From 34ed62461ae4970695974afb9a60ac3df0086830 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Mon, 7 Jan 2013 13:37:42 -0800 Subject: rcu: Remove restrictions on no-CBs CPUs Currently, CPU 0 is constrained to not be a no-CBs CPU, and furthermore at least one no-CBs CPU must remain online at any given time. These restrictions are problematic in some situations, such as cases where all CPUs must run a real-time workload that needs to be insulated from OS jitter and latencies due to RCU callback invocation. This commit therefore provides no-CBs CPUs a (very crude and energy-inefficient) way to start and to wait for grace periods independently of the normal RCU callback mechanisms. This approach allows any or all of the CPUs to be designated as no-CBs CPUs, and allows any proper subset of the CPUs (whether no-CBs CPUs or not) to be offlined. This commit also provides a fix for a locking bug spotted by Xie ChanglongX . Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney --- init/Kconfig | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'init') diff --git a/init/Kconfig b/init/Kconfig index 22616cd..c8bd349 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -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" + bool "Offload RCU callback processing from boot-selected CPUs (EXPERIMENTAL" depends on TREE_RCU || TREE_PREEMPT_RCU default n help @@ -673,7 +673,7 @@ config RCU_NOCB_CPU callback, and (2) affinity or cgroups can be used to force the kthreads to run on whatever set of CPUs is desired. - Say Y here if you want reduced OS jitter on selected CPUs. + Say Y here if you want to help to debug reduced OS jitter. Say N here if you are unsure. endmenu # "RCU Subsystem" -- cgit v1.1 From 3d374d09f16f64ab4d71704cbe621514d36cd0b1 Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Wed, 6 Mar 2013 12:35:31 -0800 Subject: final removal of CONFIG_EXPERIMENTAL Remove "config EXPERIMENTAL" itself, now that every "depends on" it has been removed from the tree. Signed-off-by: Kees Cook Signed-off-by: Greg Kroah-Hartman --- init/Kconfig | 4 ---- 1 file changed, 4 deletions(-) (limited to 'init') diff --git a/init/Kconfig b/init/Kconfig index 22616cd..5341d72 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -28,10 +28,6 @@ config BUILDTIME_EXTABLE_SORT menu "General setup" -config EXPERIMENTAL - bool - default y - config BROKEN bool -- cgit v1.1 From 911af505ef407c2511106c224dd640f882f0f590 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Mon, 11 Feb 2013 10:23:27 -0800 Subject: rcu: Provide compile-time control for no-CBs CPUs Currently, the only way to specify no-CBs CPUs is via the rcu_nocbs kernel command-line parameter. This is inconvenient in some cases, particularly for randconfig testing, so this commit adds a new set of kernel configuration parameters. CONFIG_RCU_NOCB_CPU_NONE (the default) retains the old behavior, CONFIG_RCU_NOCB_CPU_ZERO offloads callback processing from CPU 0 (along with any other CPUs specified by the rcu_nocbs boot-time parameter), and CONFIG_RCU_NOCB_CPU_ALL offloads callback processing from all CPUs. Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney --- init/Kconfig | 39 +++++++++++++++++++++++++++++++++++++++ 1 file changed, 39 insertions(+) (limited to 'init') diff --git a/init/Kconfig b/init/Kconfig index c8bd349..3074997 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -676,6 +676,45 @@ config RCU_NOCB_CPU Say Y here if you want to help to debug reduced OS jitter. Say N here if you are unsure. +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. + +config RCU_NOCB_CPU_NONE + bool "No build_forced no-CBs CPUs" + depends on RCU_NOCB_CPU + 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. + +config RCU_NOCB_CPU_ZERO + bool "CPU 0 is a build_forced no-CBs CPU" + depends on RCU_NOCB_CPU + 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. + + Select this if CPU 0 needs to be a no-CBs CPU for real-time + or energy-efficiency reasons. + +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. + + Select this if all CPUs need to be no-CBs CPUs for real-time + or energy-efficiency reasons. + +endchoice + endmenu # "RCU Subsystem" config IKCONFIG -- cgit v1.1 From a488985851cf2facd2227bd982cc2c251df56268 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Mon, 3 Dec 2012 08:16:28 -0800 Subject: rcu: Distinguish "rcuo" kthreads by RCU flavor Currently, the per-no-CBs-CPU kthreads are named "rcuo" followed by the CPU number, for example, "rcuo". This is problematic given that there are either two or three RCU flavors, each of which gets a per-CPU kthread with exactly the same name. This commit therefore introduces a one-letter abbreviation for each RCU flavor, namely 'b' for RCU-bh, 'p' for RCU-preempt, and 's' for RCU-sched. This abbreviation is used to distinguish the "rcuo" kthreads, for example, for CPU 0 we would have "rcuob/0", "rcuop/0", and "rcuos/0". Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney Tested-by: Dietmar Eggemann --- init/Kconfig | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) (limited to 'init') diff --git a/init/Kconfig b/init/Kconfig index 3074997..7175840 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -666,12 +666,13 @@ config RCU_NOCB_CPU This option offloads callback invocation from the set of CPUs specified at boot time by the rcu_nocbs parameter. - For each such CPU, a kthread ("rcuoN") will be created to - invoke callbacks, where the "N" is the CPU being offloaded. - Nothing prevents this kthread from running on the specified - CPUs, but (1) the kthreads may be preempted between each - callback, and (2) affinity or cgroups can be used to force - the kthreads to run on whatever set of CPUs is desired. + For each such CPU, a kthread ("rcuox/N") will be created to + invoke callbacks, where the "N" is the CPU being offloaded, + and where the "x" is "b" for RCU-bh, "p" for RCU-preempt, and + "s" for RCU-sched. Nothing prevents this kthread from running + on the specified CPUs, but (1) the kthreads may be preempted + between each callback, and (2) affinity or cgroups can be used + to force the kthreads to run on whatever set of CPUs is desired. Say Y here if you want to help to debug reduced OS jitter. Say N here if you are unsure. -- cgit v1.1 From c0f4dfd4f90f1667d234d21f15153ea09a2eaa66 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Fri, 28 Dec 2012 11:30:36 -0800 Subject: rcu: Make RCU_FAST_NO_HZ take advantage of numbered callbacks Because RCU callbacks are now associated with the number of the grace period that they must wait for, CPUs can now take advance callbacks corresponding to grace periods that ended while a given CPU was in dyntick-idle mode. This eliminates the need to try forcing the RCU state machine while entering idle, thus reducing the CPU intensiveness of RCU_FAST_NO_HZ, which should increase its energy efficiency. Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney --- init/Kconfig | 17 ++++++++++------- 1 file changed, 10 insertions(+), 7 deletions(-) (limited to 'init') diff --git a/init/Kconfig b/init/Kconfig index 7175840..a3a2304 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -582,13 +582,16 @@ config RCU_FAST_NO_HZ depends on NO_HZ && SMP default n help - This option causes RCU to attempt to accelerate grace periods in - order to allow CPUs to enter dynticks-idle state more quickly. - On the other hand, this option increases the overhead of the - dynticks-idle checking, thus degrading scheduling latency. - - Say Y if energy efficiency is critically important, and you don't - care about real-time response. + This option permits CPUs to enter dynticks-idle state even if + they have RCU callbacks queued, and prevents RCU from waking + these CPUs up more than roughly once every four jiffies (by + default, you can adjust this using the rcutree.rcu_idle_gp_delay + parameter), thus improving energy efficiency. On the other + hand, this option increases the duration of RCU grace periods, + for example, slowing down synchronize_rcu(). + + Say Y if energy efficiency is critically important, and you + don't care about increased grace-period durations. Say N if you are unsure. -- cgit v1.1 From a1a04ec3c7c27a682473fd9beb2c996316a64649 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Thu, 21 Mar 2013 22:49:34 +0100 Subject: idle: Provide a generic entry point for the idle code For now this calls cpu_idle(), but in the long run we want to move the cpu bringup code to the core and therefor we add a state argument. Signed-off-by: Thomas Gleixner Cc: Linus Torvalds Cc: Rusty Russell Cc: Paul McKenney Cc: Peter Zijlstra Reviewed-by: Cc: Srivatsa S. Bhat Cc: Magnus Damm Link: http://lkml.kernel.org/r/20130321215233.583190032@linutronix.de Signed-off-by: Thomas Gleixner --- init/main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'init') diff --git a/init/main.c b/init/main.c index 63534a1..adb179d 100644 --- a/init/main.c +++ b/init/main.c @@ -384,7 +384,7 @@ static noinline void __init_refok rest_init(void) init_idle_bootup_task(current); schedule_preempt_disabled(); /* Call into cpu_idle with preempt disabled */ - cpu_idle(); + cpu_startup_entry(CPUHP_ONLINE); } /* Check for early params. */ -- cgit v1.1 From f91eb62f71b31e69e405663ff8d047bc3b9f7525 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Mon, 29 Apr 2013 16:18:18 -0700 Subject: init: scream bloody murder if interrupts are enabled too early As I was testing a lot of my code recently, and having several "successes", I accidentally noticed in the dmesg this little line: start_kernel(): bug: interrupts were enabled *very* early, fixing it Sure enough, one of my patches two commits ago enabled interrupts early. The sad part here is that I never noticed it, and I ran several tests with ktest too, and ktest did not notice this line. What ktest looks for (and so does many other automated testing scripts) is a back trace produced by a WARN_ON() or BUG(). As a back trace was never produced, my buggy patch could have slipped into linux-next, or even worse, mainline. Adding a WARN(!irqs_disabled()) makes this bug a little more obvious: PID hash table entries: 4096 (order: 3, 32768 bytes) __ex_table already sorted, skipping sort Checking aperture... No AGP bridge found Calgary: detecting Calgary via BIOS EBDA area Calgary: Unable to locate Rio Grande table in EBDA - bailing! Memory: 2003252k/2054848k available (4857k kernel code, 460k absent, 51136k reserved, 6210k data, 1096k init) ------------[ cut here ]------------ WARNING: at /home/rostedt/work/git/linux-trace.git/init/main.c:543 start_kernel+0x21e/0x415() Hardware name: To Be Filled By O.E.M. Interrupts were enabled *very* early, fixing it Modules linked in: Pid: 0, comm: swapper/0 Not tainted 3.8.0-test+ #286 Call Trace: warn_slowpath_common+0x83/0x9b warn_slowpath_fmt+0x46/0x48 start_kernel+0x21e/0x415 x86_64_start_reservations+0x10e/0x112 x86_64_start_kernel+0x102/0x111 ---[ end trace 007d8b0491b4f5d8 ]--- Preemptible hierarchical RCU implementation. RCU restricting CPUs from NR_CPUS=8 to nr_cpu_ids=4. NR_IRQS:4352 nr_irqs:712 16 Console: colour VGA+ 80x25 console [ttyS0] enabled, bootconsole disabled Do you see it? The original version of this patch just slapped a WARN_ON() in there and kept the printk(). Ard van Breemen suggested using the WARN() interface, which makes the code a bit cleaner. Also, while examining other warnings in init/main.c, I found two other locations that deserve a bloody murder scream if their conditions are hit, and updated them accordingly. Signed-off-by: Steven Rostedt Cc: Ard van Breemen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- init/main.c | 13 +++---------- 1 file changed, 3 insertions(+), 10 deletions(-) (limited to 'init') diff --git a/init/main.c b/init/main.c index 63534a1..26cd398 100644 --- a/init/main.c +++ b/init/main.c @@ -539,11 +539,8 @@ asmlinkage void __init start_kernel(void) * fragile until we cpu_idle() for the first time. */ preempt_disable(); - if (!irqs_disabled()) { - printk(KERN_WARNING "start_kernel(): bug: interrupts were " - "enabled *very* early, fixing it\n"); + if (WARN(!irqs_disabled(), "Interrupts were enabled *very* early, fixing it\n")) local_irq_disable(); - } idr_init_cache(); perf_event_init(); rcu_init(); @@ -558,9 +555,7 @@ asmlinkage void __init start_kernel(void) time_init(); profile_init(); call_function_init(); - if (!irqs_disabled()) - printk(KERN_CRIT "start_kernel(): bug: interrupts were " - "enabled early\n"); + WARN(!irqs_disabled(), "Interrupts were enabled early\n"); early_boot_irqs_disabled = false; local_irq_enable(); @@ -702,9 +697,7 @@ int __init_or_module do_one_initcall(initcall_t fn) strlcat(msgbuf, "disabled interrupts ", sizeof(msgbuf)); local_irq_enable(); } - if (msgbuf[0]) { - printk("initcall %pF returned with %s\n", fn, msgbuf); - } + WARN(msgbuf[0], "initcall %pF returned with %s\n", fn, msgbuf); return ret; } -- cgit v1.1 From c2409b004ac4757ac5121851f8a58e0bcbcf7a3c Mon Sep 17 00:00:00 2001 From: Richard Weinberger Date: Mon, 29 Apr 2013 16:18:19 -0700 Subject: init: raise log level If the kernel was booted with the "quiet" boot option we have currently no chance to see why an initrd fails. Change KERN_WARNING to KERN_ERR to see what is going on. Signed-off-by: Richard Weinberger Cc: "H. Peter Anvin" Cc: Rusty Russell Cc: Jim Cromie Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- init/main.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'init') diff --git a/init/main.c b/init/main.c index 26cd398..e1c87e1 100644 --- a/init/main.c +++ b/init/main.c @@ -825,7 +825,7 @@ static int __ref kernel_init(void *unused) if (ramdisk_execute_command) { if (!run_init_process(ramdisk_execute_command)) return 0; - printk(KERN_WARNING "Failed to execute %s\n", + printk(KERN_ERR "Failed to execute %s\n", ramdisk_execute_command); } @@ -838,7 +838,7 @@ static int __ref kernel_init(void *unused) if (execute_command) { if (!run_init_process(execute_command)) return 0; - printk(KERN_WARNING "Failed to execute %s. Attempting " + printk(KERN_ERR "Failed to execute %s. Attempting " "defaults...\n", execute_command); } if (!run_init_process("/sbin/init") || @@ -884,7 +884,7 @@ static noinline void __init kernel_init_freeable(void) /* Open the /dev/console on the rootfs, this should never fail */ if (sys_open((const char __user *) "/dev/console", O_RDWR, 0) < 0) - printk(KERN_WARNING "Warning: unable to open an initial console.\n"); + printk(KERN_ERR "Warning: unable to open an initial console.\n"); (void) sys_dup(0); (void) sys_dup(0); -- cgit v1.1 From ea676e846a8171b8e215627259f485a4e70328cb Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Mon, 29 Apr 2013 16:18:20 -0700 Subject: init/main.c: convert to pr_foo() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Also enables cleanup of some 80-col trickery. Cc: Richard Weinberger Cc: Uwe Kleine-König Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- init/main.c | 31 +++++++++++++++---------------- 1 file changed, 15 insertions(+), 16 deletions(-) (limited to 'init') diff --git a/init/main.c b/init/main.c index e1c87e1..ea73e1e 100644 --- a/init/main.c +++ b/init/main.c @@ -9,6 +9,8 @@ * Simplified starting of init: Michael A. Griffith */ +#define DEBUG /* Enable initcall_debug */ + #include #include #include @@ -174,8 +176,8 @@ static int __init obsolete_checksetup(char *line) if (line[n] == '\0' || line[n] == '=') had_early_param = 1; } else if (!p->setup_func) { - printk(KERN_WARNING "Parameter %s is obsolete," - " ignored\n", p->str); + pr_warn("Parameter %s is obsolete, ignored\n", + p->str); return 1; } else if (p->setup_func(line + n)) return 1; @@ -398,8 +400,7 @@ static int __init do_early_param(char *param, char *val, const char *unused) strcmp(p->str, "earlycon") == 0) ) { if (p->setup_func(val) != 0) - printk(KERN_WARNING - "Malformed early option '%s'\n", param); + pr_warn("Malformed early option '%s'\n", param); } } /* We accept everything at this stage. */ @@ -497,7 +498,7 @@ asmlinkage void __init start_kernel(void) tick_init(); boot_cpu_init(); page_address_init(); - printk(KERN_NOTICE "%s", linux_banner); + pr_notice("%s", linux_banner); setup_arch(&command_line); mm_init_owner(&init_mm, &init_task); mm_init_cpumask(&init_mm); @@ -509,7 +510,7 @@ asmlinkage void __init start_kernel(void) build_all_zonelists(NULL, NULL); page_alloc_init(); - printk(KERN_NOTICE "Kernel command line: %s\n", boot_command_line); + pr_notice("Kernel command line: %s\n", boot_command_line); parse_early_param(); parse_args("Booting kernel", static_command_line, __start___param, __stop___param - __start___param, @@ -582,8 +583,7 @@ asmlinkage void __init start_kernel(void) #ifdef CONFIG_BLK_DEV_INITRD if (initrd_start && !initrd_below_start_ok && page_to_pfn(virt_to_page((void *)initrd_start)) < min_low_pfn) { - printk(KERN_CRIT "initrd overwritten (0x%08lx < 0x%08lx) - " - "disabling it.\n", + pr_crit("initrd overwritten (0x%08lx < 0x%08lx) - disabling it.\n", page_to_pfn(virt_to_page((void *)initrd_start)), min_low_pfn); initrd_start = 0; @@ -662,14 +662,14 @@ static int __init_or_module do_one_initcall_debug(initcall_t fn) unsigned long long duration; int ret; - printk(KERN_DEBUG "calling %pF @ %i\n", fn, task_pid_nr(current)); + pr_debug("calling %pF @ %i\n", fn, task_pid_nr(current)); calltime = ktime_get(); ret = fn(); rettime = ktime_get(); delta = ktime_sub(rettime, calltime); duration = (unsigned long long) ktime_to_ns(delta) >> 10; - printk(KERN_DEBUG "initcall %pF returned %d after %lld usecs\n", fn, - ret, duration); + pr_debug("initcall %pF returned %d after %lld usecs\n", + fn, ret, duration); return ret; } @@ -825,8 +825,7 @@ static int __ref kernel_init(void *unused) if (ramdisk_execute_command) { if (!run_init_process(ramdisk_execute_command)) return 0; - printk(KERN_ERR "Failed to execute %s\n", - ramdisk_execute_command); + pr_err("Failed to execute %s\n", ramdisk_execute_command); } /* @@ -838,8 +837,8 @@ static int __ref kernel_init(void *unused) if (execute_command) { if (!run_init_process(execute_command)) return 0; - printk(KERN_ERR "Failed to execute %s. Attempting " - "defaults...\n", execute_command); + pr_err("Failed to execute %s. Attempting defaults...\n", + execute_command); } if (!run_init_process("/sbin/init") || !run_init_process("/etc/init") || @@ -884,7 +883,7 @@ static noinline void __init kernel_init_freeable(void) /* Open the /dev/console on the rootfs, this should never fail */ if (sys_open((const char __user *) "/dev/console", O_RDWR, 0) < 0) - printk(KERN_ERR "Warning: unable to open an initial console.\n"); + pr_err("Warning: unable to open an initial console.\n"); (void) sys_dup(0); (void) sys_dup(0); -- cgit v1.1