From a9d9baa1e819b2f92f9cfa5240f766c535e636a6 Mon Sep 17 00:00:00 2001 From: Ashok Raj Date: Mon, 28 Nov 2005 13:43:46 -0800 Subject: [PATCH] clean up lock_cpu_hotplug() in cpufreq There are some callers in cpufreq hotplug notify path that the lowest function calls lock_cpu_hotplug(). The lock is already held during cpu_up() and cpu_down() calls when the notify calls are broadcast to registered clients. Ideally if possible, we could disable_preempt() at the highest caller and make sure we dont sleep in the path down in cpufreq->driver_target() calls but the calls are so intertwined and cumbersome to cleanup. Hence we consistently use lock_cpu_hotplug() and unlock_cpu_hotplug() in all places. - Removed export of cpucontrol semaphore and made it static. - removed explicit uses of up/down with lock_cpu_hotplug() so we can keep track of the the callers in same thread context and just keep refcounts without calling a down() that causes a deadlock. - Removed current_in_hotplug() uses - Removed PF_HOTPLUG_CPU in sched.h introduced for the current_in_hotplug() temporary workaround. Tested with insmod of cpufreq_stat.ko, and logical online/offline to make sure we dont have any hang situations. Signed-off-by: Ashok Raj Cc: Zwane Mwaikambo Cc: Shaohua Li Cc: "Siddha, Suresh B" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/cpu.c | 83 +++++++++++++++++++++++++++++++++++------------------------- 1 file changed, 49 insertions(+), 34 deletions(-) (limited to 'kernel') diff --git a/kernel/cpu.c b/kernel/cpu.c index d61ba88f34e5..e882c6babf41 100644 --- a/kernel/cpu.c +++ b/kernel/cpu.c @@ -16,47 +16,76 @@ #include /* This protects CPUs going up and down... */ -DECLARE_MUTEX(cpucontrol); -EXPORT_SYMBOL_GPL(cpucontrol); +static DECLARE_MUTEX(cpucontrol); static struct notifier_block *cpu_chain; -/* - * Used to check by callers if they need to acquire the cpucontrol - * or not to protect a cpu from being removed. Its sometimes required to - * call these functions both for normal operations, and in response to - * a cpu being added/removed. If the context of the call is in the same - * thread context as a CPU hotplug thread, we dont need to take the lock - * since its already protected - * check drivers/cpufreq/cpufreq.c for its usage - Ashok Raj - */ +#ifdef CONFIG_HOTPLUG_CPU +static struct task_struct *lock_cpu_hotplug_owner; +static int lock_cpu_hotplug_depth; -int current_in_cpu_hotplug(void) +static int __lock_cpu_hotplug(int interruptible) { - return (current->flags & PF_HOTPLUG_CPU); + int ret = 0; + + if (lock_cpu_hotplug_owner != current) { + if (interruptible) + ret = down_interruptible(&cpucontrol); + else + down(&cpucontrol); + } + + /* + * Set only if we succeed in locking + */ + if (!ret) { + lock_cpu_hotplug_depth++; + lock_cpu_hotplug_owner = current; + } + + return ret; } -EXPORT_SYMBOL_GPL(current_in_cpu_hotplug); +void lock_cpu_hotplug(void) +{ + __lock_cpu_hotplug(0); +} +EXPORT_SYMBOL_GPL(lock_cpu_hotplug); +void unlock_cpu_hotplug(void) +{ + if (--lock_cpu_hotplug_depth == 0) { + lock_cpu_hotplug_owner = NULL; + up(&cpucontrol); + } +} +EXPORT_SYMBOL_GPL(unlock_cpu_hotplug); + +int lock_cpu_hotplug_interruptible(void) +{ + return __lock_cpu_hotplug(1); +} +EXPORT_SYMBOL_GPL(lock_cpu_hotplug_interruptible); +#endif /* CONFIG_HOTPLUG_CPU */ /* Need to know about CPUs going up/down? */ int register_cpu_notifier(struct notifier_block *nb) { int ret; - if ((ret = down_interruptible(&cpucontrol)) != 0) + if ((ret = lock_cpu_hotplug_interruptible()) != 0) return ret; ret = notifier_chain_register(&cpu_chain, nb); - up(&cpucontrol); + unlock_cpu_hotplug(); return ret; } EXPORT_SYMBOL(register_cpu_notifier); void unregister_cpu_notifier(struct notifier_block *nb) { - down(&cpucontrol); + lock_cpu_hotplug(); notifier_chain_unregister(&cpu_chain, nb); - up(&cpucontrol); + unlock_cpu_hotplug(); } EXPORT_SYMBOL(unregister_cpu_notifier); @@ -112,13 +141,6 @@ int cpu_down(unsigned int cpu) goto out; } - /* - * Leave a trace in current->flags indicating we are already in - * process of performing CPU hotplug. Callers can check if cpucontrol - * is already acquired by current thread, and if so not cause - * a dead lock by not acquiring the lock - */ - current->flags |= PF_HOTPLUG_CPU; err = notifier_call_chain(&cpu_chain, CPU_DOWN_PREPARE, (void *)(long)cpu); if (err == NOTIFY_BAD) { @@ -171,7 +193,6 @@ out_thread: out_allowed: set_cpus_allowed(current, old_allowed); out: - current->flags &= ~PF_HOTPLUG_CPU; unlock_cpu_hotplug(); return err; } @@ -182,7 +203,7 @@ int __devinit cpu_up(unsigned int cpu) int ret; void *hcpu = (void *)(long)cpu; - if ((ret = down_interruptible(&cpucontrol)) != 0) + if ((ret = lock_cpu_hotplug_interruptible()) != 0) return ret; if (cpu_online(cpu) || !cpu_present(cpu)) { @@ -190,11 +211,6 @@ int __devinit cpu_up(unsigned int cpu) goto out; } - /* - * Leave a trace in current->flags indicating we are already in - * process of performing CPU hotplug. - */ - current->flags |= PF_HOTPLUG_CPU; ret = notifier_call_chain(&cpu_chain, CPU_UP_PREPARE, hcpu); if (ret == NOTIFY_BAD) { printk("%s: attempt to bring up CPU %u failed\n", @@ -217,7 +233,6 @@ out_notify: if (ret != 0) notifier_call_chain(&cpu_chain, CPU_UP_CANCELED, hcpu); out: - current->flags &= ~PF_HOTPLUG_CPU; - up(&cpucontrol); + unlock_cpu_hotplug(); return ret; } -- cgit v1.3-14-g43fede From c13cf856cbe16aec3007604dc013cbf3a16c6686 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Mon, 28 Nov 2005 13:43:48 -0800 Subject: [PATCH] fork.c: proc_fork_connector() called under write_lock() Don't do that - it does GFP_KERNEL allocations, for a start. (Reported by Guillaume Thouvenin ) Acked-by: Matt Helsley Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/fork.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/fork.c b/kernel/fork.c index 1c1cf8dc396b..d0d49879ab7c 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -1135,13 +1135,13 @@ static task_t *copy_process(unsigned long clone_flags, __get_cpu_var(process_counts)++; } - proc_fork_connector(p); if (!current->signal->tty && p->signal->tty) p->signal->tty = NULL; nr_threads++; total_forks++; write_unlock_irq(&tasklist_lock); + proc_fork_connector(p); retval = 0; fork_out: -- cgit v1.3-14-g43fede From ee500f274914653a7d3dfca7d0140a3d21658e32 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Mon, 28 Nov 2005 13:43:55 -0800 Subject: [PATCH] fix 32bit overflow in timespec_to_sample() fix 32bit overflow in timespec_to_sample() Signed-off-by: Oleg Nesterov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/posix-cpu-timers.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/posix-cpu-timers.c b/kernel/posix-cpu-timers.c index 84af54c39e1b..cae4f5728997 100644 --- a/kernel/posix-cpu-timers.c +++ b/kernel/posix-cpu-timers.c @@ -36,7 +36,7 @@ timespec_to_sample(clockid_t which_clock, const struct timespec *tp) union cpu_time_count ret; ret.sched = 0; /* high half always zero when .cpu used */ if (CPUCLOCK_WHICH(which_clock) == CPUCLOCK_SCHED) { - ret.sched = tp->tv_sec * NSEC_PER_SEC + tp->tv_nsec; + ret.sched = (unsigned long long)tp->tv_sec * NSEC_PER_SEC + tp->tv_nsec; } else { ret.cpu = timespec_to_cputime(tp); } -- cgit v1.3-14-g43fede From bce61dd49d6ba7799be2de17c772e4c701558f14 Mon Sep 17 00:00:00 2001 From: Ben Collins Date: Mon, 28 Nov 2005 13:43:56 -0800 Subject: [PATCH] Fix hardcoded cpu=0 in workqueue for per_cpu_ptr() calls Tracked this down on an Ultra Enterprise 3000. It's a 6-way machine. Odd thing about this machine (and it's good for finding bugs like this) is that the CPU id's are not 0 based. For instance, on my machine the CPU's are 6/7/10/11/14/15. This caused some NULL pointer dereference in kernel/workqueue.c because for single_threaded workqueue's, it hardcoded the cpu to 0. I changed the 0's to any_online_cpu(cpu_online_mask), which cpumask.h claims is "First cpu in mask". So this fits the same usage. Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/workqueue.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) (limited to 'kernel') diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 42df83d7fad2..2bd5aee1c736 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -102,7 +102,7 @@ int fastcall queue_work(struct workqueue_struct *wq, struct work_struct *work) if (!test_and_set_bit(0, &work->pending)) { if (unlikely(is_single_threaded(wq))) - cpu = 0; + cpu = any_online_cpu(cpu_online_map); BUG_ON(!list_empty(&work->entry)); __queue_work(per_cpu_ptr(wq->cpu_wq, cpu), work); ret = 1; @@ -118,7 +118,7 @@ static void delayed_work_timer_fn(unsigned long __data) int cpu = smp_processor_id(); if (unlikely(is_single_threaded(wq))) - cpu = 0; + cpu = any_online_cpu(cpu_online_map); __queue_work(per_cpu_ptr(wq->cpu_wq, cpu), work); } @@ -266,8 +266,8 @@ void fastcall flush_workqueue(struct workqueue_struct *wq) might_sleep(); if (is_single_threaded(wq)) { - /* Always use cpu 0's area. */ - flush_cpu_workqueue(per_cpu_ptr(wq->cpu_wq, 0)); + /* Always use first cpu's area. */ + flush_cpu_workqueue(per_cpu_ptr(wq->cpu_wq, any_online_cpu(cpu_online_map))); } else { int cpu; @@ -320,7 +320,7 @@ struct workqueue_struct *__create_workqueue(const char *name, lock_cpu_hotplug(); if (singlethread) { INIT_LIST_HEAD(&wq->list); - p = create_workqueue_thread(wq, 0); + p = create_workqueue_thread(wq, any_online_cpu(cpu_online_map)); if (!p) destroy = 1; else @@ -374,7 +374,7 @@ void destroy_workqueue(struct workqueue_struct *wq) /* We don't need the distraction of CPUs appearing and vanishing. */ lock_cpu_hotplug(); if (is_single_threaded(wq)) - cleanup_workqueue_thread(wq, 0); + cleanup_workqueue_thread(wq, any_online_cpu(cpu_online_map)); else { for_each_online_cpu(cpu) cleanup_workqueue_thread(wq, cpu); -- cgit v1.3-14-g43fede From 8c4b8add83c93306b07d78469fd351dc462e4b66 Mon Sep 17 00:00:00 2001 From: Paul Jackson Date: Mon, 28 Nov 2005 13:44:05 -0800 Subject: [PATCH] cpuset fork locking fix Move the cpuset_fork() call below the write_unlock_irq call in kernel/fork.c copy_process(). Since the cpuset-dual-semaphore-locking-overhaul.patch, the cpuset_fork() routine acquires task_lock(), so cannot be called while holding the tasklist_lock for write. Signed-off-by: Paul Jackson Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/fork.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/fork.c b/kernel/fork.c index d0d49879ab7c..fb8572a42297 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -1124,8 +1124,6 @@ static task_t *copy_process(unsigned long clone_flags, if (unlikely(p->ptrace & PT_PTRACED)) __ptrace_link(p, current->parent); - cpuset_fork(p); - attach_pid(p, PIDTYPE_PID, p->pid); attach_pid(p, PIDTYPE_TGID, p->tgid); if (thread_group_leader(p)) { @@ -1142,6 +1140,7 @@ static task_t *copy_process(unsigned long clone_flags, total_forks++; write_unlock_irq(&tasklist_lock); proc_fork_connector(p); + cpuset_fork(p); retval = 0; fork_out: -- cgit v1.3-14-g43fede From 5bd0190bf3d7e53043a048e809ffa29d41b9d6ac Mon Sep 17 00:00:00 2001 From: David Gibson Date: Tue, 29 Nov 2005 19:34:32 -0800 Subject: [PATCH] Fix crash when ptrace poking hugepage areas set_page_dirty() will not cope with being handed a page * which is part of a compound page, but not the master page in that compound page. This case can occur via access_process_vm() if you attemp to write to another process's hugepage memory area using ptrace() (causing an oops or hang). This patch fixes the bug by only calling set_page_dirty() from access_process_vm() if the page is not a compound page. We already use a similar fix in bio_set_pages_dirty() for the case of direct io to hugepages. Signed-off-by: David Gibson Acked-by: William Irwin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/ptrace.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/ptrace.c b/kernel/ptrace.c index 17ee7e5a3451..656476eedb1b 100644 --- a/kernel/ptrace.c +++ b/kernel/ptrace.c @@ -241,7 +241,8 @@ int access_process_vm(struct task_struct *tsk, unsigned long addr, void *buf, in if (write) { copy_to_user_page(vma, page, addr, maddr + offset, buf, bytes); - set_page_dirty_lock(page); + if (!PageCompound(page)) + set_page_dirty_lock(page); } else { copy_from_user_page(vma, page, addr, buf, maddr + offset, bytes); -- cgit v1.3-14-g43fede From 123d3c13e2853a11b4d599d754b356acb12886e2 Mon Sep 17 00:00:00 2001 From: Pavel Machek Date: Tue, 29 Nov 2005 19:34:37 -0800 Subject: [PATCH] fix swsusp on machines not supporting S4 Fix swsusp on machines not supporting S4. With recent changes, it is not possible to trigger it using /sys filesystem. Swsusp does not really need any support from low-level code, it is possible to reboot or halt at the end of suspend. Signed-off-by: Pavel Machek Cc: "Brown, Len" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/power/main.c | 21 ++++++++++++++++----- 1 file changed, 16 insertions(+), 5 deletions(-) (limited to 'kernel') diff --git a/kernel/power/main.c b/kernel/power/main.c index 6ee2cad530e8..d253f3ae2fa5 100644 --- a/kernel/power/main.c +++ b/kernel/power/main.c @@ -24,7 +24,7 @@ DECLARE_MUTEX(pm_sem); -struct pm_ops * pm_ops = NULL; +struct pm_ops *pm_ops; suspend_disk_method_t pm_disk_mode = PM_DISK_SHUTDOWN; /** @@ -151,6 +151,18 @@ static char *pm_states[PM_SUSPEND_MAX] = { #endif }; +static inline int valid_state(suspend_state_t state) +{ + /* Suspend-to-disk does not really need low-level support. + * It can work with reboot if needed. */ + if (state == PM_SUSPEND_DISK) + return 1; + + if (pm_ops && pm_ops->valid && !pm_ops->valid(state)) + return 0; + return 1; +} + /** * enter_state - Do common work of entering low-power state. @@ -167,7 +179,7 @@ static int enter_state(suspend_state_t state) { int error; - if (pm_ops && pm_ops->valid && !pm_ops->valid(state)) + if (!valid_state(state)) return -ENODEV; if (down_trylock(&pm_sem)) return -EBUSY; @@ -238,9 +250,8 @@ static ssize_t state_show(struct subsystem * subsys, char * buf) char * s = buf; for (i = 0; i < PM_SUSPEND_MAX; i++) { - if (pm_states[i] && pm_ops && (!pm_ops->valid - ||(pm_ops->valid && pm_ops->valid(i)))) - s += sprintf(s,"%s ",pm_states[i]); + if (pm_states[i] && valid_state(i)) + s += sprintf(s,"%s ", pm_states[i]); } s += sprintf(s,"\n"); return (s - buf); -- cgit v1.3-14-g43fede