From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mail-yb1-f201.google.com (mail-yb1-f201.google.com [209.85.219.201]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 54FB218C006 for ; Mon, 9 Sep 2024 20:37:55 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.219.201 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1725914277; cv=none; b=GMgz5LzYSyj1c7QqM69I1IN0N2oufZ432aFRcaz3hShFu2ZAKzSA9LM1N0IKVUCxjdNtga1m/5y5rjjPBFWKI2MbJmsVlng6oU0iVEqGuHIJP/6DEhaQeNGlxaewVWPE9E//y/iJWJpOWoCTYnn9F2BptDlQjVfnjtHmPeKpf9U= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1725914277; c=relaxed/simple; bh=D+SSmFc5FfPdhU1r5UpCS2rCSL+zyGVRvs36nBFGJa4=; h=Date:In-Reply-To:Message-Id:Mime-Version:References:Subject:From: To:Content-Type; b=sn41Lrc2zO3RqxYSRXRBjw5+OYZJbUgPXj+c11G6DxP9zofwyuTT0Iha69HFxX8QAGkvJV611hUkXcVlsKsqwQWpa/RCr14J4tNLscad8xc2cDtdue0Vk+UOdHekZrO3QdJcSjB9EOrQCcV6jbC5Q7yt3PAuOoBDurEgis8fRQE= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dmarc=pass (p=reject dis=none) header.from=google.com; spf=pass smtp.mailfrom=flex--irogers.bounces.google.com; dkim=pass (2048-bit key) header.d=google.com header.i=@google.com header.b=L5lw3rU9; arc=none smtp.client-ip=209.85.219.201 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=reject dis=none) header.from=google.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=flex--irogers.bounces.google.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=google.com header.i=@google.com header.b="L5lw3rU9" Received: by mail-yb1-f201.google.com with SMTP id 3f1490d57ef6-e035949cc4eso10360045276.1 for ; Mon, 09 Sep 2024 13:37:55 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20230601; t=1725914274; x=1726519074; darn=vger.kernel.org; h=to:from:subject:references:mime-version:message-id:in-reply-to:date :from:to:cc:subject:date:message-id:reply-to; bh=vkNhK/smhh6RQk8jhHd3svFqj+E5OE7Int+5C0aR36c=; b=L5lw3rU996PibW3QpFtfvbwF6rZe/LLjPMakBJ7VZ/H8xGQJL/eF+/2rLZBLh4buXh Lh0rAGt8Ov8fqueNArZob+gi5uEX7cwdt61xwO7A/3I3pXWsbWR2O52rC5Q0cQUSHUEI xVmYKTyB0xkWkhVSP7nebW1zlAVliOljQC7Sh/BIYHbRV+zr5OTGTTx6Js8ykMiKJ1VG TIjyccvfpdE114nbDg+h+tPirHk6sK73uLBlqSzoBdLPkFM1QcLgLUS3+Z9Pdki6qx5f VDJ8vBw6k3B5Iel20GMUzvUf7T9kIIWKdN7QBiz0Itr8vhlxiry2R6pMltCWsco3E0h3 RfaA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1725914274; x=1726519074; h=to:from:subject:references:mime-version:message-id:in-reply-to:date :x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=vkNhK/smhh6RQk8jhHd3svFqj+E5OE7Int+5C0aR36c=; b=TKxmwXnFFOKJZ9djparVmHlyI5GsdwMJ7WWLGV3Cn4YaLBoLBbonfhcCSJqv/zprIZ AELb/BQaUqR/QRppV2ado54txygJsPnxwO1VqX81FgiDFvz/1M0A7xYinVCCDLVcI4x1 2YjDGX4/Y7AFmM+M4vGAaX1mrcXb7eUnQswxnwMi1MM0gONzSy/vAWWGgPcatSnG5Aud 1ukemnzyCIbYA3ezv6+9UWCXWL+OJW7JFTCdTnWwgBfiw6EgobkNH7+RwQ1SyHScYKXi JUJDqR66Z+MSGSgQmZ8IIncMnOp90x0WYG0b//X/dllf/itVuywxbWAkuTXKtoLBldBa HSPQ== X-Forwarded-Encrypted: i=1; AJvYcCVFKnAqb2zRSL9Y+gT/FlazW+dYBI8+51/+/p+D73f6PRTQ/VT/FJjLHmHkNwGw4ZkdulA0GZ34UAB21Q4xjdSL@vger.kernel.org X-Gm-Message-State: AOJu0Yy2TEVxoy4oQjMrM8UqnfTG4UNtJY1kxAJS/mLFtyU2fNNoBRJe ldfQbjWlfHdtL/GYVBmQhheQE7e4IbcTgCsgUkkiIU/5My1zhl2nhUikJv0uKNZ8X8icgvm1J/2 CcKO23A== X-Google-Smtp-Source: AGHT+IE4dvakATehXxGZD6jo95hQzfhx9BPhrQMVrXB21IOBzBpx6KVswv3eo/qQKzfxY5V7aZAXv0V+NPbz X-Received: from irogers.svl.corp.google.com ([2620:15c:2a3:200:b7b9:f9a0:1197:ff33]) (user=irogers job=sendgmr) by 2002:a25:d347:0:b0:e16:51f9:59da with SMTP id 3f1490d57ef6-e1d349e2dd5mr32365276.6.1725914274226; Mon, 09 Sep 2024 13:37:54 -0700 (PDT) Date: Mon, 9 Sep 2024 13:37:39 -0700 In-Reply-To: <20240909203740.143492-1-irogers@google.com> Message-Id: <20240909203740.143492-4-irogers@google.com> Precedence: bulk X-Mailing-List: linux-perf-users@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: Mime-Version: 1.0 References: <20240909203740.143492-1-irogers@google.com> X-Mailer: git-send-email 2.46.0.598.g6f2099f65c-goog Subject: [PATCH v2 3/4] perf inject: Lazy build-id mmap2 event insertion From: Ian Rogers To: Peter Zijlstra , Ingo Molnar , Arnaldo Carvalho de Melo , Namhyung Kim , Mark Rutland , Alexander Shishkin , Jiri Olsa , Ian Rogers , Adrian Hunter , Kan Liang , Colin Ian King , Casey Chen , Anne Macedo , Sun Haiyong , linux-perf-users@vger.kernel.org, linux-kernel@vger.kernel.org Content-Type: text/plain; charset="UTF-8" Add -B option that lazily inserts mmap2 events thereby dropping all mmap events without samples. This is similar to the behavior of -b where only build_id events are inserted when a dso is accessed in a sample. File size savings can be significant in system-wide mode, consider: ``` $ perf record -g -a -o perf.data sleep 1 $ perf inject -B -i perf.data -o perf.new.data $ ls -al perf.data perf.new.data 5147049 perf.data 2248493 perf.new.data ``` Give test coverage of the new option in pipe test. Acked-by: Namhyung Kim Signed-off-by: Ian Rogers --- tools/perf/builtin-inject.c | 62 +++++++++++++++++++++++------ tools/perf/tests/shell/pipe_test.sh | 1 + tools/perf/util/map.c | 1 + tools/perf/util/map.h | 11 +++++ 4 files changed, 63 insertions(+), 12 deletions(-) diff --git a/tools/perf/builtin-inject.c b/tools/perf/builtin-inject.c index 5a27fa46e93d..9eb72ff48d88 100644 --- a/tools/perf/builtin-inject.c +++ b/tools/perf/builtin-inject.c @@ -108,6 +108,7 @@ enum build_id_rewrite_style { BID_RWS__INJECT_HEADER_LAZY, BID_RWS__INJECT_HEADER_ALL, BID_RWS__MMAP2_BUILDID_ALL, + BID_RWS__MMAP2_BUILDID_LAZY, }; struct perf_inject { @@ -527,7 +528,8 @@ static int perf_event__repipe_common_mmap(const struct perf_tool *tool, * Remember the evsel for lazy build id generation. It is used * for the sample id header type. */ - if (inject->build_id_style == BID_RWS__INJECT_HEADER_LAZY && + if ((inject->build_id_style == BID_RWS__INJECT_HEADER_LAZY || + inject->build_id_style == BID_RWS__MMAP2_BUILDID_LAZY) && !inject->mmap_evsel) inject->mmap_evsel = evlist__event2evsel(inject->session->evlist, event); @@ -560,6 +562,9 @@ static int perf_event__repipe_common_mmap(const struct perf_tool *tool, } } dso__put(dso); + if (inject->build_id_style == BID_RWS__MMAP2_BUILDID_LAZY) + return 0; + return perf_event__repipe(tool, event, sample, machine); } @@ -825,7 +830,8 @@ static int tool__inject_mmap2_build_id(const struct perf_tool *tool, return 0; } -static int mark_dso_hit(const struct perf_tool *tool, +static int mark_dso_hit(const struct perf_inject *inject, + const struct perf_tool *tool, struct perf_sample *sample, struct machine *machine, const struct evsel *mmap_evsel, @@ -854,16 +860,39 @@ static int mark_dso_hit(const struct perf_tool *tool, } } dso = map__dso(map); - if (dso && !dso__hit(dso)) { - dso__set_hit(dso); - tool__inject_build_id(tool, sample, machine, - mmap_evsel, misc, dso__long_name(dso), dso, - map__flags(map)); + if (inject->build_id_style == BID_RWS__INJECT_HEADER_LAZY) { + if (dso && !dso__hit(dso)) { + dso__set_hit(dso); + tool__inject_build_id(tool, sample, machine, + mmap_evsel, misc, dso__long_name(dso), dso, + map__flags(map)); + } + } else if (inject->build_id_style == BID_RWS__MMAP2_BUILDID_LAZY) { + if (!map__hit(map)) { + const struct build_id null_bid = { .size = 0 }; + const struct build_id *bid = dso ? dso__bid(dso) : &null_bid; + const char *filename = dso ? dso__long_name(dso) : ""; + + map__set_hit(map); + perf_event__synthesize_mmap2_build_id(tool, sample, machine, + perf_event__repipe, + mmap_evsel, + misc, + sample->pid, sample->tid, + map__start(map), + map__end(map) - map__start(map), + map__pgoff(map), + bid, + map__prot(map), + map__flags(map), + filename); + } } return 0; } struct mark_dso_hit_args { + const struct perf_inject *inject; const struct perf_tool *tool; struct perf_sample *sample; struct machine *machine; @@ -875,7 +904,7 @@ static int mark_dso_hit_callback(struct callchain_cursor_node *node, void *data) struct mark_dso_hit_args *args = data; struct map *map = node->ms.map; - return mark_dso_hit(args->tool, args->sample, args->machine, + return mark_dso_hit(args->inject, args->tool, args->sample, args->machine, args->mmap_evsel, map, /*sample_in_dso=*/false); } @@ -888,6 +917,7 @@ int perf_event__inject_buildid(const struct perf_tool *tool, union perf_event *e struct thread *thread; struct perf_inject *inject = container_of(tool, struct perf_inject, tool); struct mark_dso_hit_args args = { + .inject = inject, .tool = tool, /* * Use the parsed sample data of the sample event, which will @@ -907,7 +937,7 @@ int perf_event__inject_buildid(const struct perf_tool *tool, union perf_event *e } if (thread__find_map(thread, sample->cpumode, sample->ip, &al)) { - mark_dso_hit(tool, sample, machine, args.mmap_evsel, al.map, + mark_dso_hit(inject, tool, sample, machine, args.mmap_evsel, al.map, /*sample_in_dso=*/true); } @@ -2155,7 +2185,8 @@ static int __cmd_inject(struct perf_inject *inject) #endif } - if (inject->build_id_style == BID_RWS__INJECT_HEADER_LAZY) { + if (inject->build_id_style == BID_RWS__INJECT_HEADER_LAZY || + inject->build_id_style == BID_RWS__MMAP2_BUILDID_LAZY) { inject->tool.sample = perf_event__inject_buildid; } else if (inject->sched_stat) { struct evsel *evsel; @@ -2338,6 +2369,7 @@ int cmd_inject(int argc, const char **argv) const char *known_build_ids = NULL; bool build_ids; bool build_id_all; + bool mmap2_build_ids; bool mmap2_build_id_all; struct option options[] = { @@ -2345,6 +2377,8 @@ int cmd_inject(int argc, const char **argv) "Inject build-ids into the output stream"), OPT_BOOLEAN(0, "buildid-all", &build_id_all, "Inject build-ids of all DSOs into the output stream"), + OPT_BOOLEAN('B', "mmap2-buildids", &mmap2_build_ids, + "Drop unused mmap events, make others mmap2 with build IDs"), OPT_BOOLEAN(0, "mmap2-buildid-all", &mmap2_build_id_all, "Rewrite all mmap events as mmap2 events with build IDs"), OPT_STRING(0, "known-build-ids", &known_build_ids, @@ -2443,6 +2477,8 @@ int cmd_inject(int argc, const char **argv) return -1; } } + if (mmap2_build_ids) + inject.build_id_style = BID_RWS__MMAP2_BUILDID_LAZY; if (mmap2_build_id_all) inject.build_id_style = BID_RWS__MMAP2_BUILDID_ALL; if (build_ids) @@ -2453,7 +2489,8 @@ int cmd_inject(int argc, const char **argv) data.path = inject.input_name; ordered_events = inject.jit_mode || inject.sched_stat || - (inject.build_id_style == BID_RWS__INJECT_HEADER_LAZY); + inject.build_id_style == BID_RWS__INJECT_HEADER_LAZY || + inject.build_id_style == BID_RWS__MMAP2_BUILDID_LAZY; perf_tool__init(&inject.tool, ordered_events); inject.tool.sample = perf_event__repipe_sample; inject.tool.read = perf_event__repipe_sample; @@ -2532,7 +2569,8 @@ int cmd_inject(int argc, const char **argv) } } - if (inject.build_id_style == BID_RWS__INJECT_HEADER_LAZY) { + if (inject.build_id_style == BID_RWS__INJECT_HEADER_LAZY || + inject.build_id_style == BID_RWS__MMAP2_BUILDID_LAZY) { /* * to make sure the mmap records are ordered correctly * and so that the correct especially due to jitted code diff --git a/tools/perf/tests/shell/pipe_test.sh b/tools/perf/tests/shell/pipe_test.sh index 250574cd68b6..d4c8005ce9b9 100755 --- a/tools/perf/tests/shell/pipe_test.sh +++ b/tools/perf/tests/shell/pipe_test.sh @@ -116,6 +116,7 @@ test_inject_bids() { } test_record_report +test_inject_bids -B test_inject_bids -b test_inject_bids --buildid-all test_inject_bids --mmap2-buildid-all diff --git a/tools/perf/util/map.c b/tools/perf/util/map.c index e781c8d56a9a..d729438b7d65 100644 --- a/tools/perf/util/map.c +++ b/tools/perf/util/map.c @@ -116,6 +116,7 @@ static void map__init(struct map *map, u64 start, u64 end, u64 pgoff, map__set_mapping_type(map, MAPPING_TYPE__DSO); assert(map__erange_warned(map) == false); assert(map__priv(map) == false); + assert(map__hit(map) == false); } struct map *map__new(struct machine *machine, u64 start, u64 len, diff --git a/tools/perf/util/map.h b/tools/perf/util/map.h index 6c43f31a9fe0..4262f5a143be 100644 --- a/tools/perf/util/map.h +++ b/tools/perf/util/map.h @@ -35,6 +35,7 @@ DECLARE_RC_STRUCT(map) { enum mapping_type mapping_type:8; bool erange_warned; bool priv; + bool hit; }; struct kmap; @@ -83,6 +84,11 @@ static inline bool map__priv(const struct map *map) return RC_CHK_ACCESS(map)->priv; } +static inline bool map__hit(const struct map *map) +{ + return RC_CHK_ACCESS(map)->hit; +} + static inline refcount_t *map__refcnt(struct map *map) { return &RC_CHK_ACCESS(map)->refcnt; @@ -287,6 +293,11 @@ static inline void map__set_priv(struct map *map) RC_CHK_ACCESS(map)->priv = true; } +static inline void map__set_hit(struct map *map) +{ + RC_CHK_ACCESS(map)->hit = true; +} + static inline void map__set_erange_warned(struct map *map) { RC_CHK_ACCESS(map)->erange_warned = true; -- 2.46.0.598.g6f2099f65c-goog