From nobody Thu Jan 1 09:12:24 2026 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 8284AC25B47 for ; Tue, 24 Oct 2023 22:26:04 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1344556AbjJXW0D (ORCPT ); Tue, 24 Oct 2023 18:26:03 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:45574 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1344586AbjJXWZf (ORCPT ); Tue, 24 Oct 2023 18:25:35 -0400 Received: from mail-yw1-x114a.google.com (mail-yw1-x114a.google.com [IPv6:2607:f8b0:4864:20::114a]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 32DE61FDB for ; Tue, 24 Oct 2023 15:24:52 -0700 (PDT) Received: by mail-yw1-x114a.google.com with SMTP id 00721157ae682-5a7ba10cb90so67236437b3.3 for ; Tue, 24 Oct 2023 15:24:52 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20230601; t=1698186289; x=1698791089; 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=ATYLaWcN5bU+Dpgi7a7JWdfUJz5sZd2hiBE3rpvu+hs=; b=YX73kaQOwuAStnFE3CSau5V0IlVcOWh41WNt+xkX/UsxmwQuGGrsZ4CXrp3e3AMPcK y2f9yD9kKb712LZF7vzSiQ2fvH/CLFP7tQxvWhB1/kOHI80Y9mB9qA/Fd98LJ0ZMuK3e p/FtL228WBsz1/KvF+g95duidgiIqJDbxrbQsx/wtXEMYI6jeEVjcRNCOdHN56t3Hlk+ XFUSEEUenG6MjLEJ3u3zFUDuRlVF0fnn1Is/M6KwfboSRDX+xTmNLM/ee4usdtT173Yz 7t/8+LBKGq7HlRqDHKLhxEeZgzc9sRkBXSppcO/juzkeEgFZ6f3NKpQZN7P5WButQJRD bsfg== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1698186289; x=1698791089; 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=ATYLaWcN5bU+Dpgi7a7JWdfUJz5sZd2hiBE3rpvu+hs=; b=ORCgdLjfTfjYtDhdUjCf7VKb1uhdFe4S8L7+Lt4dxllT46fcuX+0kpHzbNS00qwnCI BYL8x9JsYQ7ro63Ic83oenfDX3VjbqYYG0t9GMzhUyzQ2j5DfnRJg+0uyAx8R942CNXV olkCDeblMpIiEACHWnHmznQd2+ZYuWm/0MZ26g8BbxMQMH5BD0GrV2PCWRS8RIaW7zhL Jf8FYo3PO8OatDXc1P0VdZamtYSj+Cd/EUDmxuILWXQpXJk7uF98N7hUZpgGCR3+Ft6z CkozeVoiMttfmluqiPeZW2fXxvfeEzcNd4re9SApn9RFp+opiz+XExczpApy0P9O6gmN TBUg== X-Gm-Message-State: AOJu0YyJkLLW46IKS5mklIZn4xk746JytiWAOi3FpmJ9zUYg3HJsolV7 V1Rg5+W327nrGDHaxc8s4oY1IySV280c X-Google-Smtp-Source: AGHT+IFT+Pd5kd1I8od4cX2G1Bf5fCmCBiz79LjBb1HZ4UWsKZi/GR9vVgFu5bdBd6BAlHb7OsUquC0cyFl1 X-Received: from irogers.svl.corp.google.com ([2620:15c:2a3:200:93d2:18cc:4d63:45ba]) (user=irogers job=sendgmr) by 2002:a05:6902:1083:b0:d9a:4f4c:961b with SMTP id v3-20020a056902108300b00d9a4f4c961bmr353608ybu.1.1698186288663; Tue, 24 Oct 2023 15:24:48 -0700 (PDT) Date: Tue, 24 Oct 2023 15:23:19 -0700 In-Reply-To: <20231024222353.3024098-1-irogers@google.com> Message-Id: <20231024222353.3024098-17-irogers@google.com> Mime-Version: 1.0 References: <20231024222353.3024098-1-irogers@google.com> X-Mailer: git-send-email 2.42.0.758.gaed0368e0e-goog Subject: [PATCH v3 16/50] tools api fs: Switch filename__read_str to use io.h From: Ian Rogers To: Peter Zijlstra , Ingo Molnar , Arnaldo Carvalho de Melo , Mark Rutland , Alexander Shishkin , Jiri Olsa , Namhyung Kim , Ian Rogers , Adrian Hunter , Nick Terrell , Kan Liang , Andi Kleen , Leo Yan , Song Liu , Sandipan Das , James Clark , Anshuman Khandual , Miguel Ojeda , Liam Howlett , Yang Jihong , Athira Rajeev , Kajol Jain , K Prateek Nayak , Sean Christopherson , Yanteng Si , Ravi Bangoria , German Gomez , Changbin Du , Paolo Bonzini , Masami Hiramatsu , liuwenyu , linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" filename__read_str has its own string reading code that allocates memory before reading into it. The memory allocated is sized at BUFSIZ that is 8kb. Most strings are short and so most of this 8kb is wasted. Refactor io__getline so that the newline character can be configurable and ignored in the case of filename__read_str. Code like build_caches_for_cpu in perf's header.c will read many strings and hold them in a data structure, in this case multiple strings per cache level per CPU. Using io.h's io__getline avoids the wasted memory as strings are temporarily read into a buffer on the stack before being copied to a buffer that grows 128 bytes at a time and is never sized larger than the string. For a 16 hyperthread system the memory consumption of "perf record true" is reduced by 180kb, primarily through saving memory when reading the cache information. Signed-off-by: Ian Rogers --- tools/lib/api/fs/fs.c | 56 +++++++++++-------------------------------- tools/lib/api/io.h | 9 +++++-- 2 files changed, 21 insertions(+), 44 deletions(-) diff --git a/tools/lib/api/fs/fs.c b/tools/lib/api/fs/fs.c index 5cb0eeec2c8a..496812b5f1d2 100644 --- a/tools/lib/api/fs/fs.c +++ b/tools/lib/api/fs/fs.c @@ -16,6 +16,7 @@ #include =20 #include "fs.h" +#include "../io.h" #include "debug-internal.h" =20 #define _STR(x) #x @@ -344,53 +345,24 @@ int filename__read_ull(const char *filename, unsigned= long long *value) return filename__read_ull_base(filename, value, 0); } =20 -#define STRERR_BUFSIZE 128 /* For the buffer size of strerror_r */ - int filename__read_str(const char *filename, char **buf, size_t *sizep) { - size_t size =3D 0, alloc_size =3D 0; - void *bf =3D NULL, *nbf; - int fd, n, err =3D 0; - char sbuf[STRERR_BUFSIZE]; + struct io io; + char bf[128]; + int err; =20 - fd =3D open(filename, O_RDONLY); - if (fd < 0) + io.fd =3D open(filename, O_RDONLY); + if (io.fd < 0) return -errno; - - do { - if (size =3D=3D alloc_size) { - alloc_size +=3D BUFSIZ; - nbf =3D realloc(bf, alloc_size); - if (!nbf) { - err =3D -ENOMEM; - break; - } - - bf =3D nbf; - } - - n =3D read(fd, bf + size, alloc_size - size); - if (n < 0) { - if (size) { - pr_warn("read failed %d: %s\n", errno, - strerror_r(errno, sbuf, sizeof(sbuf))); - err =3D 0; - } else - err =3D -errno; - - break; - } - - size +=3D n; - } while (n > 0); - - if (!err) { - *sizep =3D size; - *buf =3D bf; + io__init(&io, io.fd, bf, sizeof(bf)); + *buf =3D NULL; + err =3D io__getline_nl(&io, buf, sizep, /*nl=3D*/-1); + if (err < 0) { + free(*buf); + *buf =3D NULL; } else - free(bf); - - close(fd); + err =3D 0; + close(io.fd); return err; } =20 diff --git a/tools/lib/api/io.h b/tools/lib/api/io.h index a77b74c5fb65..50d33e14fb56 100644 --- a/tools/lib/api/io.h +++ b/tools/lib/api/io.h @@ -141,7 +141,7 @@ static inline int io__get_dec(struct io *io, __u64 *dec) } =20 /* Read up to and including the first newline following the pattern of get= line. */ -static inline ssize_t io__getline(struct io *io, char **line_out, size_t *= line_len_out) +static inline ssize_t io__getline_nl(struct io *io, char **line_out, size_= t *line_len_out, int nl) { char buf[128]; int buf_pos =3D 0; @@ -151,7 +151,7 @@ static inline ssize_t io__getline(struct io *io, char *= *line_out, size_t *line_l =20 /* TODO: reuse previously allocated memory. */ free(*line_out); - while (ch !=3D '\n') { + while (ch !=3D nl) { ch =3D io__get_char(io); =20 if (ch < 0) @@ -184,4 +184,9 @@ static inline ssize_t io__getline(struct io *io, char *= *line_out, size_t *line_l return -ENOMEM; } =20 +static inline ssize_t io__getline(struct io *io, char **line_out, size_t *= line_len_out) +{ + return io__getline_nl(io, line_out, line_len_out, /*nl=3D*/'\n'); +} + #endif /* __API_IO__ */ --=20 2.42.0.758.gaed0368e0e-goog