From mboxrd@z Thu Jan 1 00:00:00 1970 Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.remote-ip=209.85.214.200 ARC-Seal: i=1; d=subspace.kernel.org; s=arc20261015; a=rsa-sha256; cv=none; t=1794590308; b=FNNTLVXyk3DKaBTzEy4FBbDvfRM6kk9RwH5WHpQBT0PCBPuNxMSHiN4IyUknVm0He6Qr v7UuvQIpt6KwUPhi4o1MCAHOhVrMCS7ixZB+FqVKhayjSiqOsdcSQzSXutJCcNUDXY3Eq am5Ucds0vWqqbXlx8msNF6ghg/kWLsZK5vxr1Z77cbZOUcnlID3NoVkqnAefCwOGqjzMi xAaesbfkyL/X2STcF5vhhW0b17dVX5tVjFKPo/M5SC/SiBeKywBH/HH9KNqI6JlMA29P3 I3xX6g3VJ4rqHUkz4iVSET5ZJyEijcEBh30/7DA+Q1EdaZobgalr941G2ZODsPsooLA== ARC-Message-Signature: i=1; d=subspace.kernel.org; s=arc20261015; a=rsa-sha256; c=relaxed/relaxed; t=1794590308; h=DMARC-Filter:Received:DKIM-Signature:X-Google-DKIM-Signature: X-Forwarded-Encrypted:X-Gm-Message-State:X-Received:Date:In-Reply-To: Mime-Version:References:X-Mailer:Message-ID:Subject:From:To:Cc: Content-Type; bh=zXnL8y02lrEho2IOUiAmfmNd24ln45CMwd8hEYyxJRc=; b=TYngaCGnQTNYAgeiKa3ejWNYG+iAcc+9m7jVPNV+pNKksDuwfgJ9BgL1KH6Z9G9YDSSM IwThB92QRbJ2J0x4vSs+w+FtmF8Vc4AmBexAE6JZ1SG+u77i8BwUd2EAud1OfYrqx+TYG y4YblQrp+gK3PDz4BpYRnBaJXjwjl3bDaanu3UsW1kwSNSgOF4TkXgBWLpPwVORRqbZ2p YFqfHnHdYNSAgtmjQ2DYEP+cgm70KxxX7dRSxkob9OmccC34qVVJR1XWIv7V1TXLqKavY ELmDnJJsOH6j/xOqD7WyLFIDV/0SCWZs7AiI6o1B8dbIvS6hytfhTQ4vwqA+8Gaje3g== ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass header.from=google.com; spf=pass smtp.mailfrom=flex--irogers.bounces.google.com; dkim=pass header.d=google.com header.i=@google.com header.a=rsa-sha256 header.s=20260707 header.b=KfSdNwRK; arc=none smtp.remote-ip=209.85.214.200 Received: from mail-pl1-f200.google.com (mail-pl1-f200.google.com [209.85.214.200]) (using TLSv1.3 with cipher TLS_AES_128_GCM_SHA256 (128/128 bits) key-exchange X25519 server-signature ECDSA (prime256v1) server-digest SHA256) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id F03E860A5BE0 for ; Fri, 13 Nov 2026 17:18:27 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.4.2 smtp.subspace.kernel.org F03E860A5BE0 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.a=rsa-sha256 header.s=20260707 header.b=KfSdNwRK Received: by mail-pl1-f200.google.com with SMTP id d9443c01a7336-2ed02a00b24so10881045ad.2 for ; Fri, 13 Nov 2026 09:18:27 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20260707; t=1794590307; x=1795195107; darn=vger.kernel.org; h=content-type:cc:to:from:subject:message-id:references:mime-version :in-reply-to:date:from:to:cc:subject:date:message-id:reply-to :content-type; bh=zXnL8y02lrEho2IOUiAmfmNd24ln45CMwd8hEYyxJRc=; b=KfSdNwRKlf4YckftWVSSJS1GaXse6vs41keJuM5+4ljWc3c7QrylKlboFjENs9VzlW KWueJaFcbmVICv9RP/uf9ZewLwPGuGvrtprSVMVq9cZBFn63zoRAWYSNorySlrdq0BtL ylFdo4hMpUHo0qcdp7XOPr+66ax2yfu1PZDyWKfXLfZcvkSyu79ATLMGY/3y79GtcaKY LON8O9GLorlvLkz6v2eXhNEoyygtUTnT9dsZDUAKqnpeNM99AfwCFBpEH2VAUwUneMFb CG0F+DUmgriubsye2tJtPLWitvu3jcttm9ogydpGXN8UHTxLaAe8UcOW8ogR9FQoePKy 355A== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20260707; t=1794590307; x=1795195107; h=content-type:cc:to:from:subject:message-id:references:mime-version :in-reply-to:date:x-gm-message-state:from:to:cc:subject:date :message-id:reply-to:content-type; bh=zXnL8y02lrEho2IOUiAmfmNd24ln45CMwd8hEYyxJRc=; b=rpGc1qzIpy4nog5JF+CP0IZU4gm9UM7vXLy2YoqOxgMtxTCgyuZjb16iFG+U/8lEGO be27NG8FqkOCLzA+4hJA6ShIHBrmnrhquhHUqt0Ma3lo4xR8oclfieIk4WtmXJj6CY22 0KSPmQN5Cd6s/u9vLv8pIf84ZPNAfNRPSzvA489AQ4T9/VxYL3gpMqCBBFXIZKNayAoB 7cOcEuKe+i6Kj1uVI8LHmbpAJlAX/e7RceLBKYixUtGHTCl4gPY87G8eUJD8yF7CIzP6 xT8plbiC+RjGkk+jeyUW/C6t+ZzhAbXIUB7IQX57DqSQ9MpsgopcALCA6eYBVWDnz2rN s5cg== X-Forwarded-Encrypted: i=1; AL+hErG6y6otBiBlZb4t4XGuu5TwC+eYrbvbiQcYtaBtPiNRRZgyeF8ScifVUeoRJ/qamTG5X+NgrVYGmFaUjI0=@vger.kernel.org X-Gm-Message-State: APuU4Z+sez2MBAaEGN/3lRfYQxsGx5tWWNEnjRmWTQ5OJ942+ehiA843 02/ioSaUgFwbAcoC44J17k5LLcEbQWA9NKTnb7QNE0JuVkoMyFMrX9qHuyqcix+8EObpccfBeBD SiqPNf7NWsA== X-Received: from plpw11.prod.google.com ([2002:a17:902:9a8b:b0:2e9:db6f:41b3]) (user=irogers job=prod-delivery.src-stubby-dispatcher) by 2002:a17:903:2443:b0:2ea:1b9b:b938 with SMTP id d9443c01a7336-2ee0dcadc39mr31584945ad.55.1794590307073; Fri, 13 Nov 2026 09:18:27 -0800 (PST) Date: Fri, 13 Nov 2026 09:17:23 -0800 In-Reply-To: <20261113171734.2188066-1-irogers@google.com> X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Post: List-Help: List-Subscribe: List-Unsubscribe: List-Owner: Precedence: list Mime-Version: 1.0 References: <20261110220803.2503416-1-irogers@google.com> <20261113171734.2188066-1-irogers@google.com> X-Mailer: git-send-email 2.56.1.703.gf848ed807c-goog Message-ID: <20261113171734.2188066-13-irogers@google.com> Subject: [PATCH v4 12/23] perf bench internals: Add symbols benchmark From: Ian Rogers To: irogers@google.com, acme@kernel.org, haghdoost@uber.com, namhyung@kernel.org Cc: adrian.hunter@intel.com, james.clark@linaro.org, jolsa@kernel.org, linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org, mingo@redhat.com, peterz@infradead.org Content-Type: text/plain; charset="UTF-8" Add a 'perf bench internals symbols' subcommand to measure symbol loading, sorting, and lookup operations directly: 1. Loads symbols from an ELF binary (defaulting to the running perf executable, or specified via -b/--binary) without sorting by name, reporting average load runtime and maximum RSS via getrusage(). 2. Sorts the loaded DSO symbols by name (dso__sort_by_name()), reporting average sort runtime and maximum RSS separately so address-only/lazy symbol loading can be evaluated independently of symbol name sorting. 3. Inserts synthetic symbols in two ascending runs (simulating .symtab and .dynsym), fixes up symbol ends and duplicates, and queries each symbol by address. 4. Incrementally appends symbols at interior addresses and immediately queries each appended symbol by address. Example output: $ perf bench internals symbols # Running 'internals/symbols' benchmark: Average DSO load: 9.174 msec (+- 0.220 msec) Max RSS (DSO load): 75640 KB Average DSO sort by name: 2.787 msec (+- 0.019 msec) Max RSS (with sort by name): 75640 KB Average bulk insert + lookup: 2030.050 usec (+- 13.376 usec) Average append + lookup: 54.650 usec (+- 1.466 usec) Signed-off-by: Ian Rogers --- tools/perf/Documentation/perf-bench.txt | 3 + tools/perf/bench/Build | 1 + tools/perf/bench/bench.h | 1 + tools/perf/bench/symbols.c | 273 ++++++++++++++++++++++++ tools/perf/builtin-bench.c | 1 + 5 files changed, 279 insertions(+) create mode 100644 tools/perf/bench/symbols.c diff --git a/tools/perf/Documentation/perf-bench.txt b/tools/perf/Documentation/perf-bench.txt index 7f820f69740f..54eb809540c4 100644 --- a/tools/perf/Documentation/perf-bench.txt +++ b/tools/perf/Documentation/perf-bench.txt @@ -351,6 +351,9 @@ SUITES FOR 'internals' *synthesize*:: Suite for evaluating perf's event synthesis performance. +*symbols*:: +Suite for evaluating ELF symbol loading, sorting, and lookup performance. + SEE ALSO -------- linkperf:perf[1] diff --git a/tools/perf/bench/Build b/tools/perf/bench/Build index c64c52468d3d..c3c0319696ae 100644 --- a/tools/perf/bench/Build +++ b/tools/perf/bench/Build @@ -20,6 +20,7 @@ perf-bench-y += evlist-open-close.o perf-bench-y += breakpoint.o perf-bench-y += pmu-scan.o perf-bench-y += uprobe.o +perf-bench-y += symbols.o perf-bench-$(CONFIG_X86_64) += mem-memcpy-x86-64-asm.o perf-bench-$(CONFIG_X86_64) += mem-memset-x86-64-asm.o diff --git a/tools/perf/bench/bench.h b/tools/perf/bench/bench.h index 85e3ac471875..e0b35d45559a 100644 --- a/tools/perf/bench/bench.h +++ b/tools/perf/bench/bench.h @@ -52,6 +52,7 @@ int bench_uprobe_trace_printk(int argc, const char **argv); int bench_uprobe_empty_ret(int argc, const char **argv); int bench_uprobe_trace_printk_ret(int argc, const char **argv); int bench_pmu_scan(int argc, const char **argv); +int bench_symbols(int argc, const char **argv); #define BENCH_FORMAT_DEFAULT_STR "default" #define BENCH_FORMAT_DEFAULT 0 diff --git a/tools/perf/bench/symbols.c b/tools/perf/bench/symbols.c new file mode 100644 index 000000000000..c5043fc08198 --- /dev/null +++ b/tools/perf/bench/symbols.c @@ -0,0 +1,273 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Benchmark symbol loading, sorting, and address/name lookup. + */ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include "bench.h" +#include "../util/debug.h" +#include "../util/dso.h" +#include "../util/map.h" +#include "../util/stat.h" +#include "../util/symbol.h" +#include "../util/util.h" + +static unsigned int iterations = 20; +static unsigned int nr_symbols = 10000; +static unsigned int nr_appends = 200; +static const char *binary_path; +static bool no_demangle; + +static const struct option options[] = { + OPT_UINTEGER('i', "iterations", &iterations, + "Number of iterations used to compute average (default: 20)"), + OPT_STRING('b', "binary", &binary_path, "path", + "Binary path for ELF symbol load benchmark (default: perf binary)"), + OPT_UINTEGER('n', "nr-symbols", &nr_symbols, + "Number of synthetic symbols for bulk benchmark (default: 10000)"), + OPT_UINTEGER('a', "nr-appends", &nr_appends, + "Number of incremental symbol appends per iteration (default: 200)"), + OPT_BOOLEAN(0, "no-demangle", &no_demangle, + "Disable symbol demangling during ELF load"), + OPT_END() +}; + +static const char *const bench_usage[] = { + "perf bench internals symbols ", + NULL +}; + +static int do_dso_load_bench(const char *path, bool *skipped) +{ + struct stats time_stats; + struct rusage rusage; + unsigned int i; + + *skipped = false; + init_stats(&time_stats); + + for (i = 0; i < iterations; i++) { + struct timeval start, end, diff; + struct map *map; + struct dso *dso; + u64 runtime_us; + int nr; + + gettimeofday(&start, NULL); + map = dso__new_map(path); + if (!map) { + pr_err("Failed to create map for %s\n", path); + return -ENOMEM; + } + dso = map__dso(map); + nr = dso__load(dso, map); + if (nr < 0) { + pr_err("Failed to load symbols from %s\n", path); + map__put(map); + return -EINVAL; + } + if (nr == 0) { + printf(" Skipping DSO load benchmark (%s has no symbols)\n", path); + map__put(map); + *skipped = true; + return 0; + } + gettimeofday(&end, NULL); + + map__put(map); + + timersub(&end, &start, &diff); + runtime_us = (u64)diff.tv_sec * USEC_PER_SEC + diff.tv_usec; + update_stats(&time_stats, runtime_us); + } + + printf(" Average DSO load: %.3f msec (+- %.3f msec)\n", + avg_stats(&time_stats) / USEC_PER_MSEC, + stddev_stats(&time_stats) / USEC_PER_MSEC); + if (getrusage(RUSAGE_SELF, &rusage) == 0) + printf(" Max RSS (DSO load): %ld KB\n", rusage.ru_maxrss); + return 0; +} + +static int do_dso_sort_by_name_bench(const char *path) +{ + struct stats time_stats; + struct rusage rusage; + unsigned int i; + + init_stats(&time_stats); + + for (i = 0; i < iterations; i++) { + struct timeval start, end, diff; + struct map *map; + struct dso *dso; + u64 runtime_us; + + map = dso__new_map(path); + if (!map) { + pr_err("Failed to create map for %s\n", path); + return -ENOMEM; + } + dso = map__dso(map); + if (dso__load(dso, map) <= 0) { + map__put(map); + return -EINVAL; + } + + gettimeofday(&start, NULL); + dso__sort_by_name(dso); + gettimeofday(&end, NULL); + + map__put(map); + + timersub(&end, &start, &diff); + runtime_us = (u64)diff.tv_sec * USEC_PER_SEC + diff.tv_usec; + update_stats(&time_stats, runtime_us); + } + + printf(" Average DSO sort by name: %.3f msec (+- %.3f msec)\n", + avg_stats(&time_stats) / USEC_PER_MSEC, + stddev_stats(&time_stats) / USEC_PER_MSEC); + if (getrusage(RUSAGE_SELF, &rusage) == 0) + printf(" Max RSS (with sort by name): %ld KB\n", rusage.ru_maxrss); + return 0; +} + +static int do_synthetic_symbols_bench(void) +{ + struct stats bulk_stats, append_stats; + unsigned int i, j, half = nr_symbols / 2; + + if (half == 0) + return 0; + + init_stats(&bulk_stats); + init_stats(&append_stats); + + for (i = 0; i < iterations; i++) { + struct timeval start, end, diff; + struct dso *dso = dso__new("[bench_symbols]"); + u64 runtime_us; + char name[32]; + + if (!dso) + return -ENOMEM; + + /* + * Phase 1: Insert symbols in two ascending runs (simulating + * .symtab + .dynsym), fixup ends and duplicates, and look up + * all symbols by address. + */ + gettimeofday(&start, NULL); + for (j = 0; j < nr_symbols; j++) { + u64 addr = (j < half) ? + (0x100000ULL + (u64)j * 0x40ULL) : + (0x100020ULL + (u64)(j - half) * 0x40ULL); + struct symbol *sym; + + scnprintf(name, sizeof(name), "bench_sym_%06u", j); + sym = symbol__new(addr, 0x18, STB_GLOBAL, STT_FUNC, name); + if (!sym) { + dso__put(dso); + return -ENOMEM; + } + symbols__insert(dso__symbols(dso), sym); + } + symbols__fixup_end(dso__symbols(dso), false); + symbols__fixup_duplicate(dso__symbols(dso)); + + for (j = 0; j < nr_symbols; j++) { + u64 addr = (j < half) ? + (0x100000ULL + (u64)j * 0x40ULL + 4) : + (0x100020ULL + (u64)(j - half) * 0x40ULL + 4); + + (void)dso__find_symbol_nocache(dso, addr); + } + gettimeofday(&end, NULL); + timersub(&end, &start, &diff); + runtime_us = (u64)diff.tv_sec * USEC_PER_SEC + diff.tv_usec; + update_stats(&bulk_stats, runtime_us); + + /* + * Phase 2: Incrementally append one symbol at a time at an + * interior address and immediately query by address. + */ + gettimeofday(&start, NULL); + for (j = 0; j < nr_appends; j++) { + u64 addr = 0x100010ULL + ((u64)(j * 197 % half)) * 0x40ULL; + struct symbol *sym; + + scnprintf(name, sizeof(name), "jit_sym_%06u", j); + sym = symbol__new(addr, 0x08, STB_GLOBAL, STT_FUNC, name); + if (!sym) { + dso__put(dso); + return -ENOMEM; + } + dso__insert_symbol(dso, sym); + (void)dso__find_symbol_nocache(dso, addr + 2); + } + gettimeofday(&end, NULL); + timersub(&end, &start, &diff); + runtime_us = (u64)diff.tv_sec * USEC_PER_SEC + diff.tv_usec; + update_stats(&append_stats, runtime_us); + + dso__put(dso); + } + + printf(" Average bulk insert + lookup: %.3f usec (+- %.3f usec)\n", + avg_stats(&bulk_stats), stddev_stats(&bulk_stats)); + printf(" Average append + lookup: %.3f usec (+- %.3f usec)\n", + avg_stats(&append_stats), stddev_stats(&append_stats)); + return 0; +} + +int bench_symbols(int argc, const char **argv) +{ + char exe_buf[PATH_MAX]; + const char *path; + bool orig_demangle; + bool skipped = false; + int err; + + argc = parse_options(argc, argv, options, bench_usage, 0); + if (argc) { + usage_with_options(bench_usage, options); + exit(EXIT_FAILURE); + } + + if (symbol__init(NULL) < 0) + return -1; + + orig_demangle = symbol_conf.demangle; + if (no_demangle) + symbol_conf.demangle = false; + + if (binary_path) { + path = binary_path; + } else { + perf_exe(exe_buf, sizeof(exe_buf)); + path = exe_buf; + } + + err = do_dso_load_bench(path, &skipped); + if (!err && !skipped) + err = do_dso_sort_by_name_bench(path); + if (!err) + err = do_synthetic_symbols_bench(); + + symbol_conf.demangle = orig_demangle; + symbol__exit(); + return err; +} diff --git a/tools/perf/builtin-bench.c b/tools/perf/builtin-bench.c index 62bebdedd3b8..f937c08f3900 100644 --- a/tools/perf/builtin-bench.c +++ b/tools/perf/builtin-bench.c @@ -97,6 +97,7 @@ static const struct bench internals_benchmarks[] = { { "inject-build-id", "Benchmark build-id injection", bench_inject_build_id }, { "evlist-open-close", "Benchmark evlist open and close", bench_evlist_open_close }, { "pmu-scan", "Benchmark sysfs PMU info scanning", bench_pmu_scan }, + { "symbols", "Benchmark symbol loading, sorting and lookup", bench_symbols }, { NULL, NULL, NULL } }; -- 2.56.1.703.gf848ed807c-goog