aboutsummaryrefslogtreecommitdiffstats
path: root/kernel
diff options
context:
space:
mode:
Diffstat (limited to 'kernel')
-rw-r--r--kernel/Makefile4
-rw-r--r--kernel/capability.c113
-rw-r--r--kernel/exit.c3
-rw-r--r--kernel/fork.c20
-rw-r--r--kernel/hrtimer.c9
-rw-r--r--kernel/latency.c280
-rw-r--r--kernel/pm_qos_params.c425
-rw-r--r--kernel/posix-timers.c9
-rw-r--r--kernel/power/disk.c4
-rw-r--r--kernel/power/snapshot.c4
-rw-r--r--kernel/signal.c75
-rw-r--r--kernel/sys.c15
-rw-r--r--kernel/sys_ni.c7
-rw-r--r--kernel/sysctl.c53
-rw-r--r--kernel/sysctl_check.c7
15 files changed, 606 insertions, 422 deletions
diff --git a/kernel/Makefile b/kernel/Makefile
index db9af707ff5b..135a1b943446 100644
--- a/kernel/Makefile
+++ b/kernel/Makefile
@@ -8,8 +8,8 @@ obj-y = sched.o fork.o exec_domain.o panic.o printk.o profile.o \
8 signal.o sys.o kmod.o workqueue.o pid.o \ 8 signal.o sys.o kmod.o workqueue.o pid.o \
9 rcupdate.o extable.o params.o posix-timers.o \ 9 rcupdate.o extable.o params.o posix-timers.o \
10 kthread.o wait.o kfifo.o sys_ni.o posix-cpu-timers.o mutex.o \ 10 kthread.o wait.o kfifo.o sys_ni.o posix-cpu-timers.o mutex.o \
11 hrtimer.o rwsem.o latency.o nsproxy.o srcu.o \ 11 hrtimer.o rwsem.o nsproxy.o srcu.o \
12 utsname.o notifier.o ksysfs.o 12 utsname.o notifier.o ksysfs.o pm_qos_params.o
13 13
14obj-$(CONFIG_SYSCTL) += sysctl_check.o 14obj-$(CONFIG_SYSCTL) += sysctl_check.o
15obj-$(CONFIG_STACKTRACE) += stacktrace.o 15obj-$(CONFIG_STACKTRACE) += stacktrace.o
diff --git a/kernel/capability.c b/kernel/capability.c
index efbd9cdce132..39e8193b41ea 100644
--- a/kernel/capability.c
+++ b/kernel/capability.c
@@ -22,6 +22,37 @@
22static DEFINE_SPINLOCK(task_capability_lock); 22static DEFINE_SPINLOCK(task_capability_lock);
23 23
24/* 24/*
25 * Leveraged for setting/resetting capabilities
26 */
27
28const kernel_cap_t __cap_empty_set = CAP_EMPTY_SET;
29const kernel_cap_t __cap_full_set = CAP_FULL_SET;
30const kernel_cap_t __cap_init_eff_set = CAP_INIT_EFF_SET;
31
32EXPORT_SYMBOL(__cap_empty_set);
33EXPORT_SYMBOL(__cap_full_set);
34EXPORT_SYMBOL(__cap_init_eff_set);
35
36/*
37 * More recent versions of libcap are available from:
38 *
39 * http://www.kernel.org/pub/linux/libs/security/linux-privs/
40 */
41
42static void warn_legacy_capability_use(void)
43{
44 static int warned;
45 if (!warned) {
46 char name[sizeof(current->comm)];
47
48 printk(KERN_INFO "warning: `%s' uses 32-bit capabilities"
49 " (legacy support in use)\n",
50 get_task_comm(name, current));
51 warned = 1;
52 }
53}
54
55/*
25 * For sys_getproccap() and sys_setproccap(), any of the three 56 * For sys_getproccap() and sys_setproccap(), any of the three
26 * capability set pointers may be NULL -- indicating that that set is 57 * capability set pointers may be NULL -- indicating that that set is
27 * uninteresting and/or not to be changed. 58 * uninteresting and/or not to be changed.
@@ -42,12 +73,21 @@ asmlinkage long sys_capget(cap_user_header_t header, cap_user_data_t dataptr)
42 pid_t pid; 73 pid_t pid;
43 __u32 version; 74 __u32 version;
44 struct task_struct *target; 75 struct task_struct *target;
45 struct __user_cap_data_struct data; 76 unsigned tocopy;
77 kernel_cap_t pE, pI, pP;
46 78
47 if (get_user(version, &header->version)) 79 if (get_user(version, &header->version))
48 return -EFAULT; 80 return -EFAULT;
49 81
50 if (version != _LINUX_CAPABILITY_VERSION) { 82 switch (version) {
83 case _LINUX_CAPABILITY_VERSION_1:
84 warn_legacy_capability_use();
85 tocopy = _LINUX_CAPABILITY_U32S_1;
86 break;
87 case _LINUX_CAPABILITY_VERSION_2:
88 tocopy = _LINUX_CAPABILITY_U32S_2;
89 break;
90 default:
51 if (put_user(_LINUX_CAPABILITY_VERSION, &header->version)) 91 if (put_user(_LINUX_CAPABILITY_VERSION, &header->version))
52 return -EFAULT; 92 return -EFAULT;
53 return -EINVAL; 93 return -EINVAL;
@@ -71,14 +111,47 @@ asmlinkage long sys_capget(cap_user_header_t header, cap_user_data_t dataptr)
71 } else 111 } else
72 target = current; 112 target = current;
73 113
74 ret = security_capget(target, &data.effective, &data.inheritable, &data.permitted); 114 ret = security_capget(target, &pE, &pI, &pP);
75 115
76out: 116out:
77 read_unlock(&tasklist_lock); 117 read_unlock(&tasklist_lock);
78 spin_unlock(&task_capability_lock); 118 spin_unlock(&task_capability_lock);
79 119
80 if (!ret && copy_to_user(dataptr, &data, sizeof data)) 120 if (!ret) {
81 return -EFAULT; 121 struct __user_cap_data_struct kdata[_LINUX_CAPABILITY_U32S];
122 unsigned i;
123
124 for (i = 0; i < tocopy; i++) {
125 kdata[i].effective = pE.cap[i];
126 kdata[i].permitted = pP.cap[i];
127 kdata[i].inheritable = pI.cap[i];
128 }
129
130 /*
131 * Note, in the case, tocopy < _LINUX_CAPABILITY_U32S,
132 * we silently drop the upper capabilities here. This
133 * has the effect of making older libcap
134 * implementations implicitly drop upper capability
135 * bits when they perform a: capget/modify/capset
136 * sequence.
137 *
138 * This behavior is considered fail-safe
139 * behavior. Upgrading the application to a newer
140 * version of libcap will enable access to the newer
141 * capabilities.
142 *
143 * An alternative would be to return an error here
144 * (-ERANGE), but that causes legacy applications to
145 * unexpectidly fail; the capget/modify/capset aborts
146 * before modification is attempted and the application
147 * fails.
148 */
149
150 if (copy_to_user(dataptr, kdata, tocopy
151 * sizeof(struct __user_cap_data_struct))) {
152 return -EFAULT;
153 }
154 }
82 155
83 return ret; 156 return ret;
84} 157}
@@ -167,6 +240,8 @@ static inline int cap_set_all(kernel_cap_t *effective,
167 */ 240 */
168asmlinkage long sys_capset(cap_user_header_t header, const cap_user_data_t data) 241asmlinkage long sys_capset(cap_user_header_t header, const cap_user_data_t data)
169{ 242{
243 struct __user_cap_data_struct kdata[_LINUX_CAPABILITY_U32S];
244 unsigned i, tocopy;
170 kernel_cap_t inheritable, permitted, effective; 245 kernel_cap_t inheritable, permitted, effective;
171 __u32 version; 246 __u32 version;
172 struct task_struct *target; 247 struct task_struct *target;
@@ -176,7 +251,15 @@ asmlinkage long sys_capset(cap_user_header_t header, const cap_user_data_t data)
176 if (get_user(version, &header->version)) 251 if (get_user(version, &header->version))
177 return -EFAULT; 252 return -EFAULT;
178 253
179 if (version != _LINUX_CAPABILITY_VERSION) { 254 switch (version) {
255 case _LINUX_CAPABILITY_VERSION_1:
256 warn_legacy_capability_use();
257 tocopy = _LINUX_CAPABILITY_U32S_1;
258 break;
259 case _LINUX_CAPABILITY_VERSION_2:
260 tocopy = _LINUX_CAPABILITY_U32S_2;
261 break;
262 default:
180 if (put_user(_LINUX_CAPABILITY_VERSION, &header->version)) 263 if (put_user(_LINUX_CAPABILITY_VERSION, &header->version))
181 return -EFAULT; 264 return -EFAULT;
182 return -EINVAL; 265 return -EINVAL;
@@ -188,10 +271,22 @@ asmlinkage long sys_capset(cap_user_header_t header, const cap_user_data_t data)
188 if (pid && pid != task_pid_vnr(current) && !capable(CAP_SETPCAP)) 271 if (pid && pid != task_pid_vnr(current) && !capable(CAP_SETPCAP))
189 return -EPERM; 272 return -EPERM;
190 273
191 if (copy_from_user(&effective, &data->effective, sizeof(effective)) || 274 if (copy_from_user(&kdata, data, tocopy
192 copy_from_user(&inheritable, &data->inheritable, sizeof(inheritable)) || 275 * sizeof(struct __user_cap_data_struct))) {
193 copy_from_user(&permitted, &data->permitted, sizeof(permitted)))
194 return -EFAULT; 276 return -EFAULT;
277 }
278
279 for (i = 0; i < tocopy; i++) {
280 effective.cap[i] = kdata[i].effective;
281 permitted.cap[i] = kdata[i].permitted;
282 inheritable.cap[i] = kdata[i].inheritable;
283 }
284 while (i < _LINUX_CAPABILITY_U32S) {
285 effective.cap[i] = 0;
286 permitted.cap[i] = 0;
287 inheritable.cap[i] = 0;
288 i++;
289 }
195 290
196 spin_lock(&task_capability_lock); 291 spin_lock(&task_capability_lock);
197 read_lock(&tasklist_lock); 292 read_lock(&tasklist_lock);
diff --git a/kernel/exit.c b/kernel/exit.c
index 9e459fefda77..9d3d0f0b27d9 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -1083,11 +1083,12 @@ do_group_exit(int exit_code)
1083 struct signal_struct *const sig = current->signal; 1083 struct signal_struct *const sig = current->signal;
1084 struct sighand_struct *const sighand = current->sighand; 1084 struct sighand_struct *const sighand = current->sighand;
1085 spin_lock_irq(&sighand->siglock); 1085 spin_lock_irq(&sighand->siglock);
1086 if (sig->flags & SIGNAL_GROUP_EXIT) 1086 if (signal_group_exit(sig))
1087 /* Another thread got here before we took the lock. */ 1087 /* Another thread got here before we took the lock. */
1088 exit_code = sig->group_exit_code; 1088 exit_code = sig->group_exit_code;
1089 else { 1089 else {
1090 sig->group_exit_code = exit_code; 1090 sig->group_exit_code = exit_code;
1091 sig->flags = SIGNAL_GROUP_EXIT;
1091 zap_other_threads(current); 1092 zap_other_threads(current);
1092 } 1093 }
1093 spin_unlock_irq(&sighand->siglock); 1094 spin_unlock_irq(&sighand->siglock);
diff --git a/kernel/fork.c b/kernel/fork.c
index 05e0b6f4365b..2b55b74cd999 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -325,7 +325,7 @@ static inline int mm_alloc_pgd(struct mm_struct * mm)
325 325
326static inline void mm_free_pgd(struct mm_struct * mm) 326static inline void mm_free_pgd(struct mm_struct * mm)
327{ 327{
328 pgd_free(mm->pgd); 328 pgd_free(mm, mm->pgd);
329} 329}
330#else 330#else
331#define dup_mmap(mm, oldmm) (0) 331#define dup_mmap(mm, oldmm) (0)
@@ -1118,6 +1118,7 @@ static struct task_struct *copy_process(unsigned long clone_flags,
1118#ifdef CONFIG_SECURITY 1118#ifdef CONFIG_SECURITY
1119 p->security = NULL; 1119 p->security = NULL;
1120#endif 1120#endif
1121 p->cap_bset = current->cap_bset;
1121 p->io_context = NULL; 1122 p->io_context = NULL;
1122 p->audit_context = NULL; 1123 p->audit_context = NULL;
1123 cgroup_fork(p); 1124 cgroup_fork(p);
@@ -1450,6 +1451,23 @@ long do_fork(unsigned long clone_flags,
1450 int trace = 0; 1451 int trace = 0;
1451 long nr; 1452 long nr;
1452 1453
1454 /*
1455 * We hope to recycle these flags after 2.6.26
1456 */
1457 if (unlikely(clone_flags & CLONE_STOPPED)) {
1458 static int __read_mostly count = 100;
1459
1460 if (count > 0 && printk_ratelimit()) {
1461 char comm[TASK_COMM_LEN];
1462
1463 count--;
1464 printk(KERN_INFO "fork(): process `%s' used deprecated "
1465 "clone flags 0x%lx\n",
1466 get_task_comm(comm, current),
1467 clone_flags & CLONE_STOPPED);
1468 }
1469 }
1470
1453 if (unlikely(current->ptrace)) { 1471 if (unlikely(current->ptrace)) {
1454 trace = fork_traceflag (clone_flags); 1472 trace = fork_traceflag (clone_flags);
1455 if (trace) 1473 if (trace)
diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c
index 1069998fe25f..668f3967eb39 100644
--- a/kernel/hrtimer.c
+++ b/kernel/hrtimer.c
@@ -306,7 +306,7 @@ EXPORT_SYMBOL_GPL(ktime_sub_ns);
306/* 306/*
307 * Divide a ktime value by a nanosecond value 307 * Divide a ktime value by a nanosecond value
308 */ 308 */
309unsigned long ktime_divns(const ktime_t kt, s64 div) 309u64 ktime_divns(const ktime_t kt, s64 div)
310{ 310{
311 u64 dclc, inc, dns; 311 u64 dclc, inc, dns;
312 int sft = 0; 312 int sft = 0;
@@ -321,7 +321,7 @@ unsigned long ktime_divns(const ktime_t kt, s64 div)
321 dclc >>= sft; 321 dclc >>= sft;
322 do_div(dclc, (unsigned long) div); 322 do_div(dclc, (unsigned long) div);
323 323
324 return (unsigned long) dclc; 324 return dclc;
325} 325}
326#endif /* BITS_PER_LONG >= 64 */ 326#endif /* BITS_PER_LONG >= 64 */
327 327
@@ -656,10 +656,9 @@ void unlock_hrtimer_base(const struct hrtimer *timer, unsigned long *flags)
656 * Forward the timer expiry so it will expire in the future. 656 * Forward the timer expiry so it will expire in the future.
657 * Returns the number of overruns. 657 * Returns the number of overruns.
658 */ 658 */
659unsigned long 659u64 hrtimer_forward(struct hrtimer *timer, ktime_t now, ktime_t interval)
660hrtimer_forward(struct hrtimer *timer, ktime_t now, ktime_t interval)
661{ 660{
662 unsigned long orun = 1; 661 u64 orun = 1;
663 ktime_t delta; 662 ktime_t delta;
664 663
665 delta = ktime_sub(now, timer->expires); 664 delta = ktime_sub(now, timer->expires);
diff --git a/kernel/latency.c b/kernel/latency.c
deleted file mode 100644
index e63fcacb61a7..000000000000
--- a/kernel/latency.c
+++ /dev/null
@@ -1,280 +0,0 @@
1/*
2 * latency.c: Explicit system-wide latency-expectation infrastructure
3 *
4 * The purpose of this infrastructure is to allow device drivers to set
5 * latency constraint they have and to collect and summarize these
6 * expectations globally. The cummulated result can then be used by
7 * power management and similar users to make decisions that have
8 * tradoffs with a latency component.
9 *
10 * An example user of this are the x86 C-states; each higher C state saves
11 * more power, but has a higher exit latency. For the idle loop power
12 * code to make a good decision which C-state to use, information about
13 * acceptable latencies is required.
14 *
15 * An example announcer of latency is an audio driver that knowns it
16 * will get an interrupt when the hardware has 200 usec of samples
17 * left in the DMA buffer; in that case the driver can set a latency
18 * constraint of, say, 150 usec.
19 *
20 * Multiple drivers can each announce their maximum accepted latency,
21 * to keep these appart, a string based identifier is used.
22 *
23 *
24 * (C) Copyright 2006 Intel Corporation
25 * Author: Arjan van de Ven <arjan@linux.intel.com>
26 *
27 * This program is free software; you can redistribute it and/or
28 * modify it under the terms of the GNU General Public License
29 * as published by the Free Software Foundation; version 2
30 * of the License.
31 */
32
33#include <linux/latency.h>
34#include <linux/list.h>
35#include <linux/spinlock.h>
36#include <linux/slab.h>
37#include <linux/module.h>
38#include <linux/notifier.h>
39#include <linux/jiffies.h>
40#include <asm/atomic.h>
41
42struct latency_info {
43 struct list_head list;
44 int usecs;
45 char *identifier;
46};
47
48/*
49 * locking rule: all modifications to current_max_latency and
50 * latency_list need to be done while holding the latency_lock.
51 * latency_lock needs to be taken _irqsave.
52 */
53static atomic_t current_max_latency;
54static DEFINE_SPINLOCK(latency_lock);
55
56static LIST_HEAD(latency_list);
57static BLOCKING_NOTIFIER_HEAD(latency_notifier);
58
59/*
60 * This function returns the maximum latency allowed, which
61 * happens to be the minimum of all maximum latencies on the
62 * list.
63 */
64static int __find_max_latency(void)
65{
66 int min = INFINITE_LATENCY;
67 struct latency_info *info;
68
69 list_for_each_entry(info, &latency_list, list) {
70 if (info->usecs < min)
71 min = info->usecs;
72 }
73 return min;
74}
75
76/**
77 * set_acceptable_latency - sets the maximum latency acceptable
78 * @identifier: string that identifies this driver
79 * @usecs: maximum acceptable latency for this driver
80 *
81 * This function informs the kernel that this device(driver)
82 * can accept at most usecs latency. This setting is used for
83 * power management and similar tradeoffs.
84 *
85 * This function sleeps and can only be called from process
86 * context.
87 * Calling this function with an existing identifier is valid
88 * and will cause the existing latency setting to be changed.
89 */
90void set_acceptable_latency(char *identifier, int usecs)
91{
92 struct latency_info *info, *iter;
93 unsigned long flags;
94 int found_old = 0;
95
96 info = kzalloc(sizeof(struct latency_info), GFP_KERNEL);
97 if (!info)
98 return;
99 info->usecs = usecs;
100 info->identifier = kstrdup(identifier, GFP_KERNEL);
101 if (!info->identifier)
102 goto free_info;
103
104 spin_lock_irqsave(&latency_lock, flags);
105 list_for_each_entry(iter, &latency_list, list) {
106 if (strcmp(iter->identifier, identifier)==0) {
107 found_old = 1;
108 iter->usecs = usecs;
109 break;
110 }
111 }
112 if (!found_old)
113 list_add(&info->list, &latency_list);
114
115 if (usecs < atomic_read(&current_max_latency))
116 atomic_set(&current_max_latency, usecs);
117
118 spin_unlock_irqrestore(&latency_lock, flags);
119
120 blocking_notifier_call_chain(&latency_notifier,
121 atomic_read(&current_max_latency), NULL);
122
123 /*
124 * if we inserted the new one, we're done; otherwise there was
125 * an existing one so we need to free the redundant data
126 */
127 if (!found_old)
128 return;
129
130 kfree(info->identifier);
131free_info:
132 kfree(info);
133}
134EXPORT_SYMBOL_GPL(set_acceptable_latency);
135
136/**
137 * modify_acceptable_latency - changes the maximum latency acceptable
138 * @identifier: string that identifies this driver
139 * @usecs: maximum acceptable latency for this driver
140 *
141 * This function informs the kernel that this device(driver)
142 * can accept at most usecs latency. This setting is used for
143 * power management and similar tradeoffs.
144 *
145 * This function does not sleep and can be called in any context.
146 * Trying to use a non-existing identifier silently gets ignored.
147 *
148 * Due to the atomic nature of this function, the modified latency
149 * value will only be used for future decisions; past decisions
150 * can still lead to longer latencies in the near future.
151 */
152void modify_acceptable_latency(char *identifier, int usecs)
153{
154 struct latency_info *iter;
155 unsigned long flags;
156
157 spin_lock_irqsave(&latency_lock, flags);
158 list_for_each_entry(iter, &latency_list, list) {
159 if (strcmp(iter->identifier, identifier) == 0) {
160 iter->usecs = usecs;
161 break;
162 }
163 }
164 if (usecs < atomic_read(&current_max_latency))
165 atomic_set(&current_max_latency, usecs);
166 spin_unlock_irqrestore(&latency_lock, flags);
167}
168EXPORT_SYMBOL_GPL(modify_acceptable_latency);
169
170/**
171 * remove_acceptable_latency - removes the maximum latency acceptable
172 * @identifier: string that identifies this driver
173 *
174 * This function removes a previously set maximum latency setting
175 * for the driver and frees up any resources associated with the
176 * bookkeeping needed for this.
177 *
178 * This function does not sleep and can be called in any context.
179 * Trying to use a non-existing identifier silently gets ignored.
180 */
181void remove_acceptable_latency(char *identifier)
182{
183 unsigned long flags;
184 int newmax = 0;
185 struct latency_info *iter, *temp;
186
187 spin_lock_irqsave(&latency_lock, flags);
188
189 list_for_each_entry_safe(iter, temp, &latency_list, list) {
190 if (strcmp(iter->identifier, identifier) == 0) {
191 list_del(&iter->list);
192 newmax = iter->usecs;
193 kfree(iter->identifier);
194 kfree(iter);
195 break;
196 }
197 }
198
199 /* If we just deleted the system wide value, we need to
200 * recalculate with a full search
201 */
202 if (newmax == atomic_read(&current_max_latency)) {
203 newmax = __find_max_latency();
204 atomic_set(&current_max_latency, newmax);
205 }
206 spin_unlock_irqrestore(&latency_lock, flags);
207}
208EXPORT_SYMBOL_GPL(remove_acceptable_latency);
209
210/**
211 * system_latency_constraint - queries the system wide latency maximum
212 *
213 * This function returns the system wide maximum latency in
214 * microseconds.
215 *
216 * This function does not sleep and can be called in any context.
217 */
218int system_latency_constraint(void)
219{
220 return atomic_read(&current_max_latency);
221}
222EXPORT_SYMBOL_GPL(system_latency_constraint);
223
224/**
225 * synchronize_acceptable_latency - recalculates all latency decisions
226 *
227 * This function will cause a callback to various kernel pieces that
228 * will make those pieces rethink their latency decisions. This implies
229 * that if there are overlong latencies in hardware state already, those
230 * latencies get taken right now. When this call completes no overlong
231 * latency decisions should be active anymore.
232 *
233 * Typical usecase of this is after a modify_acceptable_latency() call,
234 * which in itself is non-blocking and non-synchronizing.
235 *
236 * This function blocks and should not be called with locks held.
237 */
238
239void synchronize_acceptable_latency(void)
240{
241 blocking_notifier_call_chain(&latency_notifier,
242 atomic_read(&current_max_latency), NULL);
243}
244EXPORT_SYMBOL_GPL(synchronize_acceptable_latency);
245
246/*
247 * Latency notifier: this notifier gets called when a non-atomic new
248 * latency value gets set. The expectation nof the caller of the
249 * non-atomic set is that when the call returns, future latencies
250 * are within bounds, so the functions on the notifier list are
251 * expected to take the overlong latencies immediately, inside the
252 * callback, and not make a overlong latency decision anymore.
253 *
254 * The callback gets called when the new latency value is made
255 * active so system_latency_constraint() returns the new latency.
256 */
257int register_latency_notifier(struct notifier_block * nb)
258{
259 return blocking_notifier_chain_register(&latency_notifier, nb);
260}
261EXPORT_SYMBOL_GPL(register_latency_notifier);
262
263int unregister_latency_notifier(struct notifier_block * nb)
264{
265 return blocking_notifier_chain_unregister(&latency_notifier, nb);
266}
267EXPORT_SYMBOL_GPL(unregister_latency_notifier);
268
269static __init int latency_init(void)
270{
271 atomic_set(&current_max_latency, INFINITE_LATENCY);
272 /*
273 * we don't want by default to have longer latencies than 2 ticks,
274 * since that would cause lost ticks
275 */
276 set_acceptable_latency("kernel", 2*1000000/HZ);
277 return 0;
278}
279
280module_init(latency_init);
diff --git a/kernel/pm_qos_params.c b/kernel/pm_qos_params.c
new file mode 100644
index 000000000000..0afe32be4c85
--- /dev/null
+++ b/kernel/pm_qos_params.c
@@ -0,0 +1,425 @@
1/*
2 * This module exposes the interface to kernel space for specifying
3 * QoS dependencies. It provides infrastructure for registration of:
4 *
5 * Dependents on a QoS value : register requirements
6 * Watchers of QoS value : get notified when target QoS value changes
7 *
8 * This QoS design is best effort based. Dependents register their QoS needs.
9 * Watchers register to keep track of the current QoS needs of the system.
10 *
11 * There are 3 basic classes of QoS parameter: latency, timeout, throughput
12 * each have defined units:
13 * latency: usec
14 * timeout: usec <-- currently not used.
15 * throughput: kbs (kilo byte / sec)
16 *
17 * There are lists of pm_qos_objects each one wrapping requirements, notifiers
18 *
19 * User mode requirements on a QOS parameter register themselves to the
20 * subsystem by opening the device node /dev/... and writing there request to
21 * the node. As long as the process holds a file handle open to the node the
22 * client continues to be accounted for. Upon file release the usermode
23 * requirement is removed and a new qos target is computed. This way when the
24 * requirement that the application has is cleaned up when closes the file
25 * pointer or exits the pm_qos_object will get an opportunity to clean up.
26 *
27 * mark gross mgross@linux.intel.com
28 */
29
30#include <linux/pm_qos_params.h>
31#include <linux/sched.h>
32#include <linux/spinlock.h>
33#include <linux/slab.h>
34#include <linux/time.h>
35#include <linux/fs.h>
36#include <linux/device.h>
37#include <linux/miscdevice.h>
38#include <linux/string.h>
39#include <linux/platform_device.h>
40#include <linux/init.h>
41
42#include <linux/uaccess.h>
43
44/*
45 * locking rule: all changes to target_value or requirements or notifiers lists
46 * or pm_qos_object list and pm_qos_objects need to happen with pm_qos_lock
47 * held, taken with _irqsave. One lock to rule them all
48 */
49struct requirement_list {
50 struct list_head list;
51 union {
52 s32 value;
53 s32 usec;
54 s32 kbps;
55 };
56 char *name;
57};
58
59static s32 max_compare(s32 v1, s32 v2);
60static s32 min_compare(s32 v1, s32 v2);
61
62struct pm_qos_object {
63 struct requirement_list requirements;
64 struct blocking_notifier_head *notifiers;
65 struct miscdevice pm_qos_power_miscdev;
66 char *name;
67 s32 default_value;
68 s32 target_value;
69 s32 (*comparitor)(s32, s32);
70};
71
72static struct pm_qos_object null_pm_qos;
73static BLOCKING_NOTIFIER_HEAD(cpu_dma_lat_notifier);
74static struct pm_qos_object cpu_dma_pm_qos = {
75 .requirements = {LIST_HEAD_INIT(cpu_dma_pm_qos.requirements.list)},
76 .notifiers = &cpu_dma_lat_notifier,
77 .name = "cpu_dma_latency",
78 .default_value = 2000 * USEC_PER_SEC,
79 .target_value = 2000 * USEC_PER_SEC,
80 .comparitor = min_compare
81};
82
83static BLOCKING_NOTIFIER_HEAD(network_lat_notifier);
84static struct pm_qos_object network_lat_pm_qos = {
85 .requirements = {LIST_HEAD_INIT(network_lat_pm_qos.requirements.list)},
86 .notifiers = &network_lat_notifier,
87 .name = "network_latency",
88 .default_value = 2000 * USEC_PER_SEC,
89 .target_value = 2000 * USEC_PER_SEC,
90 .comparitor = min_compare
91};
92
93
94static BLOCKING_NOTIFIER_HEAD(network_throughput_notifier);
95static struct pm_qos_object network_throughput_pm_qos = {
96 .requirements =
97 {LIST_HEAD_INIT(network_throughput_pm_qos.requirements.list)},
98 .notifiers = &network_throughput_notifier,
99 .name = "network_throughput",
100 .default_value = 0,
101 .target_value = 0,
102 .comparitor = max_compare
103};
104
105
106static struct pm_qos_object *pm_qos_array[] = {
107 &null_pm_qos,
108 &cpu_dma_pm_qos,
109 &network_lat_pm_qos,
110 &network_throughput_pm_qos
111};
112
113static DEFINE_SPINLOCK(pm_qos_lock);
114
115static ssize_t pm_qos_power_write(struct file *filp, const char __user *buf,
116 size_t count, loff_t *f_pos);
117static int pm_qos_power_open(struct inode *inode, struct file *filp);
118static int pm_qos_power_release(struct inode *inode, struct file *filp);
119
120static const struct file_operations pm_qos_power_fops = {
121 .write = pm_qos_power_write,
122 .open = pm_qos_power_open,
123 .release = pm_qos_power_release,
124};
125
126/* static helper functions */
127static s32 max_compare(s32 v1, s32 v2)
128{
129 return max(v1, v2);
130}
131
132static s32 min_compare(s32 v1, s32 v2)
133{
134 return min(v1, v2);
135}
136
137
138static void update_target(int target)
139{
140 s32 extreme_value;
141 struct requirement_list *node;
142 unsigned long flags;
143 int call_notifier = 0;
144
145 spin_lock_irqsave(&pm_qos_lock, flags);
146 extreme_value = pm_qos_array[target]->default_value;
147 list_for_each_entry(node,
148 &pm_qos_array[target]->requirements.list, list) {
149 extreme_value = pm_qos_array[target]->comparitor(
150 extreme_value, node->value);
151 }
152 if (pm_qos_array[target]->target_value != extreme_value) {
153 call_notifier = 1;
154 pm_qos_array[target]->target_value = extreme_value;
155 pr_debug(KERN_ERR "new target for qos %d is %d\n", target,
156 pm_qos_array[target]->target_value);
157 }
158 spin_unlock_irqrestore(&pm_qos_lock, flags);
159
160 if (call_notifier)
161 blocking_notifier_call_chain(pm_qos_array[target]->notifiers,
162 (unsigned long) extreme_value, NULL);
163}
164
165static int register_pm_qos_misc(struct pm_qos_object *qos)
166{
167 qos->pm_qos_power_miscdev.minor = MISC_DYNAMIC_MINOR;
168 qos->pm_qos_power_miscdev.name = qos->name;
169 qos->pm_qos_power_miscdev.fops = &pm_qos_power_fops;
170
171 return misc_register(&qos->pm_qos_power_miscdev);
172}
173
174static int find_pm_qos_object_by_minor(int minor)
175{
176 int pm_qos_class;
177
178 for (pm_qos_class = 0;
179 pm_qos_class < PM_QOS_NUM_CLASSES; pm_qos_class++) {
180 if (minor ==
181 pm_qos_array[pm_qos_class]->pm_qos_power_miscdev.minor)
182 return pm_qos_class;
183 }
184 return -1;
185}
186
187/**
188 * pm_qos_requirement - returns current system wide qos expectation
189 * @pm_qos_class: identification of which qos value is requested
190 *
191 * This function returns the current target value in an atomic manner.
192 */
193int pm_qos_requirement(int pm_qos_class)
194{
195 int ret_val;
196 unsigned long flags;
197
198 spin_lock_irqsave(&pm_qos_lock, flags);
199 ret_val = pm_qos_array[pm_qos_class]->target_value;
200 spin_unlock_irqrestore(&pm_qos_lock, flags);
201
202 return ret_val;
203}
204EXPORT_SYMBOL_GPL(pm_qos_requirement);
205
206/**
207 * pm_qos_add_requirement - inserts new qos request into the list
208 * @pm_qos_class: identifies which list of qos request to us
209 * @name: identifies the request
210 * @value: defines the qos request
211 *
212 * This function inserts a new entry in the pm_qos_class list of requested qos
213 * performance charactoistics. It recomputes the agregate QoS expectations for
214 * the pm_qos_class of parrameters.
215 */
216int pm_qos_add_requirement(int pm_qos_class, char *name, s32 value)
217{
218 struct requirement_list *dep;
219 unsigned long flags;
220
221 dep = kzalloc(sizeof(struct requirement_list), GFP_KERNEL);
222 if (dep) {
223 if (value == PM_QOS_DEFAULT_VALUE)
224 dep->value = pm_qos_array[pm_qos_class]->default_value;
225 else
226 dep->value = value;
227 dep->name = kstrdup(name, GFP_KERNEL);
228 if (!dep->name)
229 goto cleanup;
230
231 spin_lock_irqsave(&pm_qos_lock, flags);
232 list_add(&dep->list,
233 &pm_qos_array[pm_qos_class]->requirements.list);
234 spin_unlock_irqrestore(&pm_qos_lock, flags);
235 update_target(pm_qos_class);
236
237 return 0;
238 }
239
240cleanup:
241 kfree(dep);
242 return -ENOMEM;
243}
244EXPORT_SYMBOL_GPL(pm_qos_add_requirement);
245
246/**
247 * pm_qos_update_requirement - modifies an existing qos request
248 * @pm_qos_class: identifies which list of qos request to us
249 * @name: identifies the request
250 * @value: defines the qos request
251 *
252 * Updates an existing qos requierement for the pm_qos_class of parameters along
253 * with updating the target pm_qos_class value.
254 *
255 * If the named request isn't in the lest then no change is made.
256 */
257int pm_qos_update_requirement(int pm_qos_class, char *name, s32 new_value)
258{
259 unsigned long flags;
260 struct requirement_list *node;
261 int pending_update = 0;
262
263 spin_lock_irqsave(&pm_qos_lock, flags);
264 list_for_each_entry(node,
265 &pm_qos_array[pm_qos_class]->requirements.list, list) {
266 if (strcmp(node->name, name) == 0) {
267 if (new_value == PM_QOS_DEFAULT_VALUE)
268 node->value =
269 pm_qos_array[pm_qos_class]->default_value;
270 else
271 node->value = new_value;
272 pending_update = 1;
273 break;
274 }
275 }
276 spin_unlock_irqrestore(&pm_qos_lock, flags);
277 if (pending_update)
278 update_target(pm_qos_class);
279
280 return 0;
281}
282EXPORT_SYMBOL_GPL(pm_qos_update_requirement);
283
284/**
285 * pm_qos_remove_requirement - modifies an existing qos request
286 * @pm_qos_class: identifies which list of qos request to us
287 * @name: identifies the request
288 *
289 * Will remove named qos request from pm_qos_class list of parrameters and
290 * recompute the current target value for the pm_qos_class.
291 */
292void pm_qos_remove_requirement(int pm_qos_class, char *name)
293{
294 unsigned long flags;
295 struct requirement_list *node;
296 int pending_update = 0;
297
298 spin_lock_irqsave(&pm_qos_lock, flags);
299 list_for_each_entry(node,
300 &pm_qos_array[pm_qos_class]->requirements.list, list) {
301 if (strcmp(node->name, name) == 0) {
302 kfree(node->name);
303 list_del(&node->list);
304 kfree(node);
305 pending_update = 1;
306 break;
307 }
308 }
309 spin_unlock_irqrestore(&pm_qos_lock, flags);
310 if (pending_update)
311 update_target(pm_qos_class);
312}
313EXPORT_SYMBOL_GPL(pm_qos_remove_requirement);
314
315/**
316 * pm_qos_add_notifier - sets notification entry for changes to target value
317 * @pm_qos_class: identifies which qos target changes should be notified.
318 * @notifier: notifier block managed by caller.
319 *
320 * will register the notifier into a notification chain that gets called
321 * uppon changes to the pm_qos_class target value.
322 */
323 int pm_qos_add_notifier(int pm_qos_class, struct notifier_block *notifier)
324{
325 int retval;
326
327 retval = blocking_notifier_chain_register(
328 pm_qos_array[pm_qos_class]->notifiers, notifier);
329
330 return retval;
331}
332EXPORT_SYMBOL_GPL(pm_qos_add_notifier);
333
334/**
335 * pm_qos_remove_notifier - deletes notification entry from chain.
336 * @pm_qos_class: identifies which qos target changes are notified.
337 * @notifier: notifier block to be removed.
338 *
339 * will remove the notifier from the notification chain that gets called
340 * uppon changes to the pm_qos_class target value.
341 */
342int pm_qos_remove_notifier(int pm_qos_class, struct notifier_block *notifier)
343{
344 int retval;
345
346 retval = blocking_notifier_chain_unregister(
347 pm_qos_array[pm_qos_class]->notifiers, notifier);
348
349 return retval;
350}
351EXPORT_SYMBOL_GPL(pm_qos_remove_notifier);
352
353#define PID_NAME_LEN sizeof("process_1234567890")
354static char name[PID_NAME_LEN];
355
356static int pm_qos_power_open(struct inode *inode, struct file *filp)
357{
358 int ret;
359 long pm_qos_class;
360
361 pm_qos_class = find_pm_qos_object_by_minor(iminor(inode));
362 if (pm_qos_class >= 0) {
363 filp->private_data = (void *)pm_qos_class;
364 sprintf(name, "process_%d", current->pid);
365 ret = pm_qos_add_requirement(pm_qos_class, name,
366 PM_QOS_DEFAULT_VALUE);
367 if (ret >= 0)
368 return 0;
369 }
370
371 return -EPERM;
372}
373
374static int pm_qos_power_release(struct inode *inode, struct file *filp)
375{
376 int pm_qos_class;
377
378 pm_qos_class = (long)filp->private_data;
379 sprintf(name, "process_%d", current->pid);
380 pm_qos_remove_requirement(pm_qos_class, name);
381
382 return 0;
383}
384
385static ssize_t pm_qos_power_write(struct file *filp, const char __user *buf,
386 size_t count, loff_t *f_pos)
387{
388 s32 value;
389 int pm_qos_class;
390
391 pm_qos_class = (long)filp->private_data;
392 if (count != sizeof(s32))
393 return -EINVAL;
394 if (copy_from_user(&value, buf, sizeof(s32)))
395 return -EFAULT;
396 sprintf(name, "process_%d", current->pid);
397 pm_qos_update_requirement(pm_qos_class, name, value);
398
399 return sizeof(s32);
400}
401
402
403static int __init pm_qos_power_init(void)
404{
405 int ret = 0;
406
407 ret = register_pm_qos_misc(&cpu_dma_pm_qos);
408 if (ret < 0) {
409 printk(KERN_ERR "pm_qos_param: cpu_dma_latency setup failed\n");
410 return ret;
411 }
412 ret = register_pm_qos_misc(&network_lat_pm_qos);
413 if (ret < 0) {
414 printk(KERN_ERR "pm_qos_param: network_latency setup failed\n");
415 return ret;
416 }
417 ret = register_pm_qos_misc(&network_throughput_pm_qos);
418 if (ret < 0)
419 printk(KERN_ERR
420 "pm_qos_param: network_throughput setup failed\n");
421
422 return ret;
423}
424
425late_initcall(pm_qos_power_init);
diff --git a/kernel/posix-timers.c b/kernel/posix-timers.c
index 36d563fd9e3b..122d5c787fe2 100644
--- a/kernel/posix-timers.c
+++ b/kernel/posix-timers.c
@@ -256,8 +256,9 @@ static void schedule_next_timer(struct k_itimer *timr)
256 if (timr->it.real.interval.tv64 == 0) 256 if (timr->it.real.interval.tv64 == 0)
257 return; 257 return;
258 258
259 timr->it_overrun += hrtimer_forward(timer, timer->base->get_time(), 259 timr->it_overrun += (unsigned int) hrtimer_forward(timer,
260 timr->it.real.interval); 260 timer->base->get_time(),
261 timr->it.real.interval);
261 262
262 timr->it_overrun_last = timr->it_overrun; 263 timr->it_overrun_last = timr->it_overrun;
263 timr->it_overrun = -1; 264 timr->it_overrun = -1;
@@ -386,7 +387,7 @@ static enum hrtimer_restart posix_timer_fn(struct hrtimer *timer)
386 now = ktime_add(now, kj); 387 now = ktime_add(now, kj);
387 } 388 }
388#endif 389#endif
389 timr->it_overrun += 390 timr->it_overrun += (unsigned int)
390 hrtimer_forward(timer, now, 391 hrtimer_forward(timer, now,
391 timr->it.real.interval); 392 timr->it.real.interval);
392 ret = HRTIMER_RESTART; 393 ret = HRTIMER_RESTART;
@@ -662,7 +663,7 @@ common_timer_get(struct k_itimer *timr, struct itimerspec *cur_setting)
662 */ 663 */
663 if (iv.tv64 && (timr->it_requeue_pending & REQUEUE_PENDING || 664 if (iv.tv64 && (timr->it_requeue_pending & REQUEUE_PENDING ||
664 (timr->it_sigev_notify & ~SIGEV_THREAD_ID) == SIGEV_NONE)) 665 (timr->it_sigev_notify & ~SIGEV_THREAD_ID) == SIGEV_NONE))
665 timr->it_overrun += hrtimer_forward(timer, now, iv); 666 timr->it_overrun += (unsigned int) hrtimer_forward(timer, now, iv);
666 667
667 remaining = ktime_sub(timer->expires, now); 668 remaining = ktime_sub(timer->expires, now);
668 /* Return 0 only, when the timer is expired and not pending */ 669 /* Return 0 only, when the timer is expired and not pending */
diff --git a/kernel/power/disk.c b/kernel/power/disk.c
index d09da0895174..859a8e59773a 100644
--- a/kernel/power/disk.c
+++ b/kernel/power/disk.c
@@ -26,7 +26,7 @@
26 26
27 27
28static int noresume = 0; 28static int noresume = 0;
29char resume_file[256] = CONFIG_PM_STD_PARTITION; 29static char resume_file[256] = CONFIG_PM_STD_PARTITION;
30dev_t swsusp_resume_device; 30dev_t swsusp_resume_device;
31sector_t swsusp_resume_block; 31sector_t swsusp_resume_block;
32 32
@@ -185,7 +185,7 @@ static void platform_restore_cleanup(int platform_mode)
185 * reappears in this routine after a restore. 185 * reappears in this routine after a restore.
186 */ 186 */
187 187
188int create_image(int platform_mode) 188static int create_image(int platform_mode)
189{ 189{
190 int error; 190 int error;
191 191
diff --git a/kernel/power/snapshot.c b/kernel/power/snapshot.c
index f6a5df934f8d..95250d7c8d91 100644
--- a/kernel/power/snapshot.c
+++ b/kernel/power/snapshot.c
@@ -1203,7 +1203,7 @@ asmlinkage int swsusp_save(void)
1203 1203
1204 printk(KERN_INFO "PM: Creating hibernation image: \n"); 1204 printk(KERN_INFO "PM: Creating hibernation image: \n");
1205 1205
1206 drain_local_pages(); 1206 drain_local_pages(NULL);
1207 nr_pages = count_data_pages(); 1207 nr_pages = count_data_pages();
1208 nr_highmem = count_highmem_pages(); 1208 nr_highmem = count_highmem_pages();
1209 printk(KERN_INFO "PM: Need to copy %u pages\n", nr_pages + nr_highmem); 1209 printk(KERN_INFO "PM: Need to copy %u pages\n", nr_pages + nr_highmem);
@@ -1221,7 +1221,7 @@ asmlinkage int swsusp_save(void)
1221 /* During allocating of suspend pagedir, new cold pages may appear. 1221 /* During allocating of suspend pagedir, new cold pages may appear.
1222 * Kill them. 1222 * Kill them.
1223 */ 1223 */
1224 drain_local_pages(); 1224 drain_local_pages(NULL);
1225 copy_data_pages(&copy_bm, &orig_bm); 1225 copy_data_pages(&copy_bm, &orig_bm);
1226 1226
1227 /* 1227 /*
diff --git a/kernel/signal.c b/kernel/signal.c
index 4333b6dbb424..6a5f97cd337a 100644
--- a/kernel/signal.c
+++ b/kernel/signal.c
@@ -911,27 +911,6 @@ __group_complete_signal(int sig, struct task_struct *p)
911 } while_each_thread(p, t); 911 } while_each_thread(p, t);
912 return; 912 return;
913 } 913 }
914
915 /*
916 * There will be a core dump. We make all threads other
917 * than the chosen one go into a group stop so that nothing
918 * happens until it gets scheduled, takes the signal off
919 * the shared queue, and does the core dump. This is a
920 * little more complicated than strictly necessary, but it
921 * keeps the signal state that winds up in the core dump
922 * unchanged from the death state, e.g. which thread had
923 * the core-dump signal unblocked.
924 */
925 rm_from_queue(SIG_KERNEL_STOP_MASK, &t->pending);
926 rm_from_queue(SIG_KERNEL_STOP_MASK, &p->signal->shared_pending);
927 p->signal->group_stop_count = 0;
928 p->signal->group_exit_task = t;
929 p = t;
930 do {
931 p->signal->group_stop_count++;
932 signal_wake_up(t, t == p);
933 } while_each_thread(p, t);
934 return;
935 } 914 }
936 915
937 /* 916 /*
@@ -978,7 +957,6 @@ void zap_other_threads(struct task_struct *p)
978{ 957{
979 struct task_struct *t; 958 struct task_struct *t;
980 959
981 p->signal->flags = SIGNAL_GROUP_EXIT;
982 p->signal->group_stop_count = 0; 960 p->signal->group_stop_count = 0;
983 961
984 for (t = next_thread(p); t != p; t = next_thread(t)) { 962 for (t = next_thread(p); t != p; t = next_thread(t)) {
@@ -1709,9 +1687,6 @@ static int do_signal_stop(int signr)
1709 struct signal_struct *sig = current->signal; 1687 struct signal_struct *sig = current->signal;
1710 int stop_count; 1688 int stop_count;
1711 1689
1712 if (!likely(sig->flags & SIGNAL_STOP_DEQUEUED))
1713 return 0;
1714
1715 if (sig->group_stop_count > 0) { 1690 if (sig->group_stop_count > 0) {
1716 /* 1691 /*
1717 * There is a group stop in progress. We don't need to 1692 * There is a group stop in progress. We don't need to
@@ -1719,12 +1694,15 @@ static int do_signal_stop(int signr)
1719 */ 1694 */
1720 stop_count = --sig->group_stop_count; 1695 stop_count = --sig->group_stop_count;
1721 } else { 1696 } else {
1697 struct task_struct *t;
1698
1699 if (!likely(sig->flags & SIGNAL_STOP_DEQUEUED) ||
1700 unlikely(sig->group_exit_task))
1701 return 0;
1722 /* 1702 /*
1723 * There is no group stop already in progress. 1703 * There is no group stop already in progress.
1724 * We must initiate one now. 1704 * We must initiate one now.
1725 */ 1705 */
1726 struct task_struct *t;
1727
1728 sig->group_exit_code = signr; 1706 sig->group_exit_code = signr;
1729 1707
1730 stop_count = 0; 1708 stop_count = 0;
@@ -1752,47 +1730,6 @@ static int do_signal_stop(int signr)
1752 return 1; 1730 return 1;
1753} 1731}
1754 1732
1755/*
1756 * Do appropriate magic when group_stop_count > 0.
1757 * We return nonzero if we stopped, after releasing the siglock.
1758 * We return zero if we still hold the siglock and should look
1759 * for another signal without checking group_stop_count again.
1760 */
1761static int handle_group_stop(void)
1762{
1763 int stop_count;
1764
1765 if (current->signal->group_exit_task == current) {
1766 /*
1767 * Group stop is so we can do a core dump,
1768 * We are the initiating thread, so get on with it.
1769 */
1770 current->signal->group_exit_task = NULL;
1771 return 0;
1772 }
1773
1774 if (current->signal->flags & SIGNAL_GROUP_EXIT)
1775 /*
1776 * Group stop is so another thread can do a core dump,
1777 * or else we are racing against a death signal.
1778 * Just punt the stop so we can get the next signal.
1779 */
1780 return 0;
1781
1782 /*
1783 * There is a group stop in progress. We stop
1784 * without any associated signal being in our queue.
1785 */
1786 stop_count = --current->signal->group_stop_count;
1787 if (stop_count == 0)
1788 current->signal->flags = SIGNAL_STOP_STOPPED;
1789 current->exit_code = current->signal->group_exit_code;
1790 set_current_state(TASK_STOPPED);
1791 spin_unlock_irq(&current->sighand->siglock);
1792 finish_stop(stop_count);
1793 return 1;
1794}
1795
1796int get_signal_to_deliver(siginfo_t *info, struct k_sigaction *return_ka, 1733int get_signal_to_deliver(siginfo_t *info, struct k_sigaction *return_ka,
1797 struct pt_regs *regs, void *cookie) 1734 struct pt_regs *regs, void *cookie)
1798{ 1735{
@@ -1807,7 +1744,7 @@ relock:
1807 struct k_sigaction *ka; 1744 struct k_sigaction *ka;
1808 1745
1809 if (unlikely(current->signal->group_stop_count > 0) && 1746 if (unlikely(current->signal->group_stop_count > 0) &&
1810 handle_group_stop()) 1747 do_signal_stop(0))
1811 goto relock; 1748 goto relock;
1812 1749
1813 signr = dequeue_signal(current, mask, info); 1750 signr = dequeue_signal(current, mask, info);
diff --git a/kernel/sys.c b/kernel/sys.c
index d1fe71eb4546..53de35fc8245 100644
--- a/kernel/sys.c
+++ b/kernel/sys.c
@@ -315,7 +315,7 @@ static void kernel_kexec(void)
315#endif 315#endif
316} 316}
317 317
318void kernel_shutdown_prepare(enum system_states state) 318static void kernel_shutdown_prepare(enum system_states state)
319{ 319{
320 blocking_notifier_call_chain(&reboot_notifier_list, 320 blocking_notifier_call_chain(&reboot_notifier_list,
321 (state == SYSTEM_HALT)?SYS_HALT:SYS_POWER_OFF, NULL); 321 (state == SYSTEM_HALT)?SYS_HALT:SYS_POWER_OFF, NULL);
@@ -1637,7 +1637,7 @@ asmlinkage long sys_umask(int mask)
1637 mask = xchg(&current->fs->umask, mask & S_IRWXUGO); 1637 mask = xchg(&current->fs->umask, mask & S_IRWXUGO);
1638 return mask; 1638 return mask;
1639} 1639}
1640 1640
1641asmlinkage long sys_prctl(int option, unsigned long arg2, unsigned long arg3, 1641asmlinkage long sys_prctl(int option, unsigned long arg2, unsigned long arg3,
1642 unsigned long arg4, unsigned long arg5) 1642 unsigned long arg4, unsigned long arg5)
1643{ 1643{
@@ -1742,6 +1742,17 @@ asmlinkage long sys_prctl(int option, unsigned long arg2, unsigned long arg3,
1742 error = prctl_set_seccomp(arg2); 1742 error = prctl_set_seccomp(arg2);
1743 break; 1743 break;
1744 1744
1745 case PR_CAPBSET_READ:
1746 if (!cap_valid(arg2))
1747 return -EINVAL;
1748 return !!cap_raised(current->cap_bset, arg2);
1749 case PR_CAPBSET_DROP:
1750#ifdef CONFIG_SECURITY_FILE_CAPABILITIES
1751 return cap_prctl_drop(arg2);
1752#else
1753 return -EINVAL;
1754#endif
1755
1745 default: 1756 default:
1746 error = -EINVAL; 1757 error = -EINVAL;
1747 break; 1758 break;
diff --git a/kernel/sys_ni.c b/kernel/sys_ni.c
index beee5b3b68a2..5b9b467de070 100644
--- a/kernel/sys_ni.c
+++ b/kernel/sys_ni.c
@@ -154,7 +154,10 @@ cond_syscall(sys_ioprio_get);
154 154
155/* New file descriptors */ 155/* New file descriptors */
156cond_syscall(sys_signalfd); 156cond_syscall(sys_signalfd);
157cond_syscall(sys_timerfd);
158cond_syscall(compat_sys_signalfd); 157cond_syscall(compat_sys_signalfd);
159cond_syscall(compat_sys_timerfd); 158cond_syscall(sys_timerfd_create);
159cond_syscall(sys_timerfd_settime);
160cond_syscall(sys_timerfd_gettime);
161cond_syscall(compat_sys_timerfd_settime);
162cond_syscall(compat_sys_timerfd_gettime);
160cond_syscall(sys_eventfd); 163cond_syscall(sys_eventfd);
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 7cb1ac3e6fff..5e2ad5bf88e2 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -84,8 +84,11 @@ extern int sysctl_stat_interval;
84extern int latencytop_enabled; 84extern int latencytop_enabled;
85 85
86/* Constants used for minimum and maximum */ 86/* Constants used for minimum and maximum */
87#ifdef CONFIG_DETECT_SOFTLOCKUP 87#if defined(CONFIG_DETECT_SOFTLOCKUP) || defined(CONFIG_HIGHMEM)
88static int one = 1; 88static int one = 1;
89#endif
90
91#ifdef CONFIG_DETECT_SOFTLOCKUP
89static int sixty = 60; 92static int sixty = 60;
90#endif 93#endif
91 94
@@ -416,15 +419,6 @@ static struct ctl_table kern_table[] = {
416 .proc_handler = &proc_dointvec, 419 .proc_handler = &proc_dointvec,
417 }, 420 },
418#endif 421#endif
419#ifdef CONFIG_SECURITY_CAPABILITIES
420 {
421 .procname = "cap-bound",
422 .data = &cap_bset,
423 .maxlen = sizeof(kernel_cap_t),
424 .mode = 0600,
425 .proc_handler = &proc_dointvec_bset,
426 },
427#endif /* def CONFIG_SECURITY_CAPABILITIES */
428#ifdef CONFIG_BLK_DEV_INITRD 422#ifdef CONFIG_BLK_DEV_INITRD
429 { 423 {
430 .ctl_name = KERN_REALROOTDEV, 424 .ctl_name = KERN_REALROOTDEV,
@@ -1150,6 +1144,19 @@ static struct ctl_table vm_table[] = {
1150 .extra1 = &zero, 1144 .extra1 = &zero,
1151 }, 1145 },
1152#endif 1146#endif
1147#ifdef CONFIG_HIGHMEM
1148 {
1149 .ctl_name = CTL_UNNUMBERED,
1150 .procname = "highmem_is_dirtyable",
1151 .data = &vm_highmem_is_dirtyable,
1152 .maxlen = sizeof(vm_highmem_is_dirtyable),
1153 .mode = 0644,
1154 .proc_handler = &proc_dointvec_minmax,
1155 .strategy = &sysctl_intvec,
1156 .extra1 = &zero,
1157 .extra2 = &one,
1158 },
1159#endif
1153/* 1160/*
1154 * NOTE: do not add new entries to this table unless you have read 1161 * NOTE: do not add new entries to this table unless you have read
1155 * Documentation/sysctl/ctl_unnumbered.txt 1162 * Documentation/sysctl/ctl_unnumbered.txt
@@ -2080,26 +2087,6 @@ static int do_proc_dointvec_bset_conv(int *negp, unsigned long *lvalp,
2080 return 0; 2087 return 0;
2081} 2088}
2082 2089
2083#ifdef CONFIG_SECURITY_CAPABILITIES
2084/*
2085 * init may raise the set.
2086 */
2087
2088int proc_dointvec_bset(struct ctl_table *table, int write, struct file *filp,
2089 void __user *buffer, size_t *lenp, loff_t *ppos)
2090{
2091 int op;
2092
2093 if (write && !capable(CAP_SYS_MODULE)) {
2094 return -EPERM;
2095 }
2096
2097 op = is_global_init(current) ? OP_SET : OP_AND;
2098 return do_proc_dointvec(table,write,filp,buffer,lenp,ppos,
2099 do_proc_dointvec_bset_conv,&op);
2100}
2101#endif /* def CONFIG_SECURITY_CAPABILITIES */
2102
2103/* 2090/*
2104 * Taint values can only be increased 2091 * Taint values can only be increased
2105 */ 2092 */
@@ -2513,12 +2500,6 @@ int proc_dointvec(struct ctl_table *table, int write, struct file *filp,
2513 return -ENOSYS; 2500 return -ENOSYS;
2514} 2501}
2515 2502
2516int proc_dointvec_bset(struct ctl_table *table, int write, struct file *filp,
2517 void __user *buffer, size_t *lenp, loff_t *ppos)
2518{
2519 return -ENOSYS;
2520}
2521
2522int proc_dointvec_minmax(struct ctl_table *table, int write, struct file *filp, 2503int proc_dointvec_minmax(struct ctl_table *table, int write, struct file *filp,
2523 void __user *buffer, size_t *lenp, loff_t *ppos) 2504 void __user *buffer, size_t *lenp, loff_t *ppos)
2524{ 2505{
diff --git a/kernel/sysctl_check.c b/kernel/sysctl_check.c
index c3206fa50048..006365b69eaf 100644
--- a/kernel/sysctl_check.c
+++ b/kernel/sysctl_check.c
@@ -37,10 +37,6 @@ static struct trans_ctl_table trans_kern_table[] = {
37 { KERN_NODENAME, "hostname" }, 37 { KERN_NODENAME, "hostname" },
38 { KERN_DOMAINNAME, "domainname" }, 38 { KERN_DOMAINNAME, "domainname" },
39 39
40#ifdef CONFIG_SECURITY_CAPABILITIES
41 { KERN_CAP_BSET, "cap-bound" },
42#endif /* def CONFIG_SECURITY_CAPABILITIES */
43
44 { KERN_PANIC, "panic" }, 40 { KERN_PANIC, "panic" },
45 { KERN_REALROOTDEV, "real-root-dev" }, 41 { KERN_REALROOTDEV, "real-root-dev" },
46 42
@@ -1498,9 +1494,6 @@ int sysctl_check_table(struct nsproxy *namespaces, struct ctl_table *table)
1498 (table->strategy == sysctl_ms_jiffies) || 1494 (table->strategy == sysctl_ms_jiffies) ||
1499 (table->proc_handler == proc_dostring) || 1495 (table->proc_handler == proc_dostring) ||
1500 (table->proc_handler == proc_dointvec) || 1496 (table->proc_handler == proc_dointvec) ||
1501#ifdef CONFIG_SECURITY_CAPABILITIES
1502 (table->proc_handler == proc_dointvec_bset) ||
1503#endif /* def CONFIG_SECURITY_CAPABILITIES */
1504 (table->proc_handler == proc_dointvec_minmax) || 1497 (table->proc_handler == proc_dointvec_minmax) ||
1505 (table->proc_handler == proc_dointvec_jiffies) || 1498 (table->proc_handler == proc_dointvec_jiffies) ||
1506 (table->proc_handler == proc_dointvec_userhz_jiffies) || 1499 (table->proc_handler == proc_dointvec_userhz_jiffies) ||
generated by cgit v1.2.2 (git 2.25.0) at 2026-01-03 07:18:30 -0500