From 39bed6cbb842d8edf5a26b01122b391d36775b5e Mon Sep 17 00:00:00 2001 From: Matt Fleming Date: Fri, 23 Jan 2015 18:45:40 +0000 Subject: perf: Make perf_cgroup_from_task() global Move perf_cgroup_from_task() from kernel/events/ to include/linux/ along with the necessary struct definitions, so that it can be used by the PMU code. When the upcoming Intel Cache Monitoring PMU driver assigns monitoring IDs to perf events, it needs to be able to check whether any two monitoring events overlap (say, a cgroup and task event), which means we need to be able to lookup the cgroup associated with a task (if any). Signed-off-by: Matt Fleming Signed-off-by: Peter Zijlstra (Intel) Cc: Arnaldo Carvalho de Melo Cc: Arnaldo Carvalho de Melo Cc: H. Peter Anvin Cc: Jiri Olsa Cc: Kanaka Juvva Cc: Linus Torvalds Cc: Paul Mackerras Cc: Vikas Shivappa Link: http://lkml.kernel.org/r/1422038748-21397-2-git-send-email-matt@codeblueprint.co.uk Signed-off-by: Ingo Molnar --- kernel/events/core.c | 28 +--------------------------- 1 file changed, 1 insertion(+), 27 deletions(-) (limited to 'kernel/events') diff --git a/kernel/events/core.c b/kernel/events/core.c index 20cece0a7aea..072de3143244 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -34,11 +34,11 @@ #include #include #include +#include #include #include #include #include -#include #include #include #include @@ -351,32 +351,6 @@ static void perf_ctx_unlock(struct perf_cpu_context *cpuctx, #ifdef CONFIG_CGROUP_PERF -/* - * perf_cgroup_info keeps track of time_enabled for a cgroup. - * This is a per-cpu dynamically allocated data structure. - */ -struct perf_cgroup_info { - u64 time; - u64 timestamp; -}; - -struct perf_cgroup { - struct cgroup_subsys_state css; - struct perf_cgroup_info __percpu *info; -}; - -/* - * Must ensure cgroup is pinned (css_get) before calling - * this function. In other words, we cannot call this function - * if there is no cgroup event for the current CPU context. - */ -static inline struct perf_cgroup * -perf_cgroup_from_task(struct task_struct *task) -{ - return container_of(task_css(task, perf_event_cgrp_id), - struct perf_cgroup, css); -} - static inline bool perf_cgroup_match(struct perf_event *event) { -- cgit v1.2.3 From eacd3ecc34472ce3751eedfc94e44c7cc6eb6305 Mon Sep 17 00:00:00 2001 From: Matt Fleming Date: Fri, 23 Jan 2015 18:45:41 +0000 Subject: perf: Add ->count() function to read per-package counters For PMU drivers that record per-package counters, the ->count variable cannot be used to record an accurate aggregated value, since it's not possible to perform SMP cross-calls to cpus on other packages from the context in which we update ->count. Introduce a new optional ->count() accessor function that can be used to customize how values are collected. If a PMU driver doesn't provide a ->count() function, we fallback to the existing code. There is necessarily a window of staleness with this approach because the task that generated the counter value may not have been scheduled by the cpu recently. An alternative and more complex approach would be to use a hrtimer to periodically refresh the values from a more permissive scheduling context. So, we're trading off complexity for accuracy. Signed-off-by: Matt Fleming Signed-off-by: Peter Zijlstra (Intel) Cc: Arnaldo Carvalho de Melo Cc: Arnaldo Carvalho de Melo Cc: H. Peter Anvin Cc: Jiri Olsa Cc: Kanaka Juvva Cc: Linus Torvalds Cc: Vikas Shivappa Link: http://lkml.kernel.org/r/1422038748-21397-3-git-send-email-matt@codeblueprint.co.uk Signed-off-by: Ingo Molnar --- kernel/events/core.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'kernel/events') diff --git a/kernel/events/core.c b/kernel/events/core.c index 072de3143244..4e8dc596f101 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -3194,7 +3194,10 @@ static void __perf_event_read(void *info) static inline u64 perf_event_count(struct perf_event *event) { - return local64_read(&event->count) + atomic64_read(&event->child_count); + if (event->pmu->count) + return event->pmu->count(event); + + return __perf_event_count(event); } static u64 perf_event_read(struct perf_event *event) -- cgit v1.2.3 From 79dff51e900fd26a073be8b23acfbd8c15edb181 Mon Sep 17 00:00:00 2001 From: Matt Fleming Date: Fri, 23 Jan 2015 18:45:42 +0000 Subject: perf: Move cgroup init before PMU ->event_init() The Intel QoS PMU needs to know whether an event is part of a cgroup during ->event_init(), because tasks in the same cgroup share a monitoring ID. Move the cgroup initialisation before calling into the PMU driver. Signed-off-by: Matt Fleming Signed-off-by: Peter Zijlstra (Intel) Cc: Arnaldo Carvalho de Melo Cc: Arnaldo Carvalho de Melo Cc: H. Peter Anvin Cc: Jiri Olsa Cc: Kanaka Juvva Cc: Linus Torvalds Cc: Vikas Shivappa Link: http://lkml.kernel.org/r/1422038748-21397-4-git-send-email-matt@codeblueprint.co.uk Signed-off-by: Ingo Molnar --- kernel/events/core.c | 28 ++++++++++++++++------------ 1 file changed, 16 insertions(+), 12 deletions(-) (limited to 'kernel/events') diff --git a/kernel/events/core.c b/kernel/events/core.c index 4e8dc596f101..1fc3bae5904a 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -7116,7 +7116,7 @@ perf_event_alloc(struct perf_event_attr *attr, int cpu, struct perf_event *group_leader, struct perf_event *parent_event, perf_overflow_handler_t overflow_handler, - void *context) + void *context, int cgroup_fd) { struct pmu *pmu; struct perf_event *event; @@ -7212,6 +7212,12 @@ perf_event_alloc(struct perf_event_attr *attr, int cpu, if (!has_branch_stack(event)) event->attr.branch_sample_type = 0; + if (cgroup_fd != -1) { + err = perf_cgroup_connect(cgroup_fd, event, attr, group_leader); + if (err) + goto err_ns; + } + pmu = perf_init_event(event); if (!pmu) goto err_ns; @@ -7235,6 +7241,8 @@ err_pmu: event->destroy(event); module_put(pmu->module); err_ns: + if (is_cgroup_event(event)) + perf_detach_cgroup(event); if (event->ns) put_pid_ns(event->ns); kfree(event); @@ -7453,6 +7461,7 @@ SYSCALL_DEFINE5(perf_event_open, int move_group = 0; int err; int f_flags = O_RDWR; + int cgroup_fd = -1; /* for future expandability... */ if (flags & ~PERF_FLAG_ALL) @@ -7518,21 +7527,16 @@ SYSCALL_DEFINE5(perf_event_open, get_online_cpus(); + if (flags & PERF_FLAG_PID_CGROUP) + cgroup_fd = pid; + event = perf_event_alloc(&attr, cpu, task, group_leader, NULL, - NULL, NULL); + NULL, NULL, cgroup_fd); if (IS_ERR(event)) { err = PTR_ERR(event); goto err_cpus; } - if (flags & PERF_FLAG_PID_CGROUP) { - err = perf_cgroup_connect(pid, event, &attr, group_leader); - if (err) { - __free_event(event); - goto err_cpus; - } - } - if (is_sampling_event(event)) { if (event->pmu->capabilities & PERF_PMU_CAP_NO_INTERRUPT) { err = -ENOTSUPP; @@ -7769,7 +7773,7 @@ perf_event_create_kernel_counter(struct perf_event_attr *attr, int cpu, */ event = perf_event_alloc(attr, cpu, task, NULL, NULL, - overflow_handler, context); + overflow_handler, context, -1); if (IS_ERR(event)) { err = PTR_ERR(event); goto err; @@ -8130,7 +8134,7 @@ inherit_event(struct perf_event *parent_event, parent_event->cpu, child, group_leader, parent_event, - NULL, NULL); + NULL, NULL, -1); if (IS_ERR(child_event)) return child_event; -- cgit v1.2.3 From bfe1fcd2688f557a6b6a88f59ea7619228728bd7 Mon Sep 17 00:00:00 2001 From: Matt Fleming Date: Fri, 23 Jan 2015 18:45:46 +0000 Subject: perf/x86/intel: Support task events with Intel CQM Add support for task events as well as system-wide events. This change has a big impact on the way that we gather LLC occupancy values in intel_cqm_event_read(). Currently, for system-wide (per-cpu) events we defer processing to userspace which knows how to discard all but one cpu result per package. Things aren't so simple for task events because we need to do the value aggregation ourselves. To do this, we defer updating the LLC occupancy value in event->count from intel_cqm_event_read() and do an SMP cross-call to read values for all packages in intel_cqm_event_count(). We need to ensure that we only do this for one task event per cache group, otherwise we'll report duplicate values. If we're a system-wide event we want to fallback to the default perf_event_count() implementation. Refactor this into a common function so that we don't duplicate the code. Also, introduce PERF_TYPE_INTEL_CQM, since we need a way to track an event's task (if the event isn't per-cpu) inside of the Intel CQM PMU driver. This task information is only availble in the upper layers of the perf infrastructure. Other perf backends stash the target task in event->hw.*target so we need to do something similar. The task is used to determine whether events should share a cache group and an RMID. Signed-off-by: Matt Fleming Signed-off-by: Peter Zijlstra (Intel) Cc: Arnaldo Carvalho de Melo Cc: Arnaldo Carvalho de Melo Cc: H. Peter Anvin Cc: Jiri Olsa Cc: Kanaka Juvva Cc: Linus Torvalds Cc: Vikas Shivappa Cc: linux-api@vger.kernel.org Link: http://lkml.kernel.org/r/1422038748-21397-8-git-send-email-matt@codeblueprint.co.uk Signed-off-by: Ingo Molnar --- kernel/events/core.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'kernel/events') diff --git a/kernel/events/core.c b/kernel/events/core.c index 1fc3bae5904a..71109a045450 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -7181,6 +7181,8 @@ perf_event_alloc(struct perf_event_attr *attr, int cpu, else if (attr->type == PERF_TYPE_BREAKPOINT) event->hw.bp_target = task; #endif + else if (attr->type == PERF_TYPE_INTEL_CQM) + event->hw.cqm_target = task; } if (!overflow_handler && parent_event) { -- cgit v1.2.3 From 50f16a8bf9d7a92c437ed1867d0f7e1dc6a9aca9 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Thu, 5 Mar 2015 22:10:19 +0100 Subject: perf: Remove type specific target pointers The only reason CQM had to use a hard-coded pmu type was so it could use cqm_target in hw_perf_event. Do away with the {tp,bp,cqm}_target pointers and provide a non type specific one. This allows us to do away with that silly pmu type as well. Signed-off-by: Peter Zijlstra (Intel) Cc: Vince Weaver Cc: acme@kernel.org Cc: acme@redhat.com Cc: hpa@zytor.com Cc: jolsa@redhat.com Cc: kanaka.d.juvva@intel.com Cc: matt.fleming@intel.com Cc: tglx@linutronix.de Cc: torvalds@linux-foundation.org Cc: vikas.shivappa@linux.intel.com Link: http://lkml.kernel.org/r/20150305211019.GU21418@twins.programming.kicks-ass.net Signed-off-by: Ingo Molnar --- kernel/events/core.c | 14 ++++---------- kernel/events/hw_breakpoint.c | 8 ++++---- 2 files changed, 8 insertions(+), 14 deletions(-) (limited to 'kernel/events') diff --git a/kernel/events/core.c b/kernel/events/core.c index 71109a045450..525062b6fba1 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -7171,18 +7171,12 @@ perf_event_alloc(struct perf_event_attr *attr, int cpu, if (task) { event->attach_state = PERF_ATTACH_TASK; - - if (attr->type == PERF_TYPE_TRACEPOINT) - event->hw.tp_target = task; -#ifdef CONFIG_HAVE_HW_BREAKPOINT /* - * hw_breakpoint is a bit difficult here.. + * XXX pmu::event_init needs to know what task to account to + * and we cannot use the ctx information because we need the + * pmu before we get a ctx. */ - else if (attr->type == PERF_TYPE_BREAKPOINT) - event->hw.bp_target = task; -#endif - else if (attr->type == PERF_TYPE_INTEL_CQM) - event->hw.cqm_target = task; + event->hw.target = task; } if (!overflow_handler && parent_event) { diff --git a/kernel/events/hw_breakpoint.c b/kernel/events/hw_breakpoint.c index 9803a6600d49..92ce5f4ccc26 100644 --- a/kernel/events/hw_breakpoint.c +++ b/kernel/events/hw_breakpoint.c @@ -116,12 +116,12 @@ static unsigned int max_task_bp_pinned(int cpu, enum bp_type_idx type) */ static int task_bp_pinned(int cpu, struct perf_event *bp, enum bp_type_idx type) { - struct task_struct *tsk = bp->hw.bp_target; + struct task_struct *tsk = bp->hw.target; struct perf_event *iter; int count = 0; list_for_each_entry(iter, &bp_task_head, hw.bp_list) { - if (iter->hw.bp_target == tsk && + if (iter->hw.target == tsk && find_slot_idx(iter) == type && (iter->cpu < 0 || cpu == iter->cpu)) count += hw_breakpoint_weight(iter); @@ -153,7 +153,7 @@ fetch_bp_busy_slots(struct bp_busy_slots *slots, struct perf_event *bp, int nr; nr = info->cpu_pinned; - if (!bp->hw.bp_target) + if (!bp->hw.target) nr += max_task_bp_pinned(cpu, type); else nr += task_bp_pinned(cpu, bp, type); @@ -210,7 +210,7 @@ toggle_bp_slot(struct perf_event *bp, bool enable, enum bp_type_idx type, weight = -weight; /* Pinned counter cpu profiling */ - if (!bp->hw.bp_target) { + if (!bp->hw.target) { get_bp_info(bp->cpu, type)->cpu_pinned += weight; return; } -- cgit v1.2.3