From e6617c6ec28a17cf2f90262b835ec05b9b861400 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Thu, 3 Sep 2009 02:35:20 -0700 Subject: sparc64: Kill spurious NMI watchdog triggers by increasing limit to 30 seconds. This is a compromise and a temporary workaround for bootup NMI watchdog triggers some people see with qla2xxx devices present. This happens when, for example: CPU 0 is in the driver init and looping submitting mailbox commands to load the firmware, then waiting for completion. CPU 1 is receiving the device interrupts. CPU 1 is where the NMI watchdog triggers. CPU 0 is submitting mailbox commands fast enough that by the time CPU 1 returns from the device interrupt handler, a new one is pending. This sequence runs for more than 5 seconds. The problematic case is CPU 1's timer interrupt running when the barrage of device interrupts begin. Then we have: timer interrupt return for softirq checking pending, thus enable interrupts qla2xxx interrupt return qla2xxx interrupt return ... 5+ seconds pass final qla2xxx interrupt for fw load return run timer softirq return At some point in the multi-second qla2xxx interrupt storm we trigger the NMI watchdog on CPU 1 from the NMI interrupt handler. The timer softirq, once we get back to running it, is smart enough to run the timer work enough times to make up for the missed timer interrupts. However, the NMI watchdogs (both x86 and sparc) use the timer interrupt count to notice the cpu is wedged. But in the above scenerio we'll receive only one such timer interrupt even if we last all the way back to running the timer softirq. The default watchdog trigger point is only 5 seconds, which is pretty low (the softwatchdog triggers at 60 seconds). So increase it to 30 seconds for now. Signed-off-by: David S. Miller --- arch/sparc/kernel/nmi.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/sparc/kernel/nmi.c') diff --git a/arch/sparc/kernel/nmi.c b/arch/sparc/kernel/nmi.c index 2c0cc72d295b..b75bf502cd42 100644 --- a/arch/sparc/kernel/nmi.c +++ b/arch/sparc/kernel/nmi.c @@ -103,7 +103,7 @@ notrace __kprobes void perfctr_irq(int irq, struct pt_regs *regs) } if (!touched && __get_cpu_var(last_irq_sum) == sum) { local_inc(&__get_cpu_var(alert_counter)); - if (local_read(&__get_cpu_var(alert_counter)) == 5 * nmi_hz) + if (local_read(&__get_cpu_var(alert_counter)) == 30 * nmi_hz) die_nmi("BUG: NMI Watchdog detected LOCKUP", regs, panic_on_timeout); } else { -- cgit v1.2.2 From a8f22264550e64c0cd11fb6647284b0bd6407f9c Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Tue, 8 Sep 2009 23:16:06 -0700 Subject: sparc64: Manage NMI watchdog enabling like x86. Use a per-cpu 'wd_enabled' boolean and a global atomic_t count of watchdog NMI enabled cpus which is set to '-1' if something is wrong with the watchdog and it can't be used. Signed-off-by: David S. Miller --- arch/sparc/kernel/nmi.c | 60 +++++++++++++++++++++++++++++++++++-------------- 1 file changed, 43 insertions(+), 17 deletions(-) (limited to 'arch/sparc/kernel/nmi.c') diff --git a/arch/sparc/kernel/nmi.c b/arch/sparc/kernel/nmi.c index 2c0cc72d295b..d1614e8384ae 100644 --- a/arch/sparc/kernel/nmi.c +++ b/arch/sparc/kernel/nmi.c @@ -34,10 +34,17 @@ static int nmi_watchdog_active; static int panic_on_timeout; -int nmi_usable; -EXPORT_SYMBOL_GPL(nmi_usable); +/* nmi_active: + * >0: the NMI watchdog is active, but can be disabled + * <0: the NMI watchdog has not been set up, and cannot be enabled + * 0: the NMI watchdog is disabled, but can be enabled + */ +atomic_t nmi_active = ATOMIC_INIT(0); /* oprofile uses this */ +EXPORT_SYMBOL(nmi_active); static unsigned int nmi_hz = HZ; +static DEFINE_PER_CPU(short, wd_enabled); +static int endflag __initdata; static DEFINE_PER_CPU(unsigned int, last_irq_sum); static DEFINE_PER_CPU(local_t, alert_counter); @@ -110,7 +117,7 @@ notrace __kprobes void perfctr_irq(int irq, struct pt_regs *regs) __get_cpu_var(last_irq_sum) = sum; local_set(&__get_cpu_var(alert_counter), 0); } - if (nmi_usable) { + if (__get_cpu_var(wd_enabled)) { write_pic(picl_value(nmi_hz)); pcr_ops->write(pcr_enable); } @@ -121,8 +128,6 @@ static inline unsigned int get_nmi_count(int cpu) return cpu_data(cpu).__nmi_count; } -static int endflag __initdata; - static __init void nmi_cpu_busy(void *data) { local_irq_enable_in_hardirq(); @@ -143,12 +148,15 @@ static void report_broken_nmi(int cpu, int *prev_nmi_count) printk(KERN_WARNING "and attach the output of the 'dmesg' command.\n"); - nmi_usable = 0; + per_cpu(wd_enabled, cpu) = 0; + atomic_dec(&nmi_active); } -static void stop_watchdog(void *unused) +static void stop_nmi_watchdog(void *unused) { pcr_ops->write(PCR_PIC_PRIV); + __get_cpu_var(wd_enabled) = 0; + atomic_dec(&nmi_active); } static int __init check_nmi_watchdog(void) @@ -156,6 +164,9 @@ static int __init check_nmi_watchdog(void) unsigned int *prev_nmi_count; int cpu, err; + if (!atomic_read(&nmi_active)) + return 0; + prev_nmi_count = kmalloc(nr_cpu_ids * sizeof(unsigned int), GFP_KERNEL); if (!prev_nmi_count) { err = -ENOMEM; @@ -172,12 +183,15 @@ static int __init check_nmi_watchdog(void) mdelay((20 * 1000) / nmi_hz); /* wait 20 ticks */ for_each_online_cpu(cpu) { + if (!per_cpu(wd_enabled, cpu)) + continue; if (get_nmi_count(cpu) - prev_nmi_count[cpu] <= 5) report_broken_nmi(cpu, prev_nmi_count); } endflag = 1; - if (!nmi_usable) { + if (!atomic_read(&nmi_active)) { kfree(prev_nmi_count); + atomic_set(&nmi_active, -1); err = -ENODEV; goto error; } @@ -188,12 +202,26 @@ static int __init check_nmi_watchdog(void) kfree(prev_nmi_count); return 0; error: - on_each_cpu(stop_watchdog, NULL, 1); + on_each_cpu(stop_nmi_watchdog, NULL, 1); return err; } -static void start_watchdog(void *unused) +static void start_nmi_watchdog(void *unused) +{ + __get_cpu_var(wd_enabled) = 1; + atomic_inc(&nmi_active); + + pcr_ops->write(PCR_PIC_PRIV); + write_pic(picl_value(nmi_hz)); + + pcr_ops->write(pcr_enable); +} + +static void nmi_adjust_hz_one(void *unused) { + if (!__get_cpu_var(wd_enabled)) + return; + pcr_ops->write(PCR_PIC_PRIV); write_pic(picl_value(nmi_hz)); @@ -203,13 +231,13 @@ static void start_watchdog(void *unused) void nmi_adjust_hz(unsigned int new_hz) { nmi_hz = new_hz; - on_each_cpu(start_watchdog, NULL, 1); + on_each_cpu(nmi_adjust_hz_one, NULL, 1); } EXPORT_SYMBOL_GPL(nmi_adjust_hz); static int nmi_shutdown(struct notifier_block *nb, unsigned long cmd, void *p) { - on_each_cpu(stop_watchdog, NULL, 1); + on_each_cpu(stop_nmi_watchdog, NULL, 1); return 0; } @@ -221,16 +249,14 @@ int __init nmi_init(void) { int err; - nmi_usable = 1; - - on_each_cpu(start_watchdog, NULL, 1); + on_each_cpu(start_nmi_watchdog, NULL, 1); err = check_nmi_watchdog(); if (!err) { err = register_reboot_notifier(&nmi_reboot_notifier); if (err) { - nmi_usable = 0; - on_each_cpu(stop_watchdog, NULL, 1); + on_each_cpu(stop_nmi_watchdog, NULL, 1); + atomic_set(&nmi_active, -1); } } return err; -- cgit v1.2.2 From d89be56b2153ccad7e8cd9e71b5d8b686632e8c2 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Tue, 8 Sep 2009 23:29:16 -0700 Subject: sparc64: Make touch_nmi_watchdog() actually work. It guards it's actions on nmi_watchdog_active, but nothing ever sets that and it's initial value is zero. Signed-off-by: David S. Miller --- arch/sparc/kernel/nmi.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'arch/sparc/kernel/nmi.c') diff --git a/arch/sparc/kernel/nmi.c b/arch/sparc/kernel/nmi.c index d1614e8384ae..47a465fe3c38 100644 --- a/arch/sparc/kernel/nmi.c +++ b/arch/sparc/kernel/nmi.c @@ -31,7 +31,6 @@ * level 14 as our IRQ off level. */ -static int nmi_watchdog_active; static int panic_on_timeout; /* nmi_active: @@ -52,7 +51,7 @@ static DEFINE_PER_CPU(int, nmi_touch); void touch_nmi_watchdog(void) { - if (nmi_watchdog_active) { + if (atomic_read(&nmi_active)) { int cpu; for_each_present_cpu(cpu) { -- cgit v1.2.2 From 2d0740c4562493b60f59ca9b0330a2d5e01d43ec Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Thu, 10 Sep 2009 05:56:16 -0700 Subject: sparc64: Use nmi_enter() and nmi_exit(), as needed. Signed-off-by: David S. Miller --- arch/sparc/kernel/nmi.c | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'arch/sparc/kernel/nmi.c') diff --git a/arch/sparc/kernel/nmi.c b/arch/sparc/kernel/nmi.c index 47a465fe3c38..7de19dd30f40 100644 --- a/arch/sparc/kernel/nmi.c +++ b/arch/sparc/kernel/nmi.c @@ -84,6 +84,7 @@ static void die_nmi(const char *str, struct pt_regs *regs, int do_panic) if (do_panic || panic_on_oops) panic("Non maskable interrupt"); + nmi_exit(); local_irq_enable(); do_exit(SIGBUS); } @@ -98,6 +99,8 @@ notrace __kprobes void perfctr_irq(int irq, struct pt_regs *regs) local_cpu_data().__nmi_count++; + nmi_enter(); + if (notify_die(DIE_NMI, "nmi", regs, 0, pt_regs_trap_type(regs), SIGINT) == NOTIFY_STOP) touched = 1; @@ -120,6 +123,8 @@ notrace __kprobes void perfctr_irq(int irq, struct pt_regs *regs) write_pic(picl_value(nmi_hz)); pcr_ops->write(pcr_enable); } + + nmi_exit(); } static inline unsigned int get_nmi_count(int cpu) -- cgit v1.2.2 From 59abbd1e7cfd6018fb8e58a96aa562aaff8711e7 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Thu, 10 Sep 2009 06:28:20 -0700 Subject: sparc64: Initial hw perf counter support. Only supports one simple counter and only UltraSPARC-IIIi chips. Signed-off-by: David S. Miller --- arch/sparc/kernel/nmi.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) (limited to 'arch/sparc/kernel/nmi.c') diff --git a/arch/sparc/kernel/nmi.c b/arch/sparc/kernel/nmi.c index 7de19dd30f40..391a6ed9a184 100644 --- a/arch/sparc/kernel/nmi.c +++ b/arch/sparc/kernel/nmi.c @@ -19,6 +19,7 @@ #include #include +#include #include #include #include @@ -156,7 +157,7 @@ static void report_broken_nmi(int cpu, int *prev_nmi_count) atomic_dec(&nmi_active); } -static void stop_nmi_watchdog(void *unused) +void stop_nmi_watchdog(void *unused) { pcr_ops->write(PCR_PIC_PRIV); __get_cpu_var(wd_enabled) = 0; @@ -210,7 +211,7 @@ error: return err; } -static void start_nmi_watchdog(void *unused) +void start_nmi_watchdog(void *unused) { __get_cpu_var(wd_enabled) = 1; atomic_inc(&nmi_active); @@ -263,6 +264,9 @@ int __init nmi_init(void) atomic_set(&nmi_active, -1); } } + if (!err) + init_hw_perf_counters(); + return err; } -- cgit v1.2.2 From cdd6c482c9ff9c55475ee7392ec8f672eddb7be6 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Mon, 21 Sep 2009 12:02:48 +0200 Subject: perf: Do the big rename: Performance Counters -> Performance Events Bye-bye Performance Counters, welcome Performance Events! In the past few months the perfcounters subsystem has grown out its initial role of counting hardware events, and has become (and is becoming) a much broader generic event enumeration, reporting, logging, monitoring, analysis facility. Naming its core object 'perf_counter' and naming the subsystem 'perfcounters' has become more and more of a misnomer. With pending code like hw-breakpoints support the 'counter' name is less and less appropriate. All in one, we've decided to rename the subsystem to 'performance events' and to propagate this rename through all fields, variables and API names. (in an ABI compatible fashion) The word 'event' is also a bit shorter than 'counter' - which makes it slightly more convenient to write/handle as well. Thanks goes to Stephane Eranian who first observed this misnomer and suggested a rename. User-space tooling and ABI compatibility is not affected - this patch should be function-invariant. (Also, defconfigs were not touched to keep the size down.) This patch has been generated via the following script: FILES=$(find * -type f | grep -vE 'oprofile|[^K]config') sed -i \ -e 's/PERF_EVENT_/PERF_RECORD_/g' \ -e 's/PERF_COUNTER/PERF_EVENT/g' \ -e 's/perf_counter/perf_event/g' \ -e 's/nb_counters/nb_events/g' \ -e 's/swcounter/swevent/g' \ -e 's/tpcounter_event/tp_event/g' \ $FILES for N in $(find . -name perf_counter.[ch]); do M=$(echo $N | sed 's/perf_counter/perf_event/g') mv $N $M done FILES=$(find . -name perf_event.*) sed -i \ -e 's/COUNTER_MASK/REG_MASK/g' \ -e 's/COUNTER/EVENT/g' \ -e 's/\/event_id/g' \ -e 's/counter/event/g' \ -e 's/Counter/Event/g' \ $FILES ... to keep it as correct as possible. This script can also be used by anyone who has pending perfcounters patches - it converts a Linux kernel tree over to the new naming. We tried to time this change to the point in time where the amount of pending patches is the smallest: the end of the merge window. Namespace clashes were fixed up in a preparatory patch - and some stylistic fallout will be fixed up in a subsequent patch. ( NOTE: 'counters' are still the proper terminology when we deal with hardware registers - and these sed scripts are a bit over-eager in renaming them. I've undone some of that, but in case there's something left where 'counter' would be better than 'event' we can undo that on an individual basis instead of touching an otherwise nicely automated patch. ) Suggested-by: Stephane Eranian Acked-by: Peter Zijlstra Acked-by: Paul Mackerras Reviewed-by: Arjan van de Ven Cc: Mike Galbraith Cc: Arnaldo Carvalho de Melo Cc: Frederic Weisbecker Cc: Steven Rostedt Cc: Benjamin Herrenschmidt Cc: David Howells Cc: Kyle McMartin Cc: Martin Schwidefsky Cc: "David S. Miller" Cc: Thomas Gleixner Cc: "H. Peter Anvin" Cc: LKML-Reference: Signed-off-by: Ingo Molnar --- arch/sparc/kernel/nmi.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch/sparc/kernel/nmi.c') diff --git a/arch/sparc/kernel/nmi.c b/arch/sparc/kernel/nmi.c index 378eb53e0776..b129611590a4 100644 --- a/arch/sparc/kernel/nmi.c +++ b/arch/sparc/kernel/nmi.c @@ -19,7 +19,7 @@ #include #include -#include +#include #include #include #include @@ -265,7 +265,7 @@ int __init nmi_init(void) } } if (!err) - init_hw_perf_counters(); + init_hw_perf_events(); return err; } -- cgit v1.2.2