are from notification path. The solution was getting very ugly and i decided this was probably good for this iteration until someone who understands cpufreq could do a better job than me. (akpm: export cpucontrol to GPL modules: drivers/cpufreq/cpufreq_stats.c now does lock_cpu_hotplug()) Signed-off-by: Ashok Raj <ashok.raj@intel.com> Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com> Cc: Dave Jones <davej@codemonkey.org.uk> Cc: Zwane Mwaikambo <zwane@holomorphy.com> Cc: Greg KH <greg@kroah.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-31 06:59:54 +08:00
case CPU_DEAD:
case CPU_DEAD_FROZEN:
[PATCH] create and destroy cpufreq sysfs entries based on cpu notifiers cpufreq entries in sysfs should only be populated when CPU is online state. When we either boot with maxcpus=x and then boot the other cpus by echoing to sysfs online file, these entries should be created and destroyed when CPU_DEAD is notified. Same treatement as cache entries under sysfs. We place the processor in the lowest frequency, so hw managed P-State transitions can still work on the other threads to save power. Primary goal was to just make these directories appear/disapper dynamically. There is one in this patch i had to do, which i really dont like myself but probably best if someone handling the cpufreq infrastructure could give this code right treatment if this is not acceptable. I guess its probably good for the first cut. - Converting lock_cpu_hotplug()/unlock_cpu_hotplug() to disable/enable preempt. The locking was smack in the middle of the notification path, when the hotplug is already holding the lock. I tried another solution to avoid this so avoid taking locks if we know we are from notification path. The solution was getting very ugly and i decided this was probably good for this iteration until someone who understands cpufreq could do a better job than me. (akpm: export cpucontrol to GPL modules: drivers/cpufreq/cpufreq_stats.c now does lock_cpu_hotplug()) Signed-off-by: Ashok Raj <ashok.raj@intel.com> Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com> Cc: Dave Jones <davej@codemonkey.org.uk> Cc: Zwane Mwaikambo <zwane@holomorphy.com> Cc: Greg KH <greg@kroah.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-31 06:59:54 +08:00
cpufreq_stats_free_table(cpu);
break;
}
return NOTIFY_OK;
}
/* priority=1 so this will get called before cpufreq_remove_dev */
static struct notifier_block cpufreq_stat_cpu_notifier __refdata = {
[PATCH] create and destroy cpufreq sysfs entries based on cpu notifiers cpufreq entries in sysfs should only be populated when CPU is online state. When we either boot with maxcpus=x and then boot the other cpus by echoing to sysfs online file, these entries should be created and destroyed when CPU_DEAD is notified. Same treatement as cache entries under sysfs. We place the processor in the lowest frequency, so hw managed P-State transitions can still work on the other threads to save power. Primary goal was to just make these directories appear/disapper dynamically. There is one in this patch i had to do, which i really dont like myself but probably best if someone handling the cpufreq infrastructure could give this code right treatment if this is not acceptable. I guess its probably good for the first cut. - Converting lock_cpu_hotplug()/unlock_cpu_hotplug() to disable/enable preempt. The locking was smack in the middle of the notification path, when the hotplug is already holding the lock. I tried another solution to avoid this so avoid taking locks if we know we are from notification path. The solution was getting very ugly and i decided this was probably good for this iteration until someone who understands cpufreq could do a better job than me. (akpm: export cpucontrol to GPL modules: drivers/cpufreq/cpufreq_stats.c now does lock_cpu_hotplug()) Signed-off-by: Ashok Raj <ashok.raj@intel.com> Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com> Cc: Dave Jones <davej@codemonkey.org.uk> Cc: Zwane Mwaikambo <zwane@holomorphy.com> Cc: Greg KH <greg@kroah.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-31 06:59:54 +08:00
.notifier_call = cpufreq_stat_cpu_callback,
.priority = 1,
[PATCH] create and destroy cpufreq sysfs entries based on cpu notifiers cpufreq entries in sysfs should only be populated when CPU is online state. When we either boot with maxcpus=x and then boot the other cpus by echoing to sysfs online file, these entries should be created and destroyed when CPU_DEAD is notified. Same treatement as cache entries under sysfs. We place the processor in the lowest frequency, so hw managed P-State transitions can still work on the other threads to save power. Primary goal was to just make these directories appear/disapper dynamically. There is one in this patch i had to do, which i really dont like myself but probably best if someone handling the cpufreq infrastructure could give this code right treatment if this is not acceptable. I guess its probably good for the first cut. - Converting lock_cpu_hotplug()/unlock_cpu_hotplug() to disable/enable preempt. The locking was smack in the middle of the notification path, when the hotplug is already holding the lock. I tried another solution to avoid this so avoid taking locks if we know we are from notification path. The solution was getting very ugly and i decided this was probably good for this iteration until someone who understands cpufreq could do a better job than me. (akpm: export cpucontrol to GPL modules: drivers/cpufreq/cpufreq_stats.c now does lock_cpu_hotplug()) Signed-off-by: Ashok Raj <ashok.raj@intel.com> Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com> Cc: Dave Jones <davej@codemonkey.org.uk> Cc: Zwane Mwaikambo <zwane@holomorphy.com> Cc: Greg KH <greg@kroah.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-31 06:59:54 +08:00
};
static struct notifier_block notifier_policy_block = {
.notifier_call = cpufreq_stat_notifier_policy
};
static struct notifier_block notifier_trans_block = {
.notifier_call = cpufreq_stat_notifier_trans
};
static int __init cpufreq_stats_init(void)
{
int ret;
unsigned int cpu;
[PATCH] create and destroy cpufreq sysfs entries based on cpu notifiers cpufreq entries in sysfs should only be populated when CPU is online state. When we either boot with maxcpus=x and then boot the other cpus by echoing to sysfs online file, these entries should be created and destroyed when CPU_DEAD is notified. Same treatement as cache entries under sysfs. We place the processor in the lowest frequency, so hw managed P-State transitions can still work on the other threads to save power. Primary goal was to just make these directories appear/disapper dynamically. There is one in this patch i had to do, which i really dont like myself but probably best if someone handling the cpufreq infrastructure could give this code right treatment if this is not acceptable. I guess its probably good for the first cut. - Converting lock_cpu_hotplug()/unlock_cpu_hotplug() to disable/enable preempt. The locking was smack in the middle of the notification path, when the hotplug is already holding the lock. I tried another solution to avoid this so avoid taking locks if we know we are from notification path. The solution was getting very ugly and i decided this was probably good for this iteration until someone who understands cpufreq could do a better job than me. (akpm: export cpucontrol to GPL modules: drivers/cpufreq/cpufreq_stats.c now does lock_cpu_hotplug()) Signed-off-by: Ashok Raj <ashok.raj@intel.com> Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com> Cc: Dave Jones <davej@codemonkey.org.uk> Cc: Zwane Mwaikambo <zwane@holomorphy.com> Cc: Greg KH <greg@kroah.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-31 06:59:54 +08:00
spin_lock_init(&cpufreq_stats_lock);
ret = cpufreq_register_notifier(&notifier_policy_block,
CPUFREQ_POLICY_NOTIFIER);
if (ret)
return ret;
register_hotcpu_notifier(&cpufreq_stat_cpu_notifier);
for_each_online_cpu(cpu)
cpufreq_update_policy(cpu);
ret = cpufreq_register_notifier(&notifier_trans_block,
CPUFREQ_TRANSITION_NOTIFIER);
if (ret) {
cpufreq_unregister_notifier(&notifier_policy_block,
CPUFREQ_POLICY_NOTIFIER);
unregister_hotcpu_notifier(&cpufreq_stat_cpu_notifier);
for_each_online_cpu(cpu)
cpufreq_stats_free_table(cpu);
return ret;
}
return 0;
}
static void __exit cpufreq_stats_exit(void)
{
unsigned int cpu;
[PATCH] create and destroy cpufreq sysfs entries based on cpu notifiers cpufreq entries in sysfs should only be populated when CPU is online state. When we either boot with maxcpus=x and then boot the other cpus by echoing to sysfs online file, these entries should be created and destroyed when CPU_DEAD is notified. Same treatement as cache entries under sysfs. We place the processor in the lowest frequency, so hw managed P-State transitions can still work on the other threads to save power. Primary goal was to just make these directories appear/disapper dynamically. There is one in this patch i had to do, which i really dont like myself but probably best if someone handling the cpufreq infrastructure could give this code right treatment if this is not acceptable. I guess its probably good for the first cut. - Converting lock_cpu_hotplug()/unlock_cpu_hotplug() to disable/enable preempt. The locking was smack in the middle of the notification path, when the hotplug is already holding the lock. I tried another solution to avoid this so avoid taking locks if we know we are from notification path. The solution was getting very ugly and i decided this was probably good for this iteration until someone who understands cpufreq could do a better job than me. (akpm: export cpucontrol to GPL modules: drivers/cpufreq/cpufreq_stats.c now does lock_cpu_hotplug()) Signed-off-by: Ashok Raj <ashok.raj@intel.com> Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com> Cc: Dave Jones <davej@codemonkey.org.uk> Cc: Zwane Mwaikambo <zwane@holomorphy.com> Cc: Greg KH <greg@kroah.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-31 06:59:54 +08:00
cpufreq_unregister_notifier(&notifier_policy_block,
CPUFREQ_POLICY_NOTIFIER);
cpufreq_unregister_notifier(&notifier_trans_block,
CPUFREQ_TRANSITION_NOTIFIER);
unregister_hotcpu_notifier(&cpufreq_stat_cpu_notifier);
[PATCH] create and destroy cpufreq sysfs entries based on cpu notifiers cpufreq entries in sysfs should only be populated when CPU is online state. When we either boot with maxcpus=x and then boot the other cpus by echoing to sysfs online file, these entries should be created and destroyed when CPU_DEAD is notified. Same treatement as cache entries under sysfs. We place the processor in the lowest frequency, so hw managed P-State transitions can still work on the other threads to save power. Primary goal was to just make these directories appear/disapper dynamically. There is one in this patch i had to do, which i really dont like myself but probably best if someone handling the cpufreq infrastructure could give this code right treatment if this is not acceptable. I guess its probably good for the first cut. - Converting lock_cpu_hotplug()/unlock_cpu_hotplug() to disable/enable preempt. The locking was smack in the middle of the notification path, when the hotplug is already holding the lock. I tried another solution to avoid this so avoid taking locks if we know we are from notification path. The solution was getting very ugly and i decided this was probably good for this iteration until someone who understands cpufreq could do a better job than me. (akpm: export cpucontrol to GPL modules: drivers/cpufreq/cpufreq_stats.c now does lock_cpu_hotplug()) Signed-off-by: Ashok Raj <ashok.raj@intel.com> Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com> Cc: Dave Jones <davej@codemonkey.org.uk> Cc: Zwane Mwaikambo <zwane@holomorphy.com> Cc: Greg KH <greg@kroah.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-31 06:59:54 +08:00
for_each_online_cpu(cpu) {
cpufreq_stats_free_table(cpu);
cpufreq_stats_free_sysfs(cpu);
[PATCH] create and destroy cpufreq sysfs entries based on cpu notifiers cpufreq entries in sysfs should only be populated when CPU is online state. When we either boot with maxcpus=x and then boot the other cpus by echoing to sysfs online file, these entries should be created and destroyed when CPU_DEAD is notified. Same treatement as cache entries under sysfs. We place the processor in the lowest frequency, so hw managed P-State transitions can still work on the other threads to save power. Primary goal was to just make these directories appear/disapper dynamically. There is one in this patch i had to do, which i really dont like myself but probably best if someone handling the cpufreq infrastructure could give this code right treatment if this is not acceptable. I guess its probably good for the first cut. - Converting lock_cpu_hotplug()/unlock_cpu_hotplug() to disable/enable preempt. The locking was smack in the middle of the notification path, when the hotplug is already holding the lock. I tried another solution to avoid this so avoid taking locks if we know we are from notification path. The solution was getting very ugly and i decided this was probably good for this iteration until someone who understands cpufreq could do a better job than me. (akpm: export cpucontrol to GPL modules: drivers/cpufreq/cpufreq_stats.c now does lock_cpu_hotplug()) Signed-off-by: Ashok Raj <ashok.raj@intel.com> Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com> Cc: Dave Jones <davej@codemonkey.org.uk> Cc: Zwane Mwaikambo <zwane@holomorphy.com> Cc: Greg KH <greg@kroah.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-31 06:59:54 +08:00
}
}
MODULE_AUTHOR("Zou Nan hai <nanhai.zou@intel.com>");
MODULE_DESCRIPTION("'cpufreq_stats' - A driver to export cpufreq stats "
"through sysfs filesystem");
MODULE_LICENSE("GPL");
module_init(cpufreq_stats_init);
module_exit(cpufreq_stats_exit);