From nobody Sun Feb 8 00:38:52 2026 Received: from foss.arm.com (foss.arm.com [217.140.110.172]) by smtp.subspace.kernel.org (Postfix) with ESMTP id BD3AC4438E; Fri, 22 Mar 2024 13:04:32 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=217.140.110.172 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1711112674; cv=none; b=uqVGPAoCjdgPYhb1ICVswbQ/E3jBbv5yEK6PJ99mj0vPEH7nO4eAIxhyT9TvD2pEhk4bBt7HVJ/5/0Pae5SxSfgkzhHzdwJ61VOSDE6m4lh/o2XfBHII4s7TuNfridU6eGFThqD52yWRaSQ2qnHpaOqKNqWSSNmFeR4hHmpVXpU= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1711112674; c=relaxed/simple; bh=EK8nvo4TRweQgyliYjiBd+AvfZpo6ofH+nQNVcwGxD8=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=VTkrbOTfIhexpMjB1l0q8Uk70I+ZLxY2SEzAdZvjTA1aQMJjb7PxEp0y1IqTOVPcZLxnnbdn0JLZMnuivML+lXct/wmL9YR1CIce0/K6jcf/XUu1BW3KJrhC0xVE71gx5SdKS7UpN1+vwHkXUff4wLBKHvj5UMubx4JTH9VkqWw= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=arm.com; spf=pass smtp.mailfrom=arm.com; arc=none smtp.client-ip=217.140.110.172 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=arm.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=arm.com Received: from usa-sjc-imap-foss1.foss.arm.com (unknown [10.121.207.14]) by usa-sjc-mx-foss1.foss.arm.com (Postfix) with ESMTP id 14FD61042; Fri, 22 Mar 2024 06:05:06 -0700 (PDT) Received: from e126817.cambridge.arm.com (e126817.cambridge.arm.com [10.2.3.5]) by usa-sjc-imap-foss1.foss.arm.com (Postfix) with ESMTPA id 3831D3F762; Fri, 22 Mar 2024 06:04:30 -0700 (PDT) From: Ben Gainey To: peterz@infradead.org, mingo@redhat.com, acme@kernel.org, namhyung@kernel.org Cc: james.clark@arm.com, mark.rutland@arm.com, alexander.shishkin@linux.intel.com, jolsa@kernel.org, irogers@google.com, adrian.hunter@intel.com, linux-perf-users@vger.kernel.org, linux-kernel@vger.kernel.org, Ben Gainey Subject: [PATCH v3 1/4] perf: Support PERF_SAMPLE_READ with inherit_stat Date: Fri, 22 Mar 2024 13:04:11 +0000 Message-ID: <20240322130414.70186-2-ben.gainey@arm.com> X-Mailer: git-send-email 2.44.0 In-Reply-To: <20240322130414.70186-1-ben.gainey@arm.com> References: <20240322130414.70186-1-ben.gainey@arm.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" This change allows events to use PERF_SAMPLE READ with inherit so long as both inherit_stat and PERF_SAMPLE_TID are set. In this configuration, an event will be inherited into any child processes / threads, allowing convenient profiling of a multiprocess or multithreaded application, whilst allowing profiling tools to collect per-thread samples, in particular of groups of counters. Signed-off-by: Ben Gainey --- include/linux/perf_event.h | 1 + kernel/events/core.c | 62 +++++++++++++++++++++++++++++--------- 2 files changed, 48 insertions(+), 15 deletions(-) diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h index d2a15c0c6f8a..7d405dff6694 100644 --- a/include/linux/perf_event.h +++ b/include/linux/perf_event.h @@ -932,6 +932,7 @@ struct perf_event_context { =20 int nr_task_data; int nr_stat; + int nr_stat_read; int nr_freq; int rotate_disable; =20 diff --git a/kernel/events/core.c b/kernel/events/core.c index f0f0f71213a1..870a7dbd8d5f 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -1767,6 +1767,18 @@ perf_event_groups_next(struct perf_event *event, str= uct pmu *pmu) event =3D rb_entry_safe(rb_next(&event->group_node), \ typeof(*event), group_node)) =20 +/* + * Does the event attribute request inherit with PERF_SAMPLE_READ + */ +#define perf_attr_has_inherit_and_sample_read(attr) \ + ((attr)->inherit && ((attr)->sample_type & PERF_SAMPLE_READ)) + +/* + * Does the event request an attribte that requests inherit with PERF_SAMP= LE_READ + */ +#define perf_event_has_inherit_and_sample_read(event) \ + perf_attr_has_inherit_and_sample_read(&((event)->attr)) + /* * Add an event from the lists for its context. * Must be called with ctx->mutex and ctx->lock held. @@ -1795,8 +1807,11 @@ list_add_event(struct perf_event *event, struct perf= _event_context *ctx) ctx->nr_events++; if (event->hw.flags & PERF_EVENT_FLAG_USER_READ_CNT) ctx->nr_user++; - if (event->attr.inherit_stat) + if (event->attr.inherit_stat) { ctx->nr_stat++; + if (perf_event_has_inherit_and_sample_read(event)) + ctx->nr_stat_read++; + } =20 if (event->state > PERF_EVENT_STATE_OFF) perf_cgroup_event_enable(event, ctx); @@ -2019,8 +2034,11 @@ list_del_event(struct perf_event *event, struct perf= _event_context *ctx) ctx->nr_events--; if (event->hw.flags & PERF_EVENT_FLAG_USER_READ_CNT) ctx->nr_user--; - if (event->attr.inherit_stat) + if (event->attr.inherit_stat) { ctx->nr_stat--; + if (perf_event_has_inherit_and_sample_read(event)) + ctx->nr_stat_read--; + } =20 list_del_rcu(&event->event_entry); =20 @@ -3529,11 +3547,19 @@ perf_event_context_sched_out(struct task_struct *ta= sk, struct task_struct *next) perf_ctx_disable(ctx, false); =20 /* PMIs are disabled; ctx->nr_pending is stable. */ - if (local_read(&ctx->nr_pending) || + if (ctx->nr_stat_read || + next_ctx->nr_stat_read || + local_read(&ctx->nr_pending) || local_read(&next_ctx->nr_pending)) { /* * Must not swap out ctx when there's pending * events that rely on the ctx->task relation. + * + * Likewise, when a context contains inherit + + * inherit_stat + SAMPLE_READ events they + * should be switched out using the slow path + * so that they are treated as if they were + * distinct contexts. */ raw_spin_unlock(&next_ctx->lock); rcu_read_unlock(); @@ -4533,8 +4559,11 @@ static void __perf_event_read(void *info) raw_spin_unlock(&ctx->lock); } =20 -static inline u64 perf_event_count(struct perf_event *event) +static inline u64 perf_event_count(struct perf_event *event, bool self_val= ue_only) { + if (self_value_only && perf_event_has_inherit_and_sample_read(event)) + return local64_read(&event->count); + return local64_read(&event->count) + atomic64_read(&event->child_count); } =20 @@ -5454,7 +5483,7 @@ static u64 __perf_event_read_value(struct perf_event = *event, u64 *enabled, u64 * mutex_lock(&event->child_mutex); =20 (void)perf_event_read(event, false); - total +=3D perf_event_count(event); + total +=3D perf_event_count(event, false); =20 *enabled +=3D event->total_time_enabled + atomic64_read(&event->child_total_time_enabled); @@ -5463,7 +5492,7 @@ static u64 __perf_event_read_value(struct perf_event = *event, u64 *enabled, u64 * =20 list_for_each_entry(child, &event->child_list, child_list) { (void)perf_event_read(child, false); - total +=3D perf_event_count(child); + total +=3D perf_event_count(child, false); *enabled +=3D child->total_time_enabled; *running +=3D child->total_time_running; } @@ -5545,14 +5574,14 @@ static int __perf_read_group_add(struct perf_event = *leader, /* * Write {count,id} tuples for every sibling. */ - values[n++] +=3D perf_event_count(leader); + values[n++] +=3D perf_event_count(leader, false); if (read_format & PERF_FORMAT_ID) values[n++] =3D primary_event_id(leader); if (read_format & PERF_FORMAT_LOST) values[n++] =3D atomic64_read(&leader->lost_samples); =20 for_each_sibling_event(sub, leader) { - values[n++] +=3D perf_event_count(sub); + values[n++] +=3D perf_event_count(sub, false); if (read_format & PERF_FORMAT_ID) values[n++] =3D primary_event_id(sub); if (read_format & PERF_FORMAT_LOST) @@ -6132,7 +6161,7 @@ void perf_event_update_userpage(struct perf_event *ev= ent) ++userpg->lock; barrier(); userpg->index =3D perf_event_index(event); - userpg->offset =3D perf_event_count(event); + userpg->offset =3D perf_event_count(event, false); if (userpg->index) userpg->offset -=3D local64_read(&event->hw.prev_count); =20 @@ -7200,7 +7229,7 @@ static void perf_output_read_one(struct perf_output_h= andle *handle, u64 values[5]; int n =3D 0; =20 - values[n++] =3D perf_event_count(event); + values[n++] =3D perf_event_count(event, true); if (read_format & PERF_FORMAT_TOTAL_TIME_ENABLED) { values[n++] =3D enabled + atomic64_read(&event->child_total_time_enabled); @@ -7245,7 +7274,7 @@ static void perf_output_read_group(struct perf_output= _handle *handle, (leader->state =3D=3D PERF_EVENT_STATE_ACTIVE)) leader->pmu->read(leader); =20 - values[n++] =3D perf_event_count(leader); + values[n++] =3D perf_event_count(leader, true); if (read_format & PERF_FORMAT_ID) values[n++] =3D primary_event_id(leader); if (read_format & PERF_FORMAT_LOST) @@ -7260,7 +7289,7 @@ static void perf_output_read_group(struct perf_output= _handle *handle, (sub->state =3D=3D PERF_EVENT_STATE_ACTIVE)) sub->pmu->read(sub); =20 - values[n++] =3D perf_event_count(sub); + values[n++] =3D perf_event_count(sub, false); if (read_format & PERF_FORMAT_ID) values[n++] =3D primary_event_id(sub); if (read_format & PERF_FORMAT_LOST) @@ -12010,10 +12039,13 @@ perf_event_alloc(struct perf_event_attr *attr, in= t cpu, local64_set(&hwc->period_left, hwc->sample_period); =20 /* - * We currently do not support PERF_SAMPLE_READ on inherited events. + * We do not support PERF_SAMPLE_READ on inherited events unless + * inherit_stat and PERF_SAMPLE_TID are also selected, which allows + * inherited events to collect per-thread samples. * See perf_output_read(). */ - if (attr->inherit && (attr->sample_type & PERF_SAMPLE_READ)) + if (perf_attr_has_inherit_and_sample_read(attr) + && !(attr->inherit_stat && (attr->sample_type & PERF_SAMPLE_TID))) goto err_ns; =20 if (!has_branch_stack(event)) @@ -13037,7 +13069,7 @@ static void sync_child_event(struct perf_event *chi= ld_event) perf_event_read_event(child_event, task); } =20 - child_val =3D perf_event_count(child_event); + child_val =3D perf_event_count(child_event, false); =20 /* * Add back the child's count to the parent's count: --=20 2.44.0 From nobody Sun Feb 8 00:38:52 2026 Received: from foss.arm.com (foss.arm.com [217.140.110.172]) by smtp.subspace.kernel.org (Postfix) with ESMTP id 76F964502D; Fri, 22 Mar 2024 13:04:34 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=217.140.110.172 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1711112676; cv=none; b=VVVgCdhDDbhEOXzUC3w7eb0oH/BgtJ6S6gwCFkDQSR/roJnMWtBj3XxbCuPncgG0vkGFPTU2PtuUoZGdxLoRplVMhOUEY+OqhLQ8VCRW5y+WwJ1OUDfmcgtG53AYumArDpp5hoGsEtnXpafLjT4A4i9iT/5foBtuz1nCMv21mpE= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1711112676; c=relaxed/simple; bh=XQSKmN1TVfytkq7r8tlE3KspdUqFU8nJv4BPKNrYrio=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=q3VHh6fCo0xe/XzwM8jVIJ49QnWcW2DQucMYJ4EToW5iDH/aqiBn5sO86rumQthdJ5rwxdBEY3l0uT3J47AdIYWOSX2Eq6t3QBma502QQZvwf55vJ4EeBb19dJlKGMzlTJMLTSS/DX3spoxGxp05bhzQ9awTelMphN4chQYi6GI= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=arm.com; spf=pass smtp.mailfrom=arm.com; arc=none smtp.client-ip=217.140.110.172 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=arm.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=arm.com Received: from usa-sjc-imap-foss1.foss.arm.com (unknown [10.121.207.14]) by usa-sjc-mx-foss1.foss.arm.com (Postfix) with ESMTP id 0185E1063; Fri, 22 Mar 2024 06:05:08 -0700 (PDT) Received: from e126817.cambridge.arm.com (e126817.cambridge.arm.com [10.2.3.5]) by usa-sjc-imap-foss1.foss.arm.com (Postfix) with ESMTPA id 245263F762; Fri, 22 Mar 2024 06:04:32 -0700 (PDT) From: Ben Gainey To: peterz@infradead.org, mingo@redhat.com, acme@kernel.org, namhyung@kernel.org Cc: james.clark@arm.com, mark.rutland@arm.com, alexander.shishkin@linux.intel.com, jolsa@kernel.org, irogers@google.com, adrian.hunter@intel.com, linux-perf-users@vger.kernel.org, linux-kernel@vger.kernel.org, Ben Gainey Subject: [PATCH v3 2/4] tools/perf: Track where perf_sample_ids need per-thread periods Date: Fri, 22 Mar 2024 13:04:12 +0000 Message-ID: <20240322130414.70186-3-ben.gainey@arm.com> X-Mailer: git-send-email 2.44.0 In-Reply-To: <20240322130414.70186-1-ben.gainey@arm.com> References: <20240322130414.70186-1-ben.gainey@arm.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" perf_sample_ids and related evlist/evsel code are modified to track which events combine inherit+inherit_stat+PERF_SAMPLE_READ. Events with this combination of properties must be handled differently when tracking each streams sample period, as their stream-id is now only unique per thread, rather than globally unique. Signed-off-by: Ben Gainey --- tools/lib/perf/evlist.c | 1 + tools/lib/perf/evsel.c | 7 +++++++ tools/lib/perf/include/internal/evsel.h | 9 +++++++++ 3 files changed, 17 insertions(+) diff --git a/tools/lib/perf/evlist.c b/tools/lib/perf/evlist.c index 058e3ff10f9b..c585c49491a5 100644 --- a/tools/lib/perf/evlist.c +++ b/tools/lib/perf/evlist.c @@ -255,6 +255,7 @@ static void perf_evlist__id_hash(struct perf_evlist *ev= list, =20 sid->id =3D id; sid->evsel =3D evsel; + sid->period_per_thread =3D perf_evsel__attr_has_per_thread_sample_period(= evsel); hash =3D hash_64(sid->id, PERF_EVLIST__HLIST_BITS); hlist_add_head(&sid->node, &evlist->heads[hash]); } diff --git a/tools/lib/perf/evsel.c b/tools/lib/perf/evsel.c index c07160953224..dd60ee0557d8 100644 --- a/tools/lib/perf/evsel.c +++ b/tools/lib/perf/evsel.c @@ -537,6 +537,13 @@ void perf_evsel__free_id(struct perf_evsel *evsel) evsel->ids =3D 0; } =20 +bool perf_evsel__attr_has_per_thread_sample_period(struct perf_evsel *evse= l) +{ + return (evsel->attr.sample_type & PERF_SAMPLE_READ) + && evsel->attr.inherit + && evsel->attr.inherit_stat; +} + void perf_counts_values__scale(struct perf_counts_values *count, bool scale, __s8 *pscaled) { diff --git a/tools/lib/perf/include/internal/evsel.h b/tools/lib/perf/inclu= de/internal/evsel.h index 5cd220a61962..8dd58149986c 100644 --- a/tools/lib/perf/include/internal/evsel.h +++ b/tools/lib/perf/include/internal/evsel.h @@ -36,6 +36,13 @@ struct perf_sample_id { =20 /* Holds total ID period value for PERF_SAMPLE_READ processing. */ u64 period; + + /* + * When inherit+inherit_stat is combined with PERF_SAMPLE_READ, the + * period value is per (sample_id, thread) tuple, rather than per + * sample_id. + */ + bool period_per_thread; }; =20 struct perf_evsel { @@ -88,4 +95,6 @@ int perf_evsel__apply_filter(struct perf_evsel *evsel, co= nst char *filter); int perf_evsel__alloc_id(struct perf_evsel *evsel, int ncpus, int nthreads= ); void perf_evsel__free_id(struct perf_evsel *evsel); =20 +bool perf_evsel__attr_has_per_thread_sample_period(struct perf_evsel *evse= l); + #endif /* __LIBPERF_INTERNAL_EVSEL_H */ --=20 2.44.0 From nobody Sun Feb 8 00:38:52 2026 Received: from foss.arm.com (foss.arm.com [217.140.110.172]) by smtp.subspace.kernel.org (Postfix) with ESMTP id 5A5FF46546; Fri, 22 Mar 2024 13:04:36 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=217.140.110.172 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1711112678; cv=none; b=fRbJfTaeAVLasO+thGaQ7g36GKVhV5VaCT9AvGrzaH1BF4QRdcdYeLMLmuUs04E4wGLwLmcYyHZBmwzN0KBBGjQ9NcN0lXSPlNGT+DzFIt2elkt6kyDkM/IPbQlwMKqNfj8mhlOAhVJ/SqWc62uD4LfmXTR0otYWTUf2utu2PbI= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1711112678; c=relaxed/simple; bh=rfyrgSzBuP+vCe50n9zKSc1VKBs7oujHxT8ZQojJ7hs=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=uCMUNkRBBpIjwHQ7tcyA3kEz7iHq1UlJgahn+fQ+1NrFC7tCMSeihzfDaGaGmpGP4b7lNphahePROtrI/qEb5sFWTtfLnSqWsP1gbrWHdbRESHCUTeXXH2mlnik0PPRwi7GHUkSQaXcDAHRcvvn48VKwcsGQBoS3pNbuL3A09FQ= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=arm.com; spf=pass smtp.mailfrom=arm.com; arc=none smtp.client-ip=217.140.110.172 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=arm.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=arm.com Received: from usa-sjc-imap-foss1.foss.arm.com (unknown [10.121.207.14]) by usa-sjc-mx-foss1.foss.arm.com (Postfix) with ESMTP id E12261691; Fri, 22 Mar 2024 06:05:09 -0700 (PDT) Received: from e126817.cambridge.arm.com (e126817.cambridge.arm.com [10.2.3.5]) by usa-sjc-imap-foss1.foss.arm.com (Postfix) with ESMTPA id 109A83F762; Fri, 22 Mar 2024 06:04:33 -0700 (PDT) From: Ben Gainey To: peterz@infradead.org, mingo@redhat.com, acme@kernel.org, namhyung@kernel.org Cc: james.clark@arm.com, mark.rutland@arm.com, alexander.shishkin@linux.intel.com, jolsa@kernel.org, irogers@google.com, adrian.hunter@intel.com, linux-perf-users@vger.kernel.org, linux-kernel@vger.kernel.org, Ben Gainey Subject: [PATCH v3 3/4] tools/perf: Correctly calculate sample period for inherited SAMPLE_READ values Date: Fri, 22 Mar 2024 13:04:13 +0000 Message-ID: <20240322130414.70186-4-ben.gainey@arm.com> X-Mailer: git-send-email 2.44.0 In-Reply-To: <20240322130414.70186-1-ben.gainey@arm.com> References: <20240322130414.70186-1-ben.gainey@arm.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" Sample period calculation is updated to take into account the fact that the stream-id for events with inherit+inherit_stat+PERF_SAMPLE_READ is now unique per thread, rather than globally unique. perf_sample_id is modified to support tracking per-stream+per-thread values, along with the existing global per-stream values. In the per-thread case, values are stored in a hash by tid.=20 deliver_sample_value is modified to correctly locate the previous sample storage based on the attribute, stream id and thread id. Signed-off-by: Ben Gainey --- tools/lib/perf/evsel.c | 41 ++++++++++++++++++++++ tools/lib/perf/include/internal/evsel.h | 46 +++++++++++++++++++++++-- tools/perf/util/session.c | 11 ++++-- 3 files changed, 93 insertions(+), 5 deletions(-) diff --git a/tools/lib/perf/evsel.c b/tools/lib/perf/evsel.c index dd60ee0557d8..2bdba78cc43e 100644 --- a/tools/lib/perf/evsel.c +++ b/tools/lib/perf/evsel.c @@ -5,6 +5,7 @@ #include #include #include +#include #include #include #include @@ -23,6 +24,7 @@ void perf_evsel__init(struct perf_evsel *evsel, struct pe= rf_event_attr *attr, int idx) { INIT_LIST_HEAD(&evsel->node); + INIT_LIST_HEAD(&evsel->period_per_thread_periods); evsel->attr =3D *attr; evsel->idx =3D idx; evsel->leader =3D evsel; @@ -531,10 +533,17 @@ int perf_evsel__alloc_id(struct perf_evsel *evsel, in= t ncpus, int nthreads) =20 void perf_evsel__free_id(struct perf_evsel *evsel) { + struct perf_sample_id_period *pos, *n; + xyarray__delete(evsel->sample_id); evsel->sample_id =3D NULL; zfree(&evsel->id); evsel->ids =3D 0; + + perf_evsel_for_each_per_thread_period_safe(evsel, n, pos) { + list_del_init(&pos->node); + zfree(pos); + } } =20 bool perf_evsel__attr_has_per_thread_sample_period(struct perf_evsel *evse= l) @@ -544,6 +553,38 @@ bool perf_evsel__attr_has_per_thread_sample_period(str= uct perf_evsel *evsel) && evsel->attr.inherit_stat; } =20 +u64 *perf_sample_id__get_period_storage(struct perf_sample_id *sid, u32 ti= d) +{ + struct hlist_head *head; + struct perf_sample_id_period *res; + int hash; + + if (!sid->period_per_thread) + return &sid->period; + + hash =3D hash_32(tid, PERF_SAMPLE_ID__HLIST_BITS); + head =3D &sid->periods[hash]; + + hlist_for_each_entry(res, head, hnode) + if (res->tid =3D=3D tid) + return &res->period; + + if (sid->evsel =3D=3D NULL) + return NULL; + + res =3D zalloc(sizeof(struct perf_sample_id_period)); + if (res =3D=3D NULL) + return NULL; + + INIT_LIST_HEAD(&res->node); + res->tid =3D tid; + + list_add_tail(&res->node, &sid->evsel->period_per_thread_periods); + hlist_add_head(&res->hnode, &sid->periods[hash]); + + return &res->period; +} + void perf_counts_values__scale(struct perf_counts_values *count, bool scale, __s8 *pscaled) { diff --git a/tools/lib/perf/include/internal/evsel.h b/tools/lib/perf/inclu= de/internal/evsel.h index 8dd58149986c..52ec2edf628b 100644 --- a/tools/lib/perf/include/internal/evsel.h +++ b/tools/lib/perf/include/internal/evsel.h @@ -11,6 +11,32 @@ struct perf_thread_map; struct xyarray; =20 +/** + * The per-thread accumulated period storage node. + */ +struct perf_sample_id_period { + struct list_head node; + struct hlist_node hnode; + /* The thread that the values belongs to */ + u32 tid; + /* Holds total ID period value for PERF_SAMPLE_READ processing. */ + u64 period; +}; + +/** + * perf_evsel_for_each_per_thread_period_safe - safely iterate thru all the + * period_per_thread_periods + * @evlist:perf_evsel instance to iterate + * @item: struct perf_sample_id_period iterator + * @tmp: struct perf_sample_id_period temp iterator + */ +#define perf_evsel_for_each_per_thread_period_safe(evsel, tmp, item) \ + list_for_each_entry_safe(item, tmp, &(evsel)->period_per_thread_periods, = node) + + +#define PERF_SAMPLE_ID__HLIST_BITS 4 +#define PERF_SAMPLE_ID__HLIST_SIZE (1 << PERF_SAMPLE_ID__HLIST_BITS) + /* * Per fd, to map back from PERF_SAMPLE_ID to evsel, only used when there = are * more than one entry in the evlist. @@ -19,6 +45,7 @@ struct perf_sample_id { struct hlist_node node; u64 id; struct perf_evsel *evsel; + /* * 'idx' will be used for AUX area sampling. A sample will have AUX area * data that will be queued for decoding, where there are separate @@ -34,8 +61,18 @@ struct perf_sample_id { pid_t machine_pid; struct perf_cpu vcpu; =20 - /* Holds total ID period value for PERF_SAMPLE_READ processing. */ - u64 period; + union { + /* + * Holds total ID period value for PERF_SAMPLE_READ processing + * (when period is not per-thread). + */ + u64 period; + /* + * Holds total ID period value for PERF_SAMPLE_READ processing + * (when period is per-thread). + */ + struct hlist_head periods[PERF_SAMPLE_ID__HLIST_SIZE]; + }; =20 /* * When inherit+inherit_stat is combined with PERF_SAMPLE_READ, the @@ -65,6 +102,9 @@ struct perf_evsel { u32 ids; struct perf_evsel *leader; =20 + /* Where period_per_thread is true, stores the per-thread values */ + struct list_head period_per_thread_periods; + /* parse modifier helper */ int nr_members; /* @@ -97,4 +137,6 @@ void perf_evsel__free_id(struct perf_evsel *evsel); =20 bool perf_evsel__attr_has_per_thread_sample_period(struct perf_evsel *evse= l); =20 +u64 *perf_sample_id__get_period_storage(struct perf_sample_id *sid, u32 ti= d); + #endif /* __LIBPERF_INTERNAL_EVSEL_H */ diff --git a/tools/perf/util/session.c b/tools/perf/util/session.c index 199d3e8df315..2d6a92374847 100644 --- a/tools/perf/util/session.c +++ b/tools/perf/util/session.c @@ -1478,14 +1478,19 @@ static int deliver_sample_value(struct evlist *evli= st, { struct perf_sample_id *sid =3D evlist__id2sid(evlist, v->id); struct evsel *evsel; + u64 *storage =3D NULL; =20 if (sid) { + storage =3D perf_sample_id__get_period_storage(sid, sample->tid); + } + + if (storage) { sample->id =3D v->id; - sample->period =3D v->value - sid->period; - sid->period =3D v->value; + sample->period =3D v->value - *storage; + *storage =3D v->value; } =20 - if (!sid || sid->evsel =3D=3D NULL) { + if (!storage || sid->evsel =3D=3D NULL) { ++evlist->stats.nr_unknown_id; return 0; } --=20 2.44.0 From nobody Sun Feb 8 00:38:52 2026 Received: from foss.arm.com (foss.arm.com [217.140.110.172]) by smtp.subspace.kernel.org (Postfix) with ESMTP id 2A13247A6C; Fri, 22 Mar 2024 13:04:37 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=217.140.110.172 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1711112680; cv=none; b=npy0skV+FyyDdCT/EQwNSeZWCIAoGB+Mc2LSTxKOGEyTzeIFCRKIKHfeX/M/oL4O0oFqP7aaHXUOK8dhRpefNCM8Cw4jwFsMLWGqUBtHaHLU3mv9lJS6t/Zr4NB3qGMdLNGOHfLap0LCpofEzPX4LwfMRjVS6hJDHlXFPhA5u08= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1711112680; c=relaxed/simple; bh=Dv5y6LBmAVPIZvvpOpnDqf7nw2vZ7U/HBRtWSqL345I=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=uaCOlzEykKHJCcKSAOcIsmpnDIbRtcVSirhra9wOFdFeWFACpJrYC4rcgciHVYxWKMMBAyXDVQEBRtCtR0vJha7Y4IXlcQYCfW2nsQtjv6AFyreNXmId0lBEZ/AzC39ejz1fxqX/s8pXCaa82tGanx6xvSCKS4Loenx0QgoTqog= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=arm.com; spf=pass smtp.mailfrom=arm.com; arc=none smtp.client-ip=217.140.110.172 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=arm.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=arm.com Received: from usa-sjc-imap-foss1.foss.arm.com (unknown [10.121.207.14]) by usa-sjc-mx-foss1.foss.arm.com (Postfix) with ESMTP id CCEFA1692; Fri, 22 Mar 2024 06:05:11 -0700 (PDT) Received: from e126817.cambridge.arm.com (e126817.cambridge.arm.com [10.2.3.5]) by usa-sjc-imap-foss1.foss.arm.com (Postfix) with ESMTPA id F0BE13F762; Fri, 22 Mar 2024 06:04:35 -0700 (PDT) From: Ben Gainey To: peterz@infradead.org, mingo@redhat.com, acme@kernel.org, namhyung@kernel.org Cc: james.clark@arm.com, mark.rutland@arm.com, alexander.shishkin@linux.intel.com, jolsa@kernel.org, irogers@google.com, adrian.hunter@intel.com, linux-perf-users@vger.kernel.org, linux-kernel@vger.kernel.org, Ben Gainey Subject: [PATCH v3 4/4] tools/perf: Allow inherit + inherit_stat + PERF_SAMPLE_READ when opening events Date: Fri, 22 Mar 2024 13:04:14 +0000 Message-ID: <20240322130414.70186-5-ben.gainey@arm.com> X-Mailer: git-send-email 2.44.0 In-Reply-To: <20240322130414.70186-1-ben.gainey@arm.com> References: <20240322130414.70186-1-ben.gainey@arm.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" This change updates evsel to allow the combination of inherit and PERF_SAMPLE_READ. A fallback is implemented for kernel versions where this feature is not supported. The user must pass --stat option to perf record to opt into this new behaviour. Signed-off-by: Ben Gainey --- tools/perf/tests/attr/base-record | 2 +- tools/perf/tests/attr/base-stat | 2 +- tools/perf/tests/attr/system-wide-dummy | 2 +- tools/perf/tests/attr/test-record-dummy-C0 | 2 +- .../test-record-group-sampling-inherit-stat | 62 ++++++++++++++ tools/perf/util/evsel.c | 82 ++++++++++++++++++- tools/perf/util/evsel.h | 1 + 7 files changed, 147 insertions(+), 6 deletions(-) create mode 100644 tools/perf/tests/attr/test-record-group-sampling-inheri= t-stat diff --git a/tools/perf/tests/attr/base-record b/tools/perf/tests/attr/base= -record index b44e4e6e4443..dd76fbdb628f 100644 --- a/tools/perf/tests/attr/base-record +++ b/tools/perf/tests/attr/base-record @@ -5,7 +5,7 @@ group_fd=3D-1 flags=3D0|8 cpu=3D* type=3D0|1 -size=3D136 +size=3D144 config=3D0|1 sample_period=3D* sample_type=3D263 diff --git a/tools/perf/tests/attr/base-stat b/tools/perf/tests/attr/base-s= tat index fccd8ec4d1b0..d769e69bb4f5 100644 --- a/tools/perf/tests/attr/base-stat +++ b/tools/perf/tests/attr/base-stat @@ -5,7 +5,7 @@ group_fd=3D-1 flags=3D0|8 cpu=3D* type=3D0 -size=3D136 +size=3D144 config=3D0 sample_period=3D0 sample_type=3D65536 diff --git a/tools/perf/tests/attr/system-wide-dummy b/tools/perf/tests/att= r/system-wide-dummy index a1e1d6a263bf..9806de3d9c9e 100644 --- a/tools/perf/tests/attr/system-wide-dummy +++ b/tools/perf/tests/attr/system-wide-dummy @@ -7,7 +7,7 @@ cpu=3D* pid=3D-1 flags=3D8 type=3D1 -size=3D136 +size=3D144 config=3D9 sample_period=3D1 # PERF_SAMPLE_IP | PERF_SAMPLE_TID | PERF_SAMPLE_TIME | diff --git a/tools/perf/tests/attr/test-record-dummy-C0 b/tools/perf/tests/= attr/test-record-dummy-C0 index 576ec48b3aaf..87f5c8db5ba1 100644 --- a/tools/perf/tests/attr/test-record-dummy-C0 +++ b/tools/perf/tests/attr/test-record-dummy-C0 @@ -10,7 +10,7 @@ cpu=3D0 pid=3D-1 flags=3D8 type=3D1 -size=3D136 +size=3D144 config=3D9 sample_period=3D4000 # PERF_SAMPLE_IP | PERF_SAMPLE_TID | PERF_SAMPLE_TIME | diff --git a/tools/perf/tests/attr/test-record-group-sampling-inherit-stat = b/tools/perf/tests/attr/test-record-group-sampling-inherit-stat new file mode 100644 index 000000000000..281a17acb6ae --- /dev/null +++ b/tools/perf/tests/attr/test-record-group-sampling-inherit-stat @@ -0,0 +1,62 @@ +[config] +command =3D record +args =3D --stat --no-bpf-event -c 10000 -e '{cycles,cache-misses}:S' ki= ll >/dev/null 2>&1 +ret =3D 1 + +[event-1:base-record] +fd=3D1 +group_fd=3D-1 + +# cycles +type=3D0 +config=3D0 + +# default | PERF_SAMPLE_READ +sample_type=3D87 + +# PERF_FORMAT_ID | PERF_FORMAT_GROUP | PERF_FORMAT_LOST | PERF_FORMAT_TOT= AL_TIME_ENABLED | PERF_FORMAT_TOTAL_TIME_RUNNING +read_format=3D31 +task=3D1 +mmap=3D1 +comm=3D1 +enable_on_exec=3D1 +disabled=3D1 + +# inherit is enabled for group sampling, as is inherit_stat +inherit=3D1 +inherit_stat=3D1 + +# sampling disabled +sample_freq=3D0 +sample_period=3D10000 +freq=3D0 +write_backward=3D0 + +[event-2:base-record] +fd=3D2 +group_fd=3D1 + +# cache-misses +type=3D0 +config=3D3 + +# default | PERF_SAMPLE_READ +sample_type=3D87 + +# PERF_FORMAT_ID | PERF_FORMAT_GROUP | PERF_FORMAT_LOST | PERF_FORMAT_TOT= AL_TIME_ENABLED | PERF_FORMAT_TOTAL_TIME_RUNNING +read_format=3D31 +task=3D0 +mmap=3D0 +comm=3D0 +enable_on_exec=3D0 +disabled=3D0 + +# inherit is enabled for group sampling, as is inherit_stat +inherit=3D1 +inherit_stat=3D1 + +# sampling disabled +sample_freq=3D0 +sample_period=3D0 +freq=3D0 +write_backward=3D0 diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c index 6d7c9c58a9bc..aec6b4f5264e 100644 --- a/tools/perf/util/evsel.c +++ b/tools/perf/util/evsel.c @@ -1097,6 +1097,64 @@ static bool evsel__is_offcpu_event(struct evsel *evs= el) return evsel__is_bpf_output(evsel) && evsel__name_is(evsel, OFFCPU_EVENT); } =20 +static bool evsel__has_term_type(struct evsel *evsel, enum evsel_term_type= type) +{ + return __evsel__get_config_term(evsel, type) !=3D NULL; +} + +/* + * Determine whether or not an evsel can support inherit+inherit_stat+PERF= _SAMPLE_READ. + * + * In order not to break existing command line behaviour, this configurati= on + * will only be enabled if certain specific requirements are met: + * + * 1) When making a system-wide capture, there is no need to support this + * configuration. Likewise, if the user disables inherit, or does not r= equest + * inherit_stat, then the configuration is not supported. + * 2) If the user explicitly specifies 'freq' as a config term, then do not + * support this feature, as frequency counters are not compatible. + * 3) If the user explicitly specifies 'period' as a config term, then the + * feature is compatible with that event. + * 4) If neither was explicitly set, and the event is part of a group, then + * base the decision on the leader. + * 5) Otherwise base the decision on whether or not the user specified a p= eriod + * or frequency on the command line (which includes the default frequen= cy + * setting). + */ +static bool evsel__compat_with_inherit_sample_read(struct record_opts *opt= s, + struct evsel *leader, + struct evsel *evsel) +{ + struct perf_event_attr *attr =3D &evsel->core.attr; + + if (opts->target.system_wide) + return false; + + if (opts->no_inherit_set || !opts->inherit_stat) + return false; + + if (evsel__has_term_type(evsel, EVSEL__CONFIG_TERM_FREQ)) + return false; + + if (evsel__has_term_type(evsel, EVSEL__CONFIG_TERM_PERIOD)) + return true; + + if (leader && (leader !=3D evsel)) { + struct perf_event_attr *ldr_att =3D &leader->core.attr; + + if (evsel__has_term_type(leader, EVSEL__CONFIG_TERM_FREQ)) + return false; + + if (evsel__has_term_type(leader, EVSEL__CONFIG_TERM_PERIOD)) + return true; + + if (ldr_att->freq) + return false; + } + + return (!attr->freq && !opts->freq); +} + /* * The enable_on_exec/disabled value strategy: * @@ -1133,6 +1191,9 @@ void evsel__config(struct evsel *evsel, struct record= _opts *opts, int track =3D evsel->tracking; bool per_cpu =3D opts->target.default_per_cpu && !opts->target.per_thread; =20 + bool allow_inherit_stat_sample_read =3D evsel__compat_with_inherit_sample= _read( + opts, leader, evsel); + attr->sample_id_all =3D perf_missing_features.sample_id_all ? 0 : 1; attr->inherit =3D !opts->no_inherit; attr->write_backward =3D opts->overwrite ? 1 : 0; @@ -1156,7 +1217,17 @@ void evsel__config(struct evsel *evsel, struct recor= d_opts *opts, */ if (leader->core.nr_members > 1) { attr->read_format |=3D PERF_FORMAT_GROUP; - attr->inherit =3D 0; + if (!allow_inherit_stat_sample_read) + attr->inherit =3D 0; + } + + /* + * Inherit + READ requires inherit_stat, but only if freq is + * not set as the two are incompatible + */ + if (attr->inherit && allow_inherit_stat_sample_read) { + attr->inherit_stat =3D 1; + evsel__set_sample_bit(evsel, TID); } } =20 @@ -1832,6 +1903,8 @@ static int __evsel__prepare_open(struct evsel *evsel,= struct perf_cpu_map *cpus, =20 static void evsel__disable_missing_features(struct evsel *evsel) { + if (perf_missing_features.inherit_sample_read) + evsel->core.attr.inherit =3D 0; if (perf_missing_features.branch_counters) evsel->core.attr.branch_sample_type &=3D ~PERF_SAMPLE_BRANCH_COUNTERS; if (perf_missing_features.read_lost) @@ -1887,7 +1960,12 @@ bool evsel__detect_missing_features(struct evsel *ev= sel) * Must probe features in the order they were added to the * perf_event_attr interface. */ - if (!perf_missing_features.branch_counters && + if (!perf_missing_features.inherit_sample_read && + evsel->core.attr.inherit && (evsel->core.attr.sample_type & PERF_SAMP= LE_READ)) { + perf_missing_features.inherit_sample_read =3D true; + pr_debug2("Using PERF_SAMPLE_READ / :S modifier is not compatible with i= nherit, falling back to no-inherit.\n"); + return true; + } else if (!perf_missing_features.branch_counters && (evsel->core.attr.branch_sample_type & PERF_SAMPLE_BRANCH_COUNTERS)) { perf_missing_features.branch_counters =3D true; pr_debug2("switching off branch counters support\n"); diff --git a/tools/perf/util/evsel.h b/tools/perf/util/evsel.h index efbb6e848287..11cc9b8bee27 100644 --- a/tools/perf/util/evsel.h +++ b/tools/perf/util/evsel.h @@ -192,6 +192,7 @@ struct perf_missing_features { bool weight_struct; bool read_lost; bool branch_counters; + bool inherit_sample_read; }; =20 extern struct perf_missing_features perf_missing_features; --=20 2.44.0