From 9642d18eee2cd169b60c6ac0f20bda745b5a3d1e Mon Sep 17 00:00:00 2001 From: Vatika Harlalka Date: Tue, 1 Sep 2015 16:50:59 +0200 Subject: nohz: Affine unpinned timers to housekeepers The problem addressed in this patch is about affining unpinned timers. Adaptive or Full Dynticks CPUs are currently disturbed by unnecessary jitter due to firing of such timers on them. This patch will affine timers to online CPUs which are not full dynticks in NOHZ_FULL configured systems. It should not introduce overhead in nohz full off case due to static keys. Signed-off-by: Vatika Harlalka Signed-off-by: Frederic Weisbecker Reviewed-by: Preeti U Murthy Acked-by: Thomas Gleixner Cc: Chris Metcalf Cc: Christoph Lameter Cc: Linus Torvalds Cc: Paul E. McKenney Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/1441119060-2230-2-git-send-email-fweisbec@gmail.com Signed-off-by: Ingo Molnar --- include/linux/tick.h | 9 +++++++++ 1 file changed, 9 insertions(+) (limited to 'include') diff --git a/include/linux/tick.h b/include/linux/tick.h index 48d901f83f92..e312219ff823 100644 --- a/include/linux/tick.h +++ b/include/linux/tick.h @@ -147,11 +147,20 @@ static inline void tick_nohz_full_add_cpus_to(struct cpumask *mask) cpumask_or(mask, mask, tick_nohz_full_mask); } +static inline int housekeeping_any_cpu(void) +{ + return cpumask_any_and(housekeeping_mask, cpu_online_mask); +} + extern void tick_nohz_full_kick(void); extern void tick_nohz_full_kick_cpu(int cpu); extern void tick_nohz_full_kick_all(void); extern void __tick_nohz_task_switch(void); #else +static inline int housekeeping_any_cpu(void) +{ + return smp_processor_id(); +} static inline bool tick_nohz_full_enabled(void) { return false; } static inline bool tick_nohz_full_cpu(int cpu) { return false; } static inline void tick_nohz_full_add_cpus_to(struct cpumask *mask) { } -- cgit v1.2.3 From edcd591c77a48da753456f92daf8bb50fe9bac93 Mon Sep 17 00:00:00 2001 From: Jonathan Corbet Date: Mon, 7 Sep 2015 13:18:03 -0600 Subject: locking/static_keys: Fix a silly typo Commit: 412758cb2670 ("jump label, locking/static_keys: Update docs") introduced a typo that might as well get fixed. Signed-off-by: Jonathan Corbet Cc: Andrew Morton Cc: Jason Baron Cc: Linus Torvalds Cc: Paul E. McKenney Cc: Peter Zijlstra Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/20150907131803.54c027e1@lwn.net Signed-off-by: Ingo Molnar --- Documentation/static-keys.txt | 2 +- include/linux/jump_label.h | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/Documentation/static-keys.txt b/Documentation/static-keys.txt index f4cb0b2d5cd7..ec911583f6c5 100644 --- a/Documentation/static-keys.txt +++ b/Documentation/static-keys.txt @@ -16,7 +16,7 @@ The updated API replacements are: DEFINE_STATIC_KEY_TRUE(key); DEFINE_STATIC_KEY_FALSE(key); static_key_likely() -statick_key_unlikely() +static_key_unlikely() 0) Abstract diff --git a/include/linux/jump_label.h b/include/linux/jump_label.h index 7f653e8f6690..0684bd3a48fc 100644 --- a/include/linux/jump_label.h +++ b/include/linux/jump_label.h @@ -22,7 +22,7 @@ * DEFINE_STATIC_KEY_TRUE(key); * DEFINE_STATIC_KEY_FALSE(key); * static_key_likely() - * statick_key_unlikely() + * static_key_unlikely() * * Jump labels provide an interface to generate dynamic branches using * self-modifying code. Assuming toolchain and architecture support, if we -- cgit v1.2.3 From 43b3f02899f74ae9914a39547cc5492156f0027a Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Fri, 4 Sep 2015 17:25:23 +0200 Subject: locking/qspinlock/x86: Fix performance regression under unaccelerated VMs Dave ran into horrible performance on a VM without PARAVIRT_SPINLOCKS set and Linus noted that the test-and-set implementation was retarded. One should spin on the variable with a load, not a RMW. While there, remove 'queued' from the name, as the lock isn't queued at all, but a simple test-and-set. Suggested-by: Linus Torvalds Reported-by: Dave Chinner Tested-by: Dave Chinner Signed-off-by: Peter Zijlstra (Intel) Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Waiman Long Cc: stable@vger.kernel.org # v4.2+ Link: http://lkml.kernel.org/r/20150904152523.GR18673@twins.programming.kicks-ass.net Signed-off-by: Ingo Molnar --- arch/x86/include/asm/qspinlock.h | 16 ++++++++++++---- include/asm-generic/qspinlock.h | 4 ++-- kernel/locking/qspinlock.c | 2 +- 3 files changed, 15 insertions(+), 7 deletions(-) (limited to 'include') diff --git a/arch/x86/include/asm/qspinlock.h b/arch/x86/include/asm/qspinlock.h index 9d51fae1cba3..8dde3bdc4a05 100644 --- a/arch/x86/include/asm/qspinlock.h +++ b/arch/x86/include/asm/qspinlock.h @@ -39,15 +39,23 @@ static inline void queued_spin_unlock(struct qspinlock *lock) } #endif -#define virt_queued_spin_lock virt_queued_spin_lock +#define virt_spin_lock virt_spin_lock -static inline bool virt_queued_spin_lock(struct qspinlock *lock) +static inline bool virt_spin_lock(struct qspinlock *lock) { if (!static_cpu_has(X86_FEATURE_HYPERVISOR)) return false; - while (atomic_cmpxchg(&lock->val, 0, _Q_LOCKED_VAL) != 0) - cpu_relax(); + /* + * On hypervisors without PARAVIRT_SPINLOCKS support we fall + * back to a Test-and-Set spinlock, because fair locks have + * horrible lock 'holder' preemption issues. + */ + + do { + while (atomic_read(&lock->val) != 0) + cpu_relax(); + } while (atomic_cmpxchg(&lock->val, 0, _Q_LOCKED_VAL) != 0); return true; } diff --git a/include/asm-generic/qspinlock.h b/include/asm-generic/qspinlock.h index 83bfb87f5bf1..e2aadbc7151f 100644 --- a/include/asm-generic/qspinlock.h +++ b/include/asm-generic/qspinlock.h @@ -111,8 +111,8 @@ static inline void queued_spin_unlock_wait(struct qspinlock *lock) cpu_relax(); } -#ifndef virt_queued_spin_lock -static __always_inline bool virt_queued_spin_lock(struct qspinlock *lock) +#ifndef virt_spin_lock +static __always_inline bool virt_spin_lock(struct qspinlock *lock) { return false; } diff --git a/kernel/locking/qspinlock.c b/kernel/locking/qspinlock.c index 337c8818541d..87e9ce6a63c5 100644 --- a/kernel/locking/qspinlock.c +++ b/kernel/locking/qspinlock.c @@ -289,7 +289,7 @@ void queued_spin_lock_slowpath(struct qspinlock *lock, u32 val) if (pv_enabled()) goto queue; - if (virt_queued_spin_lock(lock)) + if (virt_spin_lock(lock)) return; /* -- cgit v1.2.3 From eef7635a22f6b144206b5ca2f1398f637acffc4d Mon Sep 17 00:00:00 2001 From: Viresh Kumar Date: Fri, 11 Sep 2015 09:34:26 +0530 Subject: clockevents: Remove unused set_mode() callback All users are migrated to the per-state callbacks, get rid of the unused interface and the core support code. Signed-off-by: Viresh Kumar Signed-off-by: Thomas Gleixner Cc: linaro-kernel@lists.linaro.org Cc: John Stultz Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/fd60de14cf6d125489c031207567bb255ad946f6.1441943991.git.viresh.kumar@linaro.org Signed-off-by: Ingo Molnar --- include/linux/clockchips.h | 29 +++++-------------------- kernel/time/clockevents.c | 42 +----------------------------------- kernel/time/tick-common.c | 1 - kernel/time/timer_list.c | 54 +++++++++++++++++++++------------------------- 4 files changed, 30 insertions(+), 96 deletions(-) (limited to 'include') diff --git a/include/linux/clockchips.h b/include/linux/clockchips.h index 31ce435981fe..bdcf358dfce2 100644 --- a/include/linux/clockchips.h +++ b/include/linux/clockchips.h @@ -18,15 +18,6 @@ struct clock_event_device; struct module; -/* Clock event mode commands for legacy ->set_mode(): OBSOLETE */ -enum clock_event_mode { - CLOCK_EVT_MODE_UNUSED, - CLOCK_EVT_MODE_SHUTDOWN, - CLOCK_EVT_MODE_PERIODIC, - CLOCK_EVT_MODE_ONESHOT, - CLOCK_EVT_MODE_RESUME, -}; - /* * Possible states of a clock event device. * @@ -86,16 +77,14 @@ enum clock_event_state { * @min_delta_ns: minimum delta value in ns * @mult: nanosecond to cycles multiplier * @shift: nanoseconds to cycles divisor (power of two) - * @mode: operating mode, relevant only to ->set_mode(), OBSOLETE * @state_use_accessors:current state of the device, assigned by the core code * @features: features * @retries: number of forced programming retries - * @set_mode: legacy set mode function, only for modes <= CLOCK_EVT_MODE_RESUME. - * @set_state_periodic: switch state to periodic, if !set_mode - * @set_state_oneshot: switch state to oneshot, if !set_mode - * @set_state_oneshot_stopped: switch state to oneshot_stopped, if !set_mode - * @set_state_shutdown: switch state to shutdown, if !set_mode - * @tick_resume: resume clkevt device, if !set_mode + * @set_state_periodic: switch state to periodic + * @set_state_oneshot: switch state to oneshot + * @set_state_oneshot_stopped: switch state to oneshot_stopped + * @set_state_shutdown: switch state to shutdown + * @tick_resume: resume clkevt device * @broadcast: function to broadcast events * @min_delta_ticks: minimum delta value in ticks stored for reconfiguration * @max_delta_ticks: maximum delta value in ticks stored for reconfiguration @@ -116,18 +105,10 @@ struct clock_event_device { u64 min_delta_ns; u32 mult; u32 shift; - enum clock_event_mode mode; enum clock_event_state state_use_accessors; unsigned int features; unsigned long retries; - /* - * State transition callback(s): Only one of the two groups should be - * defined: - * - set_mode(), only for modes <= CLOCK_EVT_MODE_RESUME. - * - set_state_{shutdown|periodic|oneshot|oneshot_stopped}(), tick_resume(). - */ - void (*set_mode)(enum clock_event_mode mode, struct clock_event_device *); int (*set_state_periodic)(struct clock_event_device *); int (*set_state_oneshot)(struct clock_event_device *); int (*set_state_oneshot_stopped)(struct clock_event_device *); diff --git a/kernel/time/clockevents.c b/kernel/time/clockevents.c index 50eb107f1198..a9b76a40319e 100644 --- a/kernel/time/clockevents.c +++ b/kernel/time/clockevents.c @@ -97,20 +97,6 @@ EXPORT_SYMBOL_GPL(clockevent_delta2ns); static int __clockevents_switch_state(struct clock_event_device *dev, enum clock_event_state state) { - /* Transition with legacy set_mode() callback */ - if (dev->set_mode) { - /* Legacy callback doesn't support new modes */ - if (state > CLOCK_EVT_STATE_ONESHOT) - return -ENOSYS; - /* - * 'clock_event_state' and 'clock_event_mode' have 1-to-1 - * mapping until *_ONESHOT, and so a simple cast will work. - */ - dev->set_mode((enum clock_event_mode)state, dev); - dev->mode = (enum clock_event_mode)state; - return 0; - } - if (dev->features & CLOCK_EVT_FEAT_DUMMY) return 0; @@ -204,12 +190,8 @@ int clockevents_tick_resume(struct clock_event_device *dev) { int ret = 0; - if (dev->set_mode) { - dev->set_mode(CLOCK_EVT_MODE_RESUME, dev); - dev->mode = CLOCK_EVT_MODE_RESUME; - } else if (dev->tick_resume) { + if (dev->tick_resume) ret = dev->tick_resume(dev); - } return ret; } @@ -460,26 +442,6 @@ int clockevents_unbind_device(struct clock_event_device *ced, int cpu) } EXPORT_SYMBOL_GPL(clockevents_unbind_device); -/* Sanity check of state transition callbacks */ -static int clockevents_sanity_check(struct clock_event_device *dev) -{ - /* Legacy set_mode() callback */ - if (dev->set_mode) { - /* We shouldn't be supporting new modes now */ - WARN_ON(dev->set_state_periodic || dev->set_state_oneshot || - dev->set_state_shutdown || dev->tick_resume || - dev->set_state_oneshot_stopped); - - BUG_ON(dev->mode != CLOCK_EVT_MODE_UNUSED); - return 0; - } - - if (dev->features & CLOCK_EVT_FEAT_DUMMY) - return 0; - - return 0; -} - /** * clockevents_register_device - register a clock event device * @dev: device to register @@ -488,8 +450,6 @@ void clockevents_register_device(struct clock_event_device *dev) { unsigned long flags; - BUG_ON(clockevents_sanity_check(dev)); - /* Initialize state to DETACHED */ clockevent_set_state(dev, CLOCK_EVT_STATE_DETACHED); diff --git a/kernel/time/tick-common.c b/kernel/time/tick-common.c index d11c55b6ab7d..4fcd99e12aa0 100644 --- a/kernel/time/tick-common.c +++ b/kernel/time/tick-common.c @@ -398,7 +398,6 @@ void tick_shutdown(unsigned int cpu) * the set mode function! */ clockevent_set_state(dev, CLOCK_EVT_STATE_DETACHED); - dev->mode = CLOCK_EVT_MODE_UNUSED; clockevents_exchange_device(dev, NULL); dev->event_handler = clockevents_handle_noop; td->evtdev = NULL; diff --git a/kernel/time/timer_list.c b/kernel/time/timer_list.c index 129c96033e46..f75e35b60149 100644 --- a/kernel/time/timer_list.c +++ b/kernel/time/timer_list.c @@ -225,7 +225,7 @@ print_tickdevice(struct seq_file *m, struct tick_device *td, int cpu) (unsigned long long) dev->min_delta_ns); SEQ_printf(m, " mult: %u\n", dev->mult); SEQ_printf(m, " shift: %u\n", dev->shift); - SEQ_printf(m, " mode: %d\n", dev->mode); + SEQ_printf(m, " mode: %d\n", clockevent_get_state(dev)); SEQ_printf(m, " next_event: %Ld nsecs\n", (unsigned long long) ktime_to_ns(dev->next_event)); @@ -233,40 +233,34 @@ print_tickdevice(struct seq_file *m, struct tick_device *td, int cpu) print_name_offset(m, dev->set_next_event); SEQ_printf(m, "\n"); - if (dev->set_mode) { - SEQ_printf(m, " set_mode: "); - print_name_offset(m, dev->set_mode); + if (dev->set_state_shutdown) { + SEQ_printf(m, " shutdown: "); + print_name_offset(m, dev->set_state_shutdown); SEQ_printf(m, "\n"); - } else { - if (dev->set_state_shutdown) { - SEQ_printf(m, " shutdown: "); - print_name_offset(m, dev->set_state_shutdown); - SEQ_printf(m, "\n"); - } + } - if (dev->set_state_periodic) { - SEQ_printf(m, " periodic: "); - print_name_offset(m, dev->set_state_periodic); - SEQ_printf(m, "\n"); - } + if (dev->set_state_periodic) { + SEQ_printf(m, " periodic: "); + print_name_offset(m, dev->set_state_periodic); + SEQ_printf(m, "\n"); + } - if (dev->set_state_oneshot) { - SEQ_printf(m, " oneshot: "); - print_name_offset(m, dev->set_state_oneshot); - SEQ_printf(m, "\n"); - } + if (dev->set_state_oneshot) { + SEQ_printf(m, " oneshot: "); + print_name_offset(m, dev->set_state_oneshot); + SEQ_printf(m, "\n"); + } - if (dev->set_state_oneshot_stopped) { - SEQ_printf(m, " oneshot stopped: "); - print_name_offset(m, dev->set_state_oneshot_stopped); - SEQ_printf(m, "\n"); - } + if (dev->set_state_oneshot_stopped) { + SEQ_printf(m, " oneshot stopped: "); + print_name_offset(m, dev->set_state_oneshot_stopped); + SEQ_printf(m, "\n"); + } - if (dev->tick_resume) { - SEQ_printf(m, " resume: "); - print_name_offset(m, dev->tick_resume); - SEQ_printf(m, "\n"); - } + if (dev->tick_resume) { + SEQ_printf(m, " resume: "); + print_name_offset(m, dev->tick_resume); + SEQ_printf(m, "\n"); } SEQ_printf(m, " event_handler: "); -- cgit v1.2.3 From 1975dbc276c6ab62230cf4f9df5ddc9ff0e0e473 Mon Sep 17 00:00:00 2001 From: Jonathan Corbet Date: Mon, 14 Sep 2015 17:11:05 -0600 Subject: locking/static_keys: Fix up the static keys documentation Fix a few small mistakes in the static key documentation and delete an unneeded sentence. Suggested-by: Jason Baron Signed-off-by: Jonathan Corbet Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/20150914171105.511e1e21@lwn.net Signed-off-by: Ingo Molnar --- Documentation/static-keys.txt | 4 ++-- include/linux/jump_label.h | 10 ++++------ 2 files changed, 6 insertions(+), 8 deletions(-) (limited to 'include') diff --git a/Documentation/static-keys.txt b/Documentation/static-keys.txt index ec911583f6c5..477927becacb 100644 --- a/Documentation/static-keys.txt +++ b/Documentation/static-keys.txt @@ -15,8 +15,8 @@ The updated API replacements are: DEFINE_STATIC_KEY_TRUE(key); DEFINE_STATIC_KEY_FALSE(key); -static_key_likely() -static_key_unlikely() +static_branch_likely() +static_branch_unlikely() 0) Abstract diff --git a/include/linux/jump_label.h b/include/linux/jump_label.h index 0684bd3a48fc..f1094238ab2a 100644 --- a/include/linux/jump_label.h +++ b/include/linux/jump_label.h @@ -21,8 +21,8 @@ * * DEFINE_STATIC_KEY_TRUE(key); * DEFINE_STATIC_KEY_FALSE(key); - * static_key_likely() - * static_key_unlikely() + * static_branch_likely() + * static_branch_unlikely() * * Jump labels provide an interface to generate dynamic branches using * self-modifying code. Assuming toolchain and architecture support, if we @@ -45,12 +45,10 @@ * statement, setting the key to true requires us to patch in a jump * to the out-of-line of true branch. * - * In addtion to static_branch_{enable,disable}, we can also reference count + * In addition to static_branch_{enable,disable}, we can also reference count * the key or branch direction via static_branch_{inc,dec}. Thus, * static_branch_inc() can be thought of as a 'make more true' and - * static_branch_dec() as a 'make more false'. The inc()/dec() - * interface is meant to be used exclusively from the inc()/dec() for a given - * key. + * static_branch_dec() as a 'make more false'. * * Since this relies on modifying code, the branch modifying functions * must be considered absolute slow paths (machine wide synchronization etc.). -- cgit v1.2.3 From 7f61f545657281a3a1b0faf68993165ebdecc51b Mon Sep 17 00:00:00 2001 From: Ilya Dryomov Date: Mon, 14 Sep 2015 16:01:05 +0300 Subject: libceph: don't access invalid memory in keepalive2 path This struct ceph_timespec ceph_ts; ... con_out_kvec_add(con, sizeof(ceph_ts), &ceph_ts); wraps ceph_ts into a kvec and adds it to con->out_kvec array, yet ceph_ts becomes invalid on return from prepare_write_keepalive(). As a result, we send out bogus keepalive2 stamps. Fix this by encoding into a ceph_timespec member, similar to how acks are read and written. Signed-off-by: Ilya Dryomov Reviewed-by: Yan, Zheng --- include/linux/ceph/messenger.h | 4 +++- net/ceph/messenger.c | 9 +++++---- 2 files changed, 8 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/include/linux/ceph/messenger.h b/include/linux/ceph/messenger.h index 7e1252e97a30..b2371d9b51fa 100644 --- a/include/linux/ceph/messenger.h +++ b/include/linux/ceph/messenger.h @@ -238,6 +238,8 @@ struct ceph_connection { bool out_kvec_is_msg; /* kvec refers to out_msg */ int out_more; /* there is more data after the kvecs */ __le64 out_temp_ack; /* for writing an ack */ + struct ceph_timespec out_temp_keepalive2; /* for writing keepalive2 + stamp */ /* message in temps */ struct ceph_msg_header in_hdr; @@ -248,7 +250,7 @@ struct ceph_connection { int in_base_pos; /* bytes read */ __le64 in_temp_ack; /* for reading an ack */ - struct timespec last_keepalive_ack; + struct timespec last_keepalive_ack; /* keepalive2 ack stamp */ struct delayed_work work; /* send|recv work */ unsigned long delay; /* current delay interval */ diff --git a/net/ceph/messenger.c b/net/ceph/messenger.c index 525f454f7531..b9b0e3b5da49 100644 --- a/net/ceph/messenger.c +++ b/net/ceph/messenger.c @@ -1353,11 +1353,12 @@ static void prepare_write_keepalive(struct ceph_connection *con) dout("prepare_write_keepalive %p\n", con); con_out_kvec_reset(con); if (con->peer_features & CEPH_FEATURE_MSGR_KEEPALIVE2) { - struct timespec ts = CURRENT_TIME; - struct ceph_timespec ceph_ts; - ceph_encode_timespec(&ceph_ts, &ts); + struct timespec now = CURRENT_TIME; + con_out_kvec_add(con, sizeof(tag_keepalive2), &tag_keepalive2); - con_out_kvec_add(con, sizeof(ceph_ts), &ceph_ts); + ceph_encode_timespec(&con->out_temp_keepalive2, &now); + con_out_kvec_add(con, sizeof(con->out_temp_keepalive2), + &con->out_temp_keepalive2); } else { con_out_kvec_add(con, sizeof(tag_keepalive), &tag_keepalive); } -- cgit v1.2.3 From 335c25858218e76ef47f92ecb9d22e919d36140d Mon Sep 17 00:00:00 2001 From: Ilya Dryomov Date: Mon, 14 Sep 2015 12:44:22 +0300 Subject: libceph: advertise support for keepalive2 We are the client, but advertise keepalive2 anyway - for consistency, if nothing else. In the future the server might want to know whether its clients support keepalive2. Signed-off-by: Ilya Dryomov Reviewed-by: Yan, Zheng --- include/linux/ceph/ceph_features.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include') diff --git a/include/linux/ceph/ceph_features.h b/include/linux/ceph/ceph_features.h index 4763ad64e832..f89b31d45cc8 100644 --- a/include/linux/ceph/ceph_features.h +++ b/include/linux/ceph/ceph_features.h @@ -107,6 +107,7 @@ static inline u64 ceph_sanitize_features(u64 features) CEPH_FEATURE_OSDMAP_ENC | \ CEPH_FEATURE_CRUSH_TUNABLES3 | \ CEPH_FEATURE_OSD_PRIMARY_AFFINITY | \ + CEPH_FEATURE_MSGR_KEEPALIVE2 | \ CEPH_FEATURE_CRUSH_V4) #define CEPH_FEATURES_REQUIRED_DEFAULT \ -- cgit v1.2.3