From e8e813ed268d90c1377f53460527c419eb52c67a Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Tue, 4 Jun 2013 14:21:17 +1000 Subject: powerpc: Rename PMU interrupts from CNT to PMI Back in commit 89713ed "Add timer, performance monitor and machine check counts to /proc/interrupts" we added a count of PMU interrupts to the output of /proc/interrupts. At the time we named them "CNT" to match x86. However in commit 89ccf46 "Rename 'performance counter interrupt'", the x86 guys renamed theirs from "CNT" to "PMI". Arguably changing the name could break someone's script, but I think the chance of that is minimal, and it's preferable to have a name that 1) is somewhat meaningful, and 2) matches x86. Signed-off-by: Michael Ellerman Signed-off-by: Benjamin Herrenschmidt diff --git a/arch/powerpc/kernel/irq.c b/arch/powerpc/kernel/irq.c index 2e51cde..c69440c 100644 --- a/arch/powerpc/kernel/irq.c +++ b/arch/powerpc/kernel/irq.c @@ -362,7 +362,7 @@ int arch_show_interrupts(struct seq_file *p, int prec) seq_printf(p, "%10u ", per_cpu(irq_stat, j).spurious_irqs); seq_printf(p, " Spurious interrupts\n"); - seq_printf(p, "%*s: ", prec, "CNT"); + seq_printf(p, "%*s: ", prec, "PMI"); for_each_online_cpu(j) seq_printf(p, "%10u ", per_cpu(irq_stat, j).pmu_irqs); seq_printf(p, " Performance monitoring interrupts\n"); -- cgit v0.10.2 From 8d7c55d01e4648605fd0dacc82d8d3989ead4db7 Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Tue, 23 Jul 2013 18:07:45 +1000 Subject: powerpc/perf: Export PERF_EVENT_CONFIG_EBB_SHIFT to userspace We use bit 63 of the event code for userspace to request that the event be counted using EBB (Event Based Branches). Export this value, making it part of the API - though only on processors that support EBB. Signed-off-by: Michael Ellerman Signed-off-by: Benjamin Herrenschmidt diff --git a/arch/powerpc/include/asm/perf_event_server.h b/arch/powerpc/include/asm/perf_event_server.h index 2dd7bfc..8b24926 100644 --- a/arch/powerpc/include/asm/perf_event_server.h +++ b/arch/powerpc/include/asm/perf_event_server.h @@ -12,6 +12,7 @@ #include #include #include +#include #define MAX_HWEVENTS 8 #define MAX_EVENT_ALTERNATIVES 8 @@ -69,11 +70,6 @@ struct power_pmu { #define PPMU_LIMITED_PMC_REQD 2 /* have to put this on a limited PMC */ #define PPMU_ONLY_COUNT_RUN 4 /* only counting in run state */ -/* - * We use the event config bit 63 as a flag to request EBB. - */ -#define EVENT_CONFIG_EBB_SHIFT 63 - extern int register_power_pmu(struct power_pmu *); struct pt_regs; diff --git a/arch/powerpc/include/uapi/asm/Kbuild b/arch/powerpc/include/uapi/asm/Kbuild index 5182c86..48be855 100644 --- a/arch/powerpc/include/uapi/asm/Kbuild +++ b/arch/powerpc/include/uapi/asm/Kbuild @@ -20,6 +20,7 @@ header-y += mman.h header-y += msgbuf.h header-y += nvram.h header-y += param.h +header-y += perf_event.h header-y += poll.h header-y += posix_types.h header-y += ps3fb.h diff --git a/arch/powerpc/include/uapi/asm/perf_event.h b/arch/powerpc/include/uapi/asm/perf_event.h new file mode 100644 index 0000000..80a4d40 --- /dev/null +++ b/arch/powerpc/include/uapi/asm/perf_event.h @@ -0,0 +1,18 @@ +/* + * Copyright 2013 Michael Ellerman, IBM Corp. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; version 2 of the + * License. + */ + +#ifndef _UAPI_ASM_POWERPC_PERF_EVENT_H +#define _UAPI_ASM_POWERPC_PERF_EVENT_H + +/* + * We use bit 63 of perf_event_attr.config as a flag to request EBB. + */ +#define PERF_EVENT_CONFIG_EBB_SHIFT 63 + +#endif /* _UAPI_ASM_POWERPC_PERF_EVENT_H */ diff --git a/arch/powerpc/perf/core-book3s.c b/arch/powerpc/perf/core-book3s.c index 24a45f9..eeae308 100644 --- a/arch/powerpc/perf/core-book3s.c +++ b/arch/powerpc/perf/core-book3s.c @@ -484,7 +484,7 @@ static bool is_ebb_event(struct perf_event *event) * use bit 63 of the event code for something else if they wish. */ return (ppmu->flags & PPMU_EBB) && - ((event->attr.config >> EVENT_CONFIG_EBB_SHIFT) & 1); + ((event->attr.config >> PERF_EVENT_CONFIG_EBB_SHIFT) & 1); } static int ebb_event_check(struct perf_event *event) diff --git a/arch/powerpc/perf/power8-pmu.c b/arch/powerpc/perf/power8-pmu.c index 7466374..2ee4a70 100644 --- a/arch/powerpc/perf/power8-pmu.c +++ b/arch/powerpc/perf/power8-pmu.c @@ -118,7 +118,7 @@ (EVENT_UNIT_MASK << EVENT_UNIT_SHIFT) | \ (EVENT_COMBINE_MASK << EVENT_COMBINE_SHIFT) | \ (EVENT_MARKED_MASK << EVENT_MARKED_SHIFT) | \ - (EVENT_EBB_MASK << EVENT_CONFIG_EBB_SHIFT) | \ + (EVENT_EBB_MASK << PERF_EVENT_CONFIG_EBB_SHIFT) | \ EVENT_PSEL_MASK) /* MMCRA IFM bits - POWER8 */ @@ -233,10 +233,10 @@ static int power8_get_constraint(u64 event, unsigned long *maskp, unsigned long pmc = (event >> EVENT_PMC_SHIFT) & EVENT_PMC_MASK; unit = (event >> EVENT_UNIT_SHIFT) & EVENT_UNIT_MASK; cache = (event >> EVENT_CACHE_SEL_SHIFT) & EVENT_CACHE_SEL_MASK; - ebb = (event >> EVENT_CONFIG_EBB_SHIFT) & EVENT_EBB_MASK; + ebb = (event >> PERF_EVENT_CONFIG_EBB_SHIFT) & EVENT_EBB_MASK; /* Clear the EBB bit in the event, so event checks work below */ - event &= ~(EVENT_EBB_MASK << EVENT_CONFIG_EBB_SHIFT); + event &= ~(EVENT_EBB_MASK << PERF_EVENT_CONFIG_EBB_SHIFT); if (pmc) { if (pmc > 6) -- cgit v0.10.2 From 3be7db6ab45b21345386d1a466da133b19cde5e4 Mon Sep 17 00:00:00 2001 From: Robert Jennings Date: Wed, 24 Jul 2013 20:13:21 -0500 Subject: powerpc: VPHN topology change updates all siblings When an associativity level change is found for one thread, the siblings threads need to be updated as well. This is done today for PRRN in stage_topology_update() but is missing for VPHN in update_cpu_associativity_changes_mask(). This patch will correctly update all thread siblings during a topology change. Without this patch a topology update can result in a CPU in init_sched_groups_power() getting stuck indefinitely in a loop. This loop is built in build_sched_groups(). As a result of the thread moving to a node separate from its siblings the struct sched_group will have its next pointer set to point to itself rather than the sched_group struct of the next thread. This happens because we have a domain without the SD_OVERLAP flag, which is correct, and a topology that doesn't conform with reality (threads on the same core assigned to different numa nodes). When this list is traversed by init_sched_groups_power() it will reach the thread's sched_group structure and loop indefinitely; the cpu will be stuck at this point. The bug was exposed when VPHN was enabled in commit b7abef0 (v3.9). Cc: [v3.9+] Reported-by: Jan Stancek Signed-off-by: Robert Jennings Signed-off-by: Benjamin Herrenschmidt diff --git a/arch/powerpc/include/asm/smp.h b/arch/powerpc/include/asm/smp.h index ffbaabe..48cfc85 100644 --- a/arch/powerpc/include/asm/smp.h +++ b/arch/powerpc/include/asm/smp.h @@ -145,6 +145,10 @@ extern void __cpu_die(unsigned int cpu); #define smp_setup_cpu_maps() static inline void inhibit_secondary_onlining(void) {} static inline void uninhibit_secondary_onlining(void) {} +static inline const struct cpumask *cpu_sibling_mask(int cpu) +{ + return cpumask_of(cpu); +} #endif /* CONFIG_SMP */ diff --git a/arch/powerpc/mm/numa.c b/arch/powerpc/mm/numa.c index 0839721..5850798 100644 --- a/arch/powerpc/mm/numa.c +++ b/arch/powerpc/mm/numa.c @@ -27,6 +27,7 @@ #include #include #include +#include #include #include #include @@ -1318,7 +1319,8 @@ static int update_cpu_associativity_changes_mask(void) } } if (changed) { - cpumask_set_cpu(cpu, changes); + cpumask_or(changes, changes, cpu_sibling_mask(cpu)); + cpu = cpu_last_thread_sibling(cpu); } } @@ -1426,7 +1428,7 @@ static int update_cpu_topology(void *data) if (!data) return -EINVAL; - cpu = get_cpu(); + cpu = smp_processor_id(); for (update = data; update; update = update->next) { if (cpu != update->cpu) @@ -1446,12 +1448,12 @@ static int update_cpu_topology(void *data) */ int arch_update_cpu_topology(void) { - unsigned int cpu, changed = 0; + unsigned int cpu, sibling, changed = 0; struct topology_update_data *updates, *ud; unsigned int associativity[VPHN_ASSOC_BUFSIZE] = {0}; cpumask_t updated_cpus; struct device *dev; - int weight, i = 0; + int weight, new_nid, i = 0; weight = cpumask_weight(&cpu_associativity_changes_mask); if (!weight) @@ -1464,19 +1466,46 @@ int arch_update_cpu_topology(void) cpumask_clear(&updated_cpus); for_each_cpu(cpu, &cpu_associativity_changes_mask) { - ud = &updates[i++]; - ud->cpu = cpu; - vphn_get_associativity(cpu, associativity); - ud->new_nid = associativity_to_nid(associativity); - - if (ud->new_nid < 0 || !node_online(ud->new_nid)) - ud->new_nid = first_online_node; + /* + * If siblings aren't flagged for changes, updates list + * will be too short. Skip on this update and set for next + * update. + */ + if (!cpumask_subset(cpu_sibling_mask(cpu), + &cpu_associativity_changes_mask)) { + pr_info("Sibling bits not set for associativity " + "change, cpu%d\n", cpu); + cpumask_or(&cpu_associativity_changes_mask, + &cpu_associativity_changes_mask, + cpu_sibling_mask(cpu)); + cpu = cpu_last_thread_sibling(cpu); + continue; + } - ud->old_nid = numa_cpu_lookup_table[cpu]; - cpumask_set_cpu(cpu, &updated_cpus); + /* Use associativity from first thread for all siblings */ + vphn_get_associativity(cpu, associativity); + new_nid = associativity_to_nid(associativity); + if (new_nid < 0 || !node_online(new_nid)) + new_nid = first_online_node; + + if (new_nid == numa_cpu_lookup_table[cpu]) { + cpumask_andnot(&cpu_associativity_changes_mask, + &cpu_associativity_changes_mask, + cpu_sibling_mask(cpu)); + cpu = cpu_last_thread_sibling(cpu); + continue; + } - if (i < weight) - ud->next = &updates[i]; + for_each_cpu(sibling, cpu_sibling_mask(cpu)) { + ud = &updates[i++]; + ud->cpu = sibling; + ud->new_nid = new_nid; + ud->old_nid = numa_cpu_lookup_table[sibling]; + cpumask_set_cpu(sibling, &updated_cpus); + if (i < weight) + ud->next = &updates[i]; + } + cpu = cpu_last_thread_sibling(cpu); } stop_machine(update_cpu_topology, &updates[0], &updated_cpus); -- cgit v0.10.2 From fe956a1d4081ce1a959f87df397a15e252201f10 Mon Sep 17 00:00:00 2001 From: Aaro Koskinen Date: Sun, 21 Jul 2013 03:30:11 +0300 Subject: powerpc/windfarm: Fix noisy slots-fan on Xserve (rm31) slots-fan on G5 Xserve is always running at full speed with windfarm_rm31 driver, resulting in a very high acoustic noise level. It seems the fan parameters are incorrect, and have been copied from the Drive Bay fan (RPM, not present on rm31) of the legacy therm_pm72 driver. This patch changes the parameters to match the Slots fan (PWM) of therm_pm72. With the patch, slots-fan speed drops from 99% to 19% during normal use, and slots-temp settle to ~42'C. Signed-off-by: Aaro Koskinen CC: Signed-off-by: Benjamin Herrenschmidt diff --git a/drivers/macintosh/windfarm_rm31.c b/drivers/macintosh/windfarm_rm31.c index 0b9a79b..82fc86a 100644 --- a/drivers/macintosh/windfarm_rm31.c +++ b/drivers/macintosh/windfarm_rm31.c @@ -439,15 +439,15 @@ static void backside_setup_pid(void) /* Slots fan */ static const struct wf_pid_param slots_param = { - .interval = 5, - .history_len = 2, - .gd = 30 << 20, - .gp = 5 << 20, - .gr = 0, - .itarget = 40 << 16, - .additive = 1, - .min = 300, - .max = 4000, + .interval = 1, + .history_len = 20, + .gd = 0, + .gp = 0, + .gr = 0x00100000, + .itarget = 3200000, + .additive = 0, + .min = 20, + .max = 100, }; static void slots_fan_tick(void) -- cgit v0.10.2