Merge branch 'linus' into sched-devel

4 files changed, 103 insertions, 55 deletions

diff --git a/kernel/capability.c b/kernel/capability.c
index 39e8193b41ea..cfbe44299488 100644
--- a/kernel/capability.c
+++ b/kernel/capability.c
@@ -53,6 +53,69 @@ static void warn_legacy_capability_use(void)
 }
 
 /*
+ * Version 2 capabilities worked fine, but the linux/capability.h file
+ * that accompanied their introduction encouraged their use without
+ * the necessary user-space source code changes. As such, we have
+ * created a version 3 with equivalent functionality to version 2, but
+ * with a header change to protect legacy source code from using
+ * version 2 when it wanted to use version 1. If your system has code
+ * that trips the following warning, it is using version 2 specific
+ * capabilities and may be doing so insecurely.
+ *
+ * The remedy is to either upgrade your version of libcap (to 2.10+,
+ * if the application is linked against it), or recompile your
+ * application with modern kernel headers and this warning will go
+ * away.
+ */
+
+static void warn_deprecated_v2(void)
+{
+        static int warned;
+
+        if (!warned) {
+                char name[sizeof(current->comm)];
+
+                printk(KERN_INFO "warning: `%s' uses deprecated v2"
+                       " capabilities in a way that may be insecure.\n",
+                       get_task_comm(name, current));
+                warned = 1;
+        }
+}
+
+/*
+ * Version check. Return the number of u32s in each capability flag
+ * array, or a negative value on error.
+ */
+static int cap_validate_magic(cap_user_header_t header, unsigned *tocopy)
+{
+        __u32 version;
+
+        if (get_user(version, &header->version))
+                return -EFAULT;
+
+        switch (version) {
+        case _LINUX_CAPABILITY_VERSION_1:
+                warn_legacy_capability_use();
+                *tocopy = _LINUX_CAPABILITY_U32S_1;
+                break;
+        case _LINUX_CAPABILITY_VERSION_2:
+                warn_deprecated_v2();
+                /*
+                 * fall through - v3 is otherwise equivalent to v2.
+                 */
+        case _LINUX_CAPABILITY_VERSION_3:
+                *tocopy = _LINUX_CAPABILITY_U32S_3;
+                break;
+        default:
+                if (put_user((u32)_KERNEL_CAPABILITY_VERSION, &header->version))
+                        return -EFAULT;
+                return -EINVAL;
+        }
+
+        return 0;
+}
+
+/*
  * For sys_getproccap() and sys_setproccap(), any of the three
  * capability set pointers may be NULL -- indicating that that set is
  * uninteresting and/or not to be changed.
@@ -71,27 +134,13 @@ asmlinkage long sys_capget(cap_user_header_t header, cap_user_data_t dataptr)
 {
         int ret = 0;
         pid_t pid;
-        __u32 version;
         struct task_struct *target;
         unsigned tocopy;
         kernel_cap_t pE, pI, pP;
 
-        if (get_user(version, &header->version))
-                return -EFAULT;
-
-        switch (version) {
-        case _LINUX_CAPABILITY_VERSION_1:
-                warn_legacy_capability_use();
-                tocopy = _LINUX_CAPABILITY_U32S_1;
-                break;
-        case _LINUX_CAPABILITY_VERSION_2:
-                tocopy = _LINUX_CAPABILITY_U32S_2;
-                break;
-        default:
-                if (put_user(_LINUX_CAPABILITY_VERSION, &header->version))
-                        return -EFAULT;
-                return -EINVAL;
-        }
+        ret = cap_validate_magic(header, &tocopy);
+        if (ret != 0)
+                return ret;
 
         if (get_user(pid, &header->pid))
                 return -EFAULT;
@@ -118,7 +167,7 @@ out:
         spin_unlock(&task_capability_lock);
 
         if (!ret) {
-                struct __user_cap_data_struct kdata[_LINUX_CAPABILITY_U32S];
+                struct __user_cap_data_struct kdata[_KERNEL_CAPABILITY_U32S];
                 unsigned i;
 
                 for (i = 0; i < tocopy; i++) {
@@ -128,7 +177,7 @@ out:
                 }
 
                 /*
-                 * Note, in the case, tocopy < _LINUX_CAPABILITY_U32S,
+                 * Note, in the case, tocopy < _KERNEL_CAPABILITY_U32S,
                  * we silently drop the upper capabilities here. This
                  * has the effect of making older libcap
                  * implementations implicitly drop upper capability
@@ -240,30 +289,16 @@ static inline int cap_set_all(kernel_cap_t *effective,
  */
 asmlinkage long sys_capset(cap_user_header_t header, const cap_user_data_t data)
 {
-        struct __user_cap_data_struct kdata[_LINUX_CAPABILITY_U32S];
+        struct __user_cap_data_struct kdata[_KERNEL_CAPABILITY_U32S];
         unsigned i, tocopy;
         kernel_cap_t inheritable, permitted, effective;
-        __u32 version;
         struct task_struct *target;
         int ret;
         pid_t pid;
 
-        if (get_user(version, &header->version))
-                return -EFAULT;
-
-        switch (version) {
-        case _LINUX_CAPABILITY_VERSION_1:
-                warn_legacy_capability_use();
-                tocopy = _LINUX_CAPABILITY_U32S_1;
-                break;
-        case _LINUX_CAPABILITY_VERSION_2:
-                tocopy = _LINUX_CAPABILITY_U32S_2;
-                break;
-        default:
-                if (put_user(_LINUX_CAPABILITY_VERSION, &header->version))
-                        return -EFAULT;
-                return -EINVAL;
-        }
+        ret = cap_validate_magic(header, &tocopy);
+        if (ret != 0)
+                return ret;
 
         if (get_user(pid, &header->pid))
                 return -EFAULT;
@@ -281,7 +316,7 @@ asmlinkage long sys_capset(cap_user_header_t header, const cap_user_data_t data)
                 permitted.cap[i] = kdata[i].permitted;
                 inheritable.cap[i] = kdata[i].inheritable;
         }
-        while (i < _LINUX_CAPABILITY_U32S) {
+        while (i < _KERNEL_CAPABILITY_U32S) {
                 effective.cap[i] = 0;
                 permitted.cap[i] = 0;
                 inheritable.cap[i] = 0;
diff --git a/kernel/cpuset.c b/kernel/cpuset.c
index b84354f4de36..fa9702ec1607 100644
--- a/kernel/cpuset.c
+++ b/kernel/cpuset.c
@@ -797,8 +797,10 @@ static int update_cpumask(struct cpuset *cs, char *buf)
                 retval = cpulist_parse(buf, trialcs.cpus_allowed);
                 if (retval < 0)
                         return retval;
+
+                if (!cpus_subset(trialcs.cpus_allowed, cpu_online_map))
+                        return -EINVAL;
         }
-        cpus_and(trialcs.cpus_allowed, trialcs.cpus_allowed, cpu_online_map);
         retval = validate_change(cs, &trialcs);
         if (retval < 0)
                 return retval;
@@ -932,9 +934,11 @@ static int update_nodemask(struct cpuset *cs, char *buf)
                 retval = nodelist_parse(buf, trialcs.mems_allowed);
                 if (retval < 0)
                         goto done;
+
+                if (!nodes_subset(trialcs.mems_allowed,
+                                node_states[N_HIGH_MEMORY]))
+                        return -EINVAL;
         }
-        nodes_and(trialcs.mems_allowed, trialcs.mems_allowed,
-                                                node_states[N_HIGH_MEMORY]);
         oldmem = cs->mems_allowed;
         if (nodes_equal(oldmem, trialcs.mems_allowed)) {
                 retval = 0;             /* Too easy - nothing to do */
diff --git a/kernel/kprobes.c b/kernel/kprobes.c
index 1e0250cb9486..d4998f81e229 100644
--- a/kernel/kprobes.c
+++ b/kernel/kprobes.c
@@ -699,8 +699,9 @@ static int __register_kprobes(struct kprobe **kps, int num,
                 return -EINVAL;
         for (i = 0; i < num; i++) {
                 ret = __register_kprobe(kps[i], called_from);
-                if (ret < 0 && i > 0) {
-                        unregister_kprobes(kps, i);
+                if (ret < 0) {
+                        if (i > 0)
+                                unregister_kprobes(kps, i);
                         break;
                 }
         }
@@ -776,8 +777,9 @@ static int __register_jprobes(struct jprobe **jps, int num,
                         jp->kp.break_handler = longjmp_break_handler;
                         ret = __register_kprobe(&jp->kp, called_from);
                 }
-                if (ret < 0 && i > 0) {
-                        unregister_jprobes(jps, i);
+                if (ret < 0) {
+                        if (i > 0)
+                                unregister_jprobes(jps, i);
                         break;
                 }
         }
@@ -920,8 +922,9 @@ static int __register_kretprobes(struct kretprobe **rps, int num,
                 return -EINVAL;
         for (i = 0; i < num; i++) {
                 ret = __register_kretprobe(rps[i], called_from);
-                if (ret < 0 && i > 0) {
-                        unregister_kretprobes(rps, i);
+                if (ret < 0) {
+                        if (i > 0)
+                                unregister_kretprobes(rps, i);
                         break;
                 }
         }
diff --git a/kernel/sched.c b/kernel/sched.c
index 07d5472dee99..554de4009803 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -315,12 +315,15 @@ static DEFINE_SPINLOCK(task_group_lock);
 #endif /* CONFIG_USER_SCHED */
 
 /*
- * A weight of 0, 1 or ULONG_MAX can cause arithmetics problems.
+ * A weight of 0 or 1 can cause arithmetics problems.
+ * A weight of a cfs_rq is the sum of weights of which entities
+ * are queued on this cfs_rq, so a weight of a entity should not be
+ * too large, so as the shares value of a task group.
  * (The default weight is 1024 - so there's no practical
  *  limitation from this.)
  */
 #define MIN_SHARES      2
-#define MAX_SHARES      (ULONG_MAX - 1)
+#define MAX_SHARES      (1UL << 18)
 
 static int init_task_group_load = INIT_TASK_GROUP_LOAD;
 #endif
@@ -1346,8 +1349,13 @@ calc_delta_mine(unsigned long delta_exec, unsigned long weight,
 {
         u64 tmp;
 
-        if (!lw->inv_weight)
-                lw->inv_weight = 1 + (WMULT_CONST-lw->weight/2)/(lw->weight+1);
+        if (!lw->inv_weight) {
+                if (BITS_PER_LONG > 32 && unlikely(lw->weight >= WMULT_CONST))
+                        lw->inv_weight = 1;
+                else
+                        lw->inv_weight = 1 + (WMULT_CONST-lw->weight/2)
+                                / (lw->weight+1);
+        }
 
         tmp = (u64)delta_exec * weight;
         /*
@@ -4166,12 +4174,10 @@ need_resched_nonpreemptible:
         clear_tsk_need_resched(prev);
 
         if (prev->state && !(preempt_count() & PREEMPT_ACTIVE)) {
-                if (unlikely((prev->state & TASK_INTERRUPTIBLE) &&
-                                signal_pending(prev))) {
+                if (unlikely(signal_pending_state(prev->state, prev)))
                         prev->state = TASK_RUNNING;
-                } else {
+                else
                         deactivate_task(rq, prev, 1);
-                }
                 switch_count = &prev->nvcsw;
         }