v2 RCU fixes for 3.11
cover v2 RCU fixes for 3.11
1-2. Convert remaining printk() calls to pr_*().
3. Remove "Experimental" flags from old RCU Kconfig options.
4. Automatically tune defaults for delays between attempts to
force quiescent states.
5. Merge adjacent identical #ifdefs.
6. Apply Dave Jones's NOCB Kconfig help feedback.
Changes since v2:
o Adaptive-ticks-CPU-kick patch dropped from this series and
added to Frederic Weisbecker's adaptive-ticks patchset.
o Patch to avoid allocating bootmem from rcu_init() was applied
to 3.10 due to the number of people running into the corresponding splat.
o Changed force-quiescent-state delay computation to use impulse
functions, as suggested by Josh Triplett.
o Added Dave Jones's NOCB Kconfig help patch.
Thanx, Paul
b/init/Kconfig | 36 +++++++++++++++++++++--------- b/kernel/rcupdate.c | 3 -- b/kernel/rcutree.c | 24 +++++++++++++++----- b/kernel/rcutree.h | 15 ++++++++---- b/kernel/rcutree_plugin.h | 55 ++++++++++++++++++++++------------------------ 5 files changed, 82 insertions(+), 51 deletions(-)
PATCH 1/6 rcu: Convert rcutree.c printk calls
From: "Paul E. McKenney" <[email protected]>
This commit converts printk() calls to the corresponding pr_*() calls.
--- kernel/rcutree.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
kernel/rcutree.c +3 -3
@@ -866,7 +866,7 @@ static void print_other_cpu_stall(struct rcu_state *rsp)* See Documentation/RCU/stallwarn.txt for info on how to debug * RCU CPU stall warnings. */ -printk (KERN_ERR "INFO: %s detected stalls on CPUs/tasks:" , +pr_err ("INFO: %s detected stalls on CPUs/tasks:" , rsp->name);print_cpu_stall_info_begin ();rcu_for_each_leaf_node (rsp, rnp) {
@@ -899,7 +899,7 @@ static void print_other_cpu_stall(struct rcu_state *rsp)smp_processor_id (), (long )(jiffies - rsp->gp_start), rsp->gpnum, rsp->completed, totqlen);if (ndetected ==0 ) -printk (KERN_ERR "INFO: Stall ended before state dump start\n" ); +pr_err ("INFO: Stall ended before state dump start\n" );else if (!trigger_all_cpu_backtrace ())rcu_dump_cpu_stacks (rsp);
@@ -922,7 +922,7 @@ static void print_cpu_stall(struct rcu_state *rsp)* See Documentation/RCU/stallwarn.txt for info on how to debug * RCU CPU stall warnings. */ -printk (KERN_ERR "INFO: %s self-detected stall on CPU" , rsp->name); +pr_err ("INFO: %s self-detected stall on CPU" , rsp->name);print_cpu_stall_info_begin ();print_cpu_stall_info (rsp,smp_processor_id ());print_cpu_stall_info_end ();
Signature · 1 line
1.8.1.5
PATCH 2/6 rcu: Convert rcutree_plugin.h printk calls
From: "Paul E. McKenney" <[email protected]>
This commit converts printk() calls to the corresponding pr_*() calls.
--- kernel/rcutree_plugin.h | 45 ++++++++++++++++++++++----------------------- 1 file changed, 22 insertions(+), 23 deletions(-)
kernel/rcutree_plugin.h +22 -23
@@ -53,38 +53,37 @@ static char __initdata nocb_buf[NR_CPUS * 5];static void __init rcu_bootup_announce_oddness (void ) {#ifdef CONFIG_RCU_TRACE -printk (KERN_INFO "\tRCU debugfs-based tracing is enabled.\n" ); +pr_info ("\tRCU debugfs-based tracing is enabled.\n" );#endif #if (defined(CONFIG_64BIT ) &&CONFIG_RCU_FANOUT !=64 ) || (!defined(CONFIG_64BIT ) &&CONFIG_RCU_FANOUT !=32 ) -printk (KERN_INFO "\tCONFIG_RCU_FANOUT set to non-default value of %d\n" , +pr_info ("\tCONFIG_RCU_FANOUT set to non-default value of %d\n" ,CONFIG_RCU_FANOUT );#endif #ifdef CONFIG_RCU_FANOUT_EXACT -printk (KERN_INFO "\tHierarchical RCU autobalancing is disabled.\n" ); +pr_info ("\tHierarchical RCU autobalancing is disabled.\n" );#endif #ifdef CONFIG_RCU_FAST_NO_HZ -printk (KERN_INFO -"\tRCU dyntick-idle grace-period acceleration is enabled.\n" ); +pr_info ("\tRCU dyntick-idle grace-period acceleration is enabled.\n" );#endif #ifdef CONFIG_PROVE_RCU -printk (KERN_INFO "\tRCU lockdep checking is enabled.\n" ); +pr_info ("\tRCU lockdep checking is enabled.\n" );#endif #ifdef CONFIG_RCU_TORTURE_TEST_RUNNABLE -printk (KERN_INFO "\tRCU torture testing starts during boot.\n" ); +pr_info ("\tRCU torture testing starts during boot.\n" );#endif #if defined(CONFIG_TREE_PREEMPT_RCU ) && !defined(CONFIG_RCU_CPU_STALL_VERBOSE ) -printk (KERN_INFO "\tDump stacks of tasks blocking RCU-preempt GP.\n" ); +pr_info ("\tDump stacks of tasks blocking RCU-preempt GP.\n" );#endif #if defined(CONFIG_RCU_CPU_STALL_INFO ) -printk (KERN_INFO "\tAdditional per-CPU info printed with stalls.\n" ); +pr_info ("\tAdditional per-CPU info printed with stalls.\n" );#endif #if NUM_RCU_LVL_4 !=0 -printk (KERN_INFO "\tFour-level hierarchy is enabled.\n" ); +pr_info ("\tFour-level hierarchy is enabled.\n" );#endif if (rcu_fanout_leaf !=CONFIG_RCU_FANOUT_LEAF ) -printk (KERN_INFO "\tExperimental boot-time adjustment of leaf fanout to %d.\n" , rcu_fanout_leaf); +pr_info ("\tExperimental boot-time adjustment of leaf fanout to %d.\n" , rcu_fanout_leaf);if (nr_cpu_ids !=NR_CPUS ) -printk (KERN_INFO "\tRCU restricting CPUs from NR_CPUS=%d to nr_cpu_ids=%d.\n" ,NR_CPUS , nr_cpu_ids); +pr_info ("\tRCU restricting CPUs from NR_CPUS=%d to nr_cpu_ids=%d.\n" ,NR_CPUS , nr_cpu_ids); #ifdef CONFIG_RCU_NOCB_CPU #ifndef CONFIG_RCU_NOCB_CPU_NONE if (!have_rcu_nocb_mask) {
@@ -123,7 +122,7 @@ static int rcu_preempted_readers_exp(struct rcu_node *rnp);*/ static void __init rcu_bootup_announce (void ) { -printk (KERN_INFO "Preemptible hierarchical RCU implementation.\n" ); +pr_info ("Preemptible hierarchical RCU implementation.\n" );rcu_bootup_announce_oddness (); }
@@ -490,13 +489,13 @@ static void rcu_print_detail_task_stall(struct rcu_state *rsp)static void rcu_print_task_stall_begin (struct rcu_node *rnp) { -printk (KERN_ERR "\tTasks blocked on level-%d rcu_node (CPUs %d-%d):" , +pr_err ("\tTasks blocked on level-%d rcu_node (CPUs %d-%d):" , rnp->level, rnp->grplo, rnp->grphi); }static void rcu_print_task_stall_end (void ) { -printk (KERN_CONT "\n" ); +pr_cont ("\n" ); }#else /* #ifdef CONFIG_RCU_CPU_STALL_INFO */
@@ -526,7 +525,7 @@ static int rcu_print_task_stall(struct rcu_node *rnp) t =list_entry (rnp->gp_tasks, struct task_struct, rcu_node_entry);list_for_each_entry_continue (t, &rnp->blkd_tasks, rcu_node_entry) { -printk (KERN_CONT " P%d" , t->pid); +pr_cont (" P%d" , t->pid); ndetected++; }rcu_print_task_stall_end ();
@@ -942,7 +941,7 @@ static struct rcu_state *rcu_state = &rcu_sched_state;*/ static void __init rcu_bootup_announce (void ) { -printk (KERN_INFO "Hierarchical RCU implementation.\n" ); +pr_info ("Hierarchical RCU implementation.\n" );rcu_bootup_announce_oddness (); }
@@ -1883,7 +1882,7 @@ static void print_cpu_stall_fast_no_hz(char *cp, int cpu)/* Initiate the stall-info list. */ static void print_cpu_stall_info_begin (void ) { -printk (KERN_CONT "\n" ); +pr_cont ("\n" ); }/*
@@ -1914,7 +1913,7 @@ static void print_cpu_stall_info(struct rcu_state *rsp, int cpu) ticks_value = rsp->gpnum - rdp->gpnum; }print_cpu_stall_fast_no_hz (fast_no_hz, cpu); - printk(KERN_ERR "\t%d: (%lu %s) idle=%03x/%llx/%d softirq=%u/%u %s\n", + pr_err("\t%d: (%lu %s) idle=%03x/%llx/%d softirq=%u/%u %s\n", cpu, ticks_value, ticks_title, atomic_read(&rdtp->dynticks) & 0xfff, rdtp->dynticks_nesting, rdtp->dynticks_nmi_nesting,
@@ -1925,7 +1924,7 @@ static void print_cpu_stall_info(struct rcu_state *rsp, int cpu)/* Terminate the stall-info list. */ static void print_cpu_stall_info_end (void ) { -printk (KERN_ERR "\t" ); +pr_err ("\t" ); }/* Zero ->ticks_this_gp for all flavors of RCU. */
@@ -1948,17 +1947,17 @@ static void increment_cpu_stall_ticks(void)static void print_cpu_stall_info_begin (void ) { -printk (KERN_CONT " {" ); +pr_cont (" {" ); }static void print_cpu_stall_info (struct rcu_state *rsp,int cpu) { -printk (KERN_CONT " %d" , cpu); +pr_cont (" %d" , cpu); }static void print_cpu_stall_info_end (void ) { -printk (KERN_CONT "} " ); +pr_cont ("} " ); }static void zero_cpu_stall_ticks (struct rcu_data *rdp)
Signature · 1 line
1.8.1.5
PATCH 3/6 rcu: Remove "Experimental" flags
In reply to Paul E. McKenneyFrom: "Paul E. McKenney" <[email protected]>
After a release or two, features are no longer experimental. Therefore, this commit removes the "Experimental" tag from them.
--- init/Kconfig | 2 +- kernel/rcutree_plugin.h | 10 +++++----- 2 files changed, 6 insertions(+), 6 deletions(-)
init/Kconfig +1 -1
@@ -655,7 +655,7 @@ config RCU_BOOST_DELAY Accept the default if unsure. config RCU_NOCB_CPU - bool "Offload RCU callback processing from boot-selected CPUs (EXPERIMENTAL" + bool "Offload RCU callback processing from boot-selected CPUs" depends on TREE_RCU || TREE_PREEMPT_RCU default n help
kernel/rcutree_plugin.h +5 -5
@@ -81,7 +81,7 @@ static void __init rcu_bootup_announce_oddness(void)pr_info ("\tFour-level hierarchy is enabled.\n" );#endif if (rcu_fanout_leaf !=CONFIG_RCU_FANOUT_LEAF ) -pr_info ("\tExperimental boot-time adjustment of leaf fanout to %d.\n" , rcu_fanout_leaf); +pr_info ("\tBoot-time adjustment of leaf fanout to %d.\n" , rcu_fanout_leaf);if (nr_cpu_ids !=NR_CPUS )pr_info ("\tRCU restricting CPUs from NR_CPUS=%d to nr_cpu_ids=%d.\n" ,NR_CPUS , nr_cpu_ids);#ifdef CONFIG_RCU_NOCB_CPU
@@ -91,19 +91,19 @@ static void __init rcu_bootup_announce_oddness(void) have_rcu_nocb_mask = true; }#ifdef CONFIG_RCU_NOCB_CPU_ZERO -pr_info ("\tExperimental no-CBs CPU 0\n" ); +pr_info ("\tOffload RCU callbacks from CPU 0\n" );cpumask_set_cpu (0 , rcu_nocb_mask);#endif /* #ifdef CONFIG_RCU_NOCB_CPU_ZERO */ #ifdef CONFIG_RCU_NOCB_CPU_ALL -pr_info ("\tExperimental no-CBs for all CPUs\n" ); +pr_info ("\tOffload RCU callbacks from all CPUs\n" );cpumask_setall (rcu_nocb_mask);#endif /* #ifdef CONFIG_RCU_NOCB_CPU_ALL */ #endif /* #ifndef CONFIG_RCU_NOCB_CPU_NONE */ if (have_rcu_nocb_mask) {cpulist_scnprintf (nocb_buf,sizeof (nocb_buf), rcu_nocb_mask); -pr_info ("\tExperimental no-CBs CPUs: %s.\n" , nocb_buf); +pr_info ("\tOffload RCU callbacks from CPUs: %s.\n" , nocb_buf);if (rcu_nocb_poll) -pr_info ("\tExperimental polled no-CBs CPUs.\n" ); +pr_info ("\tPoll for callbacks from no-CBs CPUs.\n" ); }#endif /* #ifdef CONFIG_RCU_NOCB_CPU */ }
Signature · 1 line
1.8.1.5
PATCH 4/6 rcu: Drive quiescent-state-forcing delay from HZ
In reply to Paul E. McKenneyFrom: "Paul E. McKenney" <[email protected]>
Systems with HZ=100 can have slow bootup times due to the default three-jiffy delays between quiescent-state forcing attempts. This commit therefore auto-tunes the RCU_JIFFIES_TILL_FORCE_QS value based on the value of HZ. However, this would break very large systems that require more time between quiescent-state forcing attempts. This commit therefore also ups the default delay by one jiffy for each 256 CPUs that might be on the system (based off of nr_cpu_ids at runtime, -not- NR_CPUS at build time).
Updated to collapse #ifdefs for RCU_JIFFIES_TILL_FORCE_QS into a step-function definition as suggested by Josh Triplett.
--- kernel/rcutree.c | 18 ++++++++++++++++-- kernel/rcutree.h | 15 ++++++++++----- 2 files changed, 26 insertions(+), 7 deletions(-)
kernel/rcutree.c +16 -2
@@ -218,8 +218,8 @@ module_param(blimit, long, 0444);module_param (qhimark, long,0444 );module_param (qlowmark, long,0444 ); -static ulong jiffies_till_first_fqs =RCU_JIFFIES_TILL_FORCE_QS ; -static ulong jiffies_till_next_fqs =RCU_JIFFIES_TILL_FORCE_QS ; +static ulong jiffies_till_first_fqs =ULONG_MAX ; +static ulong jiffies_till_next_fqs =ULONG_MAX ;module_param (jiffies_till_first_fqs, ulong,0644 );module_param (jiffies_till_next_fqs, ulong,0644 );
@@ -3252,11 +3252,25 @@ static void __init rcu_init_one(struct rcu_state *rsp,*/ static void __init rcu_init_geometry (void ) { +ulong d;int i;int j;int n = nr_cpu_ids;int rcu_capacity[MAX_RCU_LVLS +1 ]; +/* +* Initialize any unspecified boot parameters. +* The default values of jiffies_till_first_fqs and +* jiffies_till_next_fqs are set to the RCU_JIFFIES_TILL_FORCE_QS +* value, which is a function of HZ, then adding one for each +* RCU_JIFFIES_FQS_DIV CPUs that might be on the system. +*/ + d =RCU_JIFFIES_TILL_FORCE_QS + nr_cpu_ids /RCU_JIFFIES_FQS_DIV ; +if (jiffies_till_first_fqs ==ULONG_MAX ) + jiffies_till_first_fqs = d; +if (jiffies_till_next_fqs ==ULONG_MAX ) + jiffies_till_next_fqs = d; +/* If the compile-time values are accurate, just leave. */ if (rcu_fanout_leaf ==CONFIG_RCU_FANOUT_LEAF && nr_cpu_ids ==NR_CPUS )
kernel/rcutree.h +10 -5
@@ -342,12 +342,17 @@ struct rcu_data {#define RCU_FORCE_QS 3/* Need to force quiescent state. */ #define RCU_SIGNAL_INIT RCU_SAVE_DYNTICK -#define RCU_JIFFIES_TILL_FORCE_QS 3/* for rsp->jiffies_force_qs */ +#define RCU_JIFFIES_TILL_FORCE_QS (1 + (HZ > 250) + (HZ > 500)) +/* For jiffies_till_first_fqs and */ +/* and jiffies_till_next_fqs. */ -#define RCU_STALL_RAT_DELAY 2/* Allow other CPUs time */ -/* to take at least one */ -/* scheduling clock irq */ -/* before ratting on them. */ +#define RCU_JIFFIES_FQS_DIV 256/* Very large systems need more */ +/* delay between bouts of */ +/* quiescent-state forcing. */ + +#define RCU_STALL_RAT_DELAY 2/* Allow other CPUs time to take */ +/* at least one scheduling clock */ +/* irq before ratting on them. */ #define rcu_wait (cond) \ do { \
Signature · 1 line
1.8.1.5
PATCH 5/6 rcu: Merge adjacent identical ifdefs
In reply to Paul E. McKenneyFrom: "Paul E. McKenney" <[email protected]>
Two ifdefs in kernel/rcupdate.c now have identical conditions with nothing between them, so the commit merges them into a single ifdef.
--- kernel/rcupdate.c | 3 --- 1 file changed, 3 deletions(-)
kernel/rcupdate.c +0 -3
@@ -145,9 +145,6 @@ static struct lock_class_key rcu_sched_lock_key;struct lockdep_map rcu_sched_lock_map =STATIC_LOCKDEP_MAP_INIT ("rcu_read_lock_sched" , &rcu_sched_lock_key);EXPORT_SYMBOL_GPL (rcu_sched_lock_map); -#endif - -#ifdef CONFIG_DEBUG_LOCK_ALLOC int debug_lockdep_rcu_enabled (void ) {
Signature · 1 line
1.8.1.5
PATCH 6/6 rcu: Apply Dave Jones's NOCB Kconfig help feedback
In reply to Paul E. McKenneyFrom: "Paul E. McKenney" <[email protected]>
The Kconfig help text for the RCU_NOCB_CPU_NONE, RCU_NOCB_CPU_ZERO, and RCU_NOCB_CPU_ALL Kconfig options was unclear, so this commit adds a bit more detail.
--- init/Kconfig | 34 +++++++++++++++++++++++++--------- 1 file changed, 25 insertions(+), 9 deletions(-)
init/Kconfig +25 -9
@@ -681,9 +681,10 @@ choice prompt "Build-forced no-CBs CPUs" default RCU_NOCB_CPU_NONE help - This option allows no-CBs CPUs to be specified at build time. - Additional no-CBs CPUs may be specified by the rcu_nocbs= - boot parameter. + This option allows no-CBs CPUs (whose RCU callbacks are invoked + from kthreads rather than from softirq context) to be specified + at build time. Additional no-CBs CPUs may be specified by + the rcu_nocbs= boot parameter. config RCU_NOCB_CPU_NONE bool "No build_forced no-CBs CPUs"
@@ -691,25 +692,40 @@ config RCU_NOCB_CPU_NONE help This option does not force any of the CPUs to be no-CBs CPUs. Only CPUs designated by the rcu_nocbs= boot parameter will be - no-CBs CPUs. + no-CBs CPUs, whose RCU callbacks will be invoked by per-CPU + kthreads whose names begin with "rcuo". All other CPUs will + invoke their own RCU callbacks in softirq context. + + Select this option if you want to choose no-CBs CPUs at + boot time, for example, to allow testing of different no-CBs + configurations without having to rebuild the kernel each time. config RCU_NOCB_CPU_ZERO bool "CPU 0 is a build_forced no-CBs CPU" depends on RCU_NOCB_CPU && !NO_HZ_FULL help - This option forces CPU 0 to be a no-CBs CPU. Additional CPUs - may be designated as no-CBs CPUs using the rcu_nocbs= boot - parameter will be no-CBs CPUs. + This option forces CPU 0 to be a no-CBs CPU, so that its RCU + callbacks are invoked by a per-CPU kthread whose name begins + with "rcuo". Additional CPUs may be designated as no-CBs + CPUs using the rcu_nocbs= boot parameter will be no-CBs CPUs. + All other CPUs will invoke their own RCU callbacks in softirq + context. Select this if CPU 0 needs to be a no-CBs CPU for real-time - or energy-efficiency reasons. + or energy-efficiency reasons, but the real reason it exists + is to ensure that randconfig testing covers mixed systems. config RCU_NOCB_CPU_ALL bool "All CPUs are build_forced no-CBs CPUs" depends on RCU_NOCB_CPU help This option forces all CPUs to be no-CBs CPUs. The rcu_nocbs= - boot parameter will be ignored. + boot parameter will be ignored. All CPUs' RCU callbacks will + be executed in the context of per-CPU rcuo kthreads created for + this purpose. Assuming that the kthreads whose names start with + "rcuo" are bound to "housekeeping" CPUs, this reduces OS jitter + on the remaining CPUs, but might decrease memory locality during + RCU-callback invocation, thus potentially degrading throughput. Select this if all CPUs need to be no-CBs CPUs for real-time or energy-efficiency reasons.
Signature · 1 line
1.8.1.5