Patches contributed by Eötvös Lorand University
commit 99a3eb3845f034eb55640a3da73e5e28349678c6
Author: Ingo Molnar <mingo@elte.hu>
Date: Tue Dec 12 12:10:28 2006 +0100
[PATCH] lockdep: fix seqlock_init()
seqlock_init() needs to use spin_lock_init() for dynamic locks, so that
lockdep is notified about the presence of a new lock.
(this is a fallout of the recent networking merge, which started using
the so-far unused seqlock_init() API.)
This fix solves the following lockdep-internal warning on current -git:
INFO: trying to register non-static key.
the code is fine but needs lockdep annotation.
turning off the locking correctness validator.
__lock_acquire+0x10c/0x9f9
lock_acquire+0x56/0x72
_spin_lock+0x35/0x42
neigh_destroy+0x9d/0x12e
neigh_periodic_timer+0x10a/0x15c
run_timer_softirq+0x126/0x18e
__do_softirq+0x6b/0xe6
do_softirq+0x64/0xd2
ksoftirqd+0x82/0x138
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
diff --git a/include/linux/seqlock.h b/include/linux/seqlock.h
index 46000936f8f1..6b0648cfdffc 100644
--- a/include/linux/seqlock.h
+++ b/include/linux/seqlock.h
@@ -44,8 +44,11 @@ typedef struct {
#define SEQLOCK_UNLOCKED \
__SEQLOCK_UNLOCKED(old_style_seqlock_init)
-#define seqlock_init(x) \
- do { *(x) = (seqlock_t) __SEQLOCK_UNLOCKED(x); } while (0)
+#define seqlock_init(x) \
+ do { \
+ (x)->sequence = 0; \
+ spin_lock_init(&(x)->lock); \
+ } while (0)
#define DEFINE_SEQLOCK(x) \
seqlock_t x = __SEQLOCK_UNLOCKED(x)
commit 792908225064b1d841a8990b9d1d1cfc4e0e5bb2
Author: Ingo Molnar <mingo@elte.hu>
Date: Wed Dec 6 20:40:51 2006 -0800
[PATCH] add ignore_loglevel boot option
Sometimes the kernel prints something interesting while userspace bootup
keeps messages turned off via loglevel. Enable the printing of /all/
kernel messages via the "ignore_loglevel" boot option. Off by default.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 2a40d9f6ffad..6b3c3e37a277 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -650,6 +650,10 @@ and is between 256 and 4096 characters. It is defined in the file
idle= [HW]
Format: idle=poll or idle=halt
+ ignore_loglevel [KNL]
+ Ignore loglevel setting - this will print /all/
+ kernel messages to the console. Useful for debugging.
+
ihash_entries= [KNL]
Set number of hash buckets for inode cache.
diff --git a/kernel/printk.c b/kernel/printk.c
index c3d90a58e4c5..185bb45eacf7 100644
--- a/kernel/printk.c
+++ b/kernel/printk.c
@@ -333,13 +333,25 @@ static void __call_console_drivers(unsigned long start, unsigned long end)
}
}
+static int __read_mostly ignore_loglevel;
+
+int __init ignore_loglevel_setup(char *str)
+{
+ ignore_loglevel = 1;
+ printk(KERN_INFO "debug: ignoring loglevel setting.\n");
+
+ return 1;
+}
+
+__setup("ignore_loglevel", ignore_loglevel_setup);
+
/*
* Write out chars from start to end - 1 inclusive
*/
static void _call_console_drivers(unsigned long start,
unsigned long end, int msg_log_level)
{
- if (msg_log_level < console_loglevel &&
+ if ((msg_log_level < console_loglevel || ignore_loglevel) &&
console_drivers && start != end) {
if ((start & LOG_BUF_MASK) > (end & LOG_BUF_MASK)) {
/* wrapped write */
commit 70e4506765602cca047cfa31933836e354c61a63
Author: Ingo Molnar <mingo@elte.hu>
Date: Wed Dec 6 20:40:50 2006 -0800
[PATCH] lockdep: register_lock_class() fix
The hash_lock must only ever be taken with irqs disabled. This happens in
all the important places, except one codepath: register_lock_class(). The
race should trigger rarely because register_lock_class() is quite rare and
single-threaded (happens during init most of the time).
The fix is to disable irqs.
( bug found live in -rt: there preemption is alot more agressive and
preempting with the hash-lock held caused a lockup.)
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
diff --git a/kernel/lockdep.c b/kernel/lockdep.c
index 3926c3674354..62e73ce68197 100644
--- a/kernel/lockdep.c
+++ b/kernel/lockdep.c
@@ -1182,6 +1182,7 @@ register_lock_class(struct lockdep_map *lock, unsigned int subclass, int force)
struct lockdep_subclass_key *key;
struct list_head *hash_head;
struct lock_class *class;
+ unsigned long flags;
class = look_up_lock_class(lock, subclass);
if (likely(class))
@@ -1203,6 +1204,7 @@ register_lock_class(struct lockdep_map *lock, unsigned int subclass, int force)
key = lock->key->subkeys + subclass;
hash_head = classhashentry(key);
+ raw_local_irq_save(flags);
__raw_spin_lock(&hash_lock);
/*
* We have to do the hash-walk again, to avoid races
@@ -1217,6 +1219,7 @@ register_lock_class(struct lockdep_map *lock, unsigned int subclass, int force)
*/
if (nr_lock_classes >= MAX_LOCKDEP_KEYS) {
__raw_spin_unlock(&hash_lock);
+ raw_local_irq_restore(flags);
debug_locks_off();
printk("BUG: MAX_LOCKDEP_KEYS too low!\n");
printk("turning off the locking correctness validator.\n");
@@ -1239,15 +1242,18 @@ register_lock_class(struct lockdep_map *lock, unsigned int subclass, int force)
if (verbose(class)) {
__raw_spin_unlock(&hash_lock);
+ raw_local_irq_restore(flags);
printk("\nnew class %p: %s", class->key, class->name);
if (class->name_version > 1)
printk("#%d", class->name_version);
printk("\n");
dump_stack();
+ raw_local_irq_save(flags);
__raw_spin_lock(&hash_lock);
}
out_unlock_set:
__raw_spin_unlock(&hash_lock);
+ raw_local_irq_restore(flags);
if (!subclass || force)
lock->class_cache = class;
commit 2ee91f197c0bc654b24eed5831fd12aa0d566a7d
Author: Ingo Molnar <mingo@elte.hu>
Date: Wed Dec 6 20:39:32 2006 -0800
[PATCH] lockdep: show more details about self-test failures
Make the locking self-test failures (of 'FAILURE' type) easier to debug by
printing more information.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
diff --git a/include/linux/debug_locks.h b/include/linux/debug_locks.h
index 952bee79a8f3..a1c10b0c4cf0 100644
--- a/include/linux/debug_locks.h
+++ b/include/linux/debug_locks.h
@@ -24,7 +24,7 @@ extern int debug_locks_off(void);
int __ret = 0; \
\
if (unlikely(c)) { \
- if (debug_locks_off()) \
+ if (debug_locks_silent || debug_locks_off()) \
WARN_ON(1); \
__ret = 1; \
} \
diff --git a/kernel/mutex-debug.c b/kernel/mutex-debug.c
index 18651641a7b5..841539d72c55 100644
--- a/kernel/mutex-debug.c
+++ b/kernel/mutex-debug.c
@@ -77,6 +77,9 @@ void mutex_remove_waiter(struct mutex *lock, struct mutex_waiter *waiter,
void debug_mutex_unlock(struct mutex *lock)
{
+ if (unlikely(!debug_locks))
+ return;
+
DEBUG_LOCKS_WARN_ON(lock->owner != current_thread_info());
DEBUG_LOCKS_WARN_ON(lock->magic != lock);
DEBUG_LOCKS_WARN_ON(!lock->wait_list.prev && !lock->wait_list.next);
diff --git a/lib/locking-selftest.c b/lib/locking-selftest.c
index 7945787f439a..280332c1827c 100644
--- a/lib/locking-selftest.c
+++ b/lib/locking-selftest.c
@@ -963,7 +963,9 @@ static void dotest(void (*testcase_fn)(void), int expected, int lockclass_mask)
printk("failed|");
} else {
unexpected_testcase_failures++;
+
printk("FAILED|");
+ dump_stack();
}
} else {
testcase_successes++;
commit 50cc670aebf4fc64afaf533fb9fa1c8570f09d74
Author: Ingo Molnar <mingo@elte.hu>
Date: Wed Dec 6 20:39:30 2006 -0800
[PATCH] lockdep: more chains
Some have reported a chain-table overflow - double its size.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
diff --git a/kernel/lockdep_internals.h b/kernel/lockdep_internals.h
index eab043c83bb2..8ce09bc4613d 100644
--- a/kernel/lockdep_internals.h
+++ b/kernel/lockdep_internals.h
@@ -20,7 +20,7 @@
#define MAX_LOCKDEP_KEYS_BITS 11
#define MAX_LOCKDEP_KEYS (1UL << MAX_LOCKDEP_KEYS_BITS)
-#define MAX_LOCKDEP_CHAINS_BITS 13
+#define MAX_LOCKDEP_CHAINS_BITS 14
#define MAX_LOCKDEP_CHAINS (1UL << MAX_LOCKDEP_CHAINS_BITS)
/*
commit 02316067852187b8bec781bec07410e91af79627
Author: Ingo Molnar <mingo@elte.hu>
Date: Wed Dec 6 20:38:17 2006 -0800
[PATCH] hotplug CPU: clean up hotcpu_notifier() use
There was lots of #ifdef noise in the kernel due to hotcpu_notifier(fn,
prio) not correctly marking 'fn' as used in the !HOTPLUG_CPU case, and thus
generating compiler warnings of unused symbols, hence forcing people to add
#ifdefs.
the compiler can skip truly unused functions just fine:
text data bss dec hex filename
1624412 728710 3674856 6027978 5bfaca vmlinux.before
1624412 728710 3674856 6027978 5bfaca vmlinux.after
[akpm@osdl.org: topology.c fix]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
diff --git a/arch/i386/kernel/cpu/mcheck/therm_throt.c b/arch/i386/kernel/cpu/mcheck/therm_throt.c
index bad8b4420709..065005c3f168 100644
--- a/arch/i386/kernel/cpu/mcheck/therm_throt.c
+++ b/arch/i386/kernel/cpu/mcheck/therm_throt.c
@@ -116,7 +116,6 @@ static __cpuinit int thermal_throttle_add_dev(struct sys_device *sys_dev)
return sysfs_create_group(&sys_dev->kobj, &thermal_throttle_attr_group);
}
-#ifdef CONFIG_HOTPLUG_CPU
static __cpuinit void thermal_throttle_remove_dev(struct sys_device *sys_dev)
{
return sysfs_remove_group(&sys_dev->kobj, &thermal_throttle_attr_group);
@@ -153,7 +152,6 @@ static struct notifier_block thermal_throttle_cpu_notifier =
{
.notifier_call = thermal_throttle_cpu_callback,
};
-#endif /* CONFIG_HOTPLUG_CPU */
static __init int thermal_throttle_init_device(void)
{
diff --git a/arch/i386/kernel/cpuid.c b/arch/i386/kernel/cpuid.c
index ab0c327e79dc..23b2cc748d4e 100644
--- a/arch/i386/kernel/cpuid.c
+++ b/arch/i386/kernel/cpuid.c
@@ -167,7 +167,6 @@ static int cpuid_device_create(int i)
return err;
}
-#ifdef CONFIG_HOTPLUG_CPU
static int cpuid_class_cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu)
{
unsigned int cpu = (unsigned long)hcpu;
@@ -187,7 +186,6 @@ static struct notifier_block __cpuinitdata cpuid_class_cpu_notifier =
{
.notifier_call = cpuid_class_cpu_callback,
};
-#endif /* !CONFIG_HOTPLUG_CPU */
static int __init cpuid_init(void)
{
diff --git a/arch/i386/kernel/microcode.c b/arch/i386/kernel/microcode.c
index 23f5984d0654..972346604f9d 100644
--- a/arch/i386/kernel/microcode.c
+++ b/arch/i386/kernel/microcode.c
@@ -703,7 +703,6 @@ static struct sysdev_driver mc_sysdev_driver = {
.resume = mc_sysdev_resume,
};
-#ifdef CONFIG_HOTPLUG_CPU
static __cpuinit int
mc_cpu_callback(struct notifier_block *nb, unsigned long action, void *hcpu)
{
@@ -726,7 +725,6 @@ mc_cpu_callback(struct notifier_block *nb, unsigned long action, void *hcpu)
static struct notifier_block mc_cpu_notifier = {
.notifier_call = mc_cpu_callback,
};
-#endif
static int __init microcode_init (void)
{
diff --git a/arch/i386/kernel/msr.c b/arch/i386/kernel/msr.c
index a773f776c9ea..7763c67ca282 100644
--- a/arch/i386/kernel/msr.c
+++ b/arch/i386/kernel/msr.c
@@ -250,7 +250,6 @@ static int msr_device_create(int i)
return err;
}
-#ifdef CONFIG_HOTPLUG_CPU
static int msr_class_cpu_callback(struct notifier_block *nfb,
unsigned long action, void *hcpu)
{
@@ -271,7 +270,6 @@ static struct notifier_block __cpuinitdata msr_class_cpu_notifier =
{
.notifier_call = msr_class_cpu_callback,
};
-#endif
static int __init msr_init(void)
{
diff --git a/arch/ia64/kernel/palinfo.c b/arch/ia64/kernel/palinfo.c
index 0b546e2b36ac..c4c10a0b99d9 100644
--- a/arch/ia64/kernel/palinfo.c
+++ b/arch/ia64/kernel/palinfo.c
@@ -952,7 +952,6 @@ remove_palinfo_proc_entries(unsigned int hcpu)
}
}
-#ifdef CONFIG_HOTPLUG_CPU
static int palinfo_cpu_callback(struct notifier_block *nfb,
unsigned long action, void *hcpu)
{
@@ -974,7 +973,6 @@ static struct notifier_block palinfo_cpu_notifier =
.notifier_call = palinfo_cpu_callback,
.priority = 0,
};
-#endif
static int __init
palinfo_init(void)
diff --git a/arch/ia64/kernel/salinfo.c b/arch/ia64/kernel/salinfo.c
index e63b8ca5344a..fd607ca51a8d 100644
--- a/arch/ia64/kernel/salinfo.c
+++ b/arch/ia64/kernel/salinfo.c
@@ -575,7 +575,6 @@ static struct file_operations salinfo_data_fops = {
.write = salinfo_log_write,
};
-#ifdef CONFIG_HOTPLUG_CPU
static int __devinit
salinfo_cpu_callback(struct notifier_block *nb, unsigned long action, void *hcpu)
{
@@ -620,7 +619,6 @@ static struct notifier_block salinfo_cpu_notifier =
.notifier_call = salinfo_cpu_callback,
.priority = 0,
};
-#endif /* CONFIG_HOTPLUG_CPU */
static int __init
salinfo_init(void)
diff --git a/arch/s390/appldata/appldata_base.c b/arch/s390/appldata/appldata_base.c
index 67d5cf9cba83..b8c237290263 100644
--- a/arch/s390/appldata/appldata_base.c
+++ b/arch/s390/appldata/appldata_base.c
@@ -561,7 +561,6 @@ appldata_offline_cpu(int cpu)
spin_unlock(&appldata_timer_lock);
}
-#ifdef CONFIG_HOTPLUG_CPU
static int __cpuinit
appldata_cpu_notify(struct notifier_block *self,
unsigned long action, void *hcpu)
@@ -582,7 +581,6 @@ appldata_cpu_notify(struct notifier_block *self,
static struct notifier_block appldata_nb = {
.notifier_call = appldata_cpu_notify,
};
-#endif
/*
* appldata_init()
diff --git a/arch/x86_64/kernel/mce.c b/arch/x86_64/kernel/mce.c
index c7587fc39015..bc863c464a1f 100644
--- a/arch/x86_64/kernel/mce.c
+++ b/arch/x86_64/kernel/mce.c
@@ -641,7 +641,6 @@ static __cpuinit int mce_create_device(unsigned int cpu)
return err;
}
-#ifdef CONFIG_HOTPLUG_CPU
static void mce_remove_device(unsigned int cpu)
{
int i;
@@ -674,7 +673,6 @@ mce_cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu)
static struct notifier_block mce_cpu_notifier = {
.notifier_call = mce_cpu_callback,
};
-#endif
static __init int mce_init_device(void)
{
diff --git a/arch/x86_64/kernel/mce_amd.c b/arch/x86_64/kernel/mce_amd.c
index 883fe747f64c..fa09debad4b7 100644
--- a/arch/x86_64/kernel/mce_amd.c
+++ b/arch/x86_64/kernel/mce_amd.c
@@ -551,7 +551,6 @@ static __cpuinit int threshold_create_device(unsigned int cpu)
return err;
}
-#ifdef CONFIG_HOTPLUG_CPU
/*
* let's be hotplug friendly.
* in case of multiple core processors, the first core always takes ownership
@@ -594,12 +593,14 @@ static void threshold_remove_bank(unsigned int cpu, int bank)
sprintf(name, "threshold_bank%i", bank);
+#ifdef CONFIG_SMP
/* sibling symlink */
if (shared_bank[bank] && b->blocks->cpu != cpu) {
sysfs_remove_link(&per_cpu(device_mce, cpu).kobj, name);
per_cpu(threshold_banks, cpu)[bank] = NULL;
return;
}
+#endif
/* remove all sibling symlinks before unregistering */
for_each_cpu_mask(i, b->cpus) {
@@ -656,7 +657,6 @@ static int threshold_cpu_callback(struct notifier_block *nfb,
static struct notifier_block threshold_cpu_notifier = {
.notifier_call = threshold_cpu_callback,
};
-#endif /* CONFIG_HOTPLUG_CPU */
static __init int threshold_init_device(void)
{
diff --git a/arch/x86_64/kernel/vsyscall.c b/arch/x86_64/kernel/vsyscall.c
index 630036c06c75..3785e4954734 100644
--- a/arch/x86_64/kernel/vsyscall.c
+++ b/arch/x86_64/kernel/vsyscall.c
@@ -275,7 +275,6 @@ static void __cpuinit cpu_vsyscall_init(void *arg)
vsyscall_set_cpu(raw_smp_processor_id());
}
-#ifdef CONFIG_HOTPLUG_CPU
static int __cpuinit
cpu_vsyscall_notifier(struct notifier_block *n, unsigned long action, void *arg)
{
@@ -284,7 +283,6 @@ cpu_vsyscall_notifier(struct notifier_block *n, unsigned long action, void *arg)
smp_call_function_single(cpu, cpu_vsyscall_init, NULL, 0, 1);
return NOTIFY_DONE;
}
-#endif
static void __init map_vsyscall(void)
{
diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c
index a4ff3271d4a8..31512cd9f3ad 100644
--- a/block/ll_rw_blk.c
+++ b/block/ll_rw_blk.c
@@ -3459,8 +3459,6 @@ static void blk_done_softirq(struct softirq_action *h)
}
}
-#ifdef CONFIG_HOTPLUG_CPU
-
static int blk_cpu_notify(struct notifier_block *self, unsigned long action,
void *hcpu)
{
@@ -3486,8 +3484,6 @@ static struct notifier_block __devinitdata blk_cpu_notifier = {
.notifier_call = blk_cpu_notify,
};
-#endif /* CONFIG_HOTPLUG_CPU */
-
/**
* blk_complete_request - end I/O on a request
* @req: the request being processed
diff --git a/drivers/base/topology.c b/drivers/base/topology.c
index 3d12b85b0962..067a9e8bc377 100644
--- a/drivers/base/topology.c
+++ b/drivers/base/topology.c
@@ -108,7 +108,6 @@ static int __cpuinit topology_add_dev(unsigned int cpu)
return rc;
}
-#ifdef CONFIG_HOTPLUG_CPU
static void __cpuinit topology_remove_dev(unsigned int cpu)
{
struct sys_device *sys_dev = get_cpu_sysdev(cpu);
@@ -136,7 +135,6 @@ static int __cpuinit topology_cpu_callback(struct notifier_block *nfb,
}
return rc ? NOTIFY_BAD : NOTIFY_OK;
}
-#endif
static int __cpuinit topology_sysfs_init(void)
{
diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c
index 7a7c6e6dfe4f..47ab42db122a 100644
--- a/drivers/cpufreq/cpufreq.c
+++ b/drivers/cpufreq/cpufreq.c
@@ -1537,7 +1537,6 @@ int cpufreq_update_policy(unsigned int cpu)
}
EXPORT_SYMBOL(cpufreq_update_policy);
-#ifdef CONFIG_HOTPLUG_CPU
static int cpufreq_cpu_callback(struct notifier_block *nfb,
unsigned long action, void *hcpu)
{
@@ -1577,7 +1576,6 @@ static struct notifier_block __cpuinitdata cpufreq_cpu_notifier =
{
.notifier_call = cpufreq_cpu_callback,
};
-#endif /* CONFIG_HOTPLUG_CPU */
/*********************************************************************
* REGISTER / UNREGISTER CPUFREQ DRIVER *
diff --git a/fs/buffer.c b/fs/buffer.c
index a8ca0ac21488..517860f2d75b 100644
--- a/fs/buffer.c
+++ b/fs/buffer.c
@@ -2972,7 +2972,6 @@ init_buffer_head(void *data, struct kmem_cache *cachep, unsigned long flags)
}
}
-#ifdef CONFIG_HOTPLUG_CPU
static void buffer_exit_cpu(int cpu)
{
int i;
@@ -2994,7 +2993,6 @@ static int buffer_cpu_notify(struct notifier_block *self,
buffer_exit_cpu((unsigned long)hcpu);
return NOTIFY_OK;
}
-#endif /* CONFIG_HOTPLUG_CPU */
void __init buffer_init(void)
{
diff --git a/include/linux/cpu.h b/include/linux/cpu.h
index f02d71bf6894..71dc6ba4f73f 100644
--- a/include/linux/cpu.h
+++ b/include/linux/cpu.h
@@ -89,9 +89,9 @@ int cpu_down(unsigned int cpu);
#define lock_cpu_hotplug() do { } while (0)
#define unlock_cpu_hotplug() do { } while (0)
#define lock_cpu_hotplug_interruptible() 0
-#define hotcpu_notifier(fn, pri) do { } while (0)
-#define register_hotcpu_notifier(nb) do { } while (0)
-#define unregister_hotcpu_notifier(nb) do { } while (0)
+#define hotcpu_notifier(fn, pri) do { (void)(fn); } while (0)
+#define register_hotcpu_notifier(nb) do { (void)(nb); } while (0)
+#define unregister_hotcpu_notifier(nb) do { (void)(nb); } while (0)
/* CPUs don't go offline once they're online w/o CONFIG_HOTPLUG_CPU */
static inline int cpu_is_offline(int cpu) { return 0; }
diff --git a/kernel/cpuset.c b/kernel/cpuset.c
index bd1e89c4c96a..9b62b4c03ad0 100644
--- a/kernel/cpuset.c
+++ b/kernel/cpuset.c
@@ -2044,7 +2044,6 @@ int __init cpuset_init(void)
return err;
}
-#if defined(CONFIG_HOTPLUG_CPU) || defined(CONFIG_MEMORY_HOTPLUG)
/*
* If common_cpu_mem_hotplug_unplug(), below, unplugs any CPUs
* or memory nodes, we need to walk over the cpuset hierarchy,
@@ -2108,9 +2107,7 @@ static void common_cpu_mem_hotplug_unplug(void)
mutex_unlock(&callback_mutex);
mutex_unlock(&manage_mutex);
}
-#endif
-#ifdef CONFIG_HOTPLUG_CPU
/*
* The top_cpuset tracks what CPUs and Memory Nodes are online,
* period. This is necessary in order to make cpusets transparent
@@ -2127,7 +2124,6 @@ static int cpuset_handle_cpuhp(struct notifier_block *nb,
common_cpu_mem_hotplug_unplug();
return 0;
}
-#endif
#ifdef CONFIG_MEMORY_HOTPLUG
/*
diff --git a/kernel/profile.c b/kernel/profile.c
index 04fd84e8cdbe..0961d93e1d91 100644
--- a/kernel/profile.c
+++ b/kernel/profile.c
@@ -319,7 +319,6 @@ void profile_hits(int type, void *__pc, unsigned int nr_hits)
put_cpu();
}
-#ifdef CONFIG_HOTPLUG_CPU
static int __devinit profile_cpu_callback(struct notifier_block *info,
unsigned long action, void *__cpu)
{
@@ -372,10 +371,10 @@ static int __devinit profile_cpu_callback(struct notifier_block *info,
}
return NOTIFY_OK;
}
-#endif /* CONFIG_HOTPLUG_CPU */
#else /* !CONFIG_SMP */
#define profile_flip_buffers() do { } while (0)
#define profile_discard_flip_buffers() do { } while (0)
+#define profile_cpu_callback NULL
void profile_hits(int type, void *__pc, unsigned int nr_hits)
{
diff --git a/kernel/sched.c b/kernel/sched.c
index 75a005ed4eda..c83f531c2886 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -6740,8 +6740,6 @@ SYSDEV_ATTR(sched_smt_power_savings, 0644, sched_smt_power_savings_show,
sched_smt_power_savings_store);
#endif
-
-#ifdef CONFIG_HOTPLUG_CPU
/*
* Force a reinitialization of the sched domains hierarchy. The domains
* and groups cannot be updated in place without racing with the balancing
@@ -6774,7 +6772,6 @@ static int update_sched_domains(struct notifier_block *nfb,
return NOTIFY_OK;
}
-#endif
void __init sched_init_smp(void)
{
diff --git a/kernel/workqueue.c b/kernel/workqueue.c
index 5484d6e045c2..c5257316f4b9 100644
--- a/kernel/workqueue.c
+++ b/kernel/workqueue.c
@@ -655,7 +655,6 @@ int current_is_keventd(void)
}
-#ifdef CONFIG_HOTPLUG_CPU
/* Take the work from this (downed) CPU. */
static void take_over_work(struct workqueue_struct *wq, unsigned int cpu)
{
@@ -738,7 +737,6 @@ static int __devinit workqueue_cpu_callback(struct notifier_block *nfb,
return NOTIFY_OK;
}
-#endif
void init_workqueues(void)
{
diff --git a/lib/radix-tree.c b/lib/radix-tree.c
index e2cefabb5aa0..d69ddbe43865 100644
--- a/lib/radix-tree.c
+++ b/lib/radix-tree.c
@@ -996,7 +996,6 @@ static __init void radix_tree_init_maxindex(void)
height_to_maxindex[i] = __maxindex(i);
}
-#ifdef CONFIG_HOTPLUG_CPU
static int radix_tree_callback(struct notifier_block *nfb,
unsigned long action,
void *hcpu)
@@ -1016,7 +1015,6 @@ static int radix_tree_callback(struct notifier_block *nfb,
}
return NOTIFY_OK;
}
-#endif /* CONFIG_HOTPLUG_CPU */
void __init radix_tree_init(void)
{
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 2273952300d4..27ec7a1b8022 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -701,7 +701,6 @@ void drain_node_pages(int nodeid)
}
#endif
-#if defined(CONFIG_PM) || defined(CONFIG_HOTPLUG_CPU)
static void __drain_pages(unsigned int cpu)
{
unsigned long flags;
@@ -723,7 +722,6 @@ static void __drain_pages(unsigned int cpu)
}
}
}
-#endif /* CONFIG_PM || CONFIG_HOTPLUG_CPU */
#ifdef CONFIG_PM
@@ -2907,7 +2905,6 @@ void __init free_area_init(unsigned long *zones_size)
__pa(PAGE_OFFSET) >> PAGE_SHIFT, NULL);
}
-#ifdef CONFIG_HOTPLUG_CPU
static int page_alloc_cpu_notify(struct notifier_block *self,
unsigned long action, void *hcpu)
{
@@ -2922,7 +2919,6 @@ static int page_alloc_cpu_notify(struct notifier_block *self,
}
return NOTIFY_OK;
}
-#endif /* CONFIG_HOTPLUG_CPU */
void __init page_alloc_init(void)
{
diff --git a/mm/swap.c b/mm/swap.c
index 017e72ca9bbb..2ed7be39795e 100644
--- a/mm/swap.c
+++ b/mm/swap.c
@@ -514,5 +514,7 @@ void __init swap_setup(void)
* Right now other parts of the system means that we
* _really_ don't want to cluster much more
*/
+#ifdef CONFIG_HOTPLUG_CPU
hotcpu_notifier(cpu_swap_callback, 0);
+#endif
}
diff --git a/mm/vmscan.c b/mm/vmscan.c
index f6616e81fac7..093f5fe6dd77 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -1513,7 +1513,6 @@ unsigned long shrink_all_memory(unsigned long nr_pages)
}
#endif
-#ifdef CONFIG_HOTPLUG_CPU
/* It's optimal to keep kswapds on the same CPUs as their memory, but
not required for correctness. So if the last cpu in a node goes
away, we get changed to run anywhere: as the first one comes back,
@@ -1534,7 +1533,6 @@ static int __devinit cpu_callback(struct notifier_block *nfb,
}
return NOTIFY_OK;
}
-#endif /* CONFIG_HOTPLUG_CPU */
/*
* This kswapd start function will be called by init and node-hot-add.
diff --git a/net/core/dev.c b/net/core/dev.c
index 59d058a3b504..e660cb57e42a 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -3340,7 +3340,6 @@ void unregister_netdev(struct net_device *dev)
EXPORT_SYMBOL(unregister_netdev);
-#ifdef CONFIG_HOTPLUG_CPU
static int dev_cpu_callback(struct notifier_block *nfb,
unsigned long action,
void *ocpu)
@@ -3384,7 +3383,6 @@ static int dev_cpu_callback(struct notifier_block *nfb,
return NOTIFY_OK;
}
-#endif /* CONFIG_HOTPLUG_CPU */
#ifdef CONFIG_NET_DMA
/**
diff --git a/net/core/flow.c b/net/core/flow.c
index 104c25d00a1d..d137f971f97d 100644
--- a/net/core/flow.c
+++ b/net/core/flow.c
@@ -340,7 +340,6 @@ static void __devinit flow_cache_cpu_prepare(int cpu)
tasklet_init(tasklet, flow_cache_flush_tasklet, 0);
}
-#ifdef CONFIG_HOTPLUG_CPU
static int flow_cache_cpu(struct notifier_block *nfb,
unsigned long action,
void *hcpu)
@@ -349,7 +348,6 @@ static int flow_cache_cpu(struct notifier_block *nfb,
__flow_cache_shrink((unsigned long)hcpu, 0);
return NOTIFY_OK;
}
-#endif /* CONFIG_HOTPLUG_CPU */
static int __init flow_cache_init(void)
{
commit ece8a684c75df215320b4155944979e3f78c5c93
Author: Ingo Molnar <mingo@elte.hu>
Date: Wed Dec 6 20:37:24 2006 -0800
[PATCH] sleep profiling
Implement prof=sleep profiling. TASK_UNINTERRUPTIBLE sleeps will be taken
as a profile hit, and every millisecond spent sleeping causes a profile-hit
for the call site that initiated the sleep.
Sample readprofile output on i386:
306 ps2_sendbyte 1.3973
432 call_usermodehelper_keys 1.9548
484 ps2_command 0.6453
790 __driver_attach 4.7879
1593 msleep 44.2500
3976 sync_buffer 64.1290
4076 do_lookup 12.4648
8587 sync_page 122.6714
20820 total 0.0067
(NOTE: architectures need to check whether get_wchan() can be called from
deep within the wakeup path.)
akpm: we need to mark more functions __sched. lock_sock(), msleep(), others..
akpm: the contention in do_lookup() is a surprise. Presumably doing disk
reads for directory contents while holding i_mutex.
[akpm@osdl.org: various fixes]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 8fe6b834ef27..2a40d9f6ffad 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -1294,6 +1294,7 @@ and is between 256 and 4096 characters. It is defined in the file
Param: "schedule" - profile schedule points.
Param: <number> - step/bucket size as a power of 2 for
statistical time based profiling.
+ Param: "sleep" - profile D-state sleeping (millisecs)
processor.max_cstate= [HW,ACPI]
Limit processor to maximum C-state
diff --git a/include/linux/profile.h b/include/linux/profile.h
index acce53fd38b6..5670b340c4ef 100644
--- a/include/linux/profile.h
+++ b/include/linux/profile.h
@@ -6,10 +6,15 @@
#include <linux/kernel.h>
#include <linux/init.h>
#include <linux/cpumask.h>
+#include <linux/cache.h>
+
#include <asm/errno.h>
+extern int prof_on __read_mostly;
+
#define CPU_PROFILING 1
#define SCHED_PROFILING 2
+#define SLEEP_PROFILING 3
struct proc_dir_entry;
struct pt_regs;
@@ -18,7 +23,24 @@ struct notifier_block;
/* init basic kernel profiler */
void __init profile_init(void);
void profile_tick(int);
-void profile_hit(int, void *);
+
+/*
+ * Add multiple profiler hits to a given address:
+ */
+void profile_hits(int, void *ip, unsigned int nr_hits);
+
+/*
+ * Single profiler hit:
+ */
+static inline void profile_hit(int type, void *ip)
+{
+ /*
+ * Speedup for the common (no profiling enabled) case:
+ */
+ if (unlikely(prof_on == type))
+ profile_hits(type, ip, 1);
+}
+
#ifdef CONFIG_PROC_FS
void create_prof_cpu_mask(struct proc_dir_entry *);
#else
diff --git a/kernel/profile.c b/kernel/profile.c
index 15b012df4ff1..04fd84e8cdbe 100644
--- a/kernel/profile.c
+++ b/kernel/profile.c
@@ -40,7 +40,7 @@ int (*timer_hook)(struct pt_regs *) __read_mostly;
static atomic_t *prof_buffer;
static unsigned long prof_len, prof_shift;
-static int prof_on __read_mostly;
+int prof_on __read_mostly;
static cpumask_t prof_cpu_mask = CPU_MASK_ALL;
#ifdef CONFIG_SMP
static DEFINE_PER_CPU(struct profile_hit *[2], cpu_profile_hits);
@@ -51,9 +51,19 @@ static DEFINE_MUTEX(profile_flip_mutex);
static int __init profile_setup(char * str)
{
static char __initdata schedstr[] = "schedule";
+ static char __initdata sleepstr[] = "sleep";
int par;
- if (!strncmp(str, schedstr, strlen(schedstr))) {
+ if (!strncmp(str, sleepstr, strlen(sleepstr))) {
+ prof_on = SLEEP_PROFILING;
+ if (str[strlen(sleepstr)] == ',')
+ str += strlen(sleepstr) + 1;
+ if (get_option(&str, &par))
+ prof_shift = par;
+ printk(KERN_INFO
+ "kernel sleep profiling enabled (shift: %ld)\n",
+ prof_shift);
+ } else if (!strncmp(str, sleepstr, strlen(sleepstr))) {
prof_on = SCHED_PROFILING;
if (str[strlen(schedstr)] == ',')
str += strlen(schedstr) + 1;
@@ -204,7 +214,8 @@ EXPORT_SYMBOL_GPL(profile_event_unregister);
* positions to which hits are accounted during short intervals (e.g.
* several seconds) is usually very small. Exclusion from buffer
* flipping is provided by interrupt disablement (note that for
- * SCHED_PROFILING profile_hit() may be called from process context).
+ * SCHED_PROFILING or SLEEP_PROFILING profile_hit() may be called from
+ * process context).
* The hash function is meant to be lightweight as opposed to strong,
* and was vaguely inspired by ppc64 firmware-supported inverted
* pagetable hash functions, but uses a full hashtable full of finite
@@ -257,7 +268,7 @@ static void profile_discard_flip_buffers(void)
mutex_unlock(&profile_flip_mutex);
}
-void profile_hit(int type, void *__pc)
+void profile_hits(int type, void *__pc, unsigned int nr_hits)
{
unsigned long primary, secondary, flags, pc = (unsigned long)__pc;
int i, j, cpu;
@@ -274,21 +285,31 @@ void profile_hit(int type, void *__pc)
put_cpu();
return;
}
+ /*
+ * We buffer the global profiler buffer into a per-CPU
+ * queue and thus reduce the number of global (and possibly
+ * NUMA-alien) accesses. The write-queue is self-coalescing:
+ */
local_irq_save(flags);
do {
for (j = 0; j < PROFILE_GRPSZ; ++j) {
if (hits[i + j].pc == pc) {
- hits[i + j].hits++;
+ hits[i + j].hits += nr_hits;
goto out;
} else if (!hits[i + j].hits) {
hits[i + j].pc = pc;
- hits[i + j].hits = 1;
+ hits[i + j].hits = nr_hits;
goto out;
}
}
i = (i + secondary) & (NR_PROFILE_HIT - 1);
} while (i != primary);
- atomic_inc(&prof_buffer[pc]);
+
+ /*
+ * Add the current hit(s) and flush the write-queue out
+ * to the global buffer:
+ */
+ atomic_add(nr_hits, &prof_buffer[pc]);
for (i = 0; i < NR_PROFILE_HIT; ++i) {
atomic_add(hits[i].hits, &prof_buffer[hits[i].pc]);
hits[i].pc = hits[i].hits = 0;
@@ -356,14 +377,14 @@ static int __devinit profile_cpu_callback(struct notifier_block *info,
#define profile_flip_buffers() do { } while (0)
#define profile_discard_flip_buffers() do { } while (0)
-void profile_hit(int type, void *__pc)
+void profile_hits(int type, void *__pc, unsigned int nr_hits)
{
unsigned long pc;
if (prof_on != type || !prof_buffer)
return;
pc = ((unsigned long)__pc - (unsigned long)_stext) >> prof_shift;
- atomic_inc(&prof_buffer[min(pc, prof_len - 1)]);
+ atomic_add(nr_hits, &prof_buffer[min(pc, prof_len - 1)]);
}
#endif /* !CONFIG_SMP */
diff --git a/kernel/sched.c b/kernel/sched.c
index 343e1794233e..75a005ed4eda 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -948,6 +948,17 @@ static void activate_task(struct task_struct *p, struct rq *rq, int local)
}
#endif
+ /*
+ * Sleep time is in units of nanosecs, so shift by 20 to get a
+ * milliseconds-range estimation of the amount of time that the task
+ * spent sleeping:
+ */
+ if (unlikely(prof_on == SLEEP_PROFILING)) {
+ if (p->state == TASK_UNINTERRUPTIBLE)
+ profile_hits(SLEEP_PROFILING, (void *)get_wchan(p),
+ (now - p->timestamp) >> 20);
+ }
+
if (!rt_task(p))
p->prio = recalc_task_prio(p, now);
commit e59e2ae2c29700117a54e85c106017c24837119f
Author: Ingo Molnar <mingo@elte.hu>
Date: Wed Dec 6 20:35:59 2006 -0800
[PATCH] SysRq-X: show blocked tasks
Add SysRq-X support: show blocked (TASK_UNINTERRUPTIBLE) tasks only.
Useful for debugging IO stalls.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
diff --git a/drivers/char/sysrq.c b/drivers/char/sysrq.c
index c64f5bcff947..05810c8d20bc 100644
--- a/drivers/char/sysrq.c
+++ b/drivers/char/sysrq.c
@@ -182,6 +182,18 @@ static struct sysrq_key_op sysrq_showstate_op = {
.enable_mask = SYSRQ_ENABLE_DUMP,
};
+static void sysrq_handle_showstate_blocked(int key, struct tty_struct *tty)
+{
+ show_state_filter(TASK_UNINTERRUPTIBLE);
+}
+static struct sysrq_key_op sysrq_showstate_blocked_op = {
+ .handler = sysrq_handle_showstate_blocked,
+ .help_msg = "showBlockedTasks",
+ .action_msg = "Show Blocked State",
+ .enable_mask = SYSRQ_ENABLE_DUMP,
+};
+
+
static void sysrq_handle_showmem(int key, struct tty_struct *tty)
{
show_mem();
@@ -304,7 +316,7 @@ static struct sysrq_key_op *sysrq_key_table[36] = {
/* May be assigned at init time by SMP VOYAGER */
NULL, /* v */
NULL, /* w */
- NULL, /* x */
+ &sysrq_showstate_blocked_op, /* x */
NULL, /* y */
NULL /* z */
};
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 837a012f573c..0a90cefb0b0d 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -194,7 +194,16 @@ extern void init_idle(struct task_struct *idle, int cpu);
extern cpumask_t nohz_cpu_mask;
-extern void show_state(void);
+/*
+ * Only dump TASK_* tasks. (-1 for all tasks)
+ */
+extern void show_state_filter(unsigned long state_filter);
+
+static inline void show_state(void)
+{
+ show_state_filter(-1);
+}
+
extern void show_regs(struct pt_regs *);
/*
diff --git a/kernel/sched.c b/kernel/sched.c
index 12fdbef1d9bf..1848e280504d 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -4804,7 +4804,7 @@ static void show_task(struct task_struct *p)
show_stack(p, NULL);
}
-void show_state(void)
+void show_state_filter(unsigned long state_filter)
{
struct task_struct *g, *p;
@@ -4824,11 +4824,16 @@ void show_state(void)
* console might take alot of time:
*/
touch_nmi_watchdog();
- show_task(p);
+ if (p->state & state_filter)
+ show_task(p);
} while_each_thread(g, p);
read_unlock(&tasklist_lock);
- debug_show_all_locks();
+ /*
+ * Only show locks if all tasks are dumped:
+ */
+ if (state_filter == -1)
+ debug_show_all_locks();
}
/**
commit 24d7bb3396c51ceb2285e0e7b0c1bd1865652c43
Author: Ingo Molnar <mingo@elte.hu>
Date: Tue Nov 28 09:14:05 2006 +0100
[PATCH] x86_64: fix 'earlyprintk=...,keep' regression
Commit 2c8c0e6b8d7700a990da8d24eff767f9ca223b96 ("[PATCH] Convert x86-64
to early param") broke the earlyprintk=...,keep feature.
This restores that functionality. Tested on x86_64. Must-have for
v2.6.19, no risk.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
diff --git a/arch/x86_64/kernel/early_printk.c b/arch/x86_64/kernel/early_printk.c
index e22ecd54870d..47b6d90349da 100644
--- a/arch/x86_64/kernel/early_printk.c
+++ b/arch/x86_64/kernel/early_printk.c
@@ -224,7 +224,7 @@ static int __init setup_early_printk(char *buf)
return 0;
early_console_initialized = 1;
- if (!strcmp(buf,"keep"))
+ if (strstr(buf, "keep"))
keep_early = 1;
if (!strncmp(buf, "serial", 6)) {
commit 1ff5683043196b9ad628a5de6bf8eeca52ee8bfd
Author: Ingo Molnar <mingo@elte.hu>
Date: Fri Nov 17 19:57:22 2006 +0100
[PATCH] lockdep: fix static keys in module-allocated percpu areas
lockdep got confused by certain locks in modules:
INFO: trying to register non-static key.
the code is fine but needs lockdep annotation.
turning off the locking correctness validator.
Call Trace:
[<ffffffff8026f40d>] dump_trace+0xaa/0x3f2
[<ffffffff8026f78f>] show_trace+0x3a/0x60
[<ffffffff8026f9d1>] dump_stack+0x15/0x17
[<ffffffff802abfe8>] __lock_acquire+0x724/0x9bb
[<ffffffff802ac52b>] lock_acquire+0x4d/0x67
[<ffffffff80267139>] rt_spin_lock+0x3d/0x41
[<ffffffff8839ed3f>] :ip_conntrack:__ip_ct_refresh_acct+0x131/0x174
[<ffffffff883a1334>] :ip_conntrack:udp_packet+0xbf/0xcf
[<ffffffff8839f9af>] :ip_conntrack:ip_conntrack_in+0x394/0x4a7
[<ffffffff8023551f>] nf_iterate+0x41/0x7f
[<ffffffff8025946a>] nf_hook_slow+0x64/0xd5
[<ffffffff802369a2>] ip_rcv+0x24e/0x506
[...]
Steven Rostedt found the bug: static_obj() check did not take
PERCPU_ENOUGH_ROOM into account, so in-module DEFINE_PER_CPU-area locks
were triggering this message.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
diff --git a/kernel/lockdep.c b/kernel/lockdep.c
index b739be2a6dc9..c9fefdb1a7db 100644
--- a/kernel/lockdep.c
+++ b/kernel/lockdep.c
@@ -1081,7 +1081,8 @@ static int static_obj(void *obj)
*/
for_each_possible_cpu(i) {
start = (unsigned long) &__per_cpu_start + per_cpu_offset(i);
- end = (unsigned long) &__per_cpu_end + per_cpu_offset(i);
+ end = (unsigned long) &__per_cpu_start + PERCPU_ENOUGH_ROOM
+ + per_cpu_offset(i);
if ((addr >= start) && (addr < end))
return 1;