summaryrefslogtreecommitdiff
path: root/kernel
diff options
context:
space:
mode:
Diffstat (limited to 'kernel')
-rw-r--r--kernel/Makefile1
-rw-r--r--kernel/audit.c13
-rw-r--r--kernel/auditfilter.c10
-rw-r--r--kernel/auditsc.c4
-rw-r--r--kernel/cgroup.c39
-rw-r--r--kernel/cpu.c41
-rw-r--r--kernel/cpuset.c71
-rw-r--r--kernel/dma-coherent.c154
-rw-r--r--kernel/fork.c3
-rw-r--r--kernel/kgdb.c94
-rw-r--r--kernel/marker.c12
-rw-r--r--kernel/module.c33
-rw-r--r--kernel/printk.c8
-rw-r--r--kernel/rcuclassic.c4
-rw-r--r--kernel/resource.c2
-rw-r--r--kernel/sched.c22
-rw-r--r--kernel/stop_machine.c288
-rw-r--r--kernel/time/tick-common.c8
-rw-r--r--kernel/trace/ftrace.c6
-rw-r--r--kernel/trace/trace_sysprof.c4
-rw-r--r--kernel/workqueue.c13
21 files changed, 537 insertions, 293 deletions
diff --git a/kernel/Makefile b/kernel/Makefile
index 54f69837d35a..4e1d7df7c3e2 100644
--- a/kernel/Makefile
+++ b/kernel/Makefile
@@ -84,6 +84,7 @@ obj-$(CONFIG_TASK_DELAY_ACCT) += delayacct.o
obj-$(CONFIG_TASKSTATS) += taskstats.o tsacct.o
obj-$(CONFIG_MARKERS) += marker.o
obj-$(CONFIG_LATENCYTOP) += latencytop.o
+obj-$(CONFIG_HAVE_GENERIC_DMA_COHERENT) += dma-coherent.o
obj-$(CONFIG_FTRACE) += trace/
obj-$(CONFIG_TRACING) += trace/
obj-$(CONFIG_SMP) += sched_cpupri.o
diff --git a/kernel/audit.c b/kernel/audit.c
index e092f1c0ce30..4414e93d8750 100644
--- a/kernel/audit.c
+++ b/kernel/audit.c
@@ -707,12 +707,14 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh)
if (status_get->mask & AUDIT_STATUS_ENABLED) {
err = audit_set_enabled(status_get->enabled,
loginuid, sessionid, sid);
- if (err < 0) return err;
+ if (err < 0)
+ return err;
}
if (status_get->mask & AUDIT_STATUS_FAILURE) {
err = audit_set_failure(status_get->failure,
loginuid, sessionid, sid);
- if (err < 0) return err;
+ if (err < 0)
+ return err;
}
if (status_get->mask & AUDIT_STATUS_PID) {
int new_pid = status_get->pid;
@@ -725,9 +727,12 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh)
audit_pid = new_pid;
audit_nlk_pid = NETLINK_CB(skb).pid;
}
- if (status_get->mask & AUDIT_STATUS_RATE_LIMIT)
+ if (status_get->mask & AUDIT_STATUS_RATE_LIMIT) {
err = audit_set_rate_limit(status_get->rate_limit,
loginuid, sessionid, sid);
+ if (err < 0)
+ return err;
+ }
if (status_get->mask & AUDIT_STATUS_BACKLOG_LIMIT)
err = audit_set_backlog_limit(status_get->backlog_limit,
loginuid, sessionid, sid);
@@ -1366,7 +1371,7 @@ int audit_string_contains_control(const char *string, size_t len)
{
const unsigned char *p;
for (p = string; p < (const unsigned char *)string + len && *p; p++) {
- if (*p == '"' || *p < 0x21 || *p > 0x7f)
+ if (*p == '"' || *p < 0x21 || *p > 0x7e)
return 1;
}
return 0;
diff --git a/kernel/auditfilter.c b/kernel/auditfilter.c
index 98c50cc671bb..b7d354e2b0ef 100644
--- a/kernel/auditfilter.c
+++ b/kernel/auditfilter.c
@@ -1022,8 +1022,11 @@ static void audit_update_watch(struct audit_parent *parent,
struct audit_buffer *ab;
ab = audit_log_start(NULL, GFP_KERNEL,
AUDIT_CONFIG_CHANGE);
+ audit_log_format(ab, "auid=%u ses=%u",
+ audit_get_loginuid(current),
+ audit_get_sessionid(current));
audit_log_format(ab,
- "op=updated rules specifying path=");
+ " op=updated rules specifying path=");
audit_log_untrustedstring(ab, owatch->path);
audit_log_format(ab, " with dev=%u ino=%lu\n",
dev, ino);
@@ -1058,7 +1061,10 @@ static void audit_remove_parent_watches(struct audit_parent *parent)
struct audit_buffer *ab;
ab = audit_log_start(NULL, GFP_KERNEL,
AUDIT_CONFIG_CHANGE);
- audit_log_format(ab, "op=remove rule path=");
+ audit_log_format(ab, "auid=%u ses=%u",
+ audit_get_loginuid(current),
+ audit_get_sessionid(current));
+ audit_log_format(ab, " op=remove rule path=");
audit_log_untrustedstring(ab, w->path);
if (r->filterkey) {
audit_log_format(ab, " key=");
diff --git a/kernel/auditsc.c b/kernel/auditsc.c
index 4699950e65bd..496c3dd37276 100644
--- a/kernel/auditsc.c
+++ b/kernel/auditsc.c
@@ -610,7 +610,7 @@ static int audit_filter_rules(struct task_struct *tsk,
if (!result)
return 0;
}
- if (rule->filterkey)
+ if (rule->filterkey && ctx)
ctx->filterkey = kstrdup(rule->filterkey, GFP_ATOMIC);
switch (rule->action) {
case AUDIT_NEVER: *state = AUDIT_DISABLED; break;
@@ -2375,7 +2375,7 @@ int __audit_signal_info(int sig, struct task_struct *t)
struct audit_context *ctx = tsk->audit_context;
if (audit_pid && t->tgid == audit_pid) {
- if (sig == SIGTERM || sig == SIGHUP || sig == SIGUSR1) {
+ if (sig == SIGTERM || sig == SIGHUP || sig == SIGUSR1 || sig == SIGUSR2) {
audit_sig_pid = tsk->pid;
if (tsk->loginuid != -1)
audit_sig_uid = tsk->loginuid;
diff --git a/kernel/cgroup.c b/kernel/cgroup.c
index 657f8f8d93a5..13932abde159 100644
--- a/kernel/cgroup.c
+++ b/kernel/cgroup.c
@@ -355,6 +355,17 @@ static struct css_set *find_existing_css_set(
return NULL;
}
+static void free_cg_links(struct list_head *tmp)
+{
+ struct cg_cgroup_link *link;
+ struct cg_cgroup_link *saved_link;
+
+ list_for_each_entry_safe(link, saved_link, tmp, cgrp_link_list) {
+ list_del(&link->cgrp_link_list);
+ kfree(link);
+ }
+}
+
/*
* allocate_cg_links() allocates "count" cg_cgroup_link structures
* and chains them on tmp through their cgrp_link_list fields. Returns 0 on
@@ -363,17 +374,12 @@ static struct css_set *find_existing_css_set(
static int allocate_cg_links(int count, struct list_head *tmp)
{
struct cg_cgroup_link *link;
- struct cg_cgroup_link *saved_link;
int i;
INIT_LIST_HEAD(tmp);
for (i = 0; i < count; i++) {
link = kmalloc(sizeof(*link), GFP_KERNEL);
if (!link) {
- list_for_each_entry_safe(link, saved_link, tmp,
- cgrp_link_list) {
- list_del(&link->cgrp_link_list);
- kfree(link);
- }
+ free_cg_links(tmp);
return -ENOMEM;
}
list_add(&link->cgrp_link_list, tmp);
@@ -381,17 +387,6 @@ static int allocate_cg_links(int count, struct list_head *tmp)
return 0;
}
-static void free_cg_links(struct list_head *tmp)
-{
- struct cg_cgroup_link *link;
- struct cg_cgroup_link *saved_link;
-
- list_for_each_entry_safe(link, saved_link, tmp, cgrp_link_list) {
- list_del(&link->cgrp_link_list);
- kfree(link);
- }
-}
-
/*
* find_css_set() takes an existing cgroup group and a
* cgroup object, and returns a css_set object that's
@@ -956,7 +951,6 @@ static int cgroup_get_sb(struct file_system_type *fs_type,
struct super_block *sb;
struct cgroupfs_root *root;
struct list_head tmp_cg_links;
- INIT_LIST_HEAD(&tmp_cg_links);
/* First find the desired set of subsystems */
ret = parse_cgroupfs_options(data, &opts);
@@ -1424,14 +1418,17 @@ static ssize_t cgroup_write_string(struct cgroup *cgrp, struct cftype *cft,
if (buffer == NULL)
return -ENOMEM;
}
- if (nbytes && copy_from_user(buffer, userbuf, nbytes))
- return -EFAULT;
+ if (nbytes && copy_from_user(buffer, userbuf, nbytes)) {
+ retval = -EFAULT;
+ goto out;
+ }
buffer[nbytes] = 0; /* nul-terminate */
strstrip(buffer);
retval = cft->write_string(cgrp, cft, buffer);
if (!retval)
retval = nbytes;
+out:
if (buffer != local_buffer)
kfree(buffer);
return retval;
@@ -2371,7 +2368,7 @@ static int cgroup_mkdir(struct inode *dir, struct dentry *dentry, int mode)
return cgroup_create(c_parent, dentry, mode | S_IFDIR);
}
-static inline int cgroup_has_css_refs(struct cgroup *cgrp)
+static int cgroup_has_css_refs(struct cgroup *cgrp)
{
/* Check the reference count on each subsystem. Since we
* already established that there are no tasks in the
diff --git a/kernel/cpu.c b/kernel/cpu.c
index 10ba5f1004a5..e202a68d1cc1 100644
--- a/kernel/cpu.c
+++ b/kernel/cpu.c
@@ -216,7 +216,6 @@ static int __ref take_cpu_down(void *_param)
static int __ref _cpu_down(unsigned int cpu, int tasks_frozen)
{
int err, nr_calls = 0;
- struct task_struct *p;
cpumask_t old_allowed, tmp;
void *hcpu = (void *)(long)cpu;
unsigned long mod = tasks_frozen ? CPU_TASKS_FROZEN : 0;
@@ -249,21 +248,18 @@ static int __ref _cpu_down(unsigned int cpu, int tasks_frozen)
cpus_setall(tmp);
cpu_clear(cpu, tmp);
set_cpus_allowed_ptr(current, &tmp);
+ tmp = cpumask_of_cpu(cpu);
- p = __stop_machine_run(take_cpu_down, &tcd_param, cpu);
-
- if (IS_ERR(p) || cpu_online(cpu)) {
+ err = __stop_machine(take_cpu_down, &tcd_param, &tmp);
+ if (err) {
/* CPU didn't die: tell everyone. Can't complain. */
if (raw_notifier_call_chain(&cpu_chain, CPU_DOWN_FAILED | mod,
hcpu) == NOTIFY_BAD)
BUG();
- if (IS_ERR(p)) {
- err = PTR_ERR(p);
- goto out_allowed;
- }
- goto out_thread;
+ goto out_allowed;
}
+ BUG_ON(cpu_online(cpu));
/* Wait for it to sleep (leaving idle task). */
while (!idle_cpu(cpu))
@@ -279,8 +275,6 @@ static int __ref _cpu_down(unsigned int cpu, int tasks_frozen)
check_for_tasks(cpu);
-out_thread:
- err = kthread_stop(p);
out_allowed:
set_cpus_allowed_ptr(current, &old_allowed);
out_release:
@@ -461,3 +455,28 @@ out:
#endif /* CONFIG_PM_SLEEP_SMP */
#endif /* CONFIG_SMP */
+
+/*
+ * cpu_bit_bitmap[] is a special, "compressed" data structure that
+ * represents all NR_CPUS bits binary values of 1<<nr.
+ *
+ * It is used by cpumask_of_cpu() to get a constant address to a CPU
+ * mask value that has a single bit set only.
+ */
+
+/* cpu_bit_bitmap[0] is empty - so we can back into it */
+#define MASK_DECLARE_1(x) [x+1][0] = 1UL << (x)
+#define MASK_DECLARE_2(x) MASK_DECLARE_1(x), MASK_DECLARE_1(x+1)
+#define MASK_DECLARE_4(x) MASK_DECLARE_2(x), MASK_DECLARE_2(x+2)
+#define MASK_DECLARE_8(x) MASK_DECLARE_4(x), MASK_DECLARE_4(x+4)
+
+const unsigned long cpu_bit_bitmap[BITS_PER_LONG+1][BITS_TO_LONGS(NR_CPUS)] = {
+
+ MASK_DECLARE_8(0), MASK_DECLARE_8(8),
+ MASK_DECLARE_8(16), MASK_DECLARE_8(24),
+#if BITS_PER_LONG > 32
+ MASK_DECLARE_8(32), MASK_DECLARE_8(40),
+ MASK_DECLARE_8(48), MASK_DECLARE_8(56),
+#endif
+};
+EXPORT_SYMBOL_GPL(cpu_bit_bitmap);
diff --git a/kernel/cpuset.c b/kernel/cpuset.c
index 91cf85b36dd5..d5ab79cf516d 100644
--- a/kernel/cpuset.c
+++ b/kernel/cpuset.c
@@ -54,7 +54,6 @@
#include <asm/uaccess.h>
#include <asm/atomic.h>
#include <linux/mutex.h>
-#include <linux/kfifo.h>
#include <linux/workqueue.h>
#include <linux/cgroup.h>
@@ -486,13 +485,38 @@ static int cpusets_overlap(struct cpuset *a, struct cpuset *b)
static void
update_domain_attr(struct sched_domain_attr *dattr, struct cpuset *c)
{
- if (!dattr)
- return;
if (dattr->relax_domain_level < c->relax_domain_level)
dattr->relax_domain_level = c->relax_domain_level;
return;
}
+static void
+update_domain_attr_tree(struct sched_domain_attr *dattr, struct cpuset *c)
+{
+ LIST_HEAD(q);
+
+ list_add(&c->stack_list, &q);
+ while (!list_empty(&q)) {
+ struct cpuset *cp;
+ struct cgroup *cont;
+ struct cpuset *child;
+
+ cp = list_first_entry(&q, struct cpuset, stack_list);
+ list_del(q.next);
+
+ if (cpus_empty(cp->cpus_allowed))
+ continue;
+
+ if (is_sched_load_balance(cp))
+ update_domain_attr(dattr, cp);
+
+ list_for_each_entry(cont, &cp->css.cgroup->children, sibling) {
+ child = cgroup_cs(cont);
+ list_add_tail(&child->stack_list, &q);
+ }
+ }
+}
+
/*
* rebuild_sched_domains()
*
@@ -532,7 +556,7 @@ update_domain_attr(struct sched_domain_attr *dattr, struct cpuset *c)
* So the reverse nesting would risk an ABBA deadlock.
*
* The three key local variables below are:
- * q - a kfifo queue of cpuset pointers, used to implement a
+ * q - a linked-list queue of cpuset pointers, used to implement a
* top-down scan of all cpusets. This scan loads a pointer
* to each cpuset marked is_sched_load_balance into the
* array 'csa'. For our purposes, rebuilding the schedulers
@@ -567,7 +591,7 @@ update_domain_attr(struct sched_domain_attr *dattr, struct cpuset *c)
void rebuild_sched_domains(void)
{
- struct kfifo *q; /* queue of cpusets to be scanned */
+ LIST_HEAD(q); /* queue of cpusets to be scanned*/
struct cpuset *cp; /* scans q */
struct cpuset **csa; /* array of all cpuset ptrs */
int csn; /* how many cpuset ptrs in csa so far */
@@ -577,7 +601,6 @@ void rebuild_sched_domains(void)
int ndoms; /* number of sched domains in result */
int nslot; /* next empty doms[] cpumask_t slot */
- q = NULL;
csa = NULL;
doms = NULL;
dattr = NULL;
@@ -591,35 +614,42 @@ void rebuild_sched_domains(void)
dattr = kmalloc(sizeof(struct sched_domain_attr), GFP_KERNEL);
if (dattr) {
*dattr = SD_ATTR_INIT;
- update_domain_attr(dattr, &top_cpuset);
+ update_domain_attr_tree(dattr, &top_cpuset);
}
*doms = top_cpuset.cpus_allowed;
goto rebuild;
}
- q = kfifo_alloc(number_of_cpusets * sizeof(cp), GFP_KERNEL, NULL);
- if (IS_ERR(q))
- goto done;
csa = kmalloc(number_of_cpusets * sizeof(cp), GFP_KERNEL);
if (!csa)
goto done;
csn = 0;
- cp = &top_cpuset;
- __kfifo_put(q, (void *)&cp, sizeof(cp));
- while (__kfifo_get(q, (void *)&cp, sizeof(cp))) {
+ list_add(&top_cpuset.stack_list, &q);
+ while (!list_empty(&q)) {
struct cgroup *cont;
struct cpuset *child; /* scans child cpusets of cp */
+ cp = list_first_entry(&q, struct cpuset, stack_list);
+ list_del(q.next);
+
if (cpus_empty(cp->cpus_allowed))
continue;
- if (is_sched_load_balance(cp))
+ /*
+ * All child cpusets contain a subset of the parent's cpus, so
+ * just skip them, and then we call update_domain_attr_tree()
+ * to calc relax_domain_level of the corresponding sched
+ * domain.
+ */
+ if (is_sched_load_balance(cp)) {
csa[csn++] = cp;
+ continue;
+ }
list_for_each_entry(cont, &cp->css.cgroup->children, sibling) {
child = cgroup_cs(cont);
- __kfifo_put(q, (void *)&child, sizeof(cp));
+ list_add_tail(&child->stack_list, &q);
}
}
@@ -686,7 +716,7 @@ restart:
cpus_or(*dp, *dp, b->cpus_allowed);
b->pn = -1;
if (dattr)
- update_domain_attr(dattr
+ update_domain_attr_tree(dattr
+ nslot, b);
}
}
@@ -702,8 +732,6 @@ rebuild:
put_online_cpus();
done:
- if (q && !IS_ERR(q))
- kfifo_free(q);
kfree(csa);
/* Don't kfree(doms) -- partition_sched_domains() does that. */
/* Don't kfree(dattr) -- partition_sched_domains() does that. */
@@ -1833,24 +1861,21 @@ static void remove_tasks_in_empty_cpuset(struct cpuset *cs)
*/
static void scan_for_empty_cpusets(const struct cpuset *root)
{
+ LIST_HEAD(queue);
struct cpuset *cp; /* scans cpusets being updated */
struct cpuset *child; /* scans child cpusets of cp */
- struct list_head queue;
struct cgroup *cont;
nodemask_t oldmems;
- INIT_LIST_HEAD(&queue);
-
list_add_tail((struct list_head *)&root->stack_list, &queue);
while (!list_empty(&queue)) {
- cp = container_of(queue.next, struct cpuset, stack_list);
+ cp = list_first_entry(&queue, struct cpuset, stack_list);
list_del(queue.next);
list_for_each_entry(cont, &cp->css.cgroup->children, sibling) {
child = cgroup_cs(cont);
list_add_tail(&child->stack_list, &queue);
}
- cont = cp->css.cgroup;
/* Continue past cpusets with all cpus, mems online */
if (cpus_subset(cp->cpus_allowed, cpu_online_map) &&
diff --git a/kernel/dma-coherent.c b/kernel/dma-coherent.c
new file mode 100644
index 000000000000..7517115a8cce
--- /dev/null
+++ b/kernel/dma-coherent.c
@@ -0,0 +1,154 @@
+/*
+ * Coherent per-device memory handling.
+ * Borrowed from i386
+ */
+#include <linux/kernel.h>
+#include <linux/dma-mapping.h>
+
+struct dma_coherent_mem {
+ void *virt_base;
+ u32 device_base;
+ int size;
+ int flags;
+ unsigned long *bitmap;
+};
+
+int dma_declare_coherent_memory(struct device *dev, dma_addr_t bus_addr,
+ dma_addr_t device_addr, size_t size, int flags)
+{
+ void __iomem *mem_base = NULL;
+ int pages = size >> PAGE_SHIFT;
+ int bitmap_size = BITS_TO_LONGS(pages) * sizeof(long);
+
+ if ((flags & (DMA_MEMORY_MAP | DMA_MEMORY_IO)) == 0)
+ goto out;
+ if (!size)
+ goto out;
+ if (dev->dma_mem)
+ goto out;
+
+ /* FIXME: this routine just ignores DMA_MEMORY_INCLUDES_CHILDREN */
+
+ mem_base = ioremap(bus_addr, size);
+ if (!mem_base)
+ goto out;
+
+ dev->dma_mem = kzalloc(sizeof(struct dma_coherent_mem), GFP_KERNEL);
+ if (!dev->dma_mem)
+ goto out;
+ dev->dma_mem->bitmap = kzalloc(bitmap_size, GFP_KERNEL);
+ if (!dev->dma_mem->bitmap)
+ goto free1_out;
+
+ dev->dma_mem->virt_base = mem_base;
+ dev->dma_mem->device_base = device_addr;
+ dev->dma_mem->size = pages;
+ dev->dma_mem->flags = flags;
+
+ if (flags & DMA_MEMORY_MAP)
+ return DMA_MEMORY_MAP;
+
+ return DMA_MEMORY_IO;
+
+ free1_out:
+ kfree(dev->dma_mem);
+ out:
+ if (mem_base)
+ iounmap(mem_base);
+ return 0;
+}
+EXPORT_SYMBOL(dma_declare_coherent_memory);
+
+void dma_release_declared_memory(struct device *dev)
+{
+ struct dma_coherent_mem *mem = dev->dma_mem;
+
+ if (!mem)
+ return;
+ dev->dma_mem = NULL;
+ iounmap(mem->virt_base);
+ kfree(mem->bitmap);
+ kfree(mem);
+}
+EXPORT_SYMBOL(dma_release_declared_memory);
+
+void *dma_mark_declared_memory_occupied(struct device *dev,
+ dma_addr_t device_addr, size_t size)
+{
+ struct dma_coherent_mem *mem = dev->dma_mem;
+ int pos, err;
+ int pages = (size + (device_addr & ~PAGE_MASK) + PAGE_SIZE - 1);
+
+ pages >>= PAGE_SHIFT;
+
+ if (!mem)
+ return ERR_PTR(-EINVAL);
+
+ pos = (device_addr - mem->device_base) >> PAGE_SHIFT;
+ err = bitmap_allocate_region(mem->bitmap, pos, get_order(pages));
+ if (err != 0)
+ return ERR_PTR(err);
+ return mem->virt_base + (pos << PAGE_SHIFT);
+}
+EXPORT_SYMBOL(dma_mark_declared_memory_occupied);
+
+/**
+ * Try to allocate memory from the per-device coherent area.
+ *
+ * @dev: device from which we allocate memory
+ * @size: size of requested memory area
+ * @dma_handle: This will be filled with the correct dma handle
+ * @ret: This pointer will be filled with the virtual address
+ * to allocated area.
+ *
+ * This function should be only called from per-arch %dma_alloc_coherent()
+ * to support allocation from per-device coherent memory pools.
+ *
+ * Returns 0 if dma_alloc_coherent should continue with allocating from
+ * generic memory areas, or !0 if dma_alloc_coherent should return %ret.
+ */
+int dma_alloc_from_coherent(struct device *dev, ssize_t size,
+ dma_addr_t *dma_handle, void **ret)
+{
+ struct dma_coherent_mem *mem = dev ? dev->dma_mem : NULL;
+ int order = get_order(size);
+
+ if (mem) {
+ int page = bitmap_find_free_region(mem->bitmap, mem->size,
+ order);
+ if (page >= 0) {
+ *dma_handle = mem->device_base + (page << PAGE_SHIFT);
+ *ret = mem->virt_base + (page << PAGE_SHIFT);
+ memset(*ret, 0, size);
+ } else if (mem->flags & DMA_MEMORY_EXCLUSIVE)
+ *ret = NULL;
+ }
+ return (mem != NULL);
+}
+
+/**
+ * Try to free the memory allocated from per-device coherent memory pool.
+ * @dev: device from which the memory was allocated
+ * @order: the order of pages allocated
+ * @vaddr: virtual address of allocated pages
+ *
+ * This checks whether the memory was allocated from the per-device
+ * coherent memory pool and if so, releases that memory.
+ *
+ * Returns 1 if we correctly released the memory, or 0 if
+ * %dma_release_coherent() should proceed with releasing memory from
+ * generic pools.
+ */
+int dma_release_from_coherent(struct device *dev, int order, void *vaddr)
+{
+ struct dma_coherent_mem *mem = dev ? dev->dma_mem : NULL;
+
+ if (mem && vaddr >= mem->virt_base && vaddr <
+ (mem->virt_base + (mem->size << PAGE_SHIFT))) {
+ int page = (vaddr - mem->virt_base) >> PAGE_SHIFT;
+
+ bitmap_release_region(mem->bitmap, page, order);
+ return 1;
+ }
+ return 0;
+}
diff --git a/kernel/fork.c b/kernel/fork.c
index 8214ba7c8bb1..7ce2ebe84796 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -27,6 +27,7 @@
#include <linux/key.h>
#include <linux/binfmts.h>
#include <linux/mman.h>
+#include <linux/mmu_notifier.h>
#include <linux/fs.h>
#include <linux/nsproxy.h>
#include <linux/capability.h>
@@ -414,6 +415,7 @@ static struct mm_struct * mm_init(struct mm_struct * mm, struct task_struct *p)
if (likely(!mm_alloc_pgd(mm))) {
mm->def_flags = 0;
+ mmu_notifier_mm_init(mm);
return mm;
}
@@ -446,6 +448,7 @@ void __mmdrop(struct mm_struct *mm)
BUG_ON(mm == &init_mm);
mm_free_pgd(mm);
destroy_context(mm);
+ mmu_notifier_mm_destroy(mm);
free_mm(mm);
}
EXPORT_SYMBOL_GPL(__mmdrop);
diff --git a/kernel/kgdb.c b/kernel/kgdb.c
index 3ec23c3ec97f..eaa21fc9ad1d 100644
--- a/kernel/kgdb.c
+++ b/kernel/kgdb.c
@@ -56,12 +56,14 @@
static int kgdb_break_asap;
+#define KGDB_MAX_THREAD_QUERY 17
struct kgdb_state {
int ex_vector;
int signo;
int err_code;
int cpu;
int pass_exception;
+ unsigned long thr_query;
unsigned long threadid;
long kgdb_usethreadid;
struct pt_regs *linux_regs;
@@ -166,13 +168,6 @@ early_param("nokgdbroundup", opt_nokgdbroundup);
* Weak aliases for breakpoint management,
* can be overriden by architectures when needed:
*/
-int __weak kgdb_validate_break_address(unsigned long addr)
-{
- char tmp_variable[BREAK_INSTR_SIZE];
-
- return probe_kernel_read(tmp_variable, (char *)addr, BREAK_INSTR_SIZE);
-}
-
int __weak kgdb_arch_set_breakpoint(unsigned long addr, char *saved_instr)
{
int err;
@@ -191,6 +186,25 @@ int __weak kgdb_arch_remove_breakpoint(unsigned long addr, char *bundle)
(char *)bundle, BREAK_INSTR_SIZE);
}
+int __weak kgdb_validate_break_address(unsigned long addr)
+{
+ char tmp_variable[BREAK_INSTR_SIZE];
+ int err;
+ /* Validate setting the breakpoint and then removing it. In the
+ * remove fails, the kernel needs to emit a bad message because we
+ * are deep trouble not being able to put things back the way we
+ * found them.
+ */
+ err = kgdb_arch_set_breakpoint(addr, tmp_variable);
+ if (err)
+ return err;
+ err = kgdb_arch_remove_breakpoint(addr, tmp_variable);
+ if (err)
+ printk(KERN_ERR "KGDB: Critical breakpoint error, kernel "
+ "memory destroyed at: %lx", addr);
+ return err;
+}
+
unsigned long __weak kgdb_arch_pc(int exception, struct pt_regs *regs)
{
return instruction_pointer(regs);
@@ -433,9 +447,14 @@ int kgdb_hex2long(char **ptr, unsigned long *long_val)
{
int hex_val;
int num = 0;
+ int negate = 0;
*long_val = 0;
+ if (**ptr == '-') {
+ negate = 1;
+ (*ptr)++;
+ }
while (**ptr) {
hex_val = hex(**ptr);
if (hex_val < 0)
@@ -446,6 +465,9 @@ int kgdb_hex2long(char **ptr, unsigned long *long_val)
(*ptr)++;
}
+ if (negate)
+ *long_val = -*long_val;
+
return num;
}
@@ -515,10 +537,16 @@ static void int_to_threadref(unsigned char *id, int value)
static struct task_struct *getthread(struct pt_regs *regs, int tid)
{
/*
- * Non-positive TIDs are remapped idle tasks:
+ * Non-positive TIDs are remapped to the cpu shadow information
*/
- if (tid <= 0)
- return idle_task(-tid);
+ if (tid == 0 || tid == -1)
+ tid = -atomic_read(&kgdb_active) - 2;
+ if (tid < 0) {
+ if (kgdb_info[-tid - 2].task)
+ return kgdb_info[-tid - 2].task;
+ else
+ return idle_task(-tid - 2);
+ }
/*
* find_task_by_pid_ns() does not take the tasklist lock anymore
@@ -725,14 +753,15 @@ setundefined:
}
/*
- * Remap normal tasks to their real PID, idle tasks to -1 ... -NR_CPUs:
+ * Remap normal tasks to their real PID,
+ * CPU shadow threads are mapped to -CPU - 2
*/
static inline int shadow_pid(int realpid)
{
if (realpid)
return realpid;
- return -1-raw_smp_processor_id();
+ return -raw_smp_processor_id() - 2;
}
static char gdbmsgbuf[BUFMAX + 1];
@@ -826,7 +855,7 @@ static void gdb_cmd_getregs(struct kgdb_state *ks)
local_debuggerinfo = kgdb_info[ks->cpu].debuggerinfo;
} else {
local_debuggerinfo = NULL;
- for (i = 0; i < NR_CPUS; i++) {
+ for_each_online_cpu(i) {
/*
* Try to find the task on some other
* or possibly this node if we do not
@@ -960,10 +989,13 @@ static int gdb_cmd_reboot(struct kgdb_state *ks)
/* Handle the 'q' query packets */
static void gdb_cmd_query(struct kgdb_state *ks)
{
- struct task_struct *thread;
+ struct task_struct *g;
+ struct task_struct *p;
unsigned char thref[8];
char *ptr;
int i;
+ int cpu;
+ int finished = 0;
switch (remcom_in_buffer[1]) {
case 's':
@@ -973,22 +1005,34 @@ static void gdb_cmd_query(struct kgdb_state *ks)
break;
}
- if (remcom_in_buffer[1] == 'f')
- ks->threadid = 1;
-
+ i = 0;
remcom_out_buffer[0] = 'm';
ptr = remcom_out_buffer + 1;
-
- for (i = 0; i < 17; ks->threadid++) {
- thread = getthread(ks->linux_regs, ks->threadid);
- if (thread) {
- int_to_threadref(thref, ks->threadid);
+ if (remcom_in_buffer[1] == 'f') {
+ /* Each cpu is a shadow thread */
+ for_each_online_cpu(cpu) {
+ ks->thr_query = 0;
+ int_to_threadref(thref, -cpu - 2);
pack_threadid(ptr, thref);
ptr += BUF_THREAD_ID_SIZE;
*(ptr++) = ',';
i++;
}
}
+
+ do_each_thread(g, p) {
+ if (i >= ks->thr_query && !finished) {
+ int_to_threadref(thref, p->pid);
+ pack_threadid(ptr, thref);
+ ptr += BUF_THREAD_ID_SIZE;
+ *(ptr++) = ',';
+ ks->thr_query++;
+ if (ks->thr_query % KGDB_MAX_THREAD_QUERY == 0)
+ finished = 1;
+ }
+ i++;
+ } while_each_thread(g, p);
+
*(--ptr) = '\0';
break;
@@ -1011,15 +1055,15 @@ static void gdb_cmd_query(struct kgdb_state *ks)
error_packet(remcom_out_buffer, -EINVAL);
break;
}
- if (ks->threadid > 0) {
+ if ((int)ks->threadid > 0) {
kgdb_mem2hex(getthread(ks->linux_regs,
ks->threadid)->comm,
remcom_out_buffer, 16);
} else {
static char tmpstr[23 + BUF_THREAD_ID_SIZE];
- sprintf(tmpstr, "Shadow task %d for pid 0",
- (int)(-ks->threadid-1));
+ sprintf(tmpstr, "shadowCPU%d",
+ (int)(-ks->threadid - 2));
kgdb_mem2hex(tmpstr, remcom_out_buffer, strlen(tmpstr));
}
break;
diff --git a/kernel/marker.c b/kernel/marker.c
index 971da5317903..7d1faecd7a51 100644
--- a/kernel/marker.c
+++ b/kernel/marker.c
@@ -126,6 +126,11 @@ void marker_probe_cb(const struct marker *mdata, void *call_private, ...)
struct marker_probe_closure *multi;
int i;
/*
+ * Read mdata->ptype before mdata->multi.
+ */
+ smp_rmb();
+ multi = mdata->multi;
+ /*
* multi points to an array, therefore accessing the array
* depends on reading multi. However, even in this case,
* we must insure that the pointer is read _before_ the array
@@ -133,7 +138,6 @@ void marker_probe_cb(const struct marker *mdata, void *call_private, ...)
* in the fast path, so put the explicit barrier here.
*/
smp_read_barrier_depends();
- multi = mdata->multi;
for (i = 0; multi[i].func; i++) {
va_start(args, call_private);
multi[i].func(multi[i].probe_private, call_private,
@@ -175,6 +179,11 @@ void marker_probe_cb_noarg(const struct marker *mdata, void *call_private, ...)
struct marker_probe_closure *multi;
int i;
/*
+ * Read mdata->ptype before mdata->multi.
+ */
+ smp_rmb();
+ multi = mdata->multi;
+ /*
* multi points to an array, therefore accessing the array
* depends on reading multi. However, even in this case,
* we must insure that the pointer is read _before_ the array
@@ -182,7 +191,6 @@ void marker_probe_cb_noarg(const struct marker *mdata, void *call_private, ...)
* in the fast path, so put the explicit barrier here.
*/
smp_read_barrier_depends();
- multi = mdata->multi;
for (i = 0; multi[i].func; i++)
multi[i].func(multi[i].probe_private, call_private,
mdata->format, &args);
diff --git a/kernel/module.c b/kernel/module.c
index d8b5605132a0..61d212120df4 100644
--- a/kernel/module.c
+++ b/kernel/module.c
@@ -325,18 +325,6 @@ static unsigned long find_symbol(const char *name,
return -ENOENT;
}
-/* lookup symbol in given range of kernel_symbols */
-static const struct kernel_symbol *lookup_symbol(const char *name,
- const struct kernel_symbol *start,
- const struct kernel_symbol *stop)
-{
- const struct kernel_symbol *ks = start;
- for (; ks < stop; ks++)
- if (strcmp(ks->name, name) == 0)
- return ks;
- return NULL;
-}
-
/* Search for module by name: must hold module_mutex. */
static struct module *find_module(const char *name)
{
@@ -690,7 +678,7 @@ static int try_stop_module(struct module *mod, int flags, int *forced)
if (flags & O_NONBLOCK) {
struct stopref sref = { mod, flags, forced };
- return stop_machine_run(__try_stop_module, &sref, NR_CPUS);
+ return stop_machine(__try_stop_module, &sref, NULL);
} else {
/* We don't need to stop the machine for this. */
mod->state = MODULE_STATE_GOING;
@@ -1428,7 +1416,7 @@ static int __unlink_module(void *_mod)
static void free_module(struct module *mod)
{
/* Delete from various lists */
- stop_machine_run(__unlink_module, mod, NR_CPUS);
+ stop_machine(__unlink_module, mod, NULL);
remove_notes_attrs(mod);
remove_sect_attrs(mod);
mod_kobject_remove(mod);
@@ -1703,6 +1691,19 @@ static void setup_modinfo(struct module *mod, Elf_Shdr *sechdrs,
}
#ifdef CONFIG_KALLSYMS
+
+/* lookup symbol in given range of kernel_symbols */
+static const struct kernel_symbol *lookup_symbol(const char *name,
+ const struct kernel_symbol *start,
+ const struct kernel_symbol *stop)
+{
+ const struct kernel_symbol *ks = start;
+ for (; ks < stop; ks++)
+ if (strcmp(ks->name, name) == 0)
+ return ks;
+ return NULL;
+}
+
static int is_exported(const char *name, const struct module *mod)
{
if (!mod && lookup_symbol(name, __start___ksymtab, __stop___ksymtab))
@@ -2196,7 +2197,7 @@ static struct module *load_module(void __user *umod,
/* Now sew it into the lists so we can get lockdep and oops
* info during argument parsing. Noone should access us, since
* strong_try_module_get() will fail. */
- stop_machine_run(__link_module, mod, NR_CPUS);
+ stop_machine(__link_module, mod, NULL);
/* Size of section 0 is 0, so this works well if no params */
err = parse_args(mod->name, mod->args,
@@ -2230,7 +2231,7 @@ static struct module *load_module(void __user *umod,
return mod;
unlink:
- stop_machine_run(__unlink_module, mod, NR_CPUS);
+ stop_machine(__unlink_module, mod, NULL);
module_arch_cleanup(mod);
cleanup:
kobject_del(&mod->mkobj.kobj);
diff --git a/kernel/printk.c b/kernel/printk.c
index a7f7559c5f6c..b51b1567bb55 100644
--- a/kernel/printk.c
+++ b/kernel/printk.c
@@ -1309,14 +1309,14 @@ void tty_write_message(struct tty_struct *tty, char *msg)
#if defined CONFIG_PRINTK
-DEFINE_RATELIMIT_STATE(printk_ratelimit_state, 5 * HZ, 10);
/*
* printk rate limiting, lifted from the networking subsystem.
*
- * This enforces a rate limit: not more than one kernel message
- * every printk_ratelimit_jiffies to make a denial-of-service
- * attack impossible.
+ * This enforces a rate limit: not more than 10 kernel messages
+ * every 5s to make a denial-of-service attack impossible.
*/
+DEFINE_RATELIMIT_STATE(printk_ratelimit_state, 5 * HZ, 10);
+
int printk_ratelimit(void)
{
return __ratelimit(&printk_ratelimit_state);
diff --git a/kernel/rcuclassic.c b/kernel/rcuclassic.c
index 6f8696c502f4..aad93cdc9f68 100644
--- a/kernel/rcuclassic.c
+++ b/kernel/rcuclassic.c
@@ -91,8 +91,8 @@ static void force_quiescent_state(struct rcu_data *rdp,
* rdp->cpu is the current cpu.
*
* cpu_online_map is updated by the _cpu_down()
- * using stop_machine_run(). Since we're in irqs disabled
- * section, stop_machine_run() is not exectuting, hence
+ * using __stop_machine(). Since we're in irqs disabled
+ * section, __stop_machine() is not exectuting, hence
* the cpu_online_map is stable.
*
* However, a cpu might have been offlined _just_ before
diff --git a/kernel/resource.c b/kernel/resource.c
index 74af2d7cb5a1..f5b518eabefe 100644
--- a/kernel/resource.c
+++ b/kernel/resource.c
@@ -490,7 +490,7 @@ resource_size_t resource_alignment(struct resource *res)
{
switch (res->flags & (IORESOURCE_SIZEALIGN | IORESOURCE_STARTALIGN)) {
case IORESOURCE_SIZEALIGN:
- return res->end - res->start + 1;
+ return resource_size(res);
case IORESOURCE_STARTALIGN:
return res->start;
default:
diff --git a/kernel/sched.c b/kernel/sched.c
index 0236958addcb..21f7da94662e 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -7671,34 +7671,34 @@ static ssize_t sched_power_savings_store(const char *buf, size_t count, int smt)
}
#ifdef CONFIG_SCHED_MC
-static ssize_t sched_mc_power_savings_show(struct sys_device *dev,
- struct sysdev_attribute *attr, char *page)
+static ssize_t sched_mc_power_savings_show(struct sysdev_class *class,
+ char *page)
{
return sprintf(page, "%u\n", sched_mc_power_savings);
}
-static ssize_t sched_mc_power_savings_store(struct sys_device *dev,
- struct sysdev_attribute *attr,
+static ssize_t sched_mc_power_savings_store(struct sysdev_class *class,
const char *buf, size_t count)
{
return sched_power_savings_store(buf, count, 0);
}
-static SYSDEV_ATTR(sched_mc_power_savings, 0644, sched_mc_power_savings_show,
- sched_mc_power_savings_store);
+static SYSDEV_CLASS_ATTR(sched_mc_power_savings, 0644,
+ sched_mc_power_savings_show,
+ sched_mc_power_savings_store);
#endif
#ifdef CONFIG_SCHED_SMT
-static ssize_t sched_smt_power_savings_show(struct sys_device *dev,
- struct sysdev_attribute *attr, char *page)
+static ssize_t sched_smt_power_savings_show(struct sysdev_class *dev,
+ char *page)
{
return sprintf(page, "%u\n", sched_smt_power_savings);
}
-static ssize_t sched_smt_power_savings_store(struct sys_device *dev,
- struct sysdev_attribute *attr,
+static ssize_t sched_smt_power_savings_store(struct sysdev_class *dev,
const char *buf, size_t count)
{
return sched_power_savings_store(buf, count, 1);
}
-static SYSDEV_ATTR(sched_smt_power_savings, 0644, sched_smt_power_savings_show,
+static SYSDEV_CLASS_ATTR(sched_smt_power_savings, 0644,
+ sched_smt_power_savings_show,
sched_smt_power_savings_store);
#endif
diff --git a/kernel/stop_machine.c b/kernel/stop_machine.c
index 738b411ff2d3..e446c7c7d6a9 100644
--- a/kernel/stop_machine.c
+++ b/kernel/stop_machine.c
@@ -1,4 +1,4 @@
-/* Copyright 2005 Rusty Russell rusty@rustcorp.com.au IBM Corporation.
+/* Copyright 2008, 2005 Rusty Russell rusty@rustcorp.com.au IBM Corporation.
* GPL v2 and any later version.
*/
#include <linux/cpu.h>
@@ -13,204 +13,178 @@
#include <asm/atomic.h>
#include <asm/uaccess.h>
-/* Since we effect priority and affinity (both of which are visible
- * to, and settable by outside processes) we do indirection via a
- * kthread. */
-
-/* Thread to stop each CPU in user context. */
+/* This controls the threads on each CPU. */
enum stopmachine_state {
- STOPMACHINE_WAIT,
+ /* Dummy starting state for thread. */
+ STOPMACHINE_NONE,
+ /* Awaiting everyone to be scheduled. */
STOPMACHINE_PREPARE,
+ /* Disable interrupts. */
STOPMACHINE_DISABLE_IRQ,
+ /* Run the function */
+ STOPMACHINE_RUN,
+ /* Exit */
STOPMACHINE_EXIT,
};
+static enum stopmachine_state state;
-static enum stopmachine_state stopmachine_state;
-static unsigned int stopmachine_num_threads;
-static atomic_t stopmachine_thread_ack;
-
-static int stopmachine(void *cpu)
-{
- int irqs_disabled = 0;
- int prepared = 0;
- cpumask_of_cpu_ptr(cpumask, (int)(long)cpu);
-
- set_cpus_allowed_ptr(current, cpumask);
-
- /* Ack: we are alive */
- smp_mb(); /* Theoretically the ack = 0 might not be on this CPU yet. */
- atomic_inc(&stopmachine_thread_ack);
-
- /* Simple state machine */
- while (stopmachine_state != STOPMACHINE_EXIT) {
- if (stopmachine_state == STOPMACHINE_DISABLE_IRQ
- && !irqs_disabled) {
- local_irq_disable();
- hard_irq_disable();
- irqs_disabled = 1;
- /* Ack: irqs disabled. */
- smp_mb(); /* Must read state first. */
- atomic_inc(&stopmachine_thread_ack);
- } else if (stopmachine_state == STOPMACHINE_PREPARE
- && !prepared) {
- /* Everyone is in place, hold CPU. */
- preempt_disable();
- prepared = 1;
- smp_mb(); /* Must read state first. */
- atomic_inc(&stopmachine_thread_ack);
- }
- /* Yield in first stage: migration threads need to
- * help our sisters onto their CPUs. */
- if (!prepared && !irqs_disabled)
- yield();
- cpu_relax();
- }
-
- /* Ack: we are exiting. */
- smp_mb(); /* Must read state first. */
- atomic_inc(&stopmachine_thread_ack);
-
- if (irqs_disabled)
- local_irq_enable();
- if (prepared)
- preempt_enable();
+struct stop_machine_data {
+ int (*fn)(void *);
+ void *data;
+ int fnret;
+};
- return 0;
-}
+/* Like num_online_cpus(), but hotplug cpu uses us, so we need this. */
+static unsigned int num_threads;
+static atomic_t thread_ack;
+static struct completion finished;
+static DEFINE_MUTEX(lock);
-/* Change the thread state */
-static void stopmachine_set_state(enum stopmachine_state state)
+static void set_state(enum stopmachine_state newstate)
{
- atomic_set(&stopmachine_thread_ack, 0);
+ /* Reset ack counter. */
+ atomic_set(&thread_ack, num_threads);
smp_wmb();
- stopmachine_state = state;
- while (atomic_read(&stopmachine_thread_ack) != stopmachine_num_threads)
- cpu_relax();
+ state = newstate;
}
-static int stop_machine(void)
+/* Last one to ack a state moves to the next state. */
+static void ack_state(void)
{
- int i, ret = 0;
-
- atomic_set(&stopmachine_thread_ack, 0);
- stopmachine_num_threads = 0;
- stopmachine_state = STOPMACHINE_WAIT;
-
- for_each_online_cpu(i) {
- if (i == raw_smp_processor_id())
- continue;
- ret = kernel_thread(stopmachine, (void *)(long)i,CLONE_KERNEL);
- if (ret < 0)
- break;
- stopmachine_num_threads++;
- }
-
- /* Wait for them all to come to life. */
- while (atomic_read(&stopmachine_thread_ack) != stopmachine_num_threads) {
- yield();
- cpu_relax();
+ if (atomic_dec_and_test(&thread_ack)) {
+ /* If we're the last one to ack the EXIT, we're finished. */
+ if (state == STOPMACHINE_EXIT)
+ complete(&finished);
+ else
+ set_state(state + 1);
}
+}
- /* If some failed, kill them all. */
- if (ret < 0) {
- stopmachine_set_state(STOPMACHINE_EXIT);
- return ret;
- }
+/* This is the actual thread which stops the CPU. It exits by itself rather
+ * than waiting for kthread_stop(), because it's easier for hotplug CPU. */
+static int stop_cpu(struct stop_machine_data *smdata)
+{
+ enum stopmachine_state curstate = STOPMACHINE_NONE;
+ int uninitialized_var(ret);
- /* Now they are all started, make them hold the CPUs, ready. */
- preempt_disable();
- stopmachine_set_state(STOPMACHINE_PREPARE);
+ /* Simple state machine */
+ do {
+ /* Chill out and ensure we re-read stopmachine_state. */
+ cpu_relax();
+ if (state != curstate) {
+ curstate = state;
+ switch (curstate) {
+ case STOPMACHINE_DISABLE_IRQ:
+ local_irq_disable();
+ hard_irq_disable();
+ break;
+ case STOPMACHINE_RUN:
+ /* |= allows error detection if functions on
+ * multiple CPUs. */
+ smdata->fnret |= smdata->fn(smdata->data);
+ break;
+ default:
+ break;
+ }
+ ack_state();
+ }
+ } while (curstate != STOPMACHINE_EXIT);
- /* Make them disable irqs. */
- local_irq_disable();
- hard_irq_disable();
- stopmachine_set_state(STOPMACHINE_DISABLE_IRQ);
+ local_irq_enable();
+ do_exit(0);
+}
+/* Callback for CPUs which aren't supposed to do anything. */
+static int chill(void *unused)
+{
return 0;
}
-static void restart_machine(void)
+int __stop_machine(int (*fn)(void *), void *data, const cpumask_t *cpus)
{
- stopmachine_set_state(STOPMACHINE_EXIT);
- local_irq_enable();
- preempt_enable_no_resched();
-}
+ int i, err;
+ struct stop_machine_data active, idle;
+ struct task_struct **threads;
+
+ active.fn = fn;
+ active.data = data;
+ active.fnret = 0;
+ idle.fn = chill;
+ idle.data = NULL;
+
+ /* This could be too big for stack on large machines. */
+ threads = kcalloc(NR_CPUS, sizeof(threads[0]), GFP_KERNEL);
+ if (!threads)
+ return -ENOMEM;
+
+ /* Set up initial state. */
+ mutex_lock(&lock);
+ init_completion(&finished);
+ num_threads = num_online_cpus();
+ set_state(STOPMACHINE_PREPARE);
-struct stop_machine_data {
- int (*fn)(void *);
- void *data;
- struct completion done;
-};
+ for_each_online_cpu(i) {
+ struct stop_machine_data *smdata = &idle;
+ struct sched_param param = { .sched_priority = MAX_RT_PRIO-1 };
-static int do_stop(void *_smdata)
-{
- struct stop_machine_data *smdata = _smdata;
- int ret;
+ if (!cpus) {
+ if (i == first_cpu(cpu_online_map))
+ smdata = &active;
+ } else {
+ if (cpu_isset(i, *cpus))
+ smdata = &active;
+ }
- ret = stop_machine();
- if (ret == 0) {
- ret = smdata->fn(smdata->data);
- restart_machine();
- }
+ threads[i] = kthread_create((void *)stop_cpu, smdata, "kstop%u",
+ i);
+ if (IS_ERR(threads[i])) {
+ err = PTR_ERR(threads[i]);
+ threads[i] = NULL;
+ goto kill_threads;
+ }
- /* We're done: you can kthread_stop us now */
- complete(&smdata->done);
+ /* Place it onto correct cpu. */
+ kthread_bind(threads[i], i);
- /* Wait for kthread_stop */
- set_current_state(TASK_INTERRUPTIBLE);
- while (!kthread_should_stop()) {
- schedule();
- set_current_state(TASK_INTERRUPTIBLE);
+ /* Make it highest prio. */
+ if (sched_setscheduler_nocheck(threads[i], SCHED_FIFO, &param))
+ BUG();
}
- __set_current_state(TASK_RUNNING);
- return ret;
-}
-struct task_struct *__stop_machine_run(int (*fn)(void *), void *data,
- unsigned int cpu)
-{
- static DEFINE_MUTEX(stopmachine_mutex);
- struct stop_machine_data smdata;
- struct task_struct *p;
+ /* We've created all the threads. Wake them all: hold this CPU so one
+ * doesn't hit this CPU until we're ready. */
+ get_cpu();
+ for_each_online_cpu(i)
+ wake_up_process(threads[i]);
- smdata.fn = fn;
- smdata.data = data;
- init_completion(&smdata.done);
+ /* This will release the thread on our CPU. */
+ put_cpu();
+ wait_for_completion(&finished);
+ mutex_unlock(&lock);
- mutex_lock(&stopmachine_mutex);
+ kfree(threads);
- /* If they don't care which CPU fn runs on, bind to any online one. */
- if (cpu == NR_CPUS)
- cpu = raw_smp_processor_id();
+ return active.fnret;
- p = kthread_create(do_stop, &smdata, "kstopmachine");
- if (!IS_ERR(p)) {
- struct sched_param param = { .sched_priority = MAX_RT_PRIO-1 };
+kill_threads:
+ for_each_online_cpu(i)
+ if (threads[i])
+ kthread_stop(threads[i]);
+ mutex_unlock(&lock);
- /* One high-prio thread per cpu. We'll do this one. */
- sched_setscheduler_nocheck(p, SCHED_FIFO, &param);
- kthread_bind(p, cpu);
- wake_up_process(p);
- wait_for_completion(&smdata.done);
- }
- mutex_unlock(&stopmachine_mutex);
- return p;
+ kfree(threads);
+ return err;
}
-int stop_machine_run(int (*fn)(void *), void *data, unsigned int cpu)
+int stop_machine(int (*fn)(void *), void *data, const cpumask_t *cpus)
{
- struct task_struct *p;
int ret;
/* No CPUs can come up or down during this. */
get_online_cpus();
- p = __stop_machine_run(fn, data, cpu);
- if (!IS_ERR(p))
- ret = kthread_stop(p);
- else
- ret = PTR_ERR(p);
+ ret = __stop_machine(fn, data, cpus);
put_online_cpus();
return ret;
}
-EXPORT_SYMBOL_GPL(stop_machine_run);
+EXPORT_SYMBOL_GPL(stop_machine);
diff --git a/kernel/time/tick-common.c b/kernel/time/tick-common.c
index bf43284d6855..80c4336f4188 100644
--- a/kernel/time/tick-common.c
+++ b/kernel/time/tick-common.c
@@ -196,12 +196,10 @@ static int tick_check_new_device(struct clock_event_device *newdev)
struct tick_device *td;
int cpu, ret = NOTIFY_OK;
unsigned long flags;
- cpumask_of_cpu_ptr_declare(cpumask);
spin_lock_irqsave(&tick_device_lock, flags);
cpu = smp_processor_id();
- cpumask_of_cpu_ptr_next(cpumask, cpu);
if (!cpu_isset(cpu, newdev->cpumask))
goto out_bc;
@@ -209,7 +207,7 @@ static int tick_check_new_device(struct clock_event_device *newdev)
curdev = td->evtdev;
/* cpu local device ? */
- if (!cpus_equal(newdev->cpumask, *cpumask)) {
+ if (!cpus_equal(newdev->cpumask, cpumask_of_cpu(cpu))) {
/*
* If the cpu affinity of the device interrupt can not
@@ -222,7 +220,7 @@ static int tick_check_new_device(struct clock_event_device *newdev)
* If we have a cpu local device already, do not replace it
* by a non cpu local device
*/
- if (curdev && cpus_equal(curdev->cpumask, *cpumask))
+ if (curdev && cpus_equal(curdev->cpumask, cpumask_of_cpu(cpu)))
goto out_bc;
}
@@ -254,7 +252,7 @@ static int tick_check_new_device(struct clock_event_device *newdev)
curdev = NULL;
}
clockevents_exchange_device(curdev, newdev);
- tick_setup_device(td, newdev, cpu, cpumask);
+ tick_setup_device(td, newdev, cpu, &cpumask_of_cpu(cpu));
if (newdev->features & CLOCK_EVT_FEAT_ONESHOT)
tick_oneshot_notify();
diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c
index 4231a3dc224a..f6e3af31b403 100644
--- a/kernel/trace/ftrace.c
+++ b/kernel/trace/ftrace.c
@@ -587,7 +587,7 @@ static int __ftrace_modify_code(void *data)
static void ftrace_run_update_code(int command)
{
- stop_machine_run(__ftrace_modify_code, &command, NR_CPUS);
+ stop_machine(__ftrace_modify_code, &command, NULL);
}
void ftrace_disable_daemon(void)
@@ -787,7 +787,7 @@ static int ftrace_update_code(void)
!ftrace_enabled || !ftraced_trigger)
return 0;
- stop_machine_run(__ftrace_update_code, NULL, NR_CPUS);
+ stop_machine(__ftrace_update_code, NULL, NULL);
return 1;
}
@@ -1564,7 +1564,7 @@ static int __init ftrace_dynamic_init(void)
addr = (unsigned long)ftrace_record_ip;
- stop_machine_run(ftrace_dyn_arch_init, &addr, NR_CPUS);
+ stop_machine(ftrace_dyn_arch_init, &addr, NULL);
/* ftrace_dyn_arch_init places the return code in addr */
if (addr) {
diff --git a/kernel/trace/trace_sysprof.c b/kernel/trace/trace_sysprof.c
index ce2d723c10e1..bb948e52ce20 100644
--- a/kernel/trace/trace_sysprof.c
+++ b/kernel/trace/trace_sysprof.c
@@ -213,9 +213,7 @@ static void start_stack_timers(void)
int cpu;
for_each_online_cpu(cpu) {
- cpumask_of_cpu_ptr(new_mask, cpu);
-
- set_cpus_allowed_ptr(current, new_mask);
+ set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu));
start_stack_timer(cpu);
}
set_cpus_allowed_ptr(current, &saved_mask);
diff --git a/kernel/workqueue.c b/kernel/workqueue.c
index ec7e4f62aaff..4a26a1382df0 100644
--- a/kernel/workqueue.c
+++ b/kernel/workqueue.c
@@ -830,10 +830,21 @@ struct workqueue_struct *__create_workqueue_key(const char *name,
start_workqueue_thread(cwq, -1);
} else {
cpu_maps_update_begin();
+ /*
+ * We must place this wq on list even if the code below fails.
+ * cpu_down(cpu) can remove cpu from cpu_populated_map before
+ * destroy_workqueue() takes the lock, in that case we leak
+ * cwq[cpu]->thread.
+ */
spin_lock(&workqueue_lock);
list_add(&wq->list, &workqueues);
spin_unlock(&workqueue_lock);
-
+ /*
+ * We must initialize cwqs for each possible cpu even if we
+ * are going to call destroy_workqueue() finally. Otherwise
+ * cpu_up() can hit the uninitialized cwq once we drop the
+ * lock.
+ */
for_each_possible_cpu(cpu) {
cwq = init_cpu_workqueue(wq, cpu);
if (err || !cpu_online(cpu))
generated by cgit v1.2.3-58-ga151 (git 2.39.0) at 2024-12-25 03:06:15 +0000