* [RFC][PATCH 0/4] sched: Display deadline bandwidth and other SCHED_DEBUG clean up
@ 2016-02-01 20:26 Steven Rostedt
2016-02-01 20:26 ` [RFC][PATCH 1/4] sched: Move sched_feature file setup into debug.c Steven Rostedt
` (3 more replies)
0 siblings, 4 replies; 11+ messages in thread
From: Steven Rostedt @ 2016-02-01 20:26 UTC (permalink / raw)
To: linux-kernel
Cc: Ingo Molnar, Peter Zijlstra, Thomas Gleixner, Juri Lelli,
Clark Williams, Andrew Morton
I'm starting to play with SCHED_DEADLINE a bit and I'm able to cause
a bandwidth "leak". Then I realized there's no way to examine what bandwidths
are enabled on which CPUs. I decided to create a debugfs file that would
display the deadline bandwidths. I created a debugfs/sched/ directory to
place a "deadline_bw" file that displays the bandwidths of the CPUs in the
following format:
# cat /sys/kernel/debug/sched/deadline_bw
CPU[0]:
bw: 996147
total_bw: 0
CPU[1]:
bw: 996147
total_bw: 0
CPU[2]:
bw: 996147
total_bw: 0
CPU[3]:
bw: 996147
total_bw: 0
CPU[4]:
bw: 996147
total_bw: 0
CPU[5]:
bw: 996147
total_bw: 0
CPU[6]:
bw: 996147
total_bw: 0
CPU[7]:
bw: 996147
total_bw: 0
I created the debugfs/sched directory in case there's future scheduling
data we would like to add, instead of cluttering up the debugfs root.
There's already a sched_features file, but I did not want to move it
because other tools may need it in its current location.
Before adding this code, I also realized there was a bit of
SCHED_DEBUG code in the kernel/sched/core.c file, and decided to move that
to kernel/sched/debug.c to clean the core.c file up a bit. Those patches
are mostly orthognal to the deadline_bw file, but decided to group them
together here.
Steven Rostedt (Red Hat) (4):
sched: Move sched_feature file setup into debug.c
sched: Move sched_domain_sysctl to debug.c
sched: Move sched_domain_debug into debug.c
sched: Add debugfs/sched/deadline_bw file to show current bandwidths
----
kernel/sched/core.c | 434 +------------------------------------------
kernel/sched/debug.c | 511 +++++++++++++++++++++++++++++++++++++++++++++++++++
kernel/sched/sched.h | 30 +++
3 files changed, 542 insertions(+), 433 deletions(-)
^ permalink raw reply [flat|nested] 11+ messages in thread
* [RFC][PATCH 1/4] sched: Move sched_feature file setup into debug.c
2016-02-01 20:26 [RFC][PATCH 0/4] sched: Display deadline bandwidth and other SCHED_DEBUG clean up Steven Rostedt
@ 2016-02-01 20:26 ` Steven Rostedt
2016-02-01 20:26 ` [RFC][PATCH 2/4] sched: Move sched_domain_sysctl to debug.c Steven Rostedt
` (2 subsequent siblings)
3 siblings, 0 replies; 11+ messages in thread
From: Steven Rostedt @ 2016-02-01 20:26 UTC (permalink / raw)
To: linux-kernel
Cc: Ingo Molnar, Peter Zijlstra, Thomas Gleixner, Juri Lelli,
Clark Williams, Andrew Morton
[-- Attachment #1: 0001-sched-Move-sched_feature-file-setup-into-debug.c.patch --]
[-- Type: text/plain, Size: 6768 bytes --]
From: "Steven Rostedt (Red Hat)" <rostedt@goodmis.org>
As sched_feature is only created when SCHED_DEBUG is enabled, and the file
debug.c is only compiled when SCHED_DEBUG is enabled, it makes sense to move
sched_feature setup into that file and get rid of the #ifdef.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
---
kernel/sched/core.c | 133 ---------------------------------------------------
kernel/sched/debug.c | 131 ++++++++++++++++++++++++++++++++++++++++++++++++++
2 files changed, 131 insertions(+), 133 deletions(-)
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index 63d3a24e081a..c49e64c4675e 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -66,7 +66,6 @@
#include <linux/pagemap.h>
#include <linux/hrtimer.h>
#include <linux/tick.h>
-#include <linux/debugfs.h>
#include <linux/ctype.h>
#include <linux/ftrace.h>
#include <linux/slab.h>
@@ -124,138 +123,6 @@ const_debug unsigned int sysctl_sched_features =
#undef SCHED_FEAT
-#ifdef CONFIG_SCHED_DEBUG
-#define SCHED_FEAT(name, enabled) \
- #name ,
-
-static const char * const sched_feat_names[] = {
-#include "features.h"
-};
-
-#undef SCHED_FEAT
-
-static int sched_feat_show(struct seq_file *m, void *v)
-{
- int i;
-
- for (i = 0; i < __SCHED_FEAT_NR; i++) {
- if (!(sysctl_sched_features & (1UL << i)))
- seq_puts(m, "NO_");
- seq_printf(m, "%s ", sched_feat_names[i]);
- }
- seq_puts(m, "\n");
-
- return 0;
-}
-
-#ifdef HAVE_JUMP_LABEL
-
-#define jump_label_key__true STATIC_KEY_INIT_TRUE
-#define jump_label_key__false STATIC_KEY_INIT_FALSE
-
-#define SCHED_FEAT(name, enabled) \
- jump_label_key__##enabled ,
-
-struct static_key sched_feat_keys[__SCHED_FEAT_NR] = {
-#include "features.h"
-};
-
-#undef SCHED_FEAT
-
-static void sched_feat_disable(int i)
-{
- static_key_disable(&sched_feat_keys[i]);
-}
-
-static void sched_feat_enable(int i)
-{
- static_key_enable(&sched_feat_keys[i]);
-}
-#else
-static void sched_feat_disable(int i) { };
-static void sched_feat_enable(int i) { };
-#endif /* HAVE_JUMP_LABEL */
-
-static int sched_feat_set(char *cmp)
-{
- int i;
- int neg = 0;
-
- if (strncmp(cmp, "NO_", 3) == 0) {
- neg = 1;
- cmp += 3;
- }
-
- for (i = 0; i < __SCHED_FEAT_NR; i++) {
- if (strcmp(cmp, sched_feat_names[i]) == 0) {
- if (neg) {
- sysctl_sched_features &= ~(1UL << i);
- sched_feat_disable(i);
- } else {
- sysctl_sched_features |= (1UL << i);
- sched_feat_enable(i);
- }
- break;
- }
- }
-
- return i;
-}
-
-static ssize_t
-sched_feat_write(struct file *filp, const char __user *ubuf,
- size_t cnt, loff_t *ppos)
-{
- char buf[64];
- char *cmp;
- int i;
- struct inode *inode;
-
- if (cnt > 63)
- cnt = 63;
-
- if (copy_from_user(&buf, ubuf, cnt))
- return -EFAULT;
-
- buf[cnt] = 0;
- cmp = strstrip(buf);
-
- /* Ensure the static_key remains in a consistent state */
- inode = file_inode(filp);
- inode_lock(inode);
- i = sched_feat_set(cmp);
- inode_unlock(inode);
- if (i == __SCHED_FEAT_NR)
- return -EINVAL;
-
- *ppos += cnt;
-
- return cnt;
-}
-
-static int sched_feat_open(struct inode *inode, struct file *filp)
-{
- return single_open(filp, sched_feat_show, NULL);
-}
-
-static const struct file_operations sched_feat_fops = {
- .open = sched_feat_open,
- .write = sched_feat_write,
- .read = seq_read,
- .llseek = seq_lseek,
- .release = single_release,
-};
-
-static __init int sched_init_debug(void)
-{
- debugfs_create_file("sched_features", 0644, NULL, NULL,
- &sched_feat_fops);
-
- return 0;
-}
-late_initcall(sched_init_debug);
-#endif /* CONFIG_SCHED_DEBUG */
-
/*
* Number of tasks to iterate in a single balance run.
* Limited because this is done with IRQs disabled.
diff --git a/kernel/sched/debug.c b/kernel/sched/debug.c
index 641511771ae6..593178be8124 100644
--- a/kernel/sched/debug.c
+++ b/kernel/sched/debug.c
@@ -16,6 +16,7 @@
#include <linux/kallsyms.h>
#include <linux/utsname.h>
#include <linux/mempolicy.h>
+#include <linux/debugfs.h>
#include "sched.h"
@@ -58,6 +59,136 @@ static unsigned long nsec_low(unsigned long long nsec)
#define SPLIT_NS(x) nsec_high(x), nsec_low(x)
+#define SCHED_FEAT(name, enabled) \
+ #name ,
+
+static const char * const sched_feat_names[] = {
+#include "features.h"
+};
+
+#undef SCHED_FEAT
+
+static int sched_feat_show(struct seq_file *m, void *v)
+{
+ int i;
+
+ for (i = 0; i < __SCHED_FEAT_NR; i++) {
+ if (!(sysctl_sched_features & (1UL << i)))
+ seq_puts(m, "NO_");
+ seq_printf(m, "%s ", sched_feat_names[i]);
+ }
+ seq_puts(m, "\n");
+
+ return 0;
+}
+
+#ifdef HAVE_JUMP_LABEL
+
+#define jump_label_key__true STATIC_KEY_INIT_TRUE
+#define jump_label_key__false STATIC_KEY_INIT_FALSE
+
+#define SCHED_FEAT(name, enabled) \
+ jump_label_key__##enabled ,
+
+struct static_key sched_feat_keys[__SCHED_FEAT_NR] = {
+#include "features.h"
+};
+
+#undef SCHED_FEAT
+
+static void sched_feat_disable(int i)
+{
+ static_key_disable(&sched_feat_keys[i]);
+}
+
+static void sched_feat_enable(int i)
+{
+ static_key_enable(&sched_feat_keys[i]);
+}
+#else
+static void sched_feat_disable(int i) { };
+static void sched_feat_enable(int i) { };
+#endif /* HAVE_JUMP_LABEL */
+
+static int sched_feat_set(char *cmp)
+{
+ int i;
+ int neg = 0;
+
+ if (strncmp(cmp, "NO_", 3) == 0) {
+ neg = 1;
+ cmp += 3;
+ }
+
+ for (i = 0; i < __SCHED_FEAT_NR; i++) {
+ if (strcmp(cmp, sched_feat_names[i]) == 0) {
+ if (neg) {
+ sysctl_sched_features &= ~(1UL << i);
+ sched_feat_disable(i);
+ } else {
+ sysctl_sched_features |= (1UL << i);
+ sched_feat_enable(i);
+ }
+ break;
+ }
+ }
+
+ return i;
+}
+
+static ssize_t
+sched_feat_write(struct file *filp, const char __user *ubuf,
+ size_t cnt, loff_t *ppos)
+{
+ char buf[64];
+ char *cmp;
+ int i;
+ struct inode *inode;
+
+ if (cnt > 63)
+ cnt = 63;
+
+ if (copy_from_user(&buf, ubuf, cnt))
+ return -EFAULT;
+
+ buf[cnt] = 0;
+ cmp = strstrip(buf);
+
+ /* Ensure the static_key remains in a consistent state */
+ inode = file_inode(filp);
+ inode_lock(inode);
+ i = sched_feat_set(cmp);
+ inode_unlock(inode);
+ if (i == __SCHED_FEAT_NR)
+ return -EINVAL;
+
+ *ppos += cnt;
+
+ return cnt;
+}
+
+static int sched_feat_open(struct inode *inode, struct file *filp)
+{
+ return single_open(filp, sched_feat_show, NULL);
+}
+
+static const struct file_operations sched_feat_fops = {
+ .open = sched_feat_open,
+ .write = sched_feat_write,
+ .read = seq_read,
+ .llseek = seq_lseek,
+ .release = single_release,
+};
+
+static __init int sched_init_debug(void)
+{
+ debugfs_create_file("sched_features", 0644, NULL, NULL,
+ &sched_feat_fops);
+
+ return 0;
+}
+late_initcall(sched_init_debug);
+
#ifdef CONFIG_FAIR_GROUP_SCHED
static void print_cfs_group_stats(struct seq_file *m, int cpu, struct task_group *tg)
{
--
2.6.4
^ permalink raw reply related [flat|nested] 11+ messages in thread
* [RFC][PATCH 2/4] sched: Move sched_domain_sysctl to debug.c
2016-02-01 20:26 [RFC][PATCH 0/4] sched: Display deadline bandwidth and other SCHED_DEBUG clean up Steven Rostedt
2016-02-01 20:26 ` [RFC][PATCH 1/4] sched: Move sched_feature file setup into debug.c Steven Rostedt
@ 2016-02-01 20:26 ` Steven Rostedt
2016-02-01 20:26 ` [RFC][PATCH 3/4] sched: Move sched_domain_debug into debug.c Steven Rostedt
2016-02-01 20:26 ` [RFC][PATCH 4/4] sched: Add debugfs/sched/deadline_bw file to show current bandwidths Steven Rostedt
3 siblings, 0 replies; 11+ messages in thread
From: Steven Rostedt @ 2016-02-01 20:26 UTC (permalink / raw)
To: linux-kernel
Cc: Ingo Molnar, Peter Zijlstra, Thomas Gleixner, Juri Lelli,
Clark Williams, Andrew Morton
[-- Attachment #1: 0002-sched-Move-sched_domain_sysctl-to-debug.c.patch --]
[-- Type: text/plain, Size: 11900 bytes --]
From: "Steven Rostedt (Red Hat)" <rostedt@goodmis.org>
The sched_domain_sysctl setup is only enabled when SCHED_DEBUG is
configured. As debug.c is only compiled when SCHED_DEBUG is configured as
well, move the setup of sched_domain_sysctl into that file.
Note, the (un)register_sched_domain_sysctl() functions had to be changed
from static to allow access to them from core.c.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
---
kernel/sched/core.c | 178 ---------------------------------------------------
kernel/sched/debug.c | 173 +++++++++++++++++++++++++++++++++++++++++++++++++
kernel/sched/sched.h | 13 ++++
3 files changed, 186 insertions(+), 178 deletions(-)
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index c49e64c4675e..9c6d976d94f0 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -70,7 +70,6 @@
#include <linux/ftrace.h>
#include <linux/slab.h>
#include <linux/init_task.h>
-#include <linux/binfmts.h>
#include <linux/context_tracking.h>
#include <linux/compiler.h>
@@ -5272,183 +5271,6 @@ static void migrate_tasks(struct rq *dead_rq)
}
#endif /* CONFIG_HOTPLUG_CPU */
-#if defined(CONFIG_SCHED_DEBUG) && defined(CONFIG_SYSCTL)
-
-static struct ctl_table sd_ctl_dir[] = {
- {
- .procname = "sched_domain",
- .mode = 0555,
- },
- {}
-};
-
-static struct ctl_table sd_ctl_root[] = {
- {
- .procname = "kernel",
- .mode = 0555,
- .child = sd_ctl_dir,
- },
- {}
-};
-
-static struct ctl_table *sd_alloc_ctl_entry(int n)
-{
- struct ctl_table *entry =
- kcalloc(n, sizeof(struct ctl_table), GFP_KERNEL);
-
- return entry;
-}
-
-static void sd_free_ctl_entry(struct ctl_table **tablep)
-{
- struct ctl_table *entry;
-
- /*
- * In the intermediate directories, both the child directory and
- * procname are dynamically allocated and could fail but the mode
- * will always be set. In the lowest directory the names are
- * static strings and all have proc handlers.
- */
- for (entry = *tablep; entry->mode; entry++) {
- if (entry->child)
- sd_free_ctl_entry(&entry->child);
- if (entry->proc_handler == NULL)
- kfree(entry->procname);
- }
-
- kfree(*tablep);
- *tablep = NULL;
-}
-
-static int min_load_idx = 0;
-static int max_load_idx = CPU_LOAD_IDX_MAX-1;
-
-static void
-set_table_entry(struct ctl_table *entry,
- const char *procname, void *data, int maxlen,
- umode_t mode, proc_handler *proc_handler,
- bool load_idx)
-{
- entry->procname = procname;
- entry->data = data;
- entry->maxlen = maxlen;
- entry->mode = mode;
- entry->proc_handler = proc_handler;
-
- if (load_idx) {
- entry->extra1 = &min_load_idx;
- entry->extra2 = &max_load_idx;
- }
-}
-
-static struct ctl_table *
-sd_alloc_ctl_domain_table(struct sched_domain *sd)
-{
- struct ctl_table *table = sd_alloc_ctl_entry(14);
-
- if (table == NULL)
- return NULL;
-
- set_table_entry(&table[0], "min_interval", &sd->min_interval,
- sizeof(long), 0644, proc_doulongvec_minmax, false);
- set_table_entry(&table[1], "max_interval", &sd->max_interval,
- sizeof(long), 0644, proc_doulongvec_minmax, false);
- set_table_entry(&table[2], "busy_idx", &sd->busy_idx,
- sizeof(int), 0644, proc_dointvec_minmax, true);
- set_table_entry(&table[3], "idle_idx", &sd->idle_idx,
- sizeof(int), 0644, proc_dointvec_minmax, true);
- set_table_entry(&table[4], "newidle_idx", &sd->newidle_idx,
- sizeof(int), 0644, proc_dointvec_minmax, true);
- set_table_entry(&table[5], "wake_idx", &sd->wake_idx,
- sizeof(int), 0644, proc_dointvec_minmax, true);
- set_table_entry(&table[6], "forkexec_idx", &sd->forkexec_idx,
- sizeof(int), 0644, proc_dointvec_minmax, true);
- set_table_entry(&table[7], "busy_factor", &sd->busy_factor,
- sizeof(int), 0644, proc_dointvec_minmax, false);
- set_table_entry(&table[8], "imbalance_pct", &sd->imbalance_pct,
- sizeof(int), 0644, proc_dointvec_minmax, false);
- set_table_entry(&table[9], "cache_nice_tries",
- &sd->cache_nice_tries,
- sizeof(int), 0644, proc_dointvec_minmax, false);
- set_table_entry(&table[10], "flags", &sd->flags,
- sizeof(int), 0644, proc_dointvec_minmax, false);
- set_table_entry(&table[11], "max_newidle_lb_cost",
- &sd->max_newidle_lb_cost,
- sizeof(long), 0644, proc_doulongvec_minmax, false);
- set_table_entry(&table[12], "name", sd->name,
- CORENAME_MAX_SIZE, 0444, proc_dostring, false);
- /* &table[13] is terminator */
-
- return table;
-}
-
-static struct ctl_table *sd_alloc_ctl_cpu_table(int cpu)
-{
- struct ctl_table *entry, *table;
- struct sched_domain *sd;
- int domain_num = 0, i;
- char buf[32];
-
- for_each_domain(cpu, sd)
- domain_num++;
- entry = table = sd_alloc_ctl_entry(domain_num + 1);
- if (table == NULL)
- return NULL;
-
- i = 0;
- for_each_domain(cpu, sd) {
- snprintf(buf, 32, "domain%d", i);
- entry->procname = kstrdup(buf, GFP_KERNEL);
- entry->mode = 0555;
- entry->child = sd_alloc_ctl_domain_table(sd);
- entry++;
- i++;
- }
- return table;
-}
-
-static struct ctl_table_header *sd_sysctl_header;
-static void register_sched_domain_sysctl(void)
-{
- int i, cpu_num = num_possible_cpus();
- struct ctl_table *entry = sd_alloc_ctl_entry(cpu_num + 1);
- char buf[32];
-
- WARN_ON(sd_ctl_dir[0].child);
- sd_ctl_dir[0].child = entry;
-
- if (entry == NULL)
- return;
-
- for_each_possible_cpu(i) {
- snprintf(buf, 32, "cpu%d", i);
- entry->procname = kstrdup(buf, GFP_KERNEL);
- entry->mode = 0555;
- entry->child = sd_alloc_ctl_cpu_table(i);
- entry++;
- }
-
- WARN_ON(sd_sysctl_header);
- sd_sysctl_header = register_sysctl_table(sd_ctl_root);
-}
-
-/* may be called multiple times per register */
-static void unregister_sched_domain_sysctl(void)
-{
- unregister_sysctl_table(sd_sysctl_header);
- sd_sysctl_header = NULL;
- if (sd_ctl_dir[0].child)
- sd_free_ctl_entry(&sd_ctl_dir[0].child);
-}
-#else
-static void register_sched_domain_sysctl(void)
-{
-}
-static void unregister_sched_domain_sysctl(void)
-{
-}
-#endif /* CONFIG_SCHED_DEBUG && CONFIG_SYSCTL */
-
static void set_rq_online(struct rq *rq)
{
if (!rq->online) {
diff --git a/kernel/sched/debug.c b/kernel/sched/debug.c
index 593178be8124..a5625e793d15 100644
--- a/kernel/sched/debug.c
+++ b/kernel/sched/debug.c
@@ -189,6 +189,179 @@ static __init int sched_init_debug(void)
}
late_initcall(sched_init_debug);
+#ifdef CONFIG_SMP
+
+#ifdef CONFIG_SYSCTL
+
+static struct ctl_table sd_ctl_dir[] = {
+ {
+ .procname = "sched_domain",
+ .mode = 0555,
+ },
+ {}
+};
+
+static struct ctl_table sd_ctl_root[] = {
+ {
+ .procname = "kernel",
+ .mode = 0555,
+ .child = sd_ctl_dir,
+ },
+ {}
+};
+
+static struct ctl_table *sd_alloc_ctl_entry(int n)
+{
+ struct ctl_table *entry =
+ kcalloc(n, sizeof(struct ctl_table), GFP_KERNEL);
+
+ return entry;
+}
+
+static void sd_free_ctl_entry(struct ctl_table **tablep)
+{
+ struct ctl_table *entry;
+
+ /*
+ * In the intermediate directories, both the child directory and
+ * procname are dynamically allocated and could fail but the mode
+ * will always be set. In the lowest directory the names are
+ * static strings and all have proc handlers.
+ */
+ for (entry = *tablep; entry->mode; entry++) {
+ if (entry->child)
+ sd_free_ctl_entry(&entry->child);
+ if (entry->proc_handler == NULL)
+ kfree(entry->procname);
+ }
+
+ kfree(*tablep);
+ *tablep = NULL;
+}
+
+static int min_load_idx = 0;
+static int max_load_idx = CPU_LOAD_IDX_MAX-1;
+
+static void
+set_table_entry(struct ctl_table *entry,
+ const char *procname, void *data, int maxlen,
+ umode_t mode, proc_handler *proc_handler,
+ bool load_idx)
+{
+ entry->procname = procname;
+ entry->data = data;
+ entry->maxlen = maxlen;
+ entry->mode = mode;
+ entry->proc_handler = proc_handler;
+
+ if (load_idx) {
+ entry->extra1 = &min_load_idx;
+ entry->extra2 = &max_load_idx;
+ }
+}
+
+static struct ctl_table *
+sd_alloc_ctl_domain_table(struct sched_domain *sd)
+{
+ struct ctl_table *table = sd_alloc_ctl_entry(14);
+
+ if (table == NULL)
+ return NULL;
+
+ set_table_entry(&table[0], "min_interval", &sd->min_interval,
+ sizeof(long), 0644, proc_doulongvec_minmax, false);
+ set_table_entry(&table[1], "max_interval", &sd->max_interval,
+ sizeof(long), 0644, proc_doulongvec_minmax, false);
+ set_table_entry(&table[2], "busy_idx", &sd->busy_idx,
+ sizeof(int), 0644, proc_dointvec_minmax, true);
+ set_table_entry(&table[3], "idle_idx", &sd->idle_idx,
+ sizeof(int), 0644, proc_dointvec_minmax, true);
+ set_table_entry(&table[4], "newidle_idx", &sd->newidle_idx,
+ sizeof(int), 0644, proc_dointvec_minmax, true);
+ set_table_entry(&table[5], "wake_idx", &sd->wake_idx,
+ sizeof(int), 0644, proc_dointvec_minmax, true);
+ set_table_entry(&table[6], "forkexec_idx", &sd->forkexec_idx,
+ sizeof(int), 0644, proc_dointvec_minmax, true);
+ set_table_entry(&table[7], "busy_factor", &sd->busy_factor,
+ sizeof(int), 0644, proc_dointvec_minmax, false);
+ set_table_entry(&table[8], "imbalance_pct", &sd->imbalance_pct,
+ sizeof(int), 0644, proc_dointvec_minmax, false);
+ set_table_entry(&table[9], "cache_nice_tries",
+ &sd->cache_nice_tries,
+ sizeof(int), 0644, proc_dointvec_minmax, false);
+ set_table_entry(&table[10], "flags", &sd->flags,
+ sizeof(int), 0644, proc_dointvec_minmax, false);
+ set_table_entry(&table[11], "max_newidle_lb_cost",
+ &sd->max_newidle_lb_cost,
+ sizeof(long), 0644, proc_doulongvec_minmax, false);
+ set_table_entry(&table[12], "name", sd->name,
+ CORENAME_MAX_SIZE, 0444, proc_dostring, false);
+ /* &table[13] is terminator */
+
+ return table;
+}
+
+static struct ctl_table *sd_alloc_ctl_cpu_table(int cpu)
+{
+ struct ctl_table *entry, *table;
+ struct sched_domain *sd;
+ int domain_num = 0, i;
+ char buf[32];
+
+ for_each_domain(cpu, sd)
+ domain_num++;
+ entry = table = sd_alloc_ctl_entry(domain_num + 1);
+ if (table == NULL)
+ return NULL;
+
+ i = 0;
+ for_each_domain(cpu, sd) {
+ snprintf(buf, 32, "domain%d", i);
+ entry->procname = kstrdup(buf, GFP_KERNEL);
+ entry->mode = 0555;
+ entry->child = sd_alloc_ctl_domain_table(sd);
+ entry++;
+ i++;
+ }
+ return table;
+}
+
+static struct ctl_table_header *sd_sysctl_header;
+void register_sched_domain_sysctl(void)
+{
+ int i, cpu_num = num_possible_cpus();
+ struct ctl_table *entry = sd_alloc_ctl_entry(cpu_num + 1);
+ char buf[32];
+
+ WARN_ON(sd_ctl_dir[0].child);
+ sd_ctl_dir[0].child = entry;
+
+ if (entry == NULL)
+ return;
+
+ for_each_possible_cpu(i) {
+ snprintf(buf, 32, "cpu%d", i);
+ entry->procname = kstrdup(buf, GFP_KERNEL);
+ entry->mode = 0555;
+ entry->child = sd_alloc_ctl_cpu_table(i);
+ entry++;
+ }
+
+ WARN_ON(sd_sysctl_header);
+ sd_sysctl_header = register_sysctl_table(sd_ctl_root);
+}
+
+/* may be called multiple times per register */
+void unregister_sched_domain_sysctl(void)
+{
+ unregister_sysctl_table(sd_sysctl_header);
+ sd_sysctl_header = NULL;
+ if (sd_ctl_dir[0].child)
+ sd_free_ctl_entry(&sd_ctl_dir[0].child);
+}
+#endif /* CONFIG_SYSCTL */
+#endif /* CONFIG_SMP */
+
#ifdef CONFIG_FAIR_GROUP_SCHED
static void print_cfs_group_stats(struct seq_file *m, int cpu, struct task_group *tg)
{
diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
index 10f16374df7f..b146f5298cb1 100644
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -3,6 +3,7 @@
#include <linux/sched/sysctl.h>
#include <linux/sched/rt.h>
#include <linux/sched/deadline.h>
+#include <linux/binfmts.h>
#include <linux/mutex.h>
#include <linux/spinlock.h>
#include <linux/stop_machine.h>
@@ -909,6 +910,18 @@ static inline unsigned int group_first_cpu(struct sched_group *group)
extern int group_balance_cpu(struct sched_group *sg);
+#if defined(CONFIG_SCHED_DEBUG) && defined(CONFIG_SYSCTL)
+void register_sched_domain_sysctl(void);
+void unregister_sched_domain_sysctl(void);
+#else
+static inline void register_sched_domain_sysctl(void)
+{
+}
+static inline void unregister_sched_domain_sysctl(void)
+{
+}
+#endif
+
#else
static inline void sched_ttwu_pending(void) { }
--
2.6.4
^ permalink raw reply related [flat|nested] 11+ messages in thread
* [RFC][PATCH 3/4] sched: Move sched_domain_debug into debug.c
2016-02-01 20:26 [RFC][PATCH 0/4] sched: Display deadline bandwidth and other SCHED_DEBUG clean up Steven Rostedt
2016-02-01 20:26 ` [RFC][PATCH 1/4] sched: Move sched_feature file setup into debug.c Steven Rostedt
2016-02-01 20:26 ` [RFC][PATCH 2/4] sched: Move sched_domain_sysctl to debug.c Steven Rostedt
@ 2016-02-01 20:26 ` Steven Rostedt
2016-02-01 20:26 ` [RFC][PATCH 4/4] sched: Add debugfs/sched/deadline_bw file to show current bandwidths Steven Rostedt
3 siblings, 0 replies; 11+ messages in thread
From: Steven Rostedt @ 2016-02-01 20:26 UTC (permalink / raw)
To: linux-kernel
Cc: Ingo Molnar, Peter Zijlstra, Thomas Gleixner, Juri Lelli,
Clark Williams, Andrew Morton
[-- Attachment #1: 0003-sched-Move-sched_domain_debug-into-debug.c.patch --]
[-- Type: text/plain, Size: 7737 bytes --]
From: "Steven Rostedt (Red Hat)" <rostedt@goodmis.org>
The sched_domain_debug code only gets compiled if SCHED_DEBUG is configured,
and debug.c only gets compiled if SCHED_DEBUG is also configured. Move the
code into debug.c and remove the #ifdefs in core.c. Some functions and
variables need to not be static anymore, but everything is still local
within the kernel/sched directory.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
---
kernel/sched/core.c | 123 +--------------------------------------------------
kernel/sched/debug.c | 107 ++++++++++++++++++++++++++++++++++++++++++++
kernel/sched/sched.h | 17 +++++++
3 files changed, 125 insertions(+), 122 deletions(-)
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index 9c6d976d94f0..d0810e997db3 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -5431,128 +5431,7 @@ static int __init migration_init(void)
}
early_initcall(migration_init);
-static cpumask_var_t sched_domains_tmpmask; /* sched_domains_mutex */
-
-#ifdef CONFIG_SCHED_DEBUG
-
-static __read_mostly int sched_debug_enabled;
-
-static int __init sched_debug_setup(char *str)
-{
- sched_debug_enabled = 1;
-
- return 0;
-}
-early_param("sched_debug", sched_debug_setup);
-
-static inline bool sched_debug(void)
-{
- return sched_debug_enabled;
-}
-
-static int sched_domain_debug_one(struct sched_domain *sd, int cpu, int level,
- struct cpumask *groupmask)
-{
- struct sched_group *group = sd->groups;
-
- cpumask_clear(groupmask);
-
- printk(KERN_DEBUG "%*s domain %d: ", level, "", level);
-
- if (!(sd->flags & SD_LOAD_BALANCE)) {
- printk("does not load-balance\n");
- if (sd->parent)
- printk(KERN_ERR "ERROR: !SD_LOAD_BALANCE domain"
- " has parent");
- return -1;
- }
-
- printk(KERN_CONT "span %*pbl level %s\n",
- cpumask_pr_args(sched_domain_span(sd)), sd->name);
-
- if (!cpumask_test_cpu(cpu, sched_domain_span(sd))) {
- printk(KERN_ERR "ERROR: domain->span does not contain "
- "CPU%d\n", cpu);
- }
- if (!cpumask_test_cpu(cpu, sched_group_cpus(group))) {
- printk(KERN_ERR "ERROR: domain->groups does not contain"
- " CPU%d\n", cpu);
- }
-
- printk(KERN_DEBUG "%*s groups:", level + 1, "");
- do {
- if (!group) {
- printk("\n");
- printk(KERN_ERR "ERROR: group is NULL\n");
- break;
- }
-
- if (!cpumask_weight(sched_group_cpus(group))) {
- printk(KERN_CONT "\n");
- printk(KERN_ERR "ERROR: empty group\n");
- break;
- }
-
- if (!(sd->flags & SD_OVERLAP) &&
- cpumask_intersects(groupmask, sched_group_cpus(group))) {
- printk(KERN_CONT "\n");
- printk(KERN_ERR "ERROR: repeated CPUs\n");
- break;
- }
-
- cpumask_or(groupmask, groupmask, sched_group_cpus(group));
-
- printk(KERN_CONT " %*pbl",
- cpumask_pr_args(sched_group_cpus(group)));
- if (group->sgc->capacity != SCHED_CAPACITY_SCALE) {
- printk(KERN_CONT " (cpu_capacity = %d)",
- group->sgc->capacity);
- }
-
- group = group->next;
- } while (group != sd->groups);
- printk(KERN_CONT "\n");
-
- if (!cpumask_equal(sched_domain_span(sd), groupmask))
- printk(KERN_ERR "ERROR: groups don't span domain->span\n");
-
- if (sd->parent &&
- !cpumask_subset(groupmask, sched_domain_span(sd->parent)))
- printk(KERN_ERR "ERROR: parent span is not a superset "
- "of domain->span\n");
- return 0;
-}
-
-static void sched_domain_debug(struct sched_domain *sd, int cpu)
-{
- int level = 0;
-
- if (!sched_debug_enabled)
- return;
-
- if (!sd) {
- printk(KERN_DEBUG "CPU%d attaching NULL sched-domain.\n", cpu);
- return;
- }
-
- printk(KERN_DEBUG "CPU%d attaching sched-domain:\n", cpu);
-
- for (;;) {
- if (sched_domain_debug_one(sd, cpu, level, sched_domains_tmpmask))
- break;
- level++;
- sd = sd->parent;
- if (!sd)
- break;
- }
-}
-#else /* !CONFIG_SCHED_DEBUG */
-# define sched_domain_debug(sd, cpu) do { } while (0)
-static inline bool sched_debug(void)
-{
- return false;
-}
-#endif /* CONFIG_SCHED_DEBUG */
+cpumask_var_t sched_domains_tmpmask; /* sched_domains_mutex */
static int sd_degenerate(struct sched_domain *sd)
{
diff --git a/kernel/sched/debug.c b/kernel/sched/debug.c
index a5625e793d15..68838495624f 100644
--- a/kernel/sched/debug.c
+++ b/kernel/sched/debug.c
@@ -360,6 +360,113 @@ void unregister_sched_domain_sysctl(void)
sd_free_ctl_entry(&sd_ctl_dir[0].child);
}
#endif /* CONFIG_SYSCTL */
+
+int sched_debug_enabled __read_mostly;
+
+static int __init sched_debug_setup(char *str)
+{
+ sched_debug_enabled = 1;
+
+ return 0;
+}
+early_param("sched_debug", sched_debug_setup);
+
+int sched_domain_debug_one(struct sched_domain *sd, int cpu, int level,
+ struct cpumask *groupmask)
+{
+ struct sched_group *group = sd->groups;
+
+ cpumask_clear(groupmask);
+
+ printk(KERN_DEBUG "%*s domain %d: ", level, "", level);
+
+ if (!(sd->flags & SD_LOAD_BALANCE)) {
+ printk("does not load-balance\n");
+ if (sd->parent)
+ printk(KERN_ERR "ERROR: !SD_LOAD_BALANCE domain"
+ " has parent");
+ return -1;
+ }
+
+ printk(KERN_CONT "span %*pbl level %s\n",
+ cpumask_pr_args(sched_domain_span(sd)), sd->name);
+
+ if (!cpumask_test_cpu(cpu, sched_domain_span(sd))) {
+ printk(KERN_ERR "ERROR: domain->span does not contain "
+ "CPU%d\n", cpu);
+ }
+ if (!cpumask_test_cpu(cpu, sched_group_cpus(group))) {
+ printk(KERN_ERR "ERROR: domain->groups does not contain"
+ " CPU%d\n", cpu);
+ }
+
+ printk(KERN_DEBUG "%*s groups:", level + 1, "");
+ do {
+ if (!group) {
+ printk("\n");
+ printk(KERN_ERR "ERROR: group is NULL\n");
+ break;
+ }
+
+ if (!cpumask_weight(sched_group_cpus(group))) {
+ printk(KERN_CONT "\n");
+ printk(KERN_ERR "ERROR: empty group\n");
+ break;
+ }
+
+ if (!(sd->flags & SD_OVERLAP) &&
+ cpumask_intersects(groupmask, sched_group_cpus(group))) {
+ printk(KERN_CONT "\n");
+ printk(KERN_ERR "ERROR: repeated CPUs\n");
+ break;
+ }
+
+ cpumask_or(groupmask, groupmask, sched_group_cpus(group));
+
+ printk(KERN_CONT " %*pbl",
+ cpumask_pr_args(sched_group_cpus(group)));
+ if (group->sgc->capacity != SCHED_CAPACITY_SCALE) {
+ printk(KERN_CONT " (cpu_capacity = %d)",
+ group->sgc->capacity);
+ }
+
+ group = group->next;
+ } while (group != sd->groups);
+ printk(KERN_CONT "\n");
+
+ if (!cpumask_equal(sched_domain_span(sd), groupmask))
+ printk(KERN_ERR "ERROR: groups don't span domain->span\n");
+
+ if (sd->parent &&
+ !cpumask_subset(groupmask, sched_domain_span(sd->parent)))
+ printk(KERN_ERR "ERROR: parent span is not a superset "
+ "of domain->span\n");
+ return 0;
+}
+
+void sched_domain_debug(struct sched_domain *sd, int cpu)
+{
+ int level = 0;
+
+ if (!sched_debug_enabled)
+ return;
+
+ if (!sd) {
+ printk(KERN_DEBUG "CPU%d attaching NULL sched-domain.\n", cpu);
+ return;
+ }
+
+ printk(KERN_DEBUG "CPU%d attaching sched-domain:\n", cpu);
+
+ for (;;) {
+ if (sched_domain_debug_one(sd, cpu, level, sched_domains_tmpmask))
+ break;
+ level++;
+ sd = sd->parent;
+ if (!sd)
+ break;
+ }
+}
#endif /* CONFIG_SMP */
#ifdef CONFIG_FAIR_GROUP_SCHED
diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
index b146f5298cb1..bfa2dd97fe0b 100644
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -922,6 +922,23 @@ static inline void unregister_sched_domain_sysctl(void)
}
#endif
+extern cpumask_var_t sched_domains_tmpmask; /* sched_domains_mutex */
+
+#ifdef CONFIG_SCHED_DEBUG
+extern int sched_debug_enabled;
+void sched_domain_debug(struct sched_domain *sd, int cpu);
+static inline bool sched_debug(void)
+{
+ return sched_debug_enabled;
+}
+#else
+# define sched_domain_debug(sd, cpu) do { } while (0)
+static inline bool sched_debug(void)
+{
+ return false;
+}
+#endif
+
#else
static inline void sched_ttwu_pending(void) { }
--
2.6.4
^ permalink raw reply related [flat|nested] 11+ messages in thread
* [RFC][PATCH 4/4] sched: Add debugfs/sched/deadline_bw file to show current bandwidths
2016-02-01 20:26 [RFC][PATCH 0/4] sched: Display deadline bandwidth and other SCHED_DEBUG clean up Steven Rostedt
` (2 preceding siblings ...)
2016-02-01 20:26 ` [RFC][PATCH 3/4] sched: Move sched_domain_debug into debug.c Steven Rostedt
@ 2016-02-01 20:26 ` Steven Rostedt
2016-02-01 22:17 ` Peter Zijlstra
2016-02-01 22:18 ` Peter Zijlstra
3 siblings, 2 replies; 11+ messages in thread
From: Steven Rostedt @ 2016-02-01 20:26 UTC (permalink / raw)
To: linux-kernel
Cc: Ingo Molnar, Peter Zijlstra, Thomas Gleixner, Juri Lelli,
Clark Williams, Andrew Morton
[-- Attachment #1: 0004-sched-Add-debugfs-sched-deadline_bw-file-to-show-cur.patch --]
[-- Type: text/plain, Size: 3035 bytes --]
From: "Steven Rostedt (Red Hat)" <rostedt@goodmis.org>
Add a /sys/kernel/debug/sched directory and place a deadline_bw file there
that shows the current bandwidths of the CPUs for SCHED_DEADLINE tasks.
# cat /sys/kernel/debug/sched/deadline_bw
CPU[0]:
bw: 996147
total_bw: 0
CPU[1]:
bw: 996147
total_bw: 0
CPU[2]:
bw: 996147
total_bw: 0
CPU[3]:
bw: 996147
total_bw: 0
CPU[4]:
bw: 996147
total_bw: 0
CPU[5]:
bw: 996147
total_bw: 0
CPU[6]:
bw: 996147
total_bw: 0
CPU[7]:
bw: 996147
total_bw: 0
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
---
kernel/sched/debug.c | 100 +++++++++++++++++++++++++++++++++++++++++++++++++++
1 file changed, 100 insertions(+)
diff --git a/kernel/sched/debug.c b/kernel/sched/debug.c
index 68838495624f..a6ce364ba9b9 100644
--- a/kernel/sched/debug.c
+++ b/kernel/sched/debug.c
@@ -1070,3 +1070,103 @@ void proc_sched_set_task(struct task_struct *p)
memset(&p->se.statistics, 0, sizeof(p->se.statistics));
#endif
}
+
+#ifdef CONFIG_DEBUG_FS
+
+
+static void *dl_next(struct seq_file *m, void *v, loff_t *pos)
+{
+ long cpu = (long)v;
+
+ /* to keep from returning zero, v is always cpu + 1 */
+ cpu--;
+ cpu = cpumask_next(cpu, cpu_possible_mask);
+
+ (*pos)++;
+
+ if (cpu >= num_possible_cpus())
+ cpu = 0;
+ else
+ cpu++;
+
+ return (void *)cpu;
+}
+
+static void *dl_start(struct seq_file *m, loff_t *pos)
+{
+ long cpu = 1;
+ loff_t l = 0;
+
+ for (; cpu && l < *pos; cpu = (long)dl_next(m, (void *)cpu, &l))
+ ;
+ return (void *)cpu;
+}
+
+static void dl_stop(struct seq_file *m, void *p)
+{
+}
+
+static int dl_show(struct seq_file *m, void *v)
+{
+ struct rq *rq;
+ long cpu = (long)v;
+ struct dl_bw *dl_bw;
+
+ if (!cpu)
+ return 0;
+
+ /* to keep from returning zero, v is always cpu + 1 */
+ cpu--;
+ rq = cpu_rq(cpu);
+ raw_spin_lock_irq(&rq->lock);
+#ifdef CONFIG_SMP
+ dl_bw = &rq->rd->dl_bw;
+#else
+ dl_bw = &rq->dl.dl_bw;
+#endif
+ seq_printf(m, "CPU[%ld]:\n", cpu);
+ seq_printf(m, " bw: %lld\n", dl_bw->bw);
+ seq_printf(m, " total_bw: %lld\n", dl_bw->total_bw);
+ raw_spin_unlock_irq(&rq->lock);
+
+ return 0;
+}
+
+static const struct seq_operations deadline_bw_seq_ops = {
+ .start = dl_start,
+ .next = dl_next,
+ .stop = dl_stop,
+ .show = dl_show,
+};
+
+static int deadline_bw_open(struct inode *inode, struct file *file)
+{
+ int ret;
+
+ ret = seq_open(file, &deadline_bw_seq_ops);
+ return ret;
+}
+
+static const struct file_operations deadline_bw_fops = {
+ .open = deadline_bw_open,
+ .read = seq_read,
+ .llseek = no_llseek,
+};
+
+static struct dentry *sched_debugfs;
+
+static __init int sched_init_debugfs(void)
+{
+ sched_debugfs = debugfs_create_dir("sched", NULL);
+
+ if (WARN_ON(!sched_debugfs)) {
+ return 0;
+ }
+
+ debugfs_create_file("deadline_bw", 0444, sched_debugfs, NULL,
+ &deadline_bw_fops);
+ return 0;
+}
+fs_initcall(sched_init_debugfs);
+
+#endif /*CONFIG_DEBUG_FS */
--
2.6.4
^ permalink raw reply related [flat|nested] 11+ messages in thread
* Re: [RFC][PATCH 4/4] sched: Add debugfs/sched/deadline_bw file to show current bandwidths
2016-02-01 20:26 ` [RFC][PATCH 4/4] sched: Add debugfs/sched/deadline_bw file to show current bandwidths Steven Rostedt
@ 2016-02-01 22:17 ` Peter Zijlstra
2016-02-01 22:38 ` Steven Rostedt
2016-02-01 22:18 ` Peter Zijlstra
1 sibling, 1 reply; 11+ messages in thread
From: Peter Zijlstra @ 2016-02-01 22:17 UTC (permalink / raw)
To: Steven Rostedt
Cc: linux-kernel, Ingo Molnar, Thomas Gleixner, Juri Lelli,
Clark Williams, Andrew Morton
On Mon, Feb 01, 2016 at 03:26:28PM -0500, Steven Rostedt wrote:
> From: "Steven Rostedt (Red Hat)" <rostedt@goodmis.org>
>
> Add a /sys/kernel/debug/sched directory and place a deadline_bw file there
> that shows the current bandwidths of the CPUs for SCHED_DEADLINE tasks.
>
> # cat /sys/kernel/debug/sched/deadline_bw
> CPU[0]:
> bw: 996147
> total_bw: 0
> CPU[1]:
> bw: 996147
> total_bw: 0
> CPU[2]:
> bw: 996147
> total_bw: 0
> CPU[3]:
> bw: 996147
> total_bw: 0
> CPU[4]:
> bw: 996147
> total_bw: 0
> CPU[5]:
> bw: 996147
> total_bw: 0
> CPU[6]:
> bw: 996147
> total_bw: 0
> CPU[7]:
> bw: 996147
> total_bw: 0
Why not add this to print_dl_rq() ?
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [RFC][PATCH 4/4] sched: Add debugfs/sched/deadline_bw file to show current bandwidths
2016-02-01 20:26 ` [RFC][PATCH 4/4] sched: Add debugfs/sched/deadline_bw file to show current bandwidths Steven Rostedt
2016-02-01 22:17 ` Peter Zijlstra
@ 2016-02-01 22:18 ` Peter Zijlstra
1 sibling, 0 replies; 11+ messages in thread
From: Peter Zijlstra @ 2016-02-01 22:18 UTC (permalink / raw)
To: Steven Rostedt
Cc: linux-kernel, Ingo Molnar, Thomas Gleixner, Juri Lelli,
Clark Williams, Andrew Morton
On Mon, Feb 01, 2016 at 03:26:28PM -0500, Steven Rostedt wrote:
> +static void *dl_next(struct seq_file *m, void *v, loff_t *pos)
> +{
> + long cpu = (long)v;
> +
> + /* to keep from returning zero, v is always cpu + 1 */
> + cpu--;
> + cpu = cpumask_next(cpu, cpu_possible_mask);
> +
> + (*pos)++;
> +
> + if (cpu >= num_possible_cpus())
And that wants to be: nr_cpu_ids, imagine the bitmap having holes in.
> + cpu = 0;
> + else
> + cpu++;
> +
> + return (void *)cpu;
> +}
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [RFC][PATCH 4/4] sched: Add debugfs/sched/deadline_bw file to show current bandwidths
2016-02-01 22:17 ` Peter Zijlstra
@ 2016-02-01 22:38 ` Steven Rostedt
2016-02-01 22:41 ` Peter Zijlstra
0 siblings, 1 reply; 11+ messages in thread
From: Steven Rostedt @ 2016-02-01 22:38 UTC (permalink / raw)
To: Peter Zijlstra
Cc: linux-kernel, Ingo Molnar, Thomas Gleixner, Juri Lelli,
Clark Williams, Andrew Morton
On Mon, 1 Feb 2016 23:17:20 +0100
Peter Zijlstra <peterz@infradead.org> wrote:
> On Mon, Feb 01, 2016 at 03:26:28PM -0500, Steven Rostedt wrote:
> > From: "Steven Rostedt (Red Hat)" <rostedt@goodmis.org>
> >
> > Add a /sys/kernel/debug/sched directory and place a deadline_bw file there
> > that shows the current bandwidths of the CPUs for SCHED_DEADLINE tasks.
> >
> > # cat /sys/kernel/debug/sched/deadline_bw
> > CPU[0]:
> > bw: 996147
> > total_bw: 0
> > CPU[1]:
> > bw: 996147
> > total_bw: 0
> > CPU[2]:
> > bw: 996147
> > total_bw: 0
> > CPU[3]:
> > bw: 996147
> > total_bw: 0
> > CPU[4]:
> > bw: 996147
> > total_bw: 0
> > CPU[5]:
> > bw: 996147
> > total_bw: 0
> > CPU[6]:
> > bw: 996147
> > total_bw: 0
> > CPU[7]:
> > bw: 996147
> > total_bw: 0
>
> Why not add this to print_dl_rq() ?
Because I didn't see that ;-)
That could work, although, Damn there's a lot of info there already.
I'll play with that.
Regardless of this final patch, what do you think of the first three? I
can send them as a separate series, just to get more of the SCHED_DEBUG
code into debug.c.
-- Steve
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [RFC][PATCH 4/4] sched: Add debugfs/sched/deadline_bw file to show current bandwidths
2016-02-01 22:38 ` Steven Rostedt
@ 2016-02-01 22:41 ` Peter Zijlstra
2016-02-03 10:12 ` Ingo Molnar
0 siblings, 1 reply; 11+ messages in thread
From: Peter Zijlstra @ 2016-02-01 22:41 UTC (permalink / raw)
To: Steven Rostedt
Cc: linux-kernel, Ingo Molnar, Thomas Gleixner, Juri Lelli,
Clark Williams, Andrew Morton
On Mon, Feb 01, 2016 at 05:38:09PM -0500, Steven Rostedt wrote:
> Regardless of this final patch, what do you think of the first three? I
> can send them as a separate series, just to get more of the SCHED_DEBUG
> code into debug.c.
The first two looked ok, the third I had a vague niggle it might be part
of something bigger and tearing it up like this might make it harder.
Other than that, I didn't really look at them for more than a few
seconds. I'll give them more of a look tomorrow.
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [RFC][PATCH 4/4] sched: Add debugfs/sched/deadline_bw file to show current bandwidths
2016-02-01 22:41 ` Peter Zijlstra
@ 2016-02-03 10:12 ` Ingo Molnar
2016-02-03 13:21 ` Steven Rostedt
0 siblings, 1 reply; 11+ messages in thread
From: Ingo Molnar @ 2016-02-03 10:12 UTC (permalink / raw)
To: Peter Zijlstra
Cc: Steven Rostedt, linux-kernel, Thomas Gleixner, Juri Lelli,
Clark Williams, Andrew Morton
* Peter Zijlstra <peterz@infradead.org> wrote:
> On Mon, Feb 01, 2016 at 05:38:09PM -0500, Steven Rostedt wrote:
>
> > Regardless of this final patch, what do you think of the first three? I
> > can send them as a separate series, just to get more of the SCHED_DEBUG
> > code into debug.c.
>
> The first two looked ok, the third I had a vague niggle it might be part
> of something bigger and tearing it up like this might make it harder.
Yeah, so the purpose of kernel/sched/debug.c is mostly meant as a filesystem
interface to display various internal scheduler details. It mostly deals with
/proc/sched_debug details: hence the name.
Note that 'sched domain debugging' does not fit into that cleanly, because it also
does sanity checking of the sched-domains data. So I'm not sure we want to do
patch #3 that moves sched-domain debugging to debug.c.
So debug.c is named in a bit too generic fashion, attracting such patches.
Thanks,
Ingo
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [RFC][PATCH 4/4] sched: Add debugfs/sched/deadline_bw file to show current bandwidths
2016-02-03 10:12 ` Ingo Molnar
@ 2016-02-03 13:21 ` Steven Rostedt
0 siblings, 0 replies; 11+ messages in thread
From: Steven Rostedt @ 2016-02-03 13:21 UTC (permalink / raw)
To: Ingo Molnar
Cc: Peter Zijlstra, linux-kernel, Thomas Gleixner, Juri Lelli,
Clark Williams, Andrew Morton
On Wed, 3 Feb 2016 11:12:09 +0100
Ingo Molnar <mingo@kernel.org> wrote:
> * Peter Zijlstra <peterz@infradead.org> wrote:
>
> > On Mon, Feb 01, 2016 at 05:38:09PM -0500, Steven Rostedt wrote:
> >
> > > Regardless of this final patch, what do you think of the first three? I
> > > can send them as a separate series, just to get more of the SCHED_DEBUG
> > > code into debug.c.
> >
> > The first two looked ok, the third I had a vague niggle it might be part
> > of something bigger and tearing it up like this might make it harder.
Actually, I originally had plans to display the entire root domain
here, but decided against that.
>
> Yeah, so the purpose of kernel/sched/debug.c is mostly meant as a filesystem
> interface to display various internal scheduler details. It mostly deals with
> /proc/sched_debug details: hence the name.
>
> Note that 'sched domain debugging' does not fit into that cleanly, because it also
> does sanity checking of the sched-domains data. So I'm not sure we want to do
> patch #3 that moves sched-domain debugging to debug.c.
>
> So debug.c is named in a bit too generic fashion, attracting such patches.
>
It was actually more the #ifdef SCHED_DEBUG that attracted the patches.
The debug.c file is compiled on SCHED_DEBUG, thus it appeared natural
to have parts in core.c just moved into that file instead of having all
the #ifdef SCHED_DEBUG in it.
-- Steve
^ permalink raw reply [flat|nested] 11+ messages in thread
end of thread, other threads:[~2016-02-03 13:21 UTC | newest]
Thread overview: 11+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2016-02-01 20:26 [RFC][PATCH 0/4] sched: Display deadline bandwidth and other SCHED_DEBUG clean up Steven Rostedt
2016-02-01 20:26 ` [RFC][PATCH 1/4] sched: Move sched_feature file setup into debug.c Steven Rostedt
2016-02-01 20:26 ` [RFC][PATCH 2/4] sched: Move sched_domain_sysctl to debug.c Steven Rostedt
2016-02-01 20:26 ` [RFC][PATCH 3/4] sched: Move sched_domain_debug into debug.c Steven Rostedt
2016-02-01 20:26 ` [RFC][PATCH 4/4] sched: Add debugfs/sched/deadline_bw file to show current bandwidths Steven Rostedt
2016-02-01 22:17 ` Peter Zijlstra
2016-02-01 22:38 ` Steven Rostedt
2016-02-01 22:41 ` Peter Zijlstra
2016-02-03 10:12 ` Ingo Molnar
2016-02-03 13:21 ` Steven Rostedt
2016-02-01 22:18 ` Peter Zijlstra
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.